Grunfeld According to Svidler [Svidler, 2014]

Hi, my name is Peter Svidler and I've recorded a series of videos for chess24 dedicated to my favourite opening, the Grünfeld Defence. I picked it up as a young kid who was obviously hugely impressed by Kasparov's treatment of the opening in those years. The children who were just beginning to play chess back then were greatly influenced by the clash between Karpov and Kasparov. Most of us picked a side, and I picked Kasparov - he was perhaps my earliest idol and I based my repertoire mainly on his openings. A lot of time has passed since then, obviously, and I think I've contributed quite a bit to the opening myself. In this series of videos I'll try to pass on my years of experience and some of my knowledge to you, the viewers. I'll try to explain why I think the Grünfeld is a very good opening for those of you who are striving to get double-edged positions with a lot of counterchances. These days in almost any opening White can get a risk-free position - and the same applies to the Grünfeld, of course - but in many cases you'll have the chance to assume the initiative and start playing for a win. The Grünfeld is a sharp, double-edged and very interesting opening in which play often becomes very nonstandard. These videos will have a very wide range, obviously, as I'll try to cover pretty much the entire opening. 1.

d4 ♘f6 2. c4 g6 3. ♘c3 d5

21... ♗xg4 Over the course of this eBook you'll see positions like this one, which occurs in a video on the 3.f3 variation... 18... ♕a3 ...or this one - which is maybe even crazier than the previous example. This appears in a certain line of the 7.Be3 variation. It's a very interesting looking position because it's not that easy to figure out whose move it is. Seemingly the side to start first just wins on the spot, but that isn't in fact the case. 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6. ♕xc4 OO 7. e4 a6 8. e5 b5 9. ♕b3 ♘fd7 10. ♘g5 ♘c6 11. ♗e3 ♘a5 12.♕d1 ♘b6 13. h4 h6 14. ♘ge4 ♘ac4 15. h5 f5 The Grünfeld in general, even if you discount curious positions like the first two I've shown you, tends to lead to very unbalanced positions like the one you can see here, which can appear in a Qb3 line where Black very, very early on starts an incredibly double-edged and sharp fight. The position becomes very unbalanced and interesting for those of you who are searching for a dynamic opening. 14... e6 Positions like this one, which can arise from 7.Bc4, are also very typical - a position where Black first gives up control of the centre and then starts fighting for it tooth and nail. Play becomes very, very sharp. 7... ♗g7 Obviously over the course of the eBook there will be novelties - some of which are more important than others - and some of them I'm very proud of, like the one I'll show you here. This is a position after move 7 in the Bg5 line where I found an idea which hasn't been played before and which I think is extremely interesting. 1. d4 ♘f6 2. c4 g6 3. g3 c6 4. ♗g2 d5 5. cxd5 cxd5 6. ♘f3 ♗g7 Of course no opening, including the Grünfeld, can be covered without every now and then encountering a position like this one where you can't really hope for much with Black apart from equalising with precise play, but I will try and steer away from these positions and whenever possible I'll try to show you ways of reaching complex and double-edged positions with Black. I hope you enjoy this eBook!

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Nf3 4.e3 Bg7 5.Bd2 4.e3 Bg7 5.Nf3 4.e3 Bg7 5.Qb3

4.Nf3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. cxd5 ♘xd5 6. ♗g5This strange line exists in two versions. The one with the knight on g1 is actually a serious alternative and is covered elsewhere, whereas this one really deserves only a very brief mention. 6... c5 7. ♖c1 ♘xc3 8. bxc3 ♕a5 9. ♕d2 Pretty much forced. If White is now allowed to play e4 he'll have a very decent position. It also seems as though Black has problems castling because the e7-pawn is hanging, but he has a very strong plan in this position which solves both problems. 9... ♗f5 Stopping e4 dead in its tracks. 10. e3 ♘d7 and it turns out there's really very little White can do against the plan of Nf6-e4. A blitz game between Rychagov and Grischuk in 2004 continued 11. ♗e2 ♘f6 12. O-O ♘e4 13. ♕b2 and despite Black losing the right to castle he's at the very least perfectly fine. After a pretty much forced sequence of moves 13... ♘xg5 14. ♗b5+ ♔f8 15. ♘xg5 ♗f6 16. ♘f3 ♔g7⇆ Black was ready to connect the rooks, with one going to c8 and the other to d8. Black has very comfortable play here. Rychagov-Grischuk, Moscow 2004

4.e3 Bg7 5.Bd2 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. e3 ♗g7 5. ♗d2⁈ This has been played on a couple of occasions by the Belarussian GM Alexey Aleksandrov. 5... O-O 6. ♖c1 An unambitious try, and in fact if White wants to at least pretend he's fighting for an advantage he'll get a very double-edged position that should suit any Black player. 6... c5 As usual Black breaks in the centre at the first opportunity. 7. dxc5 ♘a6 8. ♘xd5 8. cxd5 ♘xc5 White's pieces are completely misplaced here and the fact that the bishop is on d2 makes it even harder for White to protect the pawn on d5. 9. ♗c4 ♗f5 10. ♘ge2 ♗d3 11. ♗b3 ♗a6 12. O-O ♖c8 Cusi-Yermolinsky, Chicago 2003 8... ♘e4 Avoiding the exchange of knights for now and attacking the b2-pawn. White pretty much has to play 9. f3 ♘xd2 10. ♕xd2 e6 11. ♘c3 This position was reached in a couple of games by Aleksandrov, one against Ivanchuk and one against Mikkelsen a couple of years later. Vassily played 11... ♕a5 which is a very natural and decent move. 11... ♕h4+ in the second game was also a pretty decent choice. In general the c5-pawn will fall and the pawn on f3 ensures Black lasting compensation almost regardless of what he does. This position is very

comfortable for Black and sacrificing a pawn like this shouldn't alarm any Grünfeld players. This is a line you're very unlikely to encounter, but if you do end up in a position like this you should feel confident Black has excellent compensation. 12. g3 ♕f6 13. ♗d3 ♖d8 14. ♕e2 ♘xc5 15.♗b1 AleksandrovMikkelsen, Vladimir 2008 15... a5=/∞ 12. a3 ♖d8 13. ♕c2 ♗d7 14. ♔f2 ♖ac8 15. ♗e2 ♕xc5 16. b4 ♕e717. h4 ♘c7 18. h5 ♘d5 AleksandrovIvanchuk, Chalkidiki 2002

4.e3 Bg7 5.Nf3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. e3 ♗g7 5. ♘f3 O-O 6. ♗e2 A sideline that became very fashionable in 2012 mainly and almost solely due to the efforts of Shakhriyar Mamedyarov. I myself lost a game to him in the Tashkent Grand Prix. 6. b4 This has been played on a regular basis for many, many years, but simply put, after 6... c6 Black gets an incredible version of the Schlechter Slav (a lot of people are also now playing the classical Schlechter itself, so an awareness of how little White has in this type of structure has led people to realise they can't play b4 as early as this and still hope for an advantage). In general, White needs to finish his development on the kingside before playing b4, as the pawn on b4 here makes Black's play very comfortable and straightforward. 7. ♗b2 a5Levon Aronian once played this line against me in Linares, and this was my choice. Although I'm not sure it's stronger than 7... Bg4 I thought it was a more concrete approach and I always look for ways to get concrete counterplay in the Grünfeld. I think there's some merit in what I did. ( 7...♗g4 would be the classical Slav approach, to be followed up by trading off the bishop and playing Nbd7, e6, or maybe dxc4 and e5 in some positions. It is indeed very comfortable for Black. ) 8. b5 a4 a) White can accept the pawn sacrifice with 9. bxc6 bxc6 10. ♘xa4but after 10... ♕a5+ 11. ♘c3 Black has very decent compensation for the pawn. 11... ♗a6 Trading the bishops isn't forced, but this is the simplest way to proceed, and illustrates how Black gets very good counterplay here even if there are no direct threats. ( 11... ♘a6 12.♕d2 ♖b8 13. ♘e5 ♕b6 14. ♖b1 ♗f5 15. ♗d3 ♗xd3 16. c5 ♕c717. ♘xd3 ♘e4 18. ♕c2 e5 19. O-O+/= Malaniuk-Iordachescu, Galatzi 2007 ) 12. cxd5 ♗xf1 13. ♔xf1 cxd5 White now needs to waste some time to get the king to safety and after, for instance, 14.g3 ♘bd7 15. ♔g2 ♖fc8 16. ♖c1 ♘b6 Black has fantastic compensation for the pawn because the bishop on b2 is a sorry sight and the c4-square is a constant source of worry for White. Generally speaking, he has a very passive position whereas Black can consider playing e6 and maybe getting the bishop into play by playing Bf8 and Bb4 later on. b) 9. ♗a3 cxb5 10. cxb5 ♗g4 11. ♗b4 ♘bd7 12. ♗e2 ( 12. ♘xa4e5! 13. ♗e2 ♖e8 14. dxe5 ♘x e5 15. ♘c5 ♘xf3+ 16. gxf3 ♗h3Akobian-Perelshteyn, Stillwater 2007 ) 12... ♗xf3 13. ♗xf3 ♘b6= c) 9. ♖c1 9... ♕a5 c1) 10. ♗d3 ♗g4 I finally switched back to the classical plan. (10... dxc4 11. ♗xc4 cxb5 12. ♗xb5 a3 13. ♗a1 ♘c6 14. OO♗e6 15. ♘d2 ♖fc8 16. ♘b3 ♕b6 17. ♘c5 ) 11. h3 ♗xf3 12.♕xf3 But here I got a little bit carried away. Jet lag had its say, as this took place in the years when Linares followed Morelia and the double jet lag did strange things to people in the second leg of the tournament.

c11) I played 12... e5 and what's even stranger is that Levon didn't take that pawn. After 13. dxe5 ( 13. cxd5 exd4 14.exd4 ♖e8+ 15. ♔f1 cxd5 16. g3 1/2 Aronian-Svidler, Morelia/Linares 2007 ) 13... ♘fd7 14. cxd5 ♘xe5 15. ♕e2♘xd3+ 16. ♕xd3 cxb5 17. OO b4 18. ♘d1 I believe White is better because his central passer will outweigh my strange passers on the queenside. c12) 12... cxb5 would have been much better, forcing White to take with the pawn. After 13. cxb5 ♘bd7 14. O-O e6∞ (or the immediate 14... ♘b6 there's a clear target on c4 for the knight and Black should be doing very well here. ) c2) 10. ♕d2 changes little. 10... a3 11. ♗a1 ♗f5 Preparing to play Ne4 in some positions, while there's also the obvious plan of Nbd7-b6 to put some pressure on the c4-pawn. I think Black is very comfortable here and shouldn't be suffering. 12. bxc6 bxc6 13.♗d3 ♗xd3 14. ♕xd3 ♘a6 ( 14... ♕a6 15. ♕e2 ♘bd7 16. c xd5cxd5 17. ♕xa6 ♖xa6 18. ♔e2 e6 19. ♘d2 ♖b8 20. ♖b1 ♖c821. ♖hc1 ♖ac6 22. ♘b3 ♗f8 23. f3+/= Tomashevsky) -Zinchenko, Saratov 7th N. Aratovsky Mem 2007 ) 15. O-O ♘b4 16.♕e2 ♘d7 17. ♖b1 e6∞ 6... c5 The most direct approach. My play in that game was frankly a long way from exemplary. I started by playing 6... e6 which is already a concession of sorts because it means I'm not fighting for direct equality (for reasons unexplained) and I got a slightly passive position. I then miscalculated and was for all intents and purposes lost by move 20. Not my most glorious hour. 7. O-O b6 8. b4dxc4 9. ♗xc4 ♗b7 10. a4 ♘bd7 11. ♗a3 ♖e8 12. ♕e2 My move here was a fatal miscalculation: 12... c5? (A few weeks later Anish Giri improved on my play: 12... ♘d5 13. ♘xd5 exd5 14. ♗d3 c6 15. ♖ac1a6 , and Black was solid but passive in Mamedyarov-Giri, Beijing SportAccord Rapid 2012 ) 13. dxc5 bxc5 14. bxc5 ♗f8 (I missed that14... ♘g4 is met with devastating effect by 15. ♘d4!± ) 15. ♗b5 ♖c816. ♖fd1 ♗xc5 17. ♘e5± Mamedyarov-Svidler, Tashkent FIDE GP 2012 7. dxc5 Here Black has a choice. 7... dxc4 The most forceful way of fighting for immediate and complete equality, and I think it suffices. 7... ♘a6 deserves a brief mention. It was played by Gelfand against Mamedyarov at the London Grand Prix. 8. cxd5 ♘xc5 9. O-O If you recall we had a similar position in the line with Bd2 and Rc1, but of course in this line White has made much more useful moves - Be2 and 0-0 - so here Black has no guarantee of equality. Gelfand played 9... a6 ( 9...♗f5⁉ ) 10. a4 ♗f5 and eventually drew the game after opening up the centre for his bishops, but this is at least a line in which you can see how White might be better. 11. ♘h4 ♗d7 12. ♖a3 ♖c8 13. a5 ♘ce4 14.♗d2 ♘xd2 15. ♕xd2 e6 16. ♘f3 exd5 17. ♘xd5 ♘x d5 18. ♕xd5♗c6=/∞ Mamedyarov-Gelfand, London 1st FIDE GP 2012 7... ♕a5 8. O-O ( 8. cxd5 An additional option for White. 8... ♘xd5 9.♕xd5 ♗xc3+ 10. ♗d2 ♖d8 11. ♕xd8+⁉ ♕xd8 12. ♗xc3 I don't believe White is better here, but at least he gets very interesting compensation for the queen. The position is extremely non-standard, which I think will appeal to a lot of people bored and tired with the correct official theory.12... ♕c7 13. b4 a5 14. ♗e5 ♕c6 15. b5 ♕xc5 16. OO ♗e6 17.♖fc1 ♕a3 18. ♖c3 ♕a4 19. a3 ♘d7 20. ♗d4 ♗d5 21. ♖b1=/∞Sumets-Le Quang Liem, Fujairah Masters 2012 ) 8... dxc4 9. ♕d4transposes to 7...dxc4 8.0-0 Qa5.

8. O-O Here Black has a choice between two perfectly decent moves. 8. ♕xd8 ♖xd8 9. ♗xc4 ♘bd7 a) 10. ♘g5 ♘e5 11. ♗e2 h6 12. ♘f3 ( 12. ♘ge4 ♘xe4 13. ♘xe4♗f5=/∞ ) 12... ♘ed7 13. c6 bxc6= b) 10. c6 10... bxc6 11. O-O ♘b6= 8... ♘a6 This move is also fine and has the additional bonus of perhaps giving White the opportunity to win a pawn somewhere, but allowing fantastic counterplay. 8... ♕a5⁉ a) On some occasions White tried 9. ♗xc4 ♕xc5 10. ♗b3 Even Laurent Fressinet once played this with White, but after 10... ♘c6 11.e4 ♗g4 you get an absolutely unbelievable version of the Bf4 Grünfeld. In general Black has to pay good money to get a position as good as this in that line. If anyone's fighting for an advantage here it's Black, so I think this really was a one-off occurrence. 12. ♗e3 ♕a513. h3 ♖ad8 14. ♕e2 ♗xf3 15. ♕xf3 ♘d7 ( 15... ♘d4= ) 16.♖fd1 e6 17. ♕e2 1/2 Fressinet-Dorfman, Belfort 2010 17... ♘c5=/+ b) 9. ♕d4 The only way to continue. 9... ♗e6 It's better to start with this move. (In a recent game between Sasikiran and Gupta Black played 9... ♘c6 10. ♕xc4 ♗e6 but after 11. ♕b5 he failed to achieve immediate equality and went on to lose the game. 11...♕xb5 12. ♗xb5 ♘d5 13. ♘a4 ♘db4 14. a3 ♘c2 15. ♖b1 ♖ad816. ♗d2 ♘e5 17. ♗ c3 ♘xf3+ 18. gxf3 a6 19. ♗e2 ♗b3 20. ♗xg7♔xg7 21. ♘c3 f5 22. ♖bc1 ♖d2 23. ♗d1 ♖fd8 24 . ♘b1+−Sasikiran-Gupta, New Delhi AICF-AAI Cup 2012 ) b1) 10. ♗xc4 gives Black a choice between 10... ♘fd7 , perhaps even trying to play for an advantage, and ( 10... ♘c6 11. ♕h4♗g4= where Black has no problems. ) b2) 10. ♘g5 is better, and after 10... ♘c6 11. ♕h4 ♗d7 White has a choice. 12. ♗d2 ( 12. ♕xc4 is an immediate draw after 12...♘e5 13. ♕d4 ♘c6= ) 12... h6 13. ♘ge4 ♘xe4 14. ♘xe4 ♕a415. ♗c3 ♗e6∞ A very double-edged position where I think Black is fine. 9. ♕d4 This is what Mamedyarov played against Anish Giri in Beijing. 9. c6 ♕c7 ( 9... ♕b6 10. cxb7 ♗xb7 11. ♗xc4 ♘c5 12. ♕e2♖ad8=/∞ ) 10. ♕a4 ( 10. cxb7 ♗xb7 11. ♕a 4 ♖ac8 and Black should already be thinking about an advantage because White has no comfortable way of developing his queenside, and if 12. ♘d2 Black simply plays 12... ♘c5 13. ♕xc4 ♗a6=/∞ and once again Black has terrific compensation for the pawn because White's queenside is horribly undeveloped and Black has fantastic play on the long diagonal and the c- and dfiles. ) 10... ♘c5 ( 10... ♕xc6 11. ♕xc6 bxc6 12. ♗xc4 ♘c5 13.♘e5 ♗b7 14. ♖d1 ♘fe4 15. f4 ♘d6 16. ♗e2 ♖fd8 17. ♗d2 ) 11.♕xc4 ( 11. cxb7 is pretty much impossible because of 11... ♘xa4 12.bxa8=Q ♘xc3 13. bxc3 ♗f5=/+ Black had an extra rook for a couple of moves but the queen on a8 is lost and Black is even better here because the pawn on c3 will be a target. The knight is coming to either e4 or d5. )11... ♕xc6= Black has completely equalised, at the very least. 9... ♕c7 10. ♗xc4 ♘xc5 11. ♕h4 ♗e6 12. ♗xe6 ♘xe6 13. e4 ♖fd8Black was perfectly fine. 14. ♗e3 A normal move.

As it was a blitz game Shakh continued in a very aggressive manner with14. e5⁈ ♘d5 15. ♘e4 ♗xe5 and did eventually win the game, although that was nothing to do with the opening. Black's just a pawn up here. Mamedyarov-Giri, Beijing SportAccord Blitz 2012 14... ♕a5 Black's position is perfectly comfortable here and shouldn't be a cause for concern for anyone playing Black.

4.e3 Bg7 5.Qb3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. e3 ♗g7 5. ♕b3 This line is slightly more serious than 5.Bd2. 5... e6 More ambitious than 5... c6. As usual Black wants to preserve the option of playing c5 at some point later in the game. 6. ♕a3 The whole point behind 5.Qb3, trying to stop Black from castling. Here, however, Black has a very concrete plan. He can play 6... a5 7.♘f3 c6! It's important to start with this much more precise move. In my game against Vassily Ivanchuk at the 2011 Tal Memorial I started with 7... ♘c6 believing there was no time to waste before moving the knight to b4 and castling, but after 8. ♗d2 ♘b4 9. ♕a4+ (a very important intermediate move) 9... ♗d7 10. ♕b3 White is ready to play a3, driving the knight away, and here Black might actually have some problems. 10... c5 (I chose 10... dxc4 11. ♗xc4 ♗c6 but after 12. a3♘bd5 White is clearly slightly better. He'll take on c6 and Black will be forced to ruin his pawn structure by recapturing with the pawn. 13.♘e5 O-O 14. ♘xc6 bxc6 15. ♕c2 Ivanchuk-Svidler, Moscow Tal Mem 2011 ) 11. dxc5 ♗c6 12. cxd5! I believe this is stronger than ( 12. a3♘a6 and the position was quite playable for Black in Andreikin-Fedoseev, Moscow 11th Aeroflot op A 2012 ) 12... exd5 ( 12... ♘bxd513. ♗b5+/= A lot of trades are likely to follow, which is obviously good news for White with his extra pawn. ) 13. ♘d4 O-O 14. a3 ♘a6 15.♗xa6 ♖xa6 16. OO ♘d7 17. ♘xc6 bxc6 18. ♘a4 , and Black will still have to prove he has full compensation for the pawn. 8. b3 The point of Black's move can be seen from 8. ♗e2⁇ dxc4 I was actually lucky enough to get this on the board in the World Blitz Championship in 2012 against Andreikin. White is obviously now in a lot of difficulties as he can't recapture on c4 without losing a piece on the spot. 9. b3 ( 9. ♗xc4b5−+ ) 9... b5 10. ♕b2 b4 11. ♘a4 c3 , and I was already winning in AndreikinSvidler, Astana World Blitz Final 2012 8. c5 This is one option for dealing with the threat of dxc4. 8... O-O 9.♗e2 ♘bd7 10. O-O b6⇆ It's quite clear that compared, for instance, to some Slav lines in which White plays c5 Black's pieces are very welldeveloped. In general, Black aims to put the bishop on the long diagonal in the Slav but has to waste time doing it, whereas here it's on g7 on move 4. Black has excellent counterplay on the queenside, so I think most people would prefer to be Black in this position. 8... ♘a6 Preparing to put the knight on b4 and castle comfortably.

9. ♗d2 If White plays 9. c5 here, as Viktor Korchnoi did on a number of occasions, then after 9... ♘d7 I think most people would prefer Black's position because Black has play on pretty much every part of the board. He can try playing e5, which is frankly almost impossible to stop. Loek van Wely went for a plan of expansion on the kingside first with g5 and h5. In some cases Black can also play on the queenside. 10. ♗xa6 ♖xa6 11.♘a4 g5 ( 11... O-O 12. ♗b2 ♕e7 13. OO ♖e8 14. ♖ac1 ♗f8! 15.♗a1 b5∓ Korchnoi-Kurnosov, Cheliabinsk 2007 ) 12. OO g4 13. ♘e1h5 14. ♘d3 ♕g5 15. ♕b2 e5 16. e4 ♕g6 17. ♘xe5 ♘xe5 18. dxe5♕xe4 19. ♗d2 h4⇆ Korch noi-Van Wely, Istanbul 2000 9. ♗e2 ♘e4 An important resource in many of these positions. White can almost never take on e4 because in many cases there are no good squares for the knight. 10. ♘a4 ( 10. ♘xe4 dxe4 11. ♘d2 Black can start to break in the centre here by playing c5, or maybe f5 first in this particular position. Black is very comfortable. ) 10... ♘b4 11. ♕b2 c5 12. O-OOO 13. ♖d1 cxd4 14. exd4 dxc4 15. bxc4 ♗d7∞ Korchnoi-Krasenkow, Lvov 2000 9... ♘e4! It's better to start with this move. After 9... ♘b4 10. ♕b2 Black will be driven back at some point, so you don't win that much time. 10... O-O 11. a3 ♘a6 12. ♗e2 b6 13. O-O♗b7 14. ♖fd1 ♕e7 15. ♘a4 ♘d7 16. ♖ac1 and White had a slight pull in Ibragimov-Roiz, ICC 2003 10. ♕b2 10. ♘xe4 dxe4 11. ♘g1 c5 10... O-O 11. ♖d1 ♘xd2 and Black gets a very decent version of something resembling the Moscow Variation of the Slav. In general I believe 5.Qb3 briefly became popular in 2011 due to the game Ivanchuk won against me, but what Black has to do against it has already been shown. So I think the days of the sudden popularity of this line are over.

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.e3 Bg7 5.cxd5 5.Na4 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 7.a3 7.Bg5 7.Qa4+

4.e3 Bg7 5.cxd5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. e3 ♗g7 5. cxd5 ♘xd5 6. ♘xd5♕xd5 7. ♘e2 This was briefly very fashionable around 2010, but it's hard to believe it can cause Black too many problems: 7... O-O 8. ♘c3 In this position Black has a variety of options. All of them are perfectly sound, but you still have to pick one. 8... ♕d8 The most concrete option and the one I recommend for the purposes of this series. 8... ♕d6 is very solid, and has been played by a lot of people, myself included: 9. ♗e2 c6 10. OO ♘d7 11. ♕c2 A very natural move that Kramnik played against me in Monaco 2010. ( 11. ♘e4⁉ ♕c7 12. b3 , preparing Ba3 and thus preventing e7-e5, at least for the time being, makes some sense, but Black still gets a very safe game: 12... ♖d8 13.♕c2 ♘b6 14. ♗b2 ♗f5 15. g4 ♗xe4 16. ♕xe4 e6 Postny-Giri, France 2010 ) 11... e5 12. dxe5 ♕xe5 13. e4 White retains some modicum of pressure. 13... ♘f6 14. ♗e3 After the following semi-forced sequence14... ♘g4 15. ♗xg4 ♗xg4 16. h3 ♗d7 17. ♖ad1 ♕e7 18. ♗d4 I still had to solve a few problems to make a draw in Kramnik-Svidler, Nice 2010 8... ♕a5⁉ is a slightly different set-up, aimed at early breaks in the centre. To prevent that White has to play 9. ♗d2 after which I couldn't find direct equality for Black. His position is fine, but I think White retains some hopes of an advantage. 9... c6 10. ♗c4 ( 10. ♘e4 ♕b6 11. ♗c3♘d7 12. ♗c4 c5 was immediately equal: 13. ♖c1 cxd4 14. ♗xd4♗xd4 15. ♕xd4 ♕xd4 16. exd4 ♖d8 17. ♗b3 ♘b6 in ErdosNepomniachtchi, Warsaw EU-ch Rapid 2011 ) 10... ♘d7 11. O-O ♕d8 , and now White should play 12. ♕e2⁉ ( 12. ♗b3 b6 13. ♕f3 ♗b7 14.♖fd1 e6 15. ♖ac1 ♕e7 Narciso Dublan-Hamdouchi, La Massana 2010 )12... e5 13. d5 ♘b6 14. ♖fd1 cxd5 15. ♘xd5 ♘xd5 16. ♗e1 ♗e617. ♗xd5 ♗xd5 18. e4+/= 9. ♗e2 c5 The difference between this position and the position with the queen on d6 is quite clearly the fact that White doesn't have Ne4, and therefore c5 has to be met by d5, opening up the long diagonal for the bishop on g7 - always a welcome sight for any Grünfeld player. 9... c6 10. O-O e5 11. dxe5 ♗xe5 12. e4 gives White some hopes of an edge: 12... ♕e7⁉ ( 12... ♕xd1 13. ♖xd1 ♗e6 14. f4 ♗c7 15. ♗e3♘d7 16. ♖ac1 ♗b6 17. ♔f2 ♘f6 18. h3 h5 19. ♗f3 ♖fd8 20. e5♘e8 21. b3 was quite nice for White in Gelfand-Aronian, Nice 2010 ) 13.♗e3 ♖e8 14. ♕c2 b6 15. ♖ad1 ♘a6 16. ♗d4 ♘b4 , and now 17.♕a4! ( 17. ♕b1 ♗xd4 18. ♖xd4 ♗e6 19. a3 c5 20. ♖d2 ♘c6 21.♗b5 ♖ec8 22. ♗xc6 ♖xc6 was OK in Fressinet-Kurnosov, Pamplona 2010 ) 17... ♗xd4 18. ♖xd4 a5 ( 18... c5 19. ♖d2 ♗b7 20. ♖d7 ♗c621. ♖xe7 ♗xa4 22. ♖c7!+/= ) 19. a 3 would have given Black some problems to solve.

10. d5 e6 11. e4 exd5 12. exd5 b6⁉ A very forceful option. 12... ♕b6 13. O-O ♘a6 , as recommended by Boris Avrukh, is a very decent idea and very reminiscent of the positions that arise from the Qb3-line we'll mention later. 13. O-O ♗b7 White already has to decide what to do with the pawn on d5. 14. ♗f3 The main move, of course, protecting the pawn. 14. ♕b3 ♘d7 15. ♗e3 ♗d4 This makes a lot of sense. White now pretty much has to allow Bxe3, as he lacks a sensible alternative. Play might continue 16. ♖ad1 ( 16. ♖ae1 ♗xe3 17. fxe3 Zivanic-Kritz, Brownsville 2010 17... ♕e7 18. e4 ♘e5∞ ) 16... ♗xe3 17. fxe3 ♕e7 Perhaps more precise than playing Ne5 immediately, as once the pawn is on e4 White no longer has the option of playing Ne4 later in the game. 18. e4 ♘e5Structurally Black is perfectly fine here. He only needs to transfer the bishop from the b7-square, where it's currently doing very little, to d7 in order to be absolutely comfortable. Hrant Melkumyan played 14. ♗f4⁉ in this position against Maxime Vachier-Lagrave, just giving up the pawn. 14... ♗xc3 15. bxc3 ♕xd5 16.♕xd5 ♗xd5 17. ♖fd1 ♗b7 After the sacrifice is accepted and the bishop retreats White clearly has compensation for the pawn. However, although sufficient for a draw I don't think it's any more than that. In the absence of knights and queens the weakness of the dark squares around the black king shouldn't be so telling. Black has no need to feel unduly worried. 18. ♖d6 ( 18. ♗c4 ♖e8 19. f3 ♘c6 20. ♖d7 ♖e7 21. ♖ad1♖ae8=/∞ ) 18... ♘c6 19. ♗g5 ♖ae 8 20. ♗f3 ♖e6 21. ♖xe6 fxe6 22.♗g4 ♔f7 23. ♖d1 ♗c8 24. ♖d6 ♘e7 25. c4 Melkumyan-Vachier Lagrave, BL 2012 14... ♘d7 This has been seen in a couple of games by GM Kudrin. 15. ♖e1 15. d6 ♗xf3 16. ♕xf3 ♕f6 17. ♕d5 ♕e5 ( 17... ♕d4 is imprecise, as after 18. ♖d1 Black is forced to take on d5. 18... ♕xd5 19. ♘xd5♖ae8 20. ♘e7+ ♔h8 21. ♖b1 Benjamin-Kudrin, Saint Louis 2010 ) 18.♖d1 ♖ae8= Generally speaking, in Grünfeld endgames if the pawn on d6 is safely blockaded by the knight and Black develops all his pieces naturally he should be doing very well, as he is here. 15... ♘e5 16. ♗e2 c4 Now Black always has the option of playing Nd3, or keeping it in reserve. Kudrin started some adventurous play on the kingside with 16... f5⁈ but I don't think that was necessary. 17. ♗f4 g5 18. ♗xe5 ♗xe5 19. ♗c4Lenderman-Kudrin, Saint Louis 2010. 17. ♗f4 ♖e8 18. ♕d2 ♕d7 19. ♖ad1 ♖ac8⇆ Black's position is very comfortable and he can face the future with optimism.

4.e3 Bg7 5.Na4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. ♘a4 This line was popular in the 90s but has almost disappeared from grandmaster practice since Black figured out what to do about it. The recommendation given in Avrukh's book is eminently sensible. 5. ♗d2 is a very serious move that will be covered separately.

5... ♗g7 6. e4 ♘b6 7. ♗e3 O-O This is obviously the most natural way for Black to respond to 5.Na4 8. ♘f3 ♗g4 9. ♘c5 The main move in this position. 9. ♗e2 ♘c6 10. d5 ♘e5 Black is doing very well in this position because he already has decent counterplay against the white centre. The forcing line is good for Black 11. ♘xe5 ( 11. ♘xb6 axb6 12. ♘xe5 ♗xe2 13.♕xe2 ♗xe5 14. OO ♕d7 15. a3 e6 16. ♖ad1 ♖fe8= Legky-Mikhalevski, Montreal 2003 ) 11... ♗xe2 12. ♕xe2 ♘xa4 13. f4 and now a very important move 13... f6! with f5 to follow after any of White's retreats. 14. ♘f3 ( 14. ♘c4 f5 15. ♕c2 ♘b6∞ ) 14... f5 It turns out Black is doing very well in these complications. For instance, 15. d6fxe4 16. ♕c4+ e6 and the game should end in a perpetual with best play from both sides. 17. ♘g5 ♘xb2 18. ♕xe6+ ♔h8 19. d7 ♘d3+ 20.♔e2 ♘xf4+ 21. ♗xf4 ♖xf4 22. ♘f7+ ♖xf7 23. ♕xf7 ♕g5 24. ♕e8+♗f8 25. ♕xa8 ♕xg2+= 9... ♘c6 Of course Black didn't play all this only to defend the pawn on b7. 10. ♘xb7 ♕b8 , and play becomes very sharp and forced: 11. ♗a6 11. ♘c5 offers White nothing. 11... ♗xf3 12. gxf3 Black now has a forced draw. (After 12. ♕xf3 Black is even better. 12... ♘xd4 13. ♕d1♘c4! A very nice tactical shot, creating the huge threat of 14. ♗xd4♕b4+ when White is probably already in some difficulties. ) 12... ♖d813. d5 e6 and White has nothing better than repeating moves. 14. ♘a6♕b7 15. ♘c5 ♕b8= 11... e5 Immediately attacking the centre. 12. d5 If 12. dxe5 ♘xe5 13. O-O Black has an excellent position after 13...♕c8 , creating the very important threat of Bxf3 and Qh3. 14. ♗f4Transferring the bishop to g3. 14... ♗xf3 15. gxf3 ♖b8 An important move that frees up the d8-square for the black rooks. 16. ♘c5 ♕h3 Black is doing extremely well here. 12... ♘b4 13. ♘c5 In this position Black has a forced sequence that leads to a very comfortable position for him – 13... ♘6xd5 14. exd5 White's only chance to fight for any kind of advantage. 14. ♗b7 ♘xe3 15. fxe3 a) 15... ♘a6⁉ Of course it's very difficult to resist such a fantastic-looking move when you're playing the position for the first time at the board, but objectively this is inferior to 15...Bxf3. 16. ♗xa8 ♘xc5 17.♗d5 ♕xb2 ( 17... c6 18. ♗xc6 ♗h6 19. ♕e2 ♕d6 20. ♗d5 ♘xe4 as in Kiss-Gara, Hungary 2008, is less convincing: 21. ♗xe4 ♕b4+ 22.♔f2 ♕xe4 23. ♖ac1± ) 18. O-O ♗h6=/∞ b) ♗xf3! As correctly pointed out by Avrukh. 16. gxf3 ♖d8 17. ♕b3♗f8 b1) 18. ♕xb4 a5 might even be slightly better for Black 19. ♕b5 (19. ♕b3 ♖a7=/+ ) 19... ♗xc5 20. ♕xc5 ♕xb7 21. ♕xe5♖ab8=/+

b2) 18. ♗xa8 18... ♗xc5 19. ♗d5 ♘xd5 White has a couple of positions to choose from, but neither of them offer him any advantage. 20. exd5 ( 20. ♕xb8 ♖xb8 21. exd5 ♖xb2 and Black clearly has a lot of compensation for the exchange. 22. e4 ♗f2+23. ♔d1 ♗e3= ) 20... ♕xb3 21. axb3 ♗xe3= and once again White's structure is in complete ruins whereas from b6 the bishop will cover all Black's weaknesses on the queenside. It's quite clear Black can have no problems here. 14... e4 15. ♘xe4 White should take on e4 because 15. O-O ♘xa6 16. ♘xa6 ♕c8 is fine for Black. Play might continue 17.♘xc7 ♕xc7 18. h3 ♗xf3 19. gxf3 ♖ae8=/∞ and Black has wonderful compensation for the single pawn he had to sacrifice. It's also important to know that 15. ♗b7 leads to a worse position for White after 15... ♘d3+! 16. ♘xd3 ♕xb7=/+ and Black will win back material very soon. 15... ♘xa6 15... ♗xf3 16. gxf3 ♘xa6 17. O-O ♕xb2 18. ♕a4+/= 16. O-O ♖e8 Some precision is required from Black here. For instance, 16... ♕xb2⁈ would be a mistake because of 17. ♗d4! ♗xf3 18.♗xg7!+/= and now Black is left with a knight on a6 against a knight on e4, which quite clearly isn't a very good situation to be in. 17. ♘g3 The knight really has no good squares apart from the g3-square, and here some precision is once again required. Black should start with 17... ♗xf3 The immediate 17... ♕xb2 runs into 18. ♕a4! ♗xf3 19. gxf3 and the knight on a6 is once more in trouble. 17... ♕b5 18. h3 ♗xf3 19. ♕xf3 ♖ad8 ( 19... ♘b4 20. ♖fd1 ♖ad8 )20. ♖fd1 ♕xb2 21. ♖ac1 ♕xa2 22. ♘e4→ 18. ♕xf3 If 18. gxf3 Black of course plays 18... ♕b5 preventing Qa4 and beginning to round up the pawn on d5. He should be doing extremely well here. 18... ♕xb2 19. ♖ad1 and in this position I believe it's important to make one last precise move – 19. ♘e4 ♕e5= 19... ♕b4 Black needs to make sure the knight on g3 doesn't enter the fray, so Qb4 is aimed at cutting the knight off from the e4-square. As long as the knight is stuck on g3 Black has no difficulties here. 20. ♖b1 ♕a4 21. ♖fc1 ♖ab8=

4.cxd5 5.e4 7.Ba3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 The main move. 5... ♘xc3 6. bxc3 ♗g7 We'll cover three moves in this position, 7.Ba3, 7.Bg5 and 7.Qa4+. 7. ♗a3 This was somewhat popular in the late-90s and at the beginning of the new millennium, mainly due to the efforts of Vadim Milov, but people got to grips with it and the line has almost completely disappeared from grandmaster practice. 7... ♘d7 It's important to start attacking the somewhat awkward bishop on a3 immediately. 8. ♘f3 If White tries to save on 8.Nf3 by playing something like 8. ♕b3 c5 9.♖d1 O-O 10. ♗e2 ♕a5 11. ♘f3 Black has the very strong move 11...♘f6! attacking the white centre. Black is very comfortable here. 12.♗xc5 ♘xe4 13. ♗b4 ( 13. ♗xe7 ♖e8 14. ♗b4 ♕c7 ) 13... ♕b6 8... c5 White has a choice of whether to include 9.Bc4 here. 9. ♕b3 The attempt to save on the move. White generally plays 9. ♗c4 cxd4 10. ♕b3 (If White tries to save on Qb3 here and instead castles you should take the second pawn - 10. O-Odxc3 because 11. ♗xf7+ ♔xf7 12. ♘g5+ ♔e8 is no threat to Black at all, and in fact White is almost lost here. 13. ♘e6 ♕b6 14. ♘xg7+♔f7∓ ) 10... O-O 11. cxd4 ♘b6 This position has been seen in many games. 12. ♗e2 is the main move in this position. (If 12. O-O then 12...♘xc4 13. ♕xc4 ♗g4 is quite obviously very comfortable for Black. )12... ♗e6 13. ♕b4 ♖c8! An important move after which I believe only Black can be better. He has no need to waste any time defending the e7-pawn because his counterplay against the white centre is very strong. a) 14. ♕xe7 ♕xe7 The simplest approach. ( 14... ♘c4⁉ ) 15. ♗xe7♖fe8 16. ♗d6 ♗c4 17. e5 ♗xe2 It's very difficult for White to get his king to safety. ( 17... f6⁉ can also be played first, and it's only Black who can be better. ) 18. ♔xe2 ♘c4 19. ♔d3 ( 19. ♗b4 a5 20.♗d2 f6=/+ ) 19... ♘b2+ 20. ♔e2 ♘c4= b) 14. O-O In my game against Vadim Milov I went for 14... ♗c4 15.♖fe1 ♗xe2 16. ♖xe2 ♘c4 and once again Black is very comfortable. I even went on to win the game against Milov in the 2001 World Championship, although that had a lot to do with the fact that Vadim had to play for a win after losing the first game of a knockout match.17. ♖d1 ♕d6=/+ Milov-Svidler, Moscow FIDE WCh 2001 9... O-O Black generally aims to play b5 as soon as possible is Bc4 is omitted e.g. 10. ♗e2 10. ♗d3 ♖b8 11. O-O b5⇆ and once again the bishop on a3 is left looking somewhat stupid. 10. ♖d1 ♕c7 11. ♗d3 b5! 12. ♗xb5 ♖b8 13. ♕c4 ♕a5→ 10. ♗c4 cxd4 11. cxd4 ♘b6

10... ♕c7 11. O-O ♖b8 12. e5 e6 13. ♘d2 b5 13... b6 14. ♘c4 ♗a6 15. ♖fe1 ♗xc4 16. ♗xc4 ♖fd8 17. ♗b5+/=Papaioannou-Krase 14. ♘e4 ♗b7 15. ♘xc5 ♘xc5 16. ♗xc5 ♗xe5! 17. ♗xf8 ♗xh2+ 18.♔h1 ♖xf8=/∞ Avrukh

4.cxd5 5.e4 7.Bg5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6. bxc3♗g7 7. ♗g5 A more recent and serious move, but here as well Black has managed to figure out a way to proceed after initially somewhat struggling with this strange-looking idea. 7... c5 8. ♖c1 O-O And in this position White has a choice between playing d5 immediately or starting with Nf3. 9. ♘f3 The more critical line. 9. d5 If White starts with this move Black has a very pleasant choice. a) 9... ♕d6⁉ is a very viable alternative if you don't want to enter the forcing lines after 9...f5 (those tend to end in forced draws). It carries the immediate threat of Qe5, winning the central pawn. a1) 10. ♕d2 ♘d7 ( 10... f5 11. ♗c4 ) a11) 11. ♘f3 e6 12. ♗e2 exd5 13. O-O ( 13. exd5 ♘f6 )13... ♘b6 a12) 11. c4 11... e6⇆

a2) 10. ♘f3 ♘d7 11. ♗d3 e6 a3) 10. ♗d3

a31) 10... ♘d7 11. ♘f3 ( 11. ♘e2 f5 12. exf5 ♘e5 13. OO♘xd3 14. ♕xd3 ♖xf5 15. ♗e3 ♕xd5 16. ♕xd5+ ♖xd5 17.♘f4 ♖d8∓ Housi eaux-Vachier Lagrave, Port Marly French Cup Final 2011 18. ♗xc5 e5 19. ♘e2 ♗e6∓ ) 11... e6 Black gets a lot of counterplay because it's very difficult for White to keep the centre stable while his king is still on e1. 12. dxe6 ♕xe6∞ a32) 10... e6 11. c4 exd5 12. exd5 ( 12. cxd5 f5 13. f3fxe4 14. fxe4 ♘d7→ ) 12... b5⇆

b) 9... f5 10. ♗c4 ♔h8 is considered fine by theory and I think it is indeed fine for Black. b1) 11. exf5

b11) 11... ♗xf5 12. ♕e2 ♕a5 13. ♘f3 ( 13. ♗xe7 ♕a3 ) 13...♗xc3+ 14. ♘d 2 ( 14. ♔f1 ♘d7 15. ♗xe7 ) 14... ♗f6 15.♗h6 ♖d8 16. O-O ♘d7 17. g4 b5 18. gxf5 bxc4 19. ♘xc4♕c7

b12) ♖xf5 12. ♘f3 ♘d7 13. ♗e2 ( 13. ♗b3 ♘e5 14. ♘xe5♖xe5+ 15. ♗e3 b5 ) 13... ♘f6 14. c4 ♘e4 b2) 11. ♕e2 h6 ( 11... b5 12. ♗xb5 fxe4 13. ♕xe4 ♖f5 14.♘f3 Krasenkow-Lavendelis, Warsaw EU-ch Rapid 2010 ) b3) 11. ♘e2 11... fxe4 12. ♘g3 Black has a choice. 12... ♗f5leads to a fine position for Black. ( 12... ♘d7⁉ leads to a very decent and complex position. 13. ♘xe4 ♘b6⇆ ) 13. ♕e2 ( 13.OO ♘d7 14. d6 ♗f6 15. ♗h6 ♗g7= ) 13... ♘d7 14. ♘xe4 h615. ♗h4 g5 16. ♗g3 ♘b6 1 7. ♖d1 ♘a4 Lysyj-Sutovsky, Sochi RUS-tch 2008 ( 17... ♘xc4 18. ♕xc4 b5 19. ♕e2 ♗xe4 20. ♕xe4♗xc3+ 21. ♔f1 ♕d7=/+ ) 9... ♗g4 10. d5 This is the position from which I lost my first round game against Tomi Nyback in the 2009 World Cup, but then in the rapids I was already able to demonstrate what I believe to be the correct set-up. Black starts by playing 10... ♕d6 In general this is a very important move in this line because the queen is extremely comfortable here and it also frees up the e7-pawn so you can play e6 later on in the game, if required. 11. ♗e2 This is of course the main move. 11. ♕b3 is possible here, and although the move isn't particularly critical I want to show you a very nice position for the sake of pure aesthetics.11... ♗xf3 12. gxf3 ( 12. ♕xb7⁈ isn't a good move because of the very strong 12... ♗xe4 13. ♕xa8 ♗xd5 14. ♕xa7 ♘d7 ) 12... b6 And now if White goes for ( 12... ♘d7⁉ ) 13. f4 ( 13. ♗e3 f5 14. h4 ♘d7 15. h5♘e5 16. ♗e2 f4 17. ♗d2 g5∓ ) 13... h6 14. e5 ♕c7 a) A small point, or actually an important point, is that 15. ♗h4 runs into 15... ♗xe5 and after 16. fxe5 ♕xe5+ 17. ♗e2 Black will play either 17... ♕f4∓ (or 17... ♕e4 winning the bishop on h4.)

b) 15. ♗g2 and in this position Black has the very beautiful move 15...♘c6!∓ after which White has no equality. This is the move I wanted to show you because it's not every day you get to play Nb8-c6 with a white pawn on d5 to such devastating effect. 11. ♕d2 ♘d7 12. ♗e2 ♗xf3 13. gxf3⁈ ( 13. ♗xf3 ) 13... f5 11. h3 ♗xf3 12. ♕xf3 ♘d7 13. ♗e2 f5 11... ♘d7 12. O-O Black once again has a choice. 12. h3 Tomi's choice in our rapid game. 12... ♗xf3 13. ♗xf3 b5 Black demonstrates his main idea, b5. If White now does nothing Black will play c4, put the knight on c5 and get a lot of play on the queenside. So White generally plays 14. c4 bxc4 This is more concrete than ( 14... b4 which is also fine and eventually enabled me to draw the game. 15. O-O a5 16.♗g4 ♗d4 17. ♖e1 e5 Nyback-Svidler, Khanty Mansiysk 2009) 15. ♖xc4♘e5 ( 15... ♗d4⁉ is also fine for Black.) 16. ♖a4 ( 16. ♖c1 ♘xf3+ 17.♕xf3 f5= ) 16... f5⇆

12... f5⁉ Trying to exploit the fact that the bishop is still alive on g4.

12... ♗xf3 13. ♗xf3 b5 Despite being a whole tempo down compared to the line above I believe this is fine for Black. White still needs to play 14.c4 and after ( 14. ♗e2 c4 ) 14... bxc4 ( 14... b4 15. ♗e3 ) 15. ♖xc4♗d4⇆ Black should have very comfortable play.

13. h3 Pretty much forced. 13. ♘d2 runs into the very, very strong 13... ♗xe2 14. ♕xe2 h6 and now after 15. ♘c4 ♕a6 16. ♗xe7 ♘e5 the pin along the a6-f1 diagonal wins Black some material. 17. ♗xf8 ♖xf8∓ 13. exf5 ♖xf5 14. ♗e3 ♘e5=/+ clearly isn't an attractive proposition for White. 13... ♗xf3 14. ♗xf3 ♘e5! Now Black has no problems whatsoever. 14... f4 15. ♗g4 ♘e5 16. ♗e6+ ♔h8 17. g3! fxg3 18. f4+/= 15. g3 Almost forced, as f4 is quite a serious positional threat. Black could continue on or just equalise on the spot by trading everything. 15... ♘xf3+ 16. ♕xf3 fxe4 17. ♕xe4 ♖f5 18. ♗xe7 ♕xd5 19.♕xd5+ ♖xd5= With e7 and d5 gone the game is likely to end in a draw soon.

4.cxd5 5.e4 7.Qa4+ 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6. bxc3♗g7 7. ♕a4+ This has been a resurgent line against the Grünfeld recently, and one played by a number of very strong players. I encountered it in one of the most important games of my career, from a purely sporting perspective, in the semi-final of the 2011 World Cup against Ruslan Ponomariov. 7... ♘d7 The strongest move, leaving White the choice of whether to play 8.Nf3 or try to postpone it and develop the knight to e2. I played 7... ♗d7 for some reason believing that White had to play Qb3, after which c5 gives Black a lot of counterplay. Instead, of course, 8.♕a3 followed, and after 8... ♘c6 9. ♘f3 e5 Ruslan played 10. ♗e3which led to an extremely non-standard position. That eventually worked in my favour as I won the game to progress to the World Cup final. (Objectively, however, this doesn't quite equalise for Black, and after let's say the simple 10. d5⁉ ♘e7 11. c4 I prefer White's position.; 10.♗g5⁉ f6 11. ♗e3 ) 10... exd4 11. cxd4 ♕e7 12. ♕xe7+ ♘xe7 13.♖b1 ( 13. ♖c1 ) 13... O-OO 14. ♗c4 f5 15. ♘g5 fxe4 16. ♘f7 ♘f5Ponomariov-Svidler, Khanty-Mansiysk World Cup 2011 8. ♘f3 Currently the main line. 8. ♗e3 OO 9. ♖d1 ( 9. ♖c1 c5 10. ♘f3 ♕c7 11. h3 a6 12. ♕a3b5! 13. ♗xb5 axb5 14. ♕xa8 ♗b7 15. ♕a3 ♗xe4 16. O-O ♕b7=/∞ )9... c5 10. ♗d3 Aiming to play Ne2 later. In general Black's counterplay in this variation is connected with expansion on the queenside via a6 and b5, but in this particular position I think it's better to do something else. Let me illustrate why. ( 10. ♘f3 a6 11. ♗d3 b5 12. ♕a3∞ )

a) If Black plays 10... a6⁉ it's important to start with 11. ♕a3 because otherwise b5 will come in a single tempo, as Bxb5 runs into Nb6, and White loses a piece. ( 11. ♘e2 b5 12. ♕a3 ♗b7= ) 11... ♕c7 12.♘e2 This position is still quite playable for Black, but I think it's important that the knight is on e2 and not f3, a factor which is to White's benefit. (If White plays 12. ♘f3 in this position Black has very thematic counterplay starting with 12... b5! If White accepts the exchange sacrifice with 13. ♗xb5 axb5 14. ♕xa8 ♗b7 15. ♕a3♗xe4 Black gets a lot of counterplay based on the fact that the knight on f3 is quite badly misplaced. For instance, 16. O-O ♕b7 here and the knight might have to go to e1, which is a very ugly square. Generally speaking, Black has a lot of positional counterplay for the exchange, whereas with a knight on e2 this whole line is less attractive and Black will have to play) 12... ♖b8 wasting an additional tempo before he can play b5. This is still very playable for Black, but it's a concession, so I think it's stronger to start with 10... Nb6. 13.OO ( 13. ♘f4 e6 14. d5 e5 ) 13... e6 14. f4 b5 15. f5 exf5 16.♗f4 ♕c6 17. ♗xb8 fxe4 1 8. d5 ♕b6 19. ♗xe4 ♘xb8=/∞ Goldin-Mikhalevski b) 10... ♘b6 11. ♕a3 cxd4 12. cxd4 ♕d6 Also a very thematic motif in this line, as played by Ivanchuk against Milov. In the endgame after13. ♕xd6 exd6 14. ♘e2 f5 Black's position was very comfortable. I think this is the correct way to proceed if White tries to save on the 8.Nf3 move. 15. OO fxe4 16. ♗xe4 ♗g4 17. f3 ♖ae8 18. ♘c3♗f5 19. ♗f2 ♗xe4 20. ♘xe4 d5= MilovIvanchuk, Warsaw 2005 8... O-O 9. ♗e2⁉ By process of elimination. 9. ♗e3 c5 10. ♖d1 transposes. Once again, if White plays something like 9. ♗g5 c5 10. ♖c1 Black launches the typical counterplay. 10... a6! 11. ♕a3 ( 11. h3 b5 12.♕a3 ♕b6 13. ♗d3 ♗b7 14. OO ♖fe8 15. ♗e3 ♖ac8 ) 11.. . ♕c7 The pawn on e7 doesn't need to be protected. 12. ♗d3 b5!= ( 12... e5⁉ ) 13.♗xb5 ( 13. ♗xe7 ♖e8 14. ♗h4 ♗b7=/+ Black has excellent compensation for the sacrificed pawn.; 13. O-O ♗b7 ) 13... axb5 14.♕xa8 ♗b7 15. ♕a3 ♗xe4 16. OO ♕b7=/∞ and the same compensation we saw in the previous line is again extremely strong for Black. I think he's very comfortable here. 9... c5 10. O-O Alexander Morozevich has recently played a few interesting games in this position. 10. ♖b1 ♘b6 11. ♕a3 cxd4 12. cxd4 ♗g4 13. d5 f5 10... ♘b6 My suggested way of dealing with this line. It's somewhat minimalistic, in that Black is going for a position which isn't particularly ambitious in terms of playing for a win, but at least your king is safe and you don't run the risk of getting mated. 10... a6 In the Tashkent Grand Prix Dominguez played 11. ♕a3 In the Tashkent Grand Prix Dominguez played In the Tashkent Grand Prix Dominguez playedIn the Tashkent Grand Prix Dominguez playedIn the Tashkent Grand Prix Dominguez played 11... ♕c7 against Morozevich, which isn't ideal, as after 12. ♗g5 you have to waste a tempo defending the e7-pawn. 12... ♖e8 13. ♖ad1 ♖b8 14. d5 Black was already facing quite serious problems and went on to lose the game very quickly. 14...e6 15. dxe6 ♖xe6 16. ♗h4± Morozevich-Dominguez Perez, Tashkent FIDE GP 2012 10... ♕c7⁉

a) 11. ♖b1 a6 ( 11... b6 12. e5 ♗b7 13. ♗g5 f6 14. exf6 ♘xf615. ♕b3+ ♔h8 16. ♖fe1 ♖ad8⇆ ShipovNakamura, ICC 2002 ) 12.♕a3 b5 ( 12... ♖b8 13. ♗g5 ♘f6 14. ♗d3 b5 15. ♕c1 ♖b6 16.♗f4 ♕d8 17. h3 ♗b7 18. ♕e3+ /= Aseev-Shipov, Russia 2002 ) b) 11. ♗g5 11... ♘f6 ( 11... a6 12. e5 e6 13. ♕a3 ♖e8 14. ♘d2h6 15. ♗e3 b6 16. ♘e4 ♗f8 17. ♕b2 ♗ b7 18. ♗d3+/= Milov-Avrukh, Andorra 2001 ) 12. d5! ( 12. ♗d3 ♗g4 13. ♘e5 ♗e6 14.♗e3 ♖fd8∞ ) 12... h6 13. ♗d2 ( 13. ♗h4 ♕f4 ) 13... ♖d8 b1) 14. ♖fd1 ♗d7 15. ♕c2 e6! 16. c4 ( 16. d6 ♕a5∞ ) 16...exd5 17. exd5 ♘e4! 18. ♕xe4 ♖e8 19. ♕h4 ♖xe2 20. ♗xh6♗xa1 21. ♖xa1 ♕d6 b2) 14. ♕c2! 14... b6 15. a4+/= e6 16. d6 ♖xd6 17. ♗f4+/= Much more critical is 10... cxd4 11. cxd4 ♘c5 as played by Levon Aronian against Morozevich in the Beijing SportAccord tournament. I think Black is fine in the lines that follow, but I don't really want to get into the position too deeply as I see no particular reason for Black to enter it. He has no hope of getting an advantage and in general has to be very precise not to run into some serious trouble. a) Here White doesn't, of course, play 12. ♕c2 which after 12... ♘e6gives Black a very comfortable position.( 12... ♗g4⁉ 13. e5 ♘e6 ) 13.♗e3 ( 13. ♗b2 ♘xd4 14. ♘xd4 ♗xd4 15. ♖fd1 e5 16. ♗xd4exd4 17. ♕b2 ) 13... ♘xd4 14. ♗xd4 ( 14. ♘xd4 ♗xd4 15. ♖ad1e5 16. ♗xd4 exd4 17. ♕b2 ♗e6 18. ♕xb7 ♕ b6 Miton-Ivanchuk, La Habana 2007) 14... ♗xd4 15. ♖ad1 e5 16. ♘xe5 ♕b6 is great for Black. b) 12. dxc5⁉ Black's task is very tricky: b1) 12... ♗xa1 13. e5 requires a lot of attention from Black: ( 13.♗h6 ♗g7 ; 13. ♗f4 ♗c3 ) 13... ♕c7 ( 13... ♗c3 14. ♗h6 ♕a515. ♕h4+− ) 14. ♕f4! ( 14. ♗f4 ♗c3 15. ♖c1 ♕a5 ) 14... ♗c3 , and the machine insists Black will survive, but it's very difficult to recommend for anyone. What follows are some sample lines - I make no claims that this is exhaustive analysis. I just wanted to illustrate the problems Black faces here: ( 14... f6 15. ♕c4+ e616. ♗h6 ♗xe5 17. ♗xf8 ♔xf8 18. ♖d1 , and Black will be suffering for a while because his queenside is very hard to develop.18... b5 19. ♕xb5 ♖b8 20. ♕c4+/= ) b11) 15. ♕h6 f6 16. ♗c4+ e6=/+ b12) 15. ♗c4⁉ b121) 15... ♗f5 16. g4 ♗c2 17. ♕e3 ♗b4 18. ♕h6 g5 19.♗xg5 ♔h8 20. ♗f6 + ( 20. ♗b3 ♗xb3 21. ♗f6+ exf6 22.♕xf6+ ♔g8= ) 20... exf6 21. ♕xf6+ ♔g8 22. e6 ♗g6 23.♘h4 ♗xc5 24. exf7+ ♖xf7 25. ♘xg6 hxg6 26. ♕xg6+♔f8= b122) e6 16. ♕e3 ( 16. ♖d1 f6 17. ♕e3 ♗xe5 18. ♘xe5♕xe5 19. ♕xe5 fxe5 20. ♗h6 ♖e8 21. ♗b5 ♔f7= ) b1221) 16... b5 17. ♗xb5 ♕a5 ( 17... ♖b8 18. a4 a619. ♗c4 ♗b4 20. ♕h6 ♗xc5 21. ♘g5 f5 22. ♗xe6+♗xe6 23. ♘xe6 ♕xe5 24. ♘xf 8 ♗xf8= ) 18. a4 a6 19.♗e2 ( 19. ♕h6 f6 20. ♗c6 ♖b8 21. exf6 ♗x f6 22.♘g5 ♗xg5 23. ♕xg5 ♕c7 24. ♗e4=/∞ ) 19... ♖a7 ( 19...♗b7 20. ♘g5! ♔g7 21. ♖d1!± ) 20. ♘d2 ♗xd2 21.♗xd2 ♕xa4 22. ♗f3 ♖b7! 23. ♗xb7 ♗xb7=

b1222) 16... ♕a5 17. ♕h6 f6 18. exf6 ♗xf6 19. ♗e3 (19. ♘g5 ♗xg5 20. ♕xg5 ♕c3 21. ♕h4 ♗d7 22.♕e4=/∞ ) 19... ♗d7 20. ♘g5 ♗xg5 21. ♕xg5=/∞ b13) 15. ♕e3 15... ♕a5 ( 15... ♗a5 16. ♕h6+− f6 17. exf6exf6 18. ♗c4+ ♖f 7 19. ♘g5 fxg5 20. ♗b2+− ) 16. ♕h6 f617. exf6 ♗xf6 18. ♘g5 ♗xg5 19. ♗c 4+ e6 20. ♕xg5=/∞♕c3 21. ♕h4 ♗d7 22. ♕e4 ♖ae8 ( 22... b5 23. cxb6 axb6 24. ♗b3 ♖ac8 25. ♗e3=/∞ ) 23. ♕e2 ♖f5 24. ♗b2 ♕b4 25.♖c1=/∞ b2) 12... ♗d7 13. ♕b4⁉ Perhaps even stronger than ( 13. ♕a3 , which was played by Morozevich: 13... ♗xa1 14. e5 , and after the following forced sequence: 14... ♗c6 15. ♗f4 ♗xf3 16. ♗xf3♕d4 17. g3! ♗c3 18. ♗xb7 ♗b4 19. ♕f3 ♖ad8 20. c6 ♗a5 21.♕a3 ♗b6 22. ♕xe7 ♖fe8 (Morozevich-Aronian, Beijing SportAccord Rapid 2012) Black should survive after both 22.Qg5 Qb2, as in the game, or 23. ♕a3 ♖xe5!= , and the bishop is badly misplaced on b7 ) 13... ♗xa1 14. ♗h6 ♗h8 ( 14... ♗g7? 15.♗xg7 ♔xg7 16. ♖d1 ♕c7 17. ♕d4+± ) 15. ♗xf8 ♕xf8 16.♕xb7 ♕e8 17. e5 ( 17. ♗c4 ♗c6 18. ♕c7 e6 19. ♖e1 ♕d7=/∞is fine for Black ) 17... ♗c6 18. ♕b3 ♖c8 19. ♕e3+/= , and Black will have to work very hard for the draw, e.g. 19... ♗a8 20. ♖c1♕c6 21. a3⁉ ( 21. ♘e1 ♕d5 22. f4 ♕xa2 23. ♗f3 ♗xf3 24.♘xf 3+/= ) 21... e6 22. ♘e1 ♕c7 ( 22... ♕e4 23. ♕xe4 ♗xe424. ♗f3± ; 22... ♕d5 23. f4 ) 23. f4+/= 11. ♕a3 cxd4 12. cxd4 ♗g4 13. ♗g5 If 13. ♗e3 Black can just take on f3 and d4. 13... ♗xf3 14. ♗xf3 ♗xd415. ♖ad1 e5= 13... ♗xf3 13... ♕d6 14. ♗xe7 ♕xa3 15. ♗xa3 ♗xf3 16. ♗xf3 ♗xd4 17. ♖ac1♖fc8 18. ♖fd1 ♗e5 19. ♗g4+/= 14. ♗xf3 ♕xd4 15. ♕xe7 White has to take with the queen, after which Black simply defends the b-pawn. 15. ♖ad1 ♕e5 16. ♗xe7 ♖fe8 17. ♗h4 ♕c3 18. ♕xc3 ♗xc3 19.♖d3 ♗g7 20. ♖c1 ♖ac8 21. ♖xc8 ♖xc8 22. ♗d1 ♘c4= A. Smirnov-Vorobiov, St Petersburg 2003 15. ♗xe7 ♘c4 is very comfortable for Black. He has a lot of counterplay, in many lines connected to Nd2. 16. ♕b4 a5 17. ♕b5 ♖fc8 18. ♖ad1♘d2 19. ♗e2 ♕xe4= 15... ♖ab8 16. ♖ad1 16. ♖ab1 ♖fe8 17. ♕a3 ♘c4 18. ♕a4 ♕e5 19. ♖b5 ♕e6= 16... ♕a4 I looked at this position quite carefully and I believe it should be very holdable for Black. Strategically speaking, if White could somehow transfer his bishop from f3 to b3 in one move then Black's position would be very difficult and maybe even lost, but with the bishop doing little on f3 and the other pieces awkwardly placed, I don't think it's even particularly problematic for Black. So I believe this is a more prudent and practical solution than the whole mess starting with 10...cxd4 11.cxd4 Nc5. 17. ♗f4 ♖fe8 18. ♕c7 ♘c4 19. ♕d7 ♕xd7 20. ♖xd7 ♗e5 21. ♗h6♗g7 22. ♗g5 a6 23. ♖c1 b5=

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Nf3 Bg7 5.e3 0-0 6.Bd3 and other sixth moves 6.cxd5 5.Qa4+ Bd7 6.Qb3 dxc4 7.Qxc4 0-0 8.e4 8.Bf4

5.e3 6.Bd3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. e3 O-O 6. ♗d3⁈ This move isn't very strong or critical. 6. ♗d2 here, as in most cases when Bd2 allows c5, isn't a problem for Black. 6... c5 7. dxc5 ♘a6 A very standard reaction. 8. cxd5 ♘xc5 9.♗c4 a6 a) 10. b4⁈ allows 10... ♘ce4 and after, for instance, 11. ♖c1 b5 12.♗b3 ♗b7 13. O-O ♕d6 it's Black who can perhaps entertain some thoughts of an advantage because the pawn on d5 is doomed. Black has also managed to fix a pawn on b4, which might become weak in some endgames. 14. ♕c2 ( 14. ♘xe4 ♘xe4 15. ♕c2 ♗xd5 16. ♖fd1♗xb3 17. ♕xb3 ♖ac8 KovacevicRuck, Neum 2003 ) 14... ♘xd2 15.♕xd2 ♕xb4 16. e4 ♖ac8 17. e5 ♘g4 18. ♕e1 ♕f4 19. ♘e2 ♕e420. ♘g3 ♕xe1 21. ♖fxe1 ♖xc1 22. ♖xc1 ♖d8=/+ Gurevich-Rohde, San Francisco 1987 b) 10. a4 Absolutely necessary. 10... b6⁉ is an alternative I'd like to offer you to 10...Bf5. ( 10... ♗f5 11. O-O ♖c8 12. ♕e2 ♘fe4 is covered very well by Avrukh, and indeed Black has no problems.) 11.b4 ( 11. O-O ♗b7 12. b4 ♘ce4 13. ♖c1 would transpose; 11. ♕e2is the one move with separate value, but it's just not very good, as after 11... ♗b7 12. e4 b5!=/+ White's construction begins to collapse and the a-file will be a source of worry. For instance, 13. axb5 axb514. ♖xa8 ♕xa8 15. ♗xb5 ♕a1+ is something White has no comfortable way of dealing with. The centre will start to fall apart.)11... ♘ce4 12. ♖c1 ♗b7 ( 12... ♘xd2 13. ♕xd2 ♕d6 14. b5 ♕b415. ♗d3 axb5 16. ♗xb5 ♗b7 17 . O-O ♖fd8= ) 13. O-O ♖c8 ( 13...♕d6 14. ♕b3 ♖ac8 15. ♖fd1 ♖c7 16. ♗e1 ♖fc8 17. ♘e2+/= ) b1) 14. ♕e2 If White tries this Black's simplest way to proceed is14... ♘xd2 15. ♘xd2 b5 ( 15... ♕d6 16. b5 a5 17. e4 ♘d7=/∞18. f4 ♗d4+ 19. ♔h1 ♕b4 20. ♖f3 ♘c5∞ ) 16. axb5 axb5 17.♗xb5 ♘xd5= The pawn on b4 will eventually fall. Black obviously has great compensation in the form of the two bishops and isn't worse here. b2) 14. ♘xe4 14... ♘xe4 15. ♕e2 ♗b2⁉ An important move, forcing the rook away from the cfile. ( 15... b5 16. axb5 axb5 17.♗xb5 ♕xd5 18. ♗c4 ♕f5 19. ♗d3 ♕d7=/∞ ) 16. ♖cd1 ( 16.♖c2 ru ns into 16... b5 ) 16... ♗c3 Black just wins back the material, e.g. 17. ♗xa6 ♕xd5 18. ♗xb7 Forced. ( 18. ♗xc3♘xc3 is a huge fork after which White is even worse.) 18... ♕xb719. ♗e1 The only way to preserve the pawn on b4, but after 19...♗xe1 20. ♕xe1 ♘c3 21. ♖a1 ♕e4= the b4-pawn is again under attack and a draw is just around the corner. 6... c5 7. O-O 7. dxc5 dxc4 8. ♗xc4 ♕xd1+ 9. ♔xd1 ♘bd7 is obviously good for Black.

7... cxd4 8. exd4 ♘c6 9. h3 Basically the only thing you need to know about this position is that it normally appears from the Schlecter Slav with White to move, and even in that case it's considered quite playable for Black. With Black to move there are obviously no problems, and perhaps the most natural solution is to play 9... dxc4 10. ♗xc4 ♘a5 11. ♗d3 ♗e6 when Black has very rich piece play and isn't suffering in the slightest.

5.e3 6.cxd5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. e3 O-O 6. cxd5♘xd5 7. ♗c4 This line does have to be taken seriously, and has gained popularity in recent years. 7... ♘xc3 This is the more Grünfeld-like approach that I want to recommend here. 7... ♘b6 is Black's normal response, and also what Avrukh recommends. There's nothing wrong with 8. ♗b3 c5 9. O-O cxd4 10. exd4 ♘c6 and Black should probably equalise if he's precise. 8. bxc3 c5 9. O-O I'd like to discuss this position in more general terms. What White would ideally like to accomplish is to force Black to take on d4 as early as possible. The bishop from c1 can then go to a3 and put pressure on the e7-pawn as well as simply having a very nice diagonal to operate on. One rook will go to c1 while the other comes to b1, putting a lot of pressure on the queenside before Black has any counterplay in the centre. In order to avoid that Black generally tries not to take on d4, so I want to suggest two possible approaches. 9... ♘c6⁉ This has always been known about as a possible idea but is considered slightly weak because of 10.Ba3. 9... ♕c7 is the classical approach. 10. ♕e2 a) 10... b6 is the move that pretty much everyone plays in this position, but I think it's slightly less accurate as it allows 11. e4⁉ Once again the only plan that makes any sense for White - and one that's quite ambitious - is this move followed by developing the bishop to either e3 or g5 (ideally g5, but normally it'll have to go to e3 first because the centre will come under attack). ( 11. ♗a3 ♗b7 12.♖ab1 ♘d7 13. ♖fc1 e5 14. ♗b5 ♗e4 15. ♖b3 ♗d5 16. ♗c4♗xc4 17. ♕xc4 ♖ac8 Sjugirov-Kamsky, Moscow Aeroflot Rapid 2013) a1) 11... ♘c6 12. ♗e3 ( 12. ♖d1 ) a11) 12... ♗g4 13. ♖ac1 ♘a5 ( 13... ♖fd8 ) 14. ♗b5⁉ ( 14.♗d3 ♖ad8 15. h3 ♗xf3 16. ♕xf3 e5 17. d5 ♘b7 Asefi-Sasikiran, Esfahan 2005 ) a111) 14... ♖ad8 15. ♖fd1→ a112) 14... f5 15. h3 fxe4 16. hxg4 exf3 17. gxf3 cxd4 (17... c4 18. f4! ) 18. cxd4 ♕d6 19. ♔g2+/= a113) 14... ♖fd8 15. ♖fd1 ♖ac8 16. h3 ♗d7 17. ♗a6+/= a12) 12... ♗b7 13. ♖ac1 e6 14. h4 a2) 11... cxd4 12. cxd4 ♗g4 13. ♗e3 ♗xf3 14. gxf3 ♘c6 15.♖fd1+/= a3) 11... ♗g4 12. ♗e3 ♗xf3 13. gxf3 ♘c6 14. ♖fd1 ♖fd8 (14... ♘a5 15. ♗b5 e6 16. dxc5 bxc5 17. ♖d7 ♕e5 18. ♖ad1→ )15. ♖ac1

a4) 11... ♗b7 12. ♖d1 ♘c6 13. ♗g5 b) ♘c6 I think starting with this move is more precise. b1) 11. ♗a3 b6 ( 11... ♘a5 12. ♗d3 b6 13. ♖ac1 ♖d8 14. e4♗g4 15. h3 ♗xf3 16. ♕x f3 ♕d7 17. d5 1/2 I. Sokolov-Lalic, Sibenik 2012 ) 12. ♖ac1 ♗b7 13. ♖fd1 ♖ac8 14. ♗a6 ♘a5 15.dxc5 bxc5 16. ♖b1 ♖fd8 17. ♖xd8+ ♖xd8 18. ♗xb7 ♘xb7 19.♕a6 ♕b6! Prohaszka-Kurnosov, Zurich Masters 2011 b2) 11. ♖d1! is, I think, the best move for White. 11... b6 12. e4 b21) 12... cxd4 immediately is not so good. 13. cxd4 ♗g4 14.e5 ( 14. ♗e3 e5 transposes ) 14... e6 ( 14... ♗xf3 15. gxf3♘xd4 16. ♖xd4 ♗xe5 17. ♗b2 ♗xd4 18. ♗xd4 ♖fd8 19.♗e5 ♕c5 20. ♖e1∞ ) 15. ♗e3 ♗xf3 16. gxf3 ♘e7 17. ♖ac1This is a bit awkward for Black since he has no comfortable squares for the queen. b22) ♗g4 is the better approach. White now has a choice between 13.e5 or 13.Be3. b221) 13. e5 ♗xf3 14. gxf3 Taking with the pawn is forced. ( 14. ♕xf3 is impossible because of 14... cxd4 15. cxd4♘xe5! , winning material.) 14... e6 15. ♗e3 ( 15. ♗g5 h616. ♗h4 cxd4 17. cxd4 ♘xd 4 18. ♖xd4 ♕xe5 ) 15... ♘e7This position should explain why we didn't take on d4 earlier - without the pawns on c5 and c3 White could now play Rc1, hitting the black queen. Black would have to start looking for a good square for the queen, but there aren't really any which are ideal. With the pawns that problem doesn't exist and I think this is quite comfortable for Black. b222) 13. ♗e3 If you allow White to play Rac1 and finish development it might become quite unpleasant, so you have to be very energetic. 13... cxd4 ( 13... ♖ad8 14. ♖ac1 ) 14.cxd4 e5 Forcing White's next move. 15. d5 ♘d4 16. ♗xd4exd4 This is generally very risky strategically for Black, because if White gets the pawn centre rolling he'll have a huge advantage. However, Black bases his play on the fact that after 17. ♖ac1 he can reply 17... ♕f4 , and after the probable 18. h3 ♗xf3 19. ♕xf3 ♕xf3 20. gxf3 and, for instance, 20... ♖ac8 , the black bishop will come to e5 (at the moment f4 is impossible because of ...Bh6). Once the bishop is on e5 White's whole centre is blockaded and the game should be equal. 10. ♗a3 b6 Almost forced, as if Black takes on d4 then 9...Nc6 wouldn't have been a very good move. In this position Black would be fine except for the fact that White can actually take on c5. 10... cxd4 11. cxd4 b6 ( 11... ♗f5 12. ♖c1 ♖c8 13. ♗b5 ♗e4 14.♕e2 a6 15. ♗d3 ♕d5 16. ♗xe4 ♕xe4 17. ♗c5+/= Landa-Hillarp Persson, Copenhagen 2010 ) 12. ♖c1 ♗b7 13. ♕e2 ♖c8 14. ♖fd1 ♖c715. ♘e5+/= Gajewski-Smith, Stockholm 2012

11. dxc5 This position requires some finesse, as this is an actual pawn sacrifice and Black needs to be quite precise. However, after 11... ♕c7 I think I've managed to make this work for Black. 12. ♘d4 is by far the best move. 12. ♕e2 can lead White into a very nice trap, as after 12... bxc5 13.♗d5 e6 14. ♗e4 (Pekarek-Hawelko, Zakopane 1980) Black has 14...♕a5! 15. ♗xc6 ♖b8=/+ 12... ♗b7 12... ♗d7 was played in a game between two grandmasters, but is somewhat weaker. 13. ♗d5 ♖ad8 14. cxb6 axb6 Kunte-Gupta, Aurangabad 2011 15. ♕e2+/= 13. ♗d5 The most critical reply. 13. ♗e2 ♘a5 14. cxb6 axb6 15. ♕c2 ♖fc8 16. ♖fd1 ♗f6=/∞ 13. ♘b5 ♕c8 14. ♗e2 ♖d8=/∞ 13... bxc5 Black needs to start with this move. 13... ♖fd8 14. ♗xf7+ ♔xf7 15. ♕b3++− 13... ♖ad8 was my initial idea for Black and I was happy with it until I realised that once the rook leaves the a8-square 14. cxb6! is actually quite strong. ( 14. ♗f3 bxc5 15. ♗xc5 tranposes) 14... axb6 15. ♗f3 I wasn't able to equalise in this position, so I went back to the drawing board. 15... ♖a8 16. ♘b5 ♕b8 17. ♕e2 ♖c8 18. ♖fd1 ♘e5 19.♗xb7 ♕xb7 20. ♗b4+/= 14. ♗xc5 ♖ad8 Only now. 15. ♗f3 Clearly the best move here. 15. ♗xc6 ♗xc6 is obviously a very nice position for Black, as he gets a lot of play on the queenside and the bishops are also very strong. 15... ♘a5 16. ♗b4 16. ♗xb7 ♕xc5 is an immediate draw because the pawn on c3 falls. 16... ♘c4 17. ♗xb7 ♕xb7 Black gets a very typical and I think perfectly fine position where he has more than enough compensation for the pawn. The pawn on c3 is very safely blockaded by the fantastic knight on c4 and the bishop on b4 is extremely misplaced and will get hit by a5 in almost all variations. Of course a pawn is still a pawn and I had to check this position quite carefully to make sure I'm not just recommending you an endgame a pawn down with no real play, but I think I can safely say that Black is ok. After the most likely move, 18. ♖b1 both the immediate 18... ♕c7 or 18... ♖c8=/∞ are fine for Black, as far as I could ascertain.

19. ♕a4 19. ♕e2 ♖c8 20. ♘b5 ♕d7 21. a4 a5 22. ♖fd1 ♕e6 23. ♘a7 ♖c724. ♘b5 ♖cc8= 19... ♗f6 , and Black has excellent play for the sacrificed pawn.

5.Qa4+ 8.e4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕a4+ This is an interesting variation which has been tried on a number of occasions recently. One of the main proponents of the line is Ivan Sokolov, who played it against me more than once. 5... ♗d7 6. ♕b3 dxc4 7. ♕xc4 7. ♕xb7 is, as usual, not very good after 7... ♘c6 7... O-O 8. e4 is worse than in the normal Qb3 lines, as Black has the very important additional option of 8. ♗g5 ♗e6 9. ♕b5 ♗d7 10. ♕b3 ( 10. ♕xb7 ♘c6 ) 10... c5 11.♖d1 ♕a5 8... b5! and he's just fine. 9. ♘xb5 The main move for White. 9. ♕b3 c5 10. dxc5 ( 10. ♗xb5 ♗xb5 11. ♘xb5 ♘xe4 12. O-Ocxd4=/+ Rustemov-Shomoev, Moscow Aeroflot op 2003 ) 10... ♘a6! gives Black a lot of counterplay for the pawn and in some lines it might even become very dangerous for White. 11. e5 ( 11. ♗e3 ♘g4 ) 11... ♘g412. ♗xb5 ♘xc5 13. ♕c4 ♗xb5 14. ♕xc5 ♗a6 15. h3 ♖c8 16. ♕d5 (16. ♕xa7 ♕a5 17. hxg4 ♖xc3 18. ♗d2 ♕b5 19. ♕xa6 ♕xa6 20.♗xc3 f6∓ ) 16... ♘xe5! 17. ♘xe5 ♕b6 18. f4 ♗f6‼ 19. ♕f3 ♖fd8∓

9... ♘xe4 10. ♕xc7

10. ♘xc7 ♘c6 11. ♘xa8 really can't be recommended for White, as11... ♕a5+ is extremely strong. 12. ♗d2 ♘xd2 13. ♘xd2 ♘xd4∓

10... ♘c6 11. ♗d3 ♘b4 is the old theory in this position and is considered to be - and probably is - absolutely fine. 11... ♘f6⁉ is my suggested alternative, leading to forced lines where the pawn on d4 seems to be doomed. In contrast to the old line Black tends to get very lively endgames with a lot of pieces on the board. 12. O-O ( 12.♕xd8 ♖fxd8 13. O-O ♘b4 14. ♗e2 ♘c2 15. ♖b1 ♗f5 ) a) 12... a6 13. ♕xd8 ♖fxd8 14. ♘c3 ♗g4 15. ♗e2 ♖ab8 16.♗xa6 ♗xf3 17. gxf3 ♘xd4∞ b) 12... ♕xc7 13. ♘xc7 ♖ac8 14. ♘b5 ( 14. ♘a6 ♗g4 ) 14... ♘b415. ♗e2 ♘c2 16. ♖b1 ♘d5 1 7. ♗d2 ♗xb5 18. ♗xb5 ♘xd4 19.♘xd4 ♗xd4= c) 12... ♘xd4 13. ♘fxd4 ♗xb5 14. ♕xd8 ♖fxd8 15. ♗xb5 ♖xd4=

12. ♗xe4 ♗xb5 13. ♕xd8 ♖axd8 14. ♗d2 ♘d3+ and Black gets very good compensation for the pawn. In many cases he'll win the pawn back immediately. This should be a draw, though in most cases White gets a slight plus with which he could at least hope to torture Black for a little bit. 14... f5 ?! 15. ♗xf5 ! 15... ♖xf5 16. ♗xb4+/= Giorgadze-Khalifman, BL 1997 14... ♗d3 ? 15. ♗xb4 ♗xe4 16. ♗xe7 ♖fe8 17. ♗xd8 ♗xf3+ 18.♔d2 ♗h6+ 19. ♔c3 ♗xg2 20. ♖hg1± 15. ♗xd3 ♗xd3 16. ♗c3 ♗e4= 17. ♔e2!N 17. ♖d1 ♗xf3 18. gxf3 ♖d5 19. ♔e2 ♖fd8 20. ♖d2 ♗xd4 ( 20... e621. ♖hd1 ♖h5= GiorgadzeStohl,Senec 1998. ) 21. ♖hd1 e5 22. f4 = 17... ♗xf3+ 18. ♔xf3 ♗xd4 19. ♗xd4 ♖xd4 20. ♖hd1+/= Atalik-Gustafsson, Kallithea 2003 20... ♖b4 21. b3 a5 22. ♔e3 a4 23. ♖d3 ♖fb8

5.Qa4+ 8.Bf4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕a4+ ♗d7 6. ♕b3dxc4 7. ♕xc4 O-O 8. ♗f4 is the most critical move, and gives Black a lot of options. Both ...Be6 and ...Bf5, just ignoring the fact that the pawn on c7 is hanging, are very possible. 8... c6 is, I think, the way to play here, and the move I've normally made in this position. 8... ♗c6 is also quite playable, aiming for some kind of quiet positional play after, for instance, 9. ♖d1 ♗xf3 10. gxf3 c6 Black will follow up with ...e6 and get a very stable position with decent chances, but I wanted to recommend something else. 9. e4 ♕a5⁉ This is a move recently played by Morozevich, but one I'd found earlier myself with my team when preparing for the London Candidates Tournament. This is an interesting and attractive option for Black because in many cases White actually has to be somewhat precise not to end up in strategically risky positions. 9... b5 used to be considered almost mandatory here. 10. ♕b3 ♕a5 11.♗d2 was what people used to play. ( 11. e5 Then they started to realise that 11.Bd2 wasn't forced and this move is instead possible, with 11...♗e6 followed by 12. ♕c2 This position is much harder for Black to equalise. I wouldn't say it's impossible, but I think it's potentially quite risky..) 10. b4 This is what Etienne Bacrot played against Alexander Morozevich in the Thessaloniki Grand Prix. 10. ♗d3 also transposes to what we wanted to get in the old lines: 10...b5 a) 11. ♕c5 ♘a6 12. ♕a3 ( 12. ♕xe7 b4 ) 12... ♕xa3 13. bxa3 c5 b) 11. ♕b3 11... c5 and Black is doing very well here. No particular details are needed as it's quite clear that after, for instance, 12. dxc5♘a6 something has gone a bit wrong for White because all of his pieces are suddenly under attack. 10. e5 is another option, but Black benefits from the fact that the pawn is still on b7 - in general, there's no reason to play ...b5 in these structures unless you get some immediate benefits from it.

a) 10... b5 11. ♕b3 transposes to lines we are trying to avoid. b) 10... ♗e6 11. exf6 ♗xc4 12. fxg7 ♖d8 13. ♗xc4 b1) 13... ♕b4 14. ♗b3 c5 15. ♗d2 ( 15. OO c4 16. ♗a4 ♕xb217. ♖fc1 ♘c6 18. ♗xc6 bxc6 19. ♘a4 ♕b4 20. ♘c5 c3 ) 15...c4 ( 15... cxd4 16. ♘d5 ♕d6 17. OO ♘c6 18. ♗f4+/= ♕c5 19.♘c7 ♘a5 20. ♘xa8 ♘xb3 21. axb3 ♖xa8 22 . ♖fd1+/= ) 16.♗a4 ( 16. ♘d5 ♕b5 17. ♘c3 ♕a6 18. ♗a4 ♘c6= ) 16... ♘d7 17. O-O ( 17. b3 ) 17... ♘b6 18. a3 ♕d6 19. ♗c2 ♔xg7 20.♖ae1+/= b2) c5 14. OO ( 14. d5 ♕b4−+ ; 14. dxc5 ♕b4∓ ) 14... cxd415. ♘e5! dxc3 ( 1 5... e6 16. b4! ♕xb4 17. ♘e4 ♘c6 18.♗h6+− ) 16. ♘xf7 ♔xg7 17. ♗h6+ ♔f 6 18. ♘xd8 ♕xd8 19.♖ae1+/= e5 20. ♖e3 c) 10... ♘d5 11. ♗d2 ♘a6⁉ Perhaps the simplest option in this complex position. ( 11... ♗e6⁉ ) c1) 12. ♘xd5 ♕xd5 13. ♕xd5 cxd5 14. ♗d3 ♖ac8 ( 14... ♘b815. ♖c1 ♘c6 ) 15. OO ♘c7= ( 15... ♘b8⁉ ) 16. ♗b4 ( 16.♖ac1 ♘e6 17. g3 f6= ) 16... ♖fe8 17. ♖ac1 ♗h6 18. ♖c3♗g4 c2) 12. ♕b3 12... ♕b6 13. ♘xd5 cxd5 14. ♕xb6 axb6 In this endgame the knight will now return via b8 to c6 and Black can either go for the standard play on the queenside with ...Na5, ...b5, ...Nc4, or he can also just play ...f6, trade off the e-pawn and get a perfectly safe, equal position. 15. a3 ( 15. ♗d3 ♗g4 ; 15. h3 f6 )15... ♘b8 16. ♗d3 ♘c6 17. h3 f6 ( 17... ♖fc8= ) 10. ♗d2 here is met by 10... b5 a) 11. ♕e2 b4 12. ♘d1 c5 13. e5 ♘d5 14. ♕e4 e6 15. dxc5♘a6→ b) 11. ♕d3 b4 12. ♘d1 ( 12. ♘e2 c5 13. a3 ♗b5 14. ♕b1 cxd415. ♘fxd4 ♘c6=/+ ) 12... c5 13. d5 e6 Bla ck is maybe already better. 14. ♘e3 exd5 15. exd5 ♖e8 16. ♗e2 ♘h5! is a somewhat odd-looking but perhaps even stronger move, creating the huge threat of Nf4, which isn't that easy to counter. ...Bxb2 is also an immediate threat and Black may already be better as it's very difficult for White to meet both of those threats adequately. ( 16... ♘g4 This move of Sutovsky's wasn't a mistake, as such, as Black was fine even then. Rasmussen-Sutovsky, Porto Carras 2011 ) 17. ♕c4 ( 17. ♘c4 ♗b5∓ )17... a6=/+ c) 11. ♕c5 ♘a6 12. ♕e5 ♕b6 ( 12... ♕b4 13. ♖b1 ) 13. ♕f4 ( 13.♗e2 c5 ) 13... ♖ad8 14. ♗e2 ( 14. a4 e5! 15. dxe5 ♘g4=/+ ) 14...c5=/+ d) 11. ♕b3 and we've transposed to the position we wanted to get where ...b4 is good. 11... b4 Black used to get a very decent position. For instance, an important game in this position was played between myself and Ivan Sokolov in one of the Bundesligas. 12. ♘a4 ♘xe4 13.♗xb4 ♕c7 14. ♗d3 ♗e6 d1) 15. ♕d1 ♗d5 16. O-O c5! A very nice aesthetic move to make, putting the pawn in harm's way immediately. Regardless of how White recaptures on c5 the knight on b8 comes into play with great effect via the c6-square. ( 16... ♘d6 could be met by 17.♖e1 , preventing the black knight from coming to d7 straightaway.) 17. ♘xc5 ♘xc5 18. ♗xc5 ♘c6 I was even slightly better in the game here, although it wasn't enough to win. This was thought to be extremely comfortable for Black and the whole line wasn't considered particularly dangerous. 19. ♖e1 ♗xf3 20. ♕xf3 ♘xd421. ♗xd4 ♗xd4 I. Sokolov-Svidler, BL 2004

d2) 15. ♕c2 15... ♘d6 16. O-O ♗d5 The bishop comes to d5 with tempo as it hits the knight on f3. 17. ♘e5 ♘d7 Black gets very decent counterplay. 18. f4 ♖ac8∞ 10... ♕d8 In this position it's important to note that the bishop on d7 has an important role to play. Something needs to be done immediately to counter the threat of ...a5. 11. a4 is a possible approach, and was the one adopted by Etienne. 11. ♗e2 is a normal move, but after 11... a5 12. b5 ( 12. bxa5 ♕xa513. OO b5 14. ♕d3 b4 15. ♘b1 c5 16. ♘bd2 cxd4 17. ♘xd4 ♘c6 18. ♘2b3 ♕b6 19. ♘xc6 ♗xc6 20. ♗e3 ♕b8 21. f3 ♘h5∓ Gonzalez-Swiercz, Istanbul 2012 ) 12... cxb5 White can't take with the knight because the e4-pawn is hanging and he can't take with the queen because the bishop has been driven to d7. 11. e5 is another option. a) 11... ♗e6 12. exf6 ( 12. ♕d3 ♘d5 13. ♘xd5 ♗xd5 14. ♗e2a5=/+ ) 12... ♗xc4 13. fxg7 ♔xg7 14. ♗xc4 ♘a6 15. a3 ♘c7 16.O-O+/= b) 11... b5 This is the best reaction, I think. 12. ♕b3 ( 12. ♕e2 ♘d513. ♘xd5 cxd5→ White is somewhat over-exposed on the queenside and his pawns are likely to come under sustained pressure. Black is at least not worse. ) 12... ♗e6 This is a rare position in which I'm quite happy for Black to give up all the minor pieces for the queen. b1) 13. ♕b2 ♘d5 14. ♗d2 ♘d7 ( 14... a5 15. bxa5 ♖xa5 16.♘xd5 ♗xd5 17. ♗xa5 ♕xa5+ 18. ♕d2 ♕xd2+ 19. ♘xd2♘a6=/∞ ) 15. ♘g5 ♘c7∞ b2) 13. exf6 13... ♗xb3 14. fxg7 ♔xg7 15. axb3 This never appeals to me, in general, as I'm a firm believer in three minor pieces being stronger than a queen, but in this particular position there's the very strong manoeuvre 15... ♕c8! After, for instance,16. ♗e2 ( 16. ♗d2 ♘a6 17. ♘e4 ♕e6 18. ♘g5 ♕xb3 19. ♖xa6h6!∓ ) 16... ♘a6 it turns out to be very awkward for White to protect the b4-pawn, and it pretty much has to be abandoned. 17.O-O ( 17. ♘a2 ♕e6 is very, very awkward for White to face. b3 is hanging and the bishop on e2 is now pinned, so it's difficult for White to castle..) 17... ♘xb4=/+ 11. ♖d1 a5 12. a3 ♗e6 13. ♕d3 axb4 14. axb4 ♘a6 11... ♗g4! is my suggested improvement, and was what I'd prepared before the Candidates Tournament in London. I was therefore watching the game between Bacrot and Morozevich with great interest as I wanted to see if Alexander came to the same conclusions. Seemingly not, but here's my "version of events" from this moment on. 11... a5 12. b5 cxb5 13. axb5+/= 11... b5 12. axb5 cxb5 13. ♘xb5± 11... ♗e6 was Morozevich's move, but I don't think it was the best in this instance. 12. ♕d3 a) 12... a5 13. b5 c5 ( 13... cxb5 14. axb5 ; 13... ♘bd7 14. ♗e2♘h5 15. ♗e3 c5 16. d5 ♗g4 After a lot of events in which the evaluation changed on a number of occasions this game was eventually drawn. 17. ♖c1 ♗xf3 18. ♗xf3 ♘e5 19. ♕d1 ♘xf3+ 20. ♕xf3 b621. g4 ♘f6 22. OO ♕d7 23. h3 h5 Bacrot-Morozevich, Thessaloniki GP 2013 ) 14. d5

b) 12... ♗g4 13. ♗e2 ( 13. ♘e5 a5 14. b5 c5 15. h3 cxd4 16.♘xg4 dxc3 17. ♘xf6+ ♗xf6 18. e5 ♗g7 19. ♕xc3 ♘d7 20. ♖d1♖c8 ) 13... ♘fd7 14. e5! ♘a6 15. b5 ♘b4 16. ♕b1 c5 17. O-O c) ♘a6 13. ♖b1 ♘h5 14. ♗e3 f5 15. ♘g5 fxe4 16. ♕d2 ♗d5 17.♘xd5 ♕xd5 18. ♕c2+/= I. SokolovHowell, London 2009 12. ♖d1 is perhaps the most critical reply, and here some precision is required from Black. I want to offer you a very concrete solution. 12. ♘e5 is the normal reaction for White to ...Bg4, but while the queen is still on c4 it's just not very feasible because White's pieces get attacked on every move and Black is very comfortable: 12... ♗e6 13. ♕d3 ♘fd7 12. ♗e2 is another standard reaction, but once again the fact that the queen is on c4 gives Black additional tempi. He plays 12... ♘fd7! ,threatening to take on f3 and play ...Nb6, winning the d4-pawn. a) 13. ♖d1 is met by a very strong and simple solution: 13... ♗xf3 14.♗xf3 e5 15. dxe5 ( 15. ♗e3 exd4 16. ♗xd4 ♘b6! 17. ♕d3♘a6=/+ ) 15.. . ♕e7 Once again the pawn on e5 will be won back on the next move and the fact that the pawns aren't on a2 and b2, where they'd be far away from Black's pieces and would support the knight on c3, means that Black will have constant targets to aim his play against and will be very comfortable. 16. ♕b3 ( 16. OO ♘xe5 17. ♕b3♘xf3+ 18. gxf3 ♘d7 ; 16. e6 ♕xe6! ) 16... ♘xe5 17. ♗e2 ♘a618. ♘a2 ( 18. ♗xa6 bxa6 19. OO ♖ab8 20. ♘a2 c5=/+ ) 18... c519. ♗xa6 ( 19. b5 c4! 20. ♕c2 ♘c5=/+ ) 19... bxa6 20. bxc5♖ab8 21. ♕g3 ♕xc5 22. OO ♖bd8= 23. ♘c3 ♖xd1 24. ♖xd1♖e8= b) 13. e5 ♘b6 14. ♕d3 ♘a6 c) 13. O-O ♗xf3 14. ♗xf3 ♘b6 15. ♕b3 ♕xd4 d) 13. ♗e3 is very adequately met by 13... a5 d1) 14. OO ♗xf3 15. ♗xf3 ♘b6 16. ♕b3 axb4 17. ♕xb4 ( 17.♘e2 ) 17... ♘a6 18. ♕b3 ♗xd4 d11) 19. ♖ad1 e5 20. a5 ( 20. ♘e2 ♘c5 21. ♕c2 ♘e6∓ )20... ♘d7 21. ♕xb7 ♘ac5 22. ♕xc6 ♖c8 23. ♕b5 ♗xc3 24.♗xc5 ♖xc5 25. ♕xd7 ♕xa5 d12) 19. ♗xd4 19... ♕xd4 20. ♘e2 ♕b4 21. ♖fb1 ♕xb3 22.♖xb3 ♘d7 23. ♖xb7 ♘ac5 24. ♖c7 ♖fd8 25. ♖xc6 ♖xa4 26.♖xa4 ♘xa4= d2) 14. b5 14... ♘b6 d21) 15. ♕b3 ♗xf3 16. ♗xf3 ♗xd4 17. ♖d1 e5 ( 17... c5 18.e5∞ ) 18. bxc6 bxc6!=/+ d22) 15. ♕d3 15... c5 White's centre is attacked and it's quite difficult for him to keep everything under control. 16. e5 ( 16.♖d1 is most easily met by trading off everything: 16... ♗xf317. ♗xf3 cxd4 18. ♗xd4 ♕xd4 19. ♕xd4 ♗xd4 20. ♖xd4♘8d7 =/+ The second knight is heading to c5 and the pawn on a4 is potentially quite a big problem for White. He has structural issues which will somehow have to be resolved.) 16... ♘8d7 17.dxc5 ( 17. ♕e4 f5! ) 17... ♗xf3 18. gxf3 ♘xe5 19. ♕xd8 (19. ♕c2 ♘d5 20. OOO ♘xe3 21. fxe3 ♕e8 22. ♘d5 ♖c823. ♘b6 ♖c7 '=' 24. ♘d5 ♖c8 25 . ♘b6 ♖c7= ) 19... ♖fxd820. cxb6 ( 20. OO ♘ec4 21. cxb6 ♘xe3 22. fxe3 ♗xc3

23.♖a2 ♖ac8=/+ ) 20... ♘d3+ 21. ♗xd3 ♖xd3= 22. ♔e2 ♖xc323. ♖ac1 ♖xc1 ( 23... ♖ac8 24. ♖xc3 ♖xc3 25. ♖d1± ) 24.♖xc1 ♗e5 25. f4 ♗d6= 12... ♘a6 This gives White two obvious choices. 12... a5 13. b5 cxb5 a) 14. ♕xb5 ♘c6 15. e5 ( 15. ♕xb7 ♗xf3 16. gxf3 ♘xd4 17. ♘b5♘h5∓ ) 15... ♘h5 16. ♗e3 ♘b4 17. ♖c1 ( 17. ♗e2 ♖c8 18. ♖c1b6∓ ) 17... ♖c8 18. h3 ♗xf3 19. gxf3 ♖c6!=/+ b) 14. axb5 14... ♗xf3 15. gxf3

b1) 15... a4 16. e5 ( 16. ♘xa4 ♘h5 17. ♗e3 ♕a5+ 18. ♘c3♘d7 19. ♗d2 ♖fc8 20. ♕d3 ♕d8=/∞ ; 16. ♗e2 a3 17. OO e618. d5 e5 19. ♗g5 ♘bd7 20. d6 ♘b6 ) 16... ♘h5 17. ♗e3♘d7=/+ b2) 15... ♘bd7 16. ♗c7! ♕c8 17. b6+/= b3) e6 16. ♘a4 ( 16. d5 ) 16... ♘bd7 17. ♕c7 ♕xc7 18. ♗xc7♖fc8 19. b6 ♘e8 20. ♗b5 ♘b8 21. d5→ 13. ♗e2 13. ♕b3 ♗xf3 Only now. 14. gxf3 ♘h5 15. ♗e3 ♘c7 The knight has managed to get to c7 without being captured on a6, which is obviously quite important. It's now aiming for the e6-square where it will attack the pawn on d4 and also eye the very important f4-square. Black has very decent counterplay and can even entertain some hopes of fighting for an advantage. a) 16. b5 ♘e6 ( 16... cxb5 ) 17. d5 ( 17. ♘e2 ♕a5+ ) 17... cxd518. ♘xd5 ♕a5+ 19. ♗d2 ( 19. ♖d2 ♖ad8 20. ♗g2 ♘f6 21. O-O♘xd5 22. exd5 ♘c5 23. ♕a3 ♘d7= ) 19... ♕d8 a1) 20. ♗b4 ♘d4 21. ♖xd4 ( 21. ♕c4 e6=/+ ) 21... ♗xd4 22.♗xe7 ♕c8=/+ a2) 20. ♗e3 20... ♕a5+ 21. ♗d2 ♕d8= b) 16. ♘e2 a5! 17. ♘g3 ( 17. b5 cxb5 18. axb5 a4 19. ♕c4♘e6=/+ ) 17... ♘xg3 18. hxg3 axb4 19. ♕xb4 ♕d6 20. ♕b3 b521. axb5 ( 21. ♗f4 ♕d7 22. axb5 ♘xb5 23. ♗xb5 cxb5= ) 21...♘xb5 22. ♗xb5 ♖a3 23. ♕c4 cxb5 24. ♕xb5 ♖xe3+ 25. fxe3♕xg3+ 26. ♔e2 ♕g2+ 27. ♔d3 ♗xd4‼ 28. exd4 ( 28. ♔xd4 ♖d8 +29. ♔c3 ♖c8+ 30. ♔d4 e5+ 31. ♔d5 ♖d8+ 32. ♔c6 ♕c2+ 33.♔b7 ♕c8+ 34. ♔b6 ♕b8+ 35. ♔c5 ♕a7+ 36. ♔b4 ♖b8∓ ) 28...♕xf3+ 29. ♔d2 ♕f4+= c) 16. f4 16... e6 17. ♗g2 ( 17. h4 f5 ) 17... ♕h4 18. ♘e2 ♗h6∓

13... ♕c8! A very strong idea. This move looks slightly mysterious, but its point soon becomes apparent. 14. O-O 14. ♕b3 suddenly allows Black to play 14... ♘xe4! 15. ♘xe4 ♕f5 16.♕e3 ♘xb4∓ and the combined threats of Nc2 and Nd5 mean that White will lose some material and Black is on top. 14... ♗e6 Black's second idea in this position. 14... ♘xe4 15. ♘xe4 ♕f5 16. ♘c5! ♕xf4 17. ♘xb7+/=

15. d5 cxd5 Now with the queen on c8 White is forced to enter an endgame. 16. exd5 16. ♕xc8 ♖axc8 17. ♘xd5 ( 17. exd5 ♗g4=/+ ) 17... ♗xd5 18. exd5♘xb4 19. d6 exd6 20. ♗xd6 ♖fe8 21. ♗b5 ♘c6=/+ 16... ♕xc4 17. ♗xc4 ♗g4 Once again this is a position where White would feel quite comfortable with pawns on a2 and b2, but with pawns on a4 and b4 he constantly has to pay attention to all the stuff hanging on the queenside. 18. d6 e6 19. ♗xa6 19. ♘a2 ♖fd8 20. ♗b3 ♘b8!=/+ 19... bxa6 Black clearly can't have any problems at all. 20. ♖c1 ♖fd8 21. ♖fd1 ♖ab8 22. b5 22. d7 ♖b7 22... axb5 23. ♘xb5 ♘d5 24. ♗g3 ♖d7=

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.Bd2 Nb6 6.e3 6.Nf3 6.Bg5 6.Bf4

5.Bd2 Nb6 6.e3 We're going to look at the 5.Bd2 Grünfeld, a line which has become increasingly popular in recent years. Pretty much everyone in top-flight chess has tried it at least once with White, myself included. 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. ♗d2 ♘b6 A less fashionable move right now. It used to be played quite a lot when I started off as a Grünfeld player but then attention switched to 5...Bg7. Nevertheless, I think Black is fine after this move and I'll attempt to give you an overview of what Black should do against all of White's four options. 5... ♗g7 is the main line by far. It's been extensively covered in practice and also in the Avrukh book on the Grünfeld, so I thought I'd show you something else. It helps to have a wider appreciation of Black's possibilities in various positions. 6. e3 ♗g7 7. f4 This is something I played against Boris Gelfand in the London Candidates in 2013. I can't really claim any credit as it was shown to me half an hour before the game by my team, and in particular Nikita Vitiugov, whose idea it was. I was much better and almost winning by move 15 against Boris but I quite honestly thought, and I think I even said in the press conference after the game, that I didn't expect to see this ever again because it's quite clear that you can't play like this against the Grünfeld... When I started working on my video for chess24 I thought ok, I need to work out what's going on after 6. Bf4 and 6.Bg5, because surely it'll take me about 20 minutes to refute the stupid 6.e3, 7.f4. I also honestly didn't think it would ever be played in any kind of top-level game, but by the time of the video it had already been played 5 times, including in my game and a game between Wang Hao and Grischuk in the Beijing Grand Prix, so it seems the line is blooming! In fact, there have been new developments in the short time between recording the video and finishing these notes. The most surprising part is that I'm also not sure what to recommend against it. I'll give you an overview of Black's options, but honestly I wasn't able to kill it and that fills me with mixed feelings. Firstly, I'm supposed to give you answers about what to do against everything in the Grünfeld and here I am flat out saying that I don't have the answer to this particular conundrum. On the other hand, it's something that came out of a team I was in charge of and I feel a certain sense of pride. I introduced it into the world and it's now living a life of its own and seems to be doing fine. So I'll give you some pointers, but none of the stuff I'll be showing equalises completely, in my eyes - maybe I'm just biased towards this line! 7. ♘f3 O-O 8. ♖c1 transposes to 6.Nf3 Bg7 7.Rc1. 7... O-O I think Black almost has to castle here. 7... ♘8d7 is a very interesting new attempt shown by Alexander Morozevich. 8. ♘f3 c5 9. d5N f5⁉ The only idea that could justify the previous two moves, otherwise White assumes total control in the center. White has to be very energetic now. a) The stem game went 10. ♕b3 O-O 11. ♖d1

a1) 11... ♔h8 12. a4 c4 13. ♗xc4 ♘c5 14. ♕b4 ♘xc4 15.♕xc4 b6 16. OO ♗a6 17. ♘b5 ♘e4 18 . ♗b4 ♖c8 19. ♕b3♕d7 20. ♘fd4± , and White was better in Tomashevsky-Morozevich, Tromso World Cup rapid 2013 a2) If Black wants to play c5-c4 he should do it right away: 11...c4⁉ 12. ♕a3 ( 12. ♗xc4 ♘c5 13. ♕b5 ♗d7 14. ♕b4 ♘xc4 15.♕xc4 ♖c8 16. OO ♕b6∞ ) 12... ♘f6 13. ♗e2 ♕d6 14. ♕a5♗d7 15. ♘e5 ♖ac8 16. ♗f3 ♖c5 17. ♕a3 ♖c7 18. ♕a5= a3) ♘f6! is my recommendation, and after 12. ♕a3 Black now has a very strong interposition: ( 12. d6+ e6 13. ♗c1 ♘fd5 is very good for Black.) 12... e6! 13. ♕xc5 ( 13. dxe6 ♕e7 is once again excellent for Black.) 13... exd5∞ , and only White has reasons to worry here. b) 10. ♘g5⁉ 10... ♘f6 11. ♗b5+ ( 11. ♕b3 O-O! is wrong.) 11...♔f8 , and the pawn on d5 will be taken but White has very good long-term compensation: 12. O-O b1) 12... a6 13. ♗e2 ♘fxd5 ( 13... ♘bxd5 14. ♕c2→ ) 14. ♕c1h6 15. ♘f3 ♕c7 16. ♘xd5 ♘xd5 17. ♘e5! ♗xe5 18. fxe5± b2) ♘fxd5 b21) 13. ♕c1 is much less precise: 13... c4! 14. b3 ( 14. ♖d1h6 15. ♘xd5 ♘xd5 16. ♘f3 ♗e6 17. ♕xc4 ♕b6=/+ ) 14...h6 15. ♘f3 cxb3 16. axb3 ♗e6 17. e4∞ with unclear play. b22) 13. ♕e1! b221) 13... e6 14. ♖d1 ♕e7 15. e4↑ b222) 13... ♔g8 14. ♖d1 h6 15. ♘f3 a6 16. ♗e2 ♗e6 17.♘xd5 ♕xd5 ( 17... ♘xd5 18. ♘e5 ♔h7 19. ♗f3± ) 18.♗c3 ♕xa2 19. ♗xg7 ♔xg7 20. ♕c3+ ♔h7 21. ♕xc5± b223) 13... c4 14. ♖d1 h6 15. ♘xd5 ♘xd5 16. ♗a5! b617. ♗c6± b224) 13... h6 14. ♘f3 ♗d7 15. ♗e2 ♗c6 16. ♖d1 ♕e817. ♘xd5 ♘xd5 ( 17... ♗xd5 18. ♗c3 ♗xa2 19. ♘e5 ♗d520. ♕g3 ♗xe5 21. ♗xe5↑ ) 18. ♘e5 ♗xe5 19. fxe5 ♔g720. ♗f3 , and White has obvious play against the weakened black king, but after the possible 20... e6 21. e4 ♘e7 the outcome is completely unclear. 7... c5 8. dxc5 ♘6d7 straight away is an option, but after 9. ♘e4 I don't think Black has anything better than ( 9. ♘a4 e5 is good for Black.)9... ♗xb2 10. ♖b1 ♗g7 11. ♘f3 and here Black has to castle so that after 11... O-O 12. ♗c4 we've transposed to the 7...0-0 main line. 7... ♗e6 is possible, but generally speaking I don't like developing the bishop to e6. I'm not particularly impressed by this idea and think White can deal with it quite comfortably. 8. ♘f3 OO 9. ♗e2 ( 9. ♖c1 c5 10.dxc5 ♘c4 11. b3 ♘xd2 12. ♕xd2 ♘c6 13. ♗b5 ♕a5 14. ♗xc6 bxc615. ♘d4 ♖ad8 16. ♘ce2 ♕xd2+ 17. ♔xd2 ♗d5 18. ♖hg1 ♖fe8 19.♘c3 Rusev-Grover, Golden Sands 2013) a) 9... ♘8d7 10. e4 ♗g4 11. e5 c5 12. d5+/=

b) 9... c5? 10. dxc5 ♘c4 ( 10... ♘6d7 11. b4 ) 11. ♘g5! ♗xc3 12.♗xc3 ♘xe3 13. ♕xd8 ♖xd8 14. ♔f2 ♘d5 15. ♘xe6 fxe6 16.♗d2+/= c) 9... ♘a6 10. ♗xa6 bxa6 11. ♘e4 ♘d7 12. O-O c5 13. ♗c3♗d5 8. ♘f3 c5 I think this has to be the main move here. 8... ♗g4 was what Boris played against me, but after 9. h3 ♗xf3 10.♕xf3 he made a horrendous mistake. 10... ♘8d7 Much better than Gelfand's move. ( 10... c6? 11. h4 left me close to winning in Svidler-Gelfand, London 2013. It looks ridiculous to claim that after 11 moves, but this is really an incredibly dangerous position for Black.) 11. h4Although this is clearly an improved version of the game Black still has reasons to worry. (But not 11. a4 c5 12. a5 cxd4 13. exd4 ♘c8 14.a6 ♘d6 15. axb7 ♖b8 ) 8... ♘8d7 deserves a brief mention. I think White needs to respond 9.e4! , improving the white structure as soon as possible given that the pawn on d4 isn't under attack.( 9. ♗e2 is too tepid: 9... c5 10. d5♗xc3 11. ♗xc3 ♘f6∞ ; 9. ♘e4 ♘f6 10. ♗d3 ♘xe4 11. ♗xe4 ♘c4 12.♖c1 ♗e6 13. ♘g5 ♗d5 14. ♕b3 b5! 15. ♕xb5 c6 16. ♕c5 ♘xb2 17.♕c3 ♘c4 18. ♗d3 ♘d6 19. O-O f5 was fine for Black in Tomashevsky-Morozevich, Tromso World Cup 2013) 9... ♘b8 This looks very odd, but I think it's Black's best move in the position. However, after ( 9... c5 10.d5± ) 10. ♗e3 ♗g4 11. ♗e2 ♘c6 12. d5 White gets a fantastic version of the 3.f3 or the Bd2 0-0 lines. Generally speaking, you don't get a position as nice as this in those lines so this can't be recommended either.12... ♗xf3 13. gxf3 ♘a5 14. ♗d4 ♕d6 ( 14... ♗xd4 15. ♕xd4 e6 16.♕b4 ) 15. ♗xg7 ♔xg7 16. ♕d4+ f6 17. e5 fxe5 18. fxe5 ♘c6 19.♕xb6+− 8... a5 has been played in a number of games, including by Grischuk: 9.♗e2 A much more critical reply.( 9. ♖c1 was Wang Hao's response. 9...♘a6 10. ♘a4⁈ ♘xa4 11. ♕xa4 b6 12. ♕c6 ♖a7 justified Black's play completely in Wang Hao-Grischuk, Beijing GP 2013) 9... ♘a6 10. ♘e4Hitting the pawn on a5 and also stopping ...c5 for now. To be honest, I'm not sure Black equalises here, so I can't really recommend this line in good conscience. 10... ♕d5 11. ♘f2 a4 12. OO c5 13. e4 ♕d8 14. dxc5♘d7 15. c6 ♘dc5 16. e5 bxc6 17. ♗e3+/= White was clearly ahead in Bernadskiy-Gupta, Albena 2013 8... ♘a6⁉ 9. ♗xa6 Forced, as ...c5 was now a serious threat. ( 9. ♘e4♕d5 10. ♘f2 c5 11. e4 ♕d6 is decent for Black.) 9... bxa6 10. ♖c1By far the strongest move. Surprisingly for me I wasn't able to completely equalise this position although it's interesting and gives Black decent chances for counterplay.( 10. O-O allows Black to play 10... c5 and after11. dxc5 ♘c4 I think Black has fantastic compensation for the pawn. 12.♕e2 ♗e6 13. ♖ab1 ♖c8 14. ♖fd1 ♖xc5 15. ♗e1 ♘d6∞ ; 10. b3 c511. dxc5 ♘d5=/∞ ) a) 10... ♗f5 11. b3 ♖c8 12. e4 ♗g4 13. ♗e3 ♗xf3 14. gxf3 e615. O-O ♕d7 16. ♘e2+/= b) 10... ♗b7 11. ♕c2! ( 11. ♘e2 ♖c8 12. OO ♕d6⇆ ) 11... c5 (11... ♖c8 12. ♘e4 ) 12. dxc5 ♘ d7 13. ♘e4 ( 13. b4 a5 14. a3axb4 15. axb4 a5 16. ♘e2 ♗xf3 17. gxf3 axb4 18. ♗xb4 ♕c7=/∞; 13. ♘a4 ♖c8 14. O-O e5 15. ♗c3+/= ) 13... ♗xb2 14. ♕xb2♗xe4 15. c6+/= c) 10... c5 11. dxc5 ♘c4 c1) 12. ♕c2 ♗f5 ( 12... e5 13. b3 ♘xd2 14. ♕xd2 exf4 15.♕xd8 ♖xd8 16. exf4 ♗b7 1 7. ♘a4 ♖ac8 18. ♔f2 ♗c6=/∞ ) 13.e4 ♗g4 14. ♘d5 ♘xd2 15. ♘xd2 ♗e6 16. ♕c4+/=

c2) 12. ♖c2⁉ 12... ♗g4 13. ♔f2 ( 13. ♕e2 ♘xd2 14. ♕xd2♖c8= ; 13. OO ♖c8=/∞ ) 13... ♖c8 14. ♕e2 ♘xd2 15. ♖xd2♕a5 16. ♘d5 ♕xc5 17. ♕xa6 ♗xf3 18. ♔xf3 ♗xb 2 19. ♖b1e6 20. ♖bxb2 exd5 21. ♕d3 , and White retains a small plus. 9. dxc5 ♘6d7 9... ♘d5⁈ 10. ♕b3 ♗e6 11. ♗c4± 10. ♘e4⁉ This move is untried as of right now (mid-September 2013), but I like it better than the two that were already played. 10. ♘a4 b6⁉ ( 10... e5 Black has to have decent play here because it's very hard for me to believe that you can discoordinate your pieces to this extent and still claim any kind of an advantage with White. 11. ♗c4exf4 12. OO ♘a6 13. ♗xa6 bxa6 14. exf4 ♗b7 15. ♖c1 ) 11. cxb6 (11. ♖c1 ♘xc5 12. ♘x c5 bxc5 13. b3 ♘a6 14. a3 ♕b6 15. ♕c2Rabrenovic-Boskovic, Kragujevac 2013 15... ♗e6=/+ ) 11... ♘xb6 a) 12. ♘c3 ♗a6=/∞ ( 12... ♘a6 13. ♗e2 ♘c5 14. O-O ♗e6=/∞ ) b) 12. ♖c1 12... ♘xa4 13. ♕xa4 ♗xb2 14. ♖b1 ♗g7 ( 14... ♗f615. e4⁉ ) 15. ♗e2 ♗d7 16. ♕a3 ♗f5 17. ♖b5 a6 18. ♖b3 ♘c6= 10. b4⁉ was tried in another exciting World Cup game, but Black should be fine here: 10... a5 11. a3 a) It's not so precise to play 11... axb4 12. axb4 ♖xa1 13. ♕xa1 e5, which allowed, 14. ♘e4⁉ , when Black needs to be careful: ( 14.♗c4 exf4 15. exf4 b6= ) 14... b6⁈ ( 14... exf4 15. ♗c3 ♖e8!should still hold for Black, but I can understand why Wesley didn't go for this - it's not easy to play like that when seeing this position for the first time at the board.) 15. ♗c4 ( 15. ♗c3⁉ ) 15... bxc5? (The machine says Black still holds after 15... exf4 16. ♗c3 ♗xc3+ 17.♕xc3 ♗b7 , but it's clear Black's life is far from easy.) 16. ♘fg5 exf4 17. ♗c3 ♗xc3+ 18. ♕xc3 ♕e7 19. ♘xf7 ♖xf7 20. OO ♗a621. b5 ♗xb5 22. ♗xf7+ ♕xf7 23. ♘d6! , and Black was very lucky to eventually hold in Ipatov-So, Tromso World Cup 2013 b) e5! Houdini 12. ♗c4 exf4 13. exf4 ♘c6 14. ♖b1 (14. OO axb4 15. axb4 ♖xa1 16. ♕xa1 ♘xb4= ) 14... axb4 15. axb4♖e8+ 16. ♘e2 ♘f6 17. OO ♗f5 18. ♘g5 ♘d5⇆

10... ♗xb2 I think this has to be played . 11. ♖b1 ♗g7

11... ♘f6 12. ♘eg5 ( 12. ♘xf6+ ♗xf6 13. ♕c2 ♕c7 ) 12... ♗a3 13.♕c2 , and Black lacks coordination. 12. ♗c4 ♘a6! I finally settled on this move. 12... ♕c7 13. ♕b3 ♘xc5 14. ♘xc5 ♕xc5 15. ♘g5 e6 16. ♗b4± 12... ♘f6 is probably the most obvious move in this position, but after13. ♘xf6+ ♗xf6 White has ( 13... exf6 14. O-O ♕e7 15. ♘d4+/= ) 14.♕c2⁉ A surprising alternative to 14.0-0. ( 14. OO ♕c7 15. ♕e2 ♘d716. ♖fc1 ♘xc5 17. ♗d5=/∞ ) 14... ♕c7 15. h4! Ridiculous as it might look at first glance this is actually a very serious attempt to give immediate mate and one which can't be refuted that easily, so I wasn't really satisfied with 12...Nf6.

13. ♗xa6 bxa6 At least this way we're definitely not getting mated. In this position play could continue: 14. ♖c1 A somewhat counterintuitive move as you're giving the bishop the b7-square, but otherwise Black could start an immediate attack on the c5-pawn. Rc1 prevents that. 14... ♘b8 I quite like this move for Black. 14... ♘f6 15. ♘xf6+ ♗xf6 16. O-O ♖b8 17. ♗c3+/= 15. O-O 15. ♗c3 ♕xd1+ 16. ♖xd1 ♗f5 17. ♗xg7 ♔xg7= 15. ♘f2 ♘c6⇆

15... ♕d5

15... ♘c6 16. ♕a4 16. ♘f2 ♘c6 17. ♕a4 17. ♗c3 ♕xd1 18. ♖fxd1 ♖b8= 17... ♖b8 18. ♗c3 ♗xc3 19. ♖xc3 ♖b2! It seems to me Black should hold here, but it's a reasonably close run thing because if White had an extra tempo he could start creating serious inconveniences for Black in a line like this. I have mixed feelings that I can't really provide you with immediate equality against 6.e3 and 7.f4. On the one hand I'm quite proud that this idea is so tenacious but on the other hand as a Grünfeld player I'm somewhat puzzled that something as alien and frankly ugly-looking as this cannot be immediately refuted. Such is life, though! In general I think 5...Nb6 is doing very well right now. Once again the only problem that I can see is 6.e3, 7.f4 and I suspect the only reason I can't find immediate equality against it is that I'm very biased towards this line emotionally. Perhaps something will present itself to more detached researchers! 20. ♖c2 ♖xc2 21. ♕xc2 ♗e6=

5.Bd2 Nb6 6.Nf3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. ♗d2 ♘b6 6. ♘f3The point of this move is first to develop the knight and then put the rook on c1. Only after that will White perhaps play Bg5. Generally speaking, Black aims to play ...Bg7 and ...c5 more or less straightaway against most of White's moves. By putting the knight on f3 and the rook on c1 first White attempts to prepare for that eventuality as well as he possibly can. 6... ♗g7 6... ♘c6 straightaway has been tried, forcing White to play 7. ♗f4 and then after 7... ♗g7 play 8. e3 On the one hand Black has perhaps prevented White from playing the most aggressive setup, but on the other hand putting the knight on c6 obviously makes playing ...c5 much harder, something I'm not in favour of.

7. ♖c1 ♘c6! 7... O-O was Shirov's move against Morozevich. 8. ♗g5 h6 9. ♗h4 g510. ♗g3 c5 The most thematic break. 11. d5 The game soon became very sharply poised - it's not that this is unplayable for Black, but 7...Nc6 is a simpler and more forcing solution. 11... e6 ( 11... ♗xc3+⁉ is a very solid choice if Black wants to avoid the mayhem of 11...e6. 12. ♖xc3♘xd5 13. ♖xc5 ♘f6 ) 12. d6 ♘d5 13. h4 led to a very sharp and unclear battle in Morozevich-Shirov, Pamplona 2006. 8. e3 Pretty much forced. 8. ♗f4 allows 8... ♘xd4 9. ♘xd4 e5 when Black is just fine. 10. ♘cb5 OO 11. ♗g3 c6 12. ♘xc6 bxc6 13. ♕xd8 ♖xd8 14. ♘c3 ♘d5 15.e4 ♘b4 De Jong-Sasikiran, Kemer 2007 8... O-O 9. ♗b5 The most critical move. 9. ♗e2 e5 10. dxe5 ♘xe5= 9... e5 This is what we play against almost any 9th move by White. 9... ♗g4⁉ is a rare but decent move: 10. h3 ( 10. ♗xc6 bxc6 11. ♘e2♕d5 ) 10... ♗xf3 11. ♕xf3 e5 12. d5⁉ ( 12. ♗xc6 bxc6 13. OO exd4 14. exd4 ♕xd4 15. ♗f4 ♘d5= Zhou Weiqi-Zhou Jianchao, Hefei 2011) 12... ♘e7 13. e4 a6 14. ♗e2 c6 15. O-O cxd5 16. ♘xd5♘bxd5 17. exd5 ♘f5= 10. ♗xc6 exd4 11. ♘xd4 bxc6 12. ♘ce2 The main move for White, trying to put additional pressure on the c6pawn and control the f1-a6-diagonal so that castling becomes easier. Black has tried a number of moves in this position. 12... ♕d5! The easiest way to equalise. 12... ♗a6 was played by Viktorija Cmilyte in a game against Karpov, but is less precise: 13. ♗b4 ♗xe2 14. ♘xe2 ♗xb2 15. ♕xd8 ♖fxd8 16.♖xc6 ♘d5 17. ♗a5 ♖ab8 ( 17... ♗e5 18. ♘c1 ♖ab8 19. ♘b3+/=Karpov-Cmilyte, Marianske Lazne 2008) 18. O-O ♗e5 19. ♖c5 , and White will press for a while. 12... ♕g5 was Kurnosov's move against Dreev. 13. OO c5 14. ♘f3♕e7 15. ♕c2 ( 15. ♗c3 ♖d8 16. ♕e1 ♗h6! 17. ♘e5 ♖d6 18. b3♗a6 DreevKurnosov, Dresden 2007) 15... c4 16. ♘g3 ♗b7 17. e4♖fe8 18. ♖fe1 ♖ad8 19. ♗c3+/= 13. O-O 13. ♘f4 ♕xa2 Black is fine here. 14. ♖a1 ♕c4 15. b3 ♕c5 16. ♖c1♕d6 17. ♘xc6 ♗d7 ( 17... ♗b7 18. ♗b4 ♕xd1+ 19. ♔xd1 ♖fe8 20.♘e7+ ♔h8 21. ♖xc7 ♖ab8⇆ ; 17... a5 18. ♗xa5 ) 18. ♗b4 ♕xd1+ 19.♔xd1 ♖fe8=/+

13... c5 14. ♘f4 ♕d6

14... ♕xa2⁈ can lead to a very nice trap you shouldn't fall into: 15.♖xc5 ♕xb2? 16. ♘d3 Surprisingly the queen is caught. 16... ♕a2 17.♖a5 ♕c4 It seems as though everything is fine, but.. . 18. ♘b2‼+− This is a very beautiful position. It has no particular theoretical value, but aesthetically it's very nice to see the queen on c4 with pretty much all the board and diagonals empty - but still it has no moves!

15. ♘b5 ♕d7 16. ♖xc5 ♖d8 17. ♘xc7 ♗b7 Black is fine here. 18. ♘xa8 ♗xa8 19. ♕b3 ♕xd2 20. ♖c7 ♖d7 21. ♖xd7 ♕xd7=Marcelin-Sasikiran, France 2008

5.Bd2 Nb6 6.Bg5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. ♗d2 ♘b6 6. ♗g5♗g7 7. e3 c5 A forcing solution which I think equalises on the spot. 7... O-O a) 8. ♖c1 is less studied and a bit trickier. 8... h6 9. ♗h4 g5 10.♗g3 c5 This is also quite a wellknown position which I think is fine for Black. 11. ♘f3 ( 11. dxc5 ♗xc3+ 12. ♖xc3 ♕xd1+ 13. ♔xd1♘a4 14. ♖c2 ♗f5 15. ♖d2 ♘c6 16 . ♘e2 ♘xc5 17. f3 ♗e6 was great for Black in Eljanov-Sutovsky, Sochi 2005) 11... ♘c6 ( 11...cxd4⁉ 12. ♘xd4 ♘c6 13. ♘xc6 ♕xd1+ 14. ♘xd1 bxc6 15. ♗e2PostnyBobras, Dresden 2007 15... ♗e6 16. b3 ♘d5 17. h4 a5⇆ ) 12.d5 The safer option for Black is a1) 12... ♘b4 13. d6 leads to a very messy position in which I would probably prefer to be White: a11) 13... exd6 14. a3! ( 14. ♗xd6 ♗f5 15. ♗xf8 ♕xf8 16.a3 ♖d8 17. ♘d2 ♘c6=/∞ ) 14... ♘c6 15. ♗xd6+/= a12) 13... ♗e6 14. a3 ( 14. dxe7 ♕xe7 15. ♗d6 ♕f6=/∞ )14... ♘4d5 15. ♘xd 5 ♗xd5 16. ♖xc5 e6 17. h4 g4 18.♘d4± Safin-Greenfeld, Calicut 2007 a2) 12... ♗xc3+ 13. ♖xc3 ♕xd5 14. ♕xd5 ♘xd5 15. ♖xc5 ♘f616. ♗b5 ♗d7 ( 16... ♘ e4 17. ♖c4 ♘xg3 18. hxg3 ♗e6 19. ♖c5♗xa2 20. ♘xg5+/= ) 17. O-O Black should hold after both a21) 17... ♖ac8 and 18. ♗xc6 and( 18. ♖fc1 ♘e4 ) 18...♖xc6 ( 18... ♗xc6 1 9. ♘d4 ♘e4 20. ♘xc6 ♖xc6 21. ♖xc6bxc6 22. ♗c7 ♖c8 23. ♗a5 ) 19. ♖a5 a6 20. ♘d4 ♖c4 and a22) 17... ♘e4 18. ♖d5 ♗e8 19. ♖c1 f6 20. ♖dd1 ♘xg3 21.hxg3 ♖c8 b) 8. ♘f3 This is the old theory. 8... h6 9. ♗h4 c5 10. dxc5 ♘6d7 I myself played this position in the late 90s. I think it's fine, and according to recent theory this is quite harmless and very playable. (10... g5 11. ♗g3 ♗xc3+ 12. bxc3 ♕xd1+ 13. ♖xd1 ♘6d7 14. h4g4 15. ♘e5 ♘xe5 16. ♗xe5 ♗ e6 17. ♗f4 h5 18. ♖b1 ♗xa2 19.♖xb7 ♘c6 20. e4+/= Fridman-Bobras, BL 2008) b1) 11. ♗e2 b11) 11... g5 12. ♗g3 ♘xc5 13. h4 b111) 13... ♗xc3+ 14. bxc3 ♕xd1+ 15. ♖xd1 ♘e4 16.hxg5 hxg5 17. ♗e5 f6 18. ♗d3 ♘c5 Schweda-Stohl, Austria 2007 19. ♗h7+! ♔g7 ( 19... ♔f7 20. ♗d4 ♘ba6 21.♘xg5+ fxg5 22. ♖h6± ) 20. ♗d4 ♘e6 21. ♗c2+/=

b112) g4 14. ♘d4 ♘c6 15. ♘xc6 ♕xd1+ 16. ♖xd1 bxc617. ♗c4 ♗f5 18. f3 gxf3 19. gxf3 ♖fd8 20. ♔e2 h5 21.♖hg1+/= Nyzhnyk-Gupta, Groningen op A 2010 b12) ♘c6 12. ♘d4 ♘xc5 ( 12... g5 13. ♘xc6 bxc6 14. ♗g3♘xc5 15. ♕c2 ♗a 6 16. ♗xa6 ♘xa6 17. O-O ♘b4 I. Sokolov-Leko, Dortmund 1999 18. ♕e4+/= ) 13. ♘xc6 bxc6 14. O-Og5 15. ♗g3 ♗f5= DreevMamedyarov, Gothenburg 2005 b2) 11. ♖c1 b21) 11... ♘c6⁈ 12. ♗b5 ♘xc5 13. OO g5 14. ♗g3 ♗e6 (14... ♗f5 15. ♕d5! Giri-Gupta, Wijk aan Zee 2009) 15. ♗xc6bxc6 16. ♘d4 ♗d7+/= DreevSutovsky, Sochi 2005 b22) 11... g5 12. ♗g3 ♘xc5 13. ♘d4 ♘c6 14. ♘xc6 bxc6 (14... ♕xd1+? 15. ♘xd1!+− ) b221) 15. ♗e2 ♗f5 16. OO ♘d3 ( 16... ♕a5⁈ 17. ♘d5! )17. ♗xd3 ♗xd3 18. ♖e1∞ b222) 15. ♕c2 15... ♕a5 b2221) 16. a3⁉ ♗e6 17. b4 ♕xa3 18. bxc5 ♖fd8 19.♗e2 ♗b3 20. ♕b1 ♗e6= b2222) 16. h4 ♘a4! b22221) 17. hxg5 ♗f5 18. e4 ♖fb8!↑ 19. ♗c7 ♕xc720. ♘xa4 ♕e5 21. gxh6 ♗ xe4 22. ♕c3 ♕xc3+ ( 22...♗d5+ 23. ♕xe 5 ♗xe5 24. b3 ) 23. ♘xc3 ♗xc3+ 24.bxc3 ♖b2 b22222) 17. ♗d3 g4 18. OO ♘xc3 19. bxc3 ♗e6= b22223) 17. ♗e2 b222231) 17... ♗f5 18. e4 ♗e6 19. hxg5 ♘xc3 (19... hxg5? 20. e5! ♖fd8 21. ♕xa4+− ; 19...♖fb8? 20. ♗xb8 ♖xb8 21. OO± ) 20. bxc3 hxg5 21. ♖h5→ b222232) 17... g4 18. OO ♘xc3 19. bxc3 ♗e6 20.c4 ♖ad8 21. ♖fd1 ♗f5 ! 22. e4 ♗e6= b2223) 16. ♗e2 16... ♘a4 17. OO ( 17. h4 transpo ses to 16.h4.) 17... ♘xc3 18. bxc3 ♗f5 19. ♕b3 ♖ad8 , and Black was fine in Markus-Navara, Turin 2006.

8. dxc5 ♗xc3+ 9. bxc3 ♕xd1+ 10. ♖xd1 ♘6d7 This is a very typical looking endgame which you'll encounter in many lines. Black has given up the bishop and is currently a pawn down, but on the other hand he's completely ruined White's structure on the queenside and if the pawn on c5 is won back Black is normally fine. It's also important to note that in this position with the bishop on g5 Black has an additional tempo to play f7-f6. 11. ♗b5 The main line. 11. ♖d5 f6 12. ♗h4 ( 12. ♗h6 ♘a6 ) 12... ♘a6 The simplest way for Black to proceed. 13. c6 ( 13. ♗xa6 bxa6 This isn't dangerous at all for Black because ...Bb7 is a huge threat. 14. c6 ♘e5 15. ♖xe5 fxe5 16.♘f3 e4 17. ♘e5 ♖b8 18. OO ♖b5 19. ♗g3 ♖f8 20. c4 ♖c5 21.♖b1 ♖f5 22. ♖b8 ♔d8∓ ) 13... bxc6 14. ♖a5 ♘ac5 15. ♘f3 ♘b7 16.♖a3 ♘b6 17. ♘d4 ♗d7 , and Black is fine. 11. ♘e2 ♘xc5 12. ♘d4 ♗d7 13. f3 f6 14. ♗h6 ♘c6= 11... ♘c6 12. ♖d5 f6 13. ♗h4 13. ♗f4 runs into a very strong response: 13... e6 14. ♖d6 People have been making inferior moves in this position which has been seen in a couple of high-level games. a) 14... e5 15. ♗g3 ♔e7 16. ♖d5 a6 17. ♗a4 b5 ( 17... ♘d8! ) 18.cxb6 ♗b7 19. ♖xd7+! ♔xd7 20. f4+/= Turov-Krasenkow, Dagomys 2008 b) 14... ♔e7 15. ♖xc6 e5 16. ♗xe5! ( 16. ♖d6 exf4 17. ♗xd7♗xd7 18. exf4 ♖ac8 ChuprikovDvoirys, Belgorod 2008) 16... fxe517. ♖d6 ♘xc5 18. ♖d5+/= c) 14... ♘xc5! This is a big improvement. 15. ♖xc6 bxc6 16. ♗xc6+♔e7 17. ♗xa8 ♘d3+=/+ Black is at least not worse because White can't give up the f2-pawn without heavy material losses, and after something like 18. ♔e2 Black can play 18... ♗a6 and win one of the bishops back with a continuing initiative. Black may already be better. 13... a6 First driving the bishop to a4. 14. ♗a4 ♘de5 After, for instance 15. ♘e2 15. ♖xe5 fxe5 16. ♘f3 ♗d7=/+ 15... ♗e6 16. ♖d2 ♖c8 Black is ready to play ...Kf7, ... Na5 and collect on c5. As usual in an endgame like this once the pawn on c5 falls it becomes completely obvious that Black cannot be worse.

5.Bd2 Nb6 6.Bf4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. ♗d2 ♘b6 6. ♗f4This move is trickier for Black to face than Bg5. 6... ♗g7 7. e3 O-O White has two playable options here - Rc1 or Nf3. 7... c5 8. dxc5 ♗xc3+ 9. bxc3 ♕xd1+ 10. ♖xd1 ♘6d7 In contrast to the 6.Bg5 lines the bishop is on f4 here, which gives White additional possibilities.

a) 11. ♗xb8 is possible straightaway, but Black should simply take with the rook. 11... ♖xb8! ( 11... ♘xb8 left White better: 12. ♘f3O-O 13. ♘d4+/= Nyzhnyk-Negi, Hoogeveen 2010) 12. ♗b5 e6 13.♘f3 ♔e7 Black should be fine. 14. ♗xd7 ♗xd7 15. ♘e5 ♗c6 16.O-O ♖hc8= b) White could simply ignore the pawn on c5 and play something like11. h4⁉∞ ♘xc5 12. h5 , which perhaps isn't such a stupid idea, but it's completely untested and I also don't think it's going to be the refutation of Black's setup. c) 11. ♗b5 The main move. 11... a6 ( 11... ♘c6 12. ♘f3 ♘xc5 13.♘e5 ♗d7 14. ♖d5 ♘xe5 15. ♗xe5 f6 16. ♖xc5 fxe5 17. ♖xe5♗xb5 18. ♖xb5 O-O-O 19. ♔e2± Arutinian-Kanarek, Rijeka 2010 )12. ♗a4 c1) 12... f6⁈ was played by Morozevich, but I think that was a mistake as White does get some advantage after 13. ♗xb8 ♖xb814. ♘e2 The game continued 14... e6 15. ♘d4 ♔e7 16. ♘b3 This is the point of White's play. With a knight on b3 it's very difficult for Black to untangle himself and collect on c5. Black was suffering.16... f5 17. ♖d4 ( 17. c4⁉ ) 17... ♘f6 18. ♘a5 ♘e4 19. ♖c4e5 20. f3 ♘f6 21. ♖b4 ♗e6 22. c4 ♖hc8 23. O-O ♘d7 24. c6bxc6 25. ♖xb8 ♘xb8 26. ♖b1+/= Wang Yue-Morozevich, Wijk aan Zee 2009 c2) ♘c6! Attacking the pawn on c5 straightaway. 13. ♗c7 is the other additional possibility the bishop on f4 gives White. ( 13. ♖d5is fine for Black, e.g. 13... ♘f6 14. ♗xc6+ bxc6 15. ♖d4 ♘d7 16.♘f3 ♘xc5= ) 13... OO 14. ♗b6 ♘xb6 15. cxb6 ♘e5 is fine for Black as it's impossible to hold the pawn on b6 comfortably. Black, with an advantage in development, has no problems here. 16. ♘f3♘c4 17. ♗b3 ( 17. ♖b1 ♗f5 18. ♖b4 ♖fc8 19. ♗b3 ♘d6=MuzychukBok, Wijk aan Zee 2010 ) 17... ♘xb6= 8. ♘f3 8. ♖c1 c5 9. dxc5 ♗xc3+ 10. bxc3 ( 10. ♖xc3 would of course be met by 10... ♕xd1+ 11. ♔xd1 ♘a4 which will never be any good for White.) a) 10... ♕xd1+ 11. ♖xd1 ♘6d7 is quite a dangerous endgame for Black. a1) 12. ♗b5 ♘c6 13. ♘f3 ( 13. ♖d5 ♘f6 14. ♖d1 ♘e4 15. ♘e2♘xc5 16. O-O DreevPonomariov, Moscow 2007 ) 13... ♘xc5 14.♘d4 e5 15. ♘xc6 exf4 16. ♘e7+ ♔g7 Dao Thien Hai-Gyorkos, Budapest 1993 a2) 12. ♗h6 12... ♖e8 13. ♘e2! ( 13. ♗b5 a6 14. ♗a4 ♘c6 15.♘f3 ♘xc5 16. ♗xc6 bxc6 17. ♘d4 e5= Dreev-Shirov, Moscow 2007 ) 13... ♘xc5 14. ♘f4 The position of this knight is quite unpleasant for Black due to the threat of Nd5. There's no immediate equality. 14... ♗e6 ( 14... ♘c6 15. ♘d5 ) 15. ♘xe6 ♘xe6 16.♗e2+/= b) 10... ♘6d7 It's better to play this first without exchanging queens. b1) 11. h4 ♘xc5 12. h5 ( 12. ♕f3 ♗f5→ ) 12... ♕xd1+ 13.♖xd1 ♗e6 14. hxg6 hxg6⇆ b2) 11. ♘f3 can also be played first. 11... ♘xc5 is perhaps not entirely accurate, but Black can play

( 11... ♘c6 instead.) 12.♗h6 ( 12. ♕d4 ♘e6 ) 12... ♕xd1+ 13. ♖xd1 ♖e8 14. ♘d4♗ d7= b3) 11. ♗b5 Black has a choice. 11... ♘xc5 ( 11... ♕a5⁉ ; 11...♘c6 is possible.) 12. ♗h6 It seems as though an exchange is about to be lost, but after 12... ♕a5 the complications work out fine for Black. He should hold comfortably. 13. ♕d4 ( 13. ♗xf8 ♕xb5 14.♕d8 ♘c6 15. ♕e8 ♘d4 16. ♕xb5 ♘xb5 17 . ♗xe7 ♘d3+ 18.♔d2 ♘xf2=/+ ) 13... ♘e6 14. ♕b4 ♕xb4 15. cxb4 ♖d8 16.♘f3 ( 16. ♗e2 ♘c6 17. a3 ♘e5 18. ♖c3 f6 19. ♘f3 ♘f7 20.♗f4 ♘xf4 21. exf4= ) 16... a5= b4) 11. ♗h6 11... ♖e8 b41) 12. ♕d4 runs into 12... e5 13. ♕c4 ♘a6 14. ♖d1 ♕f6→ b42) 12. ♘f3 ♘c6 13. ♗b5 ♘xc5 14. OO ♕xd1 15. ♖fxd1a6 16. ♗xc6 bxc6 1 7. ♘d4 ♗d7= b43) 12. ♗b5 would be met by 12... ♘c6 and once again Black is ready to start collecting. 13. ♘f3 ♘xc5 14. OO ( 14. ♕e2♗f5 15. OO ♗d3=/+ ) 14... ♗g4= b44) 12. ♗c4 This is perhaps the critical line. 12... ♘c6 Black is once again ok. ( 12... ♘xc5 Black has to careful as after 13.♕d4+/= he's slightly worse here..) 13. ♘f3 ♘xc5 14. ♕e2 ( 14.♘d4 ♘a5=/+ ) 14... ♗e6 15. OO ( 15. ♗b5 a6−+ ) 15...♗xc4 16. ♕xc4 b6 17. ♖fd1 ♕c7 18. ♘g5 ♘e5 8... c5 8... ♗g4⁉ is a possible but somewhat passive alternative: 9. h3 ♗xf310. ♕xf3 c6 Black is preparing ...Nd7 and ...e5 and aiming for Schlecter-type structures. 11. ♗e2 ( 11. h4⁉ ) 11... ♘8d7 12. OO e5 13. dxe5♘xe5 14. ♕e4 Rodshtein-Roiz, MaalotTarshiha 2008 9. dxc5 ♘6d7 9... ♗xc3+ leads to an endgame which is slightly risky for Black. 10.bxc3 ♕xd1+ 11. ♖xd1 ♘6d7 ( 11... ♘a4 12. ♗h6± ) 12. ♗g5 ( 12.♗h6 ♖e8 13. ♗b5 a6 14. ♗ a4 ♘c6 15. ♘d4 ♘xc5 16. ♗xc6 bxc617. O-O a5 18. c4= 1/2 Sargissian-Li Chao, Khanty-Mansiysk 2009 ) 12...♖e8 13. ♗b5 a6 14. ♗a4 a) 14... ♘c6 15. ♖d5 e6 ( 15... f6 16. ♗h4 e6 17. ♖d6 ♘xc5 18.♗xc6 bxc6 19. ♘d2 g5 20. ♗g3+/= ) 16. ♖d6 ♘xc5 17. ♗xc6bxc6 18. ♘d2+/= b) f6 15. ♗h4 ( 15. ♗h6 ♘c6 16. ♗b3+ e6 17. ♘d4 ♘xc5 18.♗c2 Khalifman-Kojovic, Budva 2009 ) 15... ♘c6 16. ♖d5+/= 9... ♕xd1+ 10. ♖xd1 ( 10. ♘xd1 ♘a4 ) 10... ♗xc3+ 11. bxc3 ♘a4?12. ♗h6!+− 10. ♘d5 Play now becomes very forced and you need to know the theory as it's easy to go wrong. If you don't go wrong, however, and you know what you're doing, I think it's pretty much a forced draw.

10. ♗c4 , just giving up the pawn on c5, leaves Black very comfortable e.g. 10... ♘xc5 11. ♕e2 ♘c6 12. O-O ♗e6 13. ♘g5 ♗xc4 14. ♕xc4♘d3 Gajewski-Melkumyan, Kavala 2009 10. ♖c1 ♘c6 11. ♗b5 ♘xc5 12. O-O ♗f5 13. ♕e2 ♗d3 14. ♗xd3♕xd3= Korchnoi-Spassky, St. Petersburg 1999 10... ♘xc5 10... e5⁈ 11. ♗g3 or even ( 11. ♗g5 is incredibly dangerous for Black. )11... ♘c6 12. ♗h4!± ( 12. ♖c1 ♘f6 13. ♘xf6+ ♕xf6 14. ♗d3 e4 15.♗xe4 ♕xb2 16. OO± Malakhov-Ponomario ) 10... ♗xb2⁈ is also very dangerous. 11. ♗c7 ♕e8 ( 11... ♗c3+? 12.♔e2 ) 12. ♖b1 ♗g7 13. ♗b5 Things don't work out the way Black would like in this position, so this isn't recommended. 13... ♘c6 a) 14. OO ♘xc5 ( 14... ♘f6 15. ♘xf6+ ♗xf6 16. ♗a4! ♗d7 17.♖xb7 ♕c8 18. ♕b1 ♘a5 19. ♗x d7 ♕xb7 20. ♕xb7 ♘xb7 21. c6♘c5 22. ♖c1+/= ) 15. ♖c1 ( 15. ♕c2 ♘e6 16. ♗e5 ♗d7 17. ♗ xg7♔xg7 18. ♖fd1 ) 15... a6 ( 15... ♘e6 16. ♖xc6!+− ) 16. ♗xc6bxc6 17. ♘xe7+ ♕xe7 18. ♗d 6 ♕e6 ( 18... ♕b7 19. ♗xc5 ♖e820. ♗d4+/= ) 19. ♗xc5+/= b) 14. ♘d4 ♗xd4! ( 14... e6 15. ♘c3 ♘xc5 16. ♗d6 e5 17. ♘xc6bxc6 18. ♗e2 e4 19. OO ♗xc3 20. ♗xc5 ♗e6 21. ♕c2 ♗f6 22.♗xf8 ♔xf8 23. ♗c4± Dreev-Lalic, Sibenik 2007 ) 15. exd4 ♘f6 16.♘e3 ( 16. ♘c3 ♗f5 17. ♖b2 ♕d7 18. ♗e5 ♖ad8 ) 16... ♕d7 17.♗e5 ♘g4= c) 14. ♗g3 14... ♕d8 15. ♘c7 ♖b8 16. O-O ♘de5 17. ♘d5± 11. ♖c1 White generally plays this move. 11. ♗c7 ♕d7 12. ♖c1 ♘e4 13. ♗xb8 ♖xb8 14. ♖c7 ♕d8 15. ♘xe7+♔h8 16. ♕xd8 ♖xd8=/+ 11. ♘c7 ♗xb2! 12. ♖c1 ♗xc1 13. ♕xc1 ♘e4 14. ♘d2 ♕xd2+ 15.♕xd2 ♘xd2 16. ♔xd2 e5! 17. ♗g5 ( 17. ♗xe5 ♘c6 ) 17... ♗e6 18.♘xa8 ♖c8= 11... ♘ba6 11... ♘bd7 12. ♗c7 ♕e8 13. ♗g3 ♕d8 14. ♗h4+/= 12. b4 ♘e4 Other moves don't really equalise. 12... ♘e6 13. ♗e5 ♗xe5 14. ♘xe5 ♕d6 15. ♗xa6 ♕xe5 16. ♗c4♖d8 17. OO+/= b5 18. ♘xe7+ ♔f8 19. ♕f3! ♔xe7 20. ♕xa8 bxc421. ♖xc4 ♗d7 22. ♕xa7+/= 12... ♗e6 13. ♘xe7+ ♕xe7 14. ♗d6 is quite unpleasant for Black. 14...♕f6 15. bxc5 ♘b4 16. ♗xf8 ♖xf8 17. ♗c4 ♖d8 18. ♘d4 ♘xa2 19.♗xa2 ♗xa2 20. O-O± 12... e6 13. ♘e7+ ♕xe7 14. ♗d6 ♕d7 15. bxc5+/= 13. ♗xa6 bxa6 14. ♖c7 ♗d7 Black is now ready to play ...e6 against almost any move. 14... ♔h8 15. O-O ♗e6 16. ♘xe7 ♕xd1 17. ♖xd1 ♘c3 18. ♖d2♘xa2 19. ♘c6±

15. ♘g5 15. ♕d3 is the other critical move here. 15... e6 16. ♕xe4 exd5 17.♕xd5 ♗b5 18. ♕xd8 ♖fxd8 19. ♗e5 ♗f8 20. a3 ♖d5! Keeping the knight rooted to f3. This is the last moment when you need to show some precision. ( 20... ♖d3 21. OO ♖xa3 22. ♖fc1+/= ; 20... a5immediately runs into 21. ♘d4 when White actually retains some pressure. 21... ♗e8 22. bxa5 ♖d5 23. f4 ♖xa5 24. ♘c2 ) a) 21. ♗b2 a5 22. bxa5 ♗d6 23. ♖c2 ♖b8 24. ♖d2 ♗c4 25. ♗c1♗xa3 26. ♖xd5 ♗xd5 27. OO ♗b4 28. ♖d1 ♖b5 b) 21. h4 b1) 21... ♖xe5 22. ♘xe5 ♗d6 23. ♖c5 ♗xc5 24. bxc5 ♖c8 25.c6 ( 25. ♔d2 ♖xc5 26. ♘f3 ♗c6 ) 25... f6 26. ♘f3 ♗xc6 27.O-O= b2) a5⁉ c) 21. ♗d4 21... a5 Now Black has perfectly decent play for the pawn as the white king is stuck on e1 and the rook on h1 can take no part in the game. 22. ♗c5 axb4 ( 22... ♖d3 23. ♗xf8 ♖xa3 24. ♔d2♔xf8 25. bxa5 ♖a2+ 26. ♖c2 ♖xa5 27. ♘d4 a6 28. ♖hc1= ) 23.♗xf8 ( 23. axb4 a5∓ 24. ♗xf8 axb4 ) 23... ♔xf8 24. axb4 a5 25.bxa5 ♖xa5 26. ♖c1 ♖a2 27. ♘d4 ♗d3= 15. O-O e6 16. ♘e7+ ♕xe7 17. ♕xd7 a) 17... ♕xb4 18. ♕c6 ♘c3 19. ♗e5 ♗xe5 20. ♘xe5 ♘e2+ ( 20...♘b5 21. ♖xf7 ) 21. ♔h1 ♖ab8! 22. g3!± ( 22. ♘d7 ♕b1! ) b) ♕xd7 18. ♖xd7 ♘c3 19. a3 ♖fd8= 15... e6 16. ♘xe4 exd5 17. ♕xd5 ♗b5 This position has been seen in a number of high-level games. 17... ♗e6 18. ♕xd8 ♖fxd8 19. ♔e2 ♗xa2 20. ♖hc1 ♗d5 21. f3+/= 18. ♕xd8 18. ♗d6 Gupta was fine in a game against Grandelius after 18... ♖c8!19. ♖xc8 ♕xc8 20. ♗xf8 ♕g4 Hitting all kinds of weak squares. 21.♕d2 ♕xg2 ( 21... ♕xe4⁉ 22. ♗xg7 ♕xg2 23. ♕d4 ♕xh1+ 24. ♔d2= )22. ♘g3 ♗xf8 It's quite obvious that Black has fantastic compensation for the exchange because White can barely move. 23. a3 a5 24. bxa5 h5→Grandelius-Gupta, Wijk aan Zee 2010 18... ♖fxd8 Once again there's a choice. 19. f3 19. ♘d6 ♗f8 20. ♘xb5 axb5 a) 21. a3 ♖d3 ( 21... a5 22. ♔e2 axb4 23. axb4 ♗xb4 24. ♖b7♗d6= ) 22. ♔e2 ♖xa3 23. ♖d1 ♗xb4 24. ♖dd7 ♖a2+ 25. ♔f3♗e1 26. ♖xf7 ♖xf2+ 27. ♔g4 ♗c3 28. ♖xh7

b) 21. ♔e2 Bacrot offered a draw here in the World Cup as Black can simply play 21...a5. Bacrot-Robson, Khanty-Mansiysk FIDE WC 201121... a5 The queenside is about to get completely annihilated. 22.♖c2 ( 22. ♗e5 axb4 23. ♖c2 ) 22... ♗xb4 23. ♖hc1= 19. a4 ♗xa4 20. O-O ♗b5 21. ♖fc1 ♗b2= 19... ♖ac8 20. ♔f2 ♖xc7 21. ♗xc7 ♖c8 Unfortunately for White 22.Rc1 runs into 22...Be5 so you can't fight for the c-file and have to retreat with the bishop. 22. ♗f4 ♖c2+ 23. ♔g3 f5! Playing this move immediately is my recommended improvement in this position. 23... ♗e2 24. a4 saw Dreev retain some pressure as White: 24... f5 25.♘d6 ♗f8 26. e4 fxe4 27. ♘xe4 ♗xb4 28. h4 Dreev-Karavade, Reykjavik 2010 23... ♖xa2 24. ♖c1 h6 25. h4 ( 25. ♘d6 g5! ) 25... ♗d3 26. ♖c8+♔h7 27. ♖c7 ♗f1 ( 27... ♔g8 28. ♗d 6 a5 29. ♖xa7 ♗xe4 30. fxe4♖d2 31. ♗e7 ♗e5+ 32. ♔f3 axb4 33. ♗xb4 ) 28. ♔h2+/= 24. ♘d6 ♗c3 With the knight occupying the d-file and the black rook not vacating the c-file there's no way for White to activate the h1-rook in time and the pawn on b4 is about to fall. 25. a3 25. ♖d1 ♗e2 26. ♖b1 ♗d3= 25... ♗b2 Once again a draw is not very far away here. 26. e4 26. ♘xb5 axb5 27. ♖d1 ♗xa3 28. ♖d7 ♗xb4 29. ♖xa7= 26... fxe4 27. ♘xe4 ♗xa3 28. ♗d6 28. ♖b1 ♗c6= ( 28... ♖b2 29. ♖c1 ) 28... ♗b2=

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 7.Bb5+ 7.Nf3 c5 8.Bb5+ Nc6 9.0-0 9.d5 8.h3 4.Nf3 Bg7 5.h4

7.Bb5+ We're now going to look at the Grünfeld sidelines in which White gives a bishop check on b5. There will also be a minor mention for a move Kramnik used to make in the early millennium but which has now gone completely out of fashion - I'll be showing you why. There's also going to be a bonus track because I realised I'd left out an interesting new sideline which I think is fine for Black and definitely deserves a special mention. 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6. bxc3♗g7 7. ♗b5+ White can give this bishop check either here or after 7.Nf3 c5. Both moves are interesting and require some precision in reply. 7... c6 I think this is the best and most critical move. 7... ♘d7 and 7... ♗d7 are nevertheless also both perfectly reasonable moves for Black. 8. ♗a4 In this position I'll be showing material which has been covered quite well, as usual, by Boris Avrukh in his book because, to be honest, the one idea I had which differs quite significantly from that book is fairly suspect. 8... O-O The main line. 8... b5 was my idea. I'll show you it anyway to give you an overview of what I thought was interesting in this position over the course of the many years during which I was debating what to do here. 9. ♗b3 ♗b7 Trying to get an unbalanced position in which Black aims for a delayed central break with ...c6-c5. 10. ♘e2 O-O ( 10... c5 is premature here, I think. )11. O-O ( 11. h4 White has the option of playing this immediately, which is hardly surprising given Black has completely abandoned the kingside and White hasn't yet castled - why not just attempt to give mate straightaway? 11... c5 12. h5 ♗xe4 13. f3 I wouldn't say Black is lost but he's in a lot of danger because hxg6 and Bh6 are coming very soon and Black will have to play very, very carefully not to get mated quickly. So 8...b5 cannot really be recommended.) 11... e6 12. ♗a3 ♖e8 This leads to a very unbalanced position. Black wants to play ...c5 right away while the pawn on e4 is hanging. 13. e5 ( 13. f4 c5 ; 13. ♕d3 ♘d7 14.♖ad1 ♕c7 ) 13... ♘d7 Black again wants to play ...c5 next move. This has the potential to be strategically dangerous but on the other hand if it works out for Black he might actually be playing for a win because the position is so unbalanced. 14. ♗d6 ( 14. f4 c5 15. dxc5 ♖c8 ) 14... c5Black has achieved the double-edged position all Grünfeld players are aiming for. 9. ♘e2 c5 This position is covered very well in Avrukh's book. I'll include the main ideas that could occur. 10. O-O

10. d5 isn't very threatening as it runs into 10... e6! , immediately undermining the white centre. 11. OO ( 11. ♗e3 is more than adequately met by 11... ♕a5 ) 11... exd5 12. exd5 ♘d7 The potential knight jumps are very difficult for White to completely control. Play might continue: 13. ♗e3 ♕a5 14. ♗b3 ♘e5 15. c4 Otherwise Black will play this himself. 15... ♗f5 16. ♘f4 Trying to control the d3-square. 16...b5 A very harmonious way of playing immediately on the queenside. (16... ♘g4 is Avrukh's suggestion and I'm not criticising it. ) 17. cxb5♕xb5 Black is fine. He wants to play ...c4 and ...Nd3 and has a lot of play in the centre. ( 17... c4⁉ is also ok, though I prefer 17...Qxb5.) 10... ♘c6 11. ♗e3 11. d5 ♘a5 Once again this is ok for Black, although after 12. ♗g5 you have to be quite energetic and immediately start with 12... f5⁉ , as Boris Avrukh correctly points out in his book. 13. ♗c2 ♘c4 Black has good counterplay. 11... ♘a5 This move is a recurring theme in this line as the bishop on a4 fails to control the c4-square, which can therefore constantly be used as a base for Black's operations. 12. ♖b1 Also a well-known move. 12. d5 b6 gives Black fantastic counterplay because once again after 13.♕d3 Black will immediately start hitting the white centre with 13... e6The fact that the knight on a5 is unopposed and is constantly looking at the c4-square makes it very difficult for White to retain his centre safely. ( 13... ♗d7 ) 14. d6 ♗d7 15. ♗c2 ♗c6 16. f4 ♕d7! was very good for Black in Dyballa-Petzold, Germany 2009. 12. dxc5 This used to be the main move in this position. 12... ♕c7 This is the better choice. Black doesn't try to win the c5-pawn back immediately but instead prepares to play for compensation. ( 12... ♘c4 This has been played by, among others, Garry Kasparov. 13. ♕xd8 ♖xd8 14. ♗g5♗d7 15. ♗b3 ♘a5 This endgame used to be considered completely equal but then Avrukh showed that White can improve on the old Kasparov games and that this position is far from an immediate draw.)13. ♗b3 was played in a recent game between Cheparinov and Caruana. (13. ♘f4 The knight shouldn't, of course, be allowed to reach d5, so Black plays 13... e6 , and after, for instance, 14. ♘d3 , as in Shariazdanov-Dvoirys, Black replies 14... b6 aiming for typical Grünfeld play against the ruined structure on the queenside. Black is fine.) 13... ♗g4 14. f3♗d7 This position is very similar to the ...Bd7 line in the classical Bc4 Grünfeld. You'll recognise this kind of structure immediately if you've seen my videos on that. Black aims to slowly coordinate his pieces and take on c5, which is what happened in the game: 15. ♕b1 e6 16. ♖d1 ♖fd8 17.♔h1 ♗e8 18. ♕b2 ♗f8 19. f4 Black is already completely prepared to take on c5 so White needs to mount some kind of an offensive on the kingside. Otherwise he'll run into difficulties. 19... ♗xc5 20. ♘d4 ♗f8Cheparinov-Caruana, Reykjavik op 2012 12... ♘c4 13. ♗g5 ♕c7 A few options have been tried by White here. Once again, you'll see a recurring motif: if the knight on c4 is attacked it has a very comfortable position on a3. 14. ♕c1 This is following one of Boris Avrukh's own games with White. 14. ♗b3 ♘a3 15. ♖c1 c4 16. ♗c2 b5 Black doesn't even need to hurry after the possible 17. ♕d2 ♗b7 18. f3 e5 , settling the situation in the centre. Black won't be taking on c2 any time soon because he wants to play ...a5 and ...b4 and the knight on a3 is very useful to stop either of the white rooks from coming to b1. It's a very interesting and unclear position.

14. ♕d3 has been tried in a number of games and is once again met by14... ♘a3 15. ♖b2 ♕a5 16. ♗d1 Here Boris correctly shows that Black doesn't need to waste any time defending e7 because he has the very strong move 16... b6! , aiming to play ...Ba6, when Black gets a lot of counterplay against the somewhat misplaced white pieces. 14... b6 15. ♗b5 ...Ba6 had been a serious positional threat. 15... ♘d6 16. ♗d3 Avrukh's opponent Mark Tseitlin here played 16... f6 17. ♗h4 e5 A lot of readers won't know the name but Tseitlin is a player who was very, very famous in the old Soviet times and had almost superhuman status. There were fantastic stories about him in the St. Petersburg chess circles but I don't think he was very well-known in the West. Since emigrating to Israel he's taught the Grünfeld to a lot of young people including Boris Avrukh himself and been instrumental in creating a new wave of Grünfeld players. Almost every single young Israeli player of the past 15 years has played the Grünfeld at one time or another and most of them looked at the opening with the help of Mark Tseitlin. In this game Boris decided he would occupy the d5-square with the knight and played 18. dxe5 18. ♗g3 is what Boris should have played instead, but after the possible18... ♕e7 Black is quite comfortable. This is the kind of very unclear and double-edged position that I think black players should welcome with open arms. 18... fxe5 19. c4 , but it turned out that after 19... ♗e6 20. ♘c3 ♕f7 21. ♘d5 ♗xd5 22. cxd5 ♗f6 23. ♗g3 ♕e7!the knight on d6 more than adequately blockades all of White's potential advances in the centre. The black passed pawns on the queenside are a more important factor and Boris eventually lost this game.

7.Nf3 c5 8.Bb5+ Nc6 9.0-0 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6. bxc3♗g7 7. ♘f3 c5 8. ♗b5+ This was played by Vishy Anand in the 2012 World Championship match, which is a stamp of quality, but Black is fine if he knows what he's doing. 8... ♘c6 The most critical and correct move. 9. O-O This is the old theory and is completely safe for Black. 9... cxd4 The main reply. 9... O-O can also be played first. 10. ♗e3 White's most likely response. (10. d5 can be met by 10... ♗xc3 as played recently by Wesley So. 11.♗h6 ♘d4 12. ♘xd4 ♗xd4 13. ♗xf8 ♕xf8 The monster on d4 obviously gives more than adequate compensation for the sacrificed exchange. Black was doing very well. 14. ♖b1 a6 15. ♗e2 b5 16. ♔h1 e5Gerzhoy-So, Montreal COQ Inv. 2012; 10. ♗a3 cxd4 11. ♗xc6 bxc6 12.cxd4 ♗g4 13. ♗c5 ♕a5 was also fine in Zhu Chen-Erdos, Gibraltar Masters 2012 ) a) 10... ♗g4 is an additional option here and those who have played it include Alexander Khalifman. Black attempts to avoid the main line which leads to an almost forced draw.

a1) The point of postponing the trade on d4 is clear from 11. ♗xc6bxc6 12. ♖c1 ♕a5 13. ♕e2 ♖fd8 14. ♖fd1 ♗xf3 15. gxf3cxd4 16. cxd4 ♗xd4!=/+ , and Black was already better in M. Ivanov-Khalifman, Minsk op 2012 a2) 11. dxc5 ♕c7⁉ ( 11... ♗xc3 12. ♖c1 ♗g7 13. ♕a4 ♕c7 14.♘d4 ♘xd4 15. ♗xd4 ♗xd4 16. ♕xd4 ♖fd8 17. ♕e3 ♗e6=Ikonnikov-Roiz, Gibraltar Masters 2011 ) 12. ♗xc6 ♕xc6 13. ♕c2♗xf3 14. gxf3 Xu Jun-Wang Yue, Ningbo CHN-tch 2011 14... b6= a3) 11. d5! I think this is the most critical move. 11... ♘e5 12.♗e2 , and White perhaps retains some pressure. 12... ♘xf3+ 13.♗xf3 ♗xf3 14. ♕xf3 ♗xc3 15. ♖ab1 b6 16. ♗xc5 b) 10... cxd4 This transposes back to the main line. 10. cxd4 O-O 11. ♗e3 ♗g4 12. ♗xc6 bxc6 13. ♖c1 ♕a5 13... ♗xf3⁈ 14. ♕xf3 ♗xd4 15. ♖fd1 e5 16. ♖xc6 is a stable and risk-free plus for White. 14. ♖xc6 14. h3 ♗xf3 15. ♕xf3 ♕xa2 ( 15... ♖fd8 ) 16. ♖xc6 a5 14. ♕d2 is more or less a forced draw, as has been known for many years. Black will play 14... ♕xd2 and after 15. ♘xd2 ♖fd8 16. ♘b3 a5all the pieces get traded off and a draw is imminent. 14... ♕xa2 15. ♖c7 a5 It's also been known for many years that the very fast-running a-pawn far outweighs the pawn Black has to sacrifice in the centre. Black is just fine here. 16. ♖xe7 a4 17. ♖b7 a3 18. ♖b1 ♖fb8 19. ♖a1 ♕b2 20. h3 20... ♗e6!→ Even stronger than 20... ♗xf3 21. ♕xf3 ♗xd4 1/2 Lugovoi-Svidler, St Petersburg 1995

7.Nf3 c5 8.Bb5+ Nc6 9.d5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6. bxc3♗g7 7. ♘f3 c5 8. ♗b5+ ♘c6 9. d5 The only critical move for White. 9... ♕a5 10. ♖b1 ♗xc3+ Precise. 10... a6 was played by Boris Gelfand against Vishy Anand in their match, but after 11. ♗xc6+ bxc6 12. OO a) 12... O-O 13. dxc6 ♗e6 ( 13... ♖d8 14. ♕b3 ♗e6 15. ♕b7 ) 14.♗f4+/= b) ♕xa2 b1) 13. ♖b2 was Vishy's choice. 13... ♕a5 14. d6 ♖a7!

b11) 15. ♗g5⁈ exd6 16. ♕xd6 ♖d7 17. ♕xc6 ( 17. ♕b8 OO) 17... ♕c7 18. ♕xc7 ♖xc7 19. ♗f4 ( 19. e5 O-O ) 19... ♖b7, and Black was even slightly better in Anand-Gelfand, Moscow WCh match 2012 b12) 15. ♗f4! 15... ♖d7 16. ♖b8 O-O 17. dxe7 ♖xe7 18.♗d6 ♖d7 19. e5=/∞ b2) 13. ♗g5⁉ would have been a much more challenging move, based on a nifty tactical trick. 13... cxd5 Forced. ( 13... h6 is the most obvious move in the position, but after 14. dxc6 hxg5 Black finds he's fallen into a beautiful trap: 15. ♖b8‼ After 15... ♖xb816. c7+− the pawn is suddenly queening.) 14. ♖a1 ♕c4 15. ♖a4♕xc3 16. ♕xd5 Although I wouldn't go as far as to say Black is lost he's in a lot of danger and the position isn't really ideal from his perspective. 11. ♗d2 a6 Only now, and I think Black makes a pretty much forced draw. 12. ♗xc6+ The main move. 12. ♗a4 is met by 12... b5 and Black is fine. 13. dxc6 bxa4 ( 13...♗xd2+ 14. ♕xd2 ♕xa4 15. O-O ♕xe4 , as in Martic-Brkic, Bol op 2011, is also playable, but the text move is more forcing.) a) 14. O-O would be met by 14... ♗xd2 15. ♘xd2 ♗e6! , stopping Nc4. Black has absolutely no problems. a1) 16. ♕c2 ♖c8 ( 16... O-O 17. ♘c4 ♗xc4 18. ♕xc4 ♖ac8 ) 17.f4 c4 a2) 16. ♖b7 16... OO 17. ♘f3 ♗c4 18. ♖e1 ♖ac8 19. ♕d2♕xd2 20. ♘xd2 ♗xa2 21. ♖a1 ♗e6 22. c7= b) 14. ♕c1 12. ♗d3 ♘d4 13. ♘xd4 cxd4 14. ♕c2 ( 14. ♗xc3 dxc3 15. O-O♕xa2∞ ) 14... e6= 12... bxc6 13. dxc6 Black has to show some precision, but his position is fine. 13. O-O first allows 13... ♗xd2 14. ♘xd2 when the easiest way to proceed is 14... cxd5 ( 14... O-O is also possible and not particularly bad for Black.) 15. ♘b3 ♕d8! Stopping Qxd5, and after e.g. 16. ♘xc5OO Black is doing very well. 17. exd5 ♗f5= 13... ♗e6! It's important to postpone castling and, generally speaking, to wait for White to castle first. 13... O-O 14. O-O ♗xd2 15. ♕xd2 ♕xd2 16. ♘xd2 ♗e6 17. ♖fc1+/= 13... ♗xd2+ 14. ♕xd2 ♕a4 15. O-O ♕xc6 16. ♕h6± 14. ♖c1 14. O-O will be met by 14... ♗xd2 15. ♕xd2 ♕xd2 16. ♘xd2 O-O-O= 14. ♗xc3 ♕xc3+ 15. ♕d2 There are many ways to equalise here for Black but the most forcing one is perhaps 15... ♕c4 , attacking the pawn on e4. 16. ♘g5 f6 17. ♖c1 ♕xa2= White has nothing better than 18.♖c2 ♕b1+ 19. ♖c1 ♕a2=

14... ♗xd2+ 15. ♕xd2 It's time for one final finesse: 15... ♕b4! Black hits the pawn on e4 and if White swaps off queens on b4 the weak pawn on c5 turns into a potential passer on b4, which is obviously not a problem for Black. 15... ♕xd2+ doesn't equalise completely: 16. ♔xd2 O-O-O 17. ♔e3 c418. ♘d4+/= 15... ♕a4 16. O-O ♕xc6 17. ♕h6→ 16. O-O ♕xd2 Only here. 17. ♘xd2 O-O-O! Black castles long, not short. He'll collect the pawn on c6 in a couple of moves and the game will be completely drawn. That was already seen in Sasikiran-Grischuk, Eilat 17th ECC 2012 17... ♗xa2 18. ♖xc5 O-O-O 19. ♖c2 ♗e6 20. ♖b1+/=

7.Nf3 c5 8.h3

1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6. bxc3♗g7 7. ♘f3 c5 8. h3 As promised I'll briefly mention this option for White. It's been played by Kramnik once or twice but has since gone completely out of fashion. 8... O-O Black is really perfectly fine in every line here because White has no way of justifying the tempo he's wasted on 8.h3. 9. ♗e2 Kramnik's move and the only sensible option in this position. 9. ♗e3 can be met by 9... ♕a5 10. ♕d2 ♘c6 and you get a fantastic version of the Be3 line. 11. ♖d1 Forced, and now the most comfortable way for Black to proceed is just to play 11... b6 , then wait for White to castle and switch to an endgame where Black's pieces are perfectly placed and there's a lot of counterplay on the queenside. 9... cxd4 10. cxd4 ♘c6 11. ♗e3 f5! You just need to know you can play this move. 12. exf5 12. ♗c4+ ♔h8 13. e5 was seen in one of Artem Timofeev's games. Black can continue: 13... f4 14. ♗c1 b5 Immediately starting to gain space on the queenside. 15. ♗b3 ♘a5 16. OO ♗b7 Black is doing very well as he has such complete control of the light squares. 12... ♕a5+ 13. ♗d2 ♕xf5 Your next two moves here as Black will be ...Be6 and ...Bd5, and once the bishop lands on d5 it will be completely obvious that only White can be worse. I think the line with 8.h3 is completely closed by now

5.h4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. h4⁉ A little bonus track. This somewhat bizarre move has already been played at a very, very high level - a couple of times by Alexander Morozevich and also famously by Grischuk against Magnus Carlsen in the London Candidates. It's actually a very serious move, and if Black doesn't play 5...c6 here but more "Romantic" moves allowing White freedom of play on the kingside Black could potentially be in a lot of danger. I'm not recommending any of those at all even though they don't, of course, really lose. 5... c6 This was Carlsen's surprising choice. 5... dxc4 6. e4 a) 6... c5 7. d5 b5 8. h5 OO ( 8... b4 9. h6 ♗f8 10. e5 bxc3 11.exf6 exf6 12. ♗xc4 cxb2 13. ♗xb 2 ♗d6 14. ♕e2+ ♕e7 15.♕xe7+ ♔xe7 16. O-O→ Fridman-Jens, Netherlands 2013) 9. hxg6fxg6 ( 9... hxg6 10. ♗h6 transposes to Mamedyarov-Kamsky, Tromso World Cup 2013.) 10. e5 ♘g4 11. d6 e6! ( 11... ♗b7 12. ♗g5 ♘c613. dxe7 ♕xd1+ 14. ♖xd1 ♖f5 15. ♘xb5! ♘gxe5 16. ♘d6 ♘xf3+17. gxf3 ♖e5+ 18. ♗e3 ♗a6 19. e8=Q+ ♖axe8 20. ♘xe8 ♖xe8 21.♖d6+− Gareev-Robson, Saint Louis 2013.) 12. ♖xh7 a1) 12... ♔xh7 13. ♘g5+ ♔g8 14. ♕xg4 ♖f5 15. ♗e3 ♘c6 16.♕h4 ♘xe5 17. OOO ♗d7 18. ♘ce4 ♘f7! ( 18... ♕a5? 19.♕h7+ ♔f8 20. ♘xc5 ♖xg5 21. ♗xg5 ♕xa2 22. ♖d4!+− Vitiugov-Ragger, Tromso World Cup 2013) a2) ♘c6! 13. ♖xg7+ ♔xg7 14. ♗g5 ♕d7 ( 14... ♕b6⁉ 15. d7♗b7 16. ♕d2∞ ) 15. ♘e4 ♘cxe5 16. ♘xe5 ♘xe5 17. ♘xc5♕c6 18. ♕d4 ♖f5 19. ♗e3 ♔g8 20. OOO ♗d7 21. g4 ( 21.♔b1 ♖c8 Morozevich-Ponomariov, Riga ACP 2013) 21... ♘f3 22.♗g2!± b) OO 7. h5 c5 8. hxg6 hxg6 9. d5 b5 10. ♗h6 ♗xh6 11. ♖xh6b4 12. ♘a4 ♔g7 ( 12... ♘xe4⁉ ; 12... ♕a5⁉ ) 13. ♖h4+/=Mamedyarov-Kamsky, Tromso World Cup 2013 5... h6⁉ 6. ♗f4 dxc4 7. e4 c5 8. d5 b5 9. e5 ♘h5 10. ♗h2 ♗g4 11.♗e2 ♘d7 12. ♘d2 ♗xe2 13. ♕xe2 b4 14. ♘a4 c3 15. ♘c4 O-O 16.O-O cxb2 17. ♘axb2 f5 Matlakov-Mamedyarov, Tromso World Cup 2013 6. cxd5 The most critical move, which was played by Morozevich against Giri in the Sport Accord tournament in Beijing. 6. ♗g5 a) 6... O-O 7. ♗xf6 ♗xf6 8. h5 ♗f5 9. e3+/= I think Carlsen was in a lot of trouble here. At one point, according to the post-match press conference, he was going for a line that was losing by force. That gives you an idea of just how dangerous his position was. 9... e6 10. hxg6 (10. ♘e5⁉ ) 10... ♗xg6 11. ♗d3 ♘d7 12. ♕e2 e5 13. ♗xg6 hxg614. dxe5 ♘xe5 15. cxd5 cxd5 16. ♖d1+/= Grischuk-Carlsen, London 2013 b) dxc4⁉ A very sensible option for Black as it's very difficult for White to justify the sacrificed pawn. It's not going to be easy to win it back and Black is ready for the potential attacks on the kingside - after all, the king isn't even on g8 yet. I'd be very interested to know what Grischuk's plan was against this move as in general I think Black is just fine. b1) 7. h5 ♘xh5 8. e4 b5∓

b2) 7. e3 even gives Black a choice. 7... ♗e6 ( 7... b5 ) 8. ♘d2♘d5 Creating the important threat of ...h6. 9. ♗xc4 ( 9. h5 h610. ♗h4 g5 11. ♗g3 OO=/+ ) 9... ♘xc3 10. bxc3 ♗xc4 11.♘xc4 ♕d5=/+ Once again Black emerges with an extra pawn and obviously shouldn't be doing that badly. b3) 7. a4 O-O 8. e4 ♗e6 9. h5 ♘xh5∓ b4) 7. ♗xf6

b41) 7... ♗xf6 8. e4 b5 9. h5 ( 9. a4 b4 10. ♘a2 ♗e6=/+ )9... ♗g4 10. e5 ♗g7 11. h6 ♗f8∞ b42) exf6⁉ 8. e3 ♗e6 9. ♘d2 b5∞ b5) 7. e4 7... b5 8. e5 ( 8. ♗e2 ♗g4 ) 8... ♘d5 This is clearly a pretty decent version of the Slav Gambit. 9. h5 ( 9. ♘e4 f6!∓ )9... h6 10. ♗d2 g5 11. a4 ♘xc3 12. bxc3 ♗e6=/+ 6... cxd5 7. ♗f4 ♘c6 8. e3 Once again Morozevich is aiming for the positions he made very popular in the Exchange Slav. Those used to be considered a complete draw but now, mainly due to his original efforts, everybody plays them and Black has to be very careful. 8... O-O 9. ♘e5 9. ♗e2 ♗g4 10. O-O e6 is pretty much completely equal: 11. ♖c1♖c8 12. ♕b3 ♕e7 13. ♘d2 ♗xe2 14. ♘xe2 ♖fd8= Melkumyan-Erdos, Berlin 2013 9... ♘xe5 This proposed improvement might look a bit cynical to some of you, but otherwise I think you might not equalise. 9... ♕b6 This was Anish Giri's somewhat imprecise move against Morozevich. 10. ♕b3! ♕xb3 11. axb3 ♘b4 ( 11... ♗f5 12. ♘xc6 bxc613. b4+/= ) 12. ♔d2 h5 13. ♘d3 ♘c 6 14. f3+/= Morozevich-Giri, Beijing 2012 9... ♗f5⁉ 10. ♗xe5 10. dxe5 ♘e4 11. ♘xe4 dxe4 12. ♗c4= 10... ♘e4 Swapping off as many pieces as possible. 11. ♗xg7 ♘xc3 12. bxc3 ♔xg7 13. h5 If White allows Black to play this himself then I simply don't see why Black should be worse. 13. ♗d3 h5 14. ♕b3= 13... ♗f5 13... ♕a5 14. ♕d2 ♗f5 15. c4+/=

14. ♗e2 ♕d6! Black is aiming to play ...Rac8, ...Rc7 and ...Rfc8. In some positions he'd also be happy to play ...e7-e5 if an opportunity arises to do so without material loss. 14... ♖c8 15. g4 ♗e6 16. ♖c1 ( 16. ♕d2 ♕d7 17. f3 ♕c7 ) 15. g4 is how the game might continue, and after 15... ♗d7 16. f4 ♖ac8 17. ♖c1 ♖c7 it looks as though White has made some advances on the kingside and will mount an impressive attack any day now. As a matter of fact, though, it's very difficult to swing any pieces over to the h-file apart from the one rook which is already there. The pawn on c3 is a constant nagging worry which will require a lot of attention. The queen will normally be stuck somewhere on d2, from which point it will be very difficult to transfer it to anywhere attacking. So I think in general Black should be fine here and we've covered all you need to know about the two Bb5+ variations and 5.h4. The latter is a very interesting and trendy recent addition to White's repertoire. I expect we'll see more of it in future but the somewhat pragmatic approach - going for symmetrical positions - should be safe enough and enable Black to equalise reasonably comfortably.

This will be a slight detour from the normal fare of my Grünfeld series. Most of the videos I'm making will be purely theoretical and very theory heavy, or at least I hope they will be, but here I want to show you a couple of my junior games. First of all, because they're very memorable, but secondly because I think they show at least glimpses of why I like this opening so much and why I stuck with it despite on occasions losing a lot of painful games and in general suffering at the hands of strong players. The games were played in the early 1990s in the World Juniors, one in 92 and the other in 94. Some theory will be mixed in but I mainly just want to show you the state of play in the early 90s and try and instil in you some of the fondness for the opening that I already had in those years. Ioannis Papaioannou 0:1 Peter Svidler World Juniors U16Duisburg, Germany, 1992 So let's begin with the game I played in 92. It was the last round of the World U16 Juniors, and there was a 3-way tie for first place between myself, Daniel Fridman, who now represents Germany, and Ronen Har-Zvi, an Israeli who many of you will know as a commentator on ICC. It was clear I needed to win because they were likely to do the same. In the final analysis even a win didn't prove enough for the title, but I'm still very happy with the way I played. I was paired with the Greek chess player Ioannis Papaioannou, who also became a strong grandmaster. Obviously I was Black: 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗g5 ♗g7 This was my signature move in those years. I can't lay claim to having been the first one ever to play it, but judging by the way they would proceed most people before me played it with the idea of blundering a pawn and losing, whereas I wanted to get compensation and a double-edged position I could play for a win perhaps, eventually. People would just take the pawn, and I think they still should, but after 5. ♗xf6 ♗xf6 6. cxd5 c6 the more sophisticated lines starting with e.g. 7.Rc1 or 7.e4 were completely unknown back then. 7. dxc6 People played this automatically and only then started to think, reasoning that if you didn't take the pawn now it might run away. 7... ♗xd4 8. cxb7 ♗xb7 This position is frankly a luxury you can't get for love or money nowadays, but in those years I think my first three or four games in the Bg7 variation got to here, with good results. 9. ♕b3 Ioannis chose this move. 9. e3 was played against me in a later junior game in a tournament in St. Petersburg in 1993 by Valery Popov, who also became a strong grandmaster. 9... ♗xc3+ 10. bxc3 ♕a5 11. ♕c2 Somewhat more circumspect. After ( 11. ♕d4 O-O 12. ♕b4 Black can just play 12...♕c7 and the knight is about to enter the fray, let's say via a6 and c5. White has only developed a single piece, and even that's a source of constant worry, so he's clearly not better here.) 11... O-O 12. ♘e2 It's important to see that after ( 12. ♖b1? Black doesn't need to play the bishop anywhere yet, because 12... ♖c8! is very strong, waiting for 13.♘e2 and only then, perhaps, 13... ♗d5=/+ ) 12... ♘d7 13. ♖b1 ( 13.♘d4 ♖ac8 14. ♖c1 ♘f6→ ) 13... ♘c5 14. f3 ♖ab8! A move I'm quite proud of. It's purely prophylactic, waiting for the knight to determine where it's going. ( 14... ♖fd8 15. ♘d4 e5 16. ♘b3∞ ; 14... ♗d5 is a move Black clearly wants to make, but if he starts with it White will reply 15. ♘f4 and now the threat of Rb5 means Black needs to start reworking his plans somewhat. I wasn't keen on that.) 15. ♘d4 ( 15.♔f2 ♗d5 ) 15... ♗d5 Only now, and at the very least Black will win back the a-pawn with quite a bit of pressure. I eventually won this game. 9... ♕c7 We'd reached a position I already had on the board back in 91 against Colin Crouch at the Lloyds Bank Masters in London.

10. e3 ♗g7 11. ♗b5+ Frankly sub-optimal, as there's no reason for White to waste a tempo on this. I think Ioannis' play was influenced by the fact he was understandably nervous, just as I was. It was the last round of the World Juniors, which was my first and probably his first as well. 11. ♘f3 This is much better than 11.Bb5+, and after 11... OO 12. ♗e2♘d7 13. OO ♖ab8 14. ♕a3 ♘e5 Black has wonderful compensation for the pawn. 15. ♖ac1 ( 15. ♘d4 is another try for White, but Black can reply with the very strong 15... ♗a8 creating the idea of Qb7, and after the possible 16. ♖fc1 ♕b7 17. ♗f1 ♖fc8=/∞ Black has fantastic compensation for the pawn and at the very least isn't worse.) 15... ♗xf316. gxf3 ♕b7 and the game soon ended in a perpetual on the kingside:17. f4 ♘f3+ 18. ♗xf3 ♕xf3 19. ♕xe7 ♖xb2 20. ♘e4 ♖xa2 21. ♘f6+♔h8 22. ♘d7 ♕g4+ Crouch-Svidler, London 1991 11... ♘c6 12. ♘f3 O-O 13. O-O ♘e5 This is clearly a much improved version of the game against Crouch because the bishop on b5 is completely misplaced and will probably have to be drawn back to e2 at the cost of a tempo. 14. ♘xe5 ♕xe5 15. e4 A move that can only be explained by his desire to offer me something which might tempt me to equalise immediately, but I wasn't really interested. I wanted to keep the pressure on. 15. ♗e2 is probably the best move here, but after 15... ♖ab8 16. ♕a3a5 Black certainly isn't worse, although I think White should hold. 17.♖fd1 ♗c6=/∞ 15... ♖ab8 This is what I went for. 16. ♕a4 ♕c5 Black doesn't need to try and win back material immediately as by now it's clear that some material will be won back at some point. 17. ♗e2 ♗a8 18. ♖ab1 ♖fd8 Black just develops naturally and it's very difficult for White to withstand the pressure, especially if in his desire to simplify and make his life at least a little bit easier he does what my opponent did. 19. ♕c4 ♕a5 20. ♕a6 ♕xa6 21. ♗xa6 Black is finally ready to start winning back some material, and soon I was completely winning. The game didn't win me the title - Ronen became World Champion on tiebreaks - but it reinforced my belief that the Grünfeld is a very good fighting opening which gives your opponents a lot of scope to go wrong. It generally leads to unbalanced and interesting positions where you can express yourself. 21... ♖d2 22. ♖fc1 ♗xc3 23. bxc3 ♖xb1 24. ♖xb1 ♗xe4 25. ♖e1♗d5 26. ♗f1 ♗e6 27. ♖c1 ♔f8 28. a4 ♖a2 29. ♗b5 ♗b3 30. h3♗xa4 31. ♗f1 ♗b3 32. c4 a5 33. ♗d3 ♖d2 34. ♗f1 a4 35. ♖a1♖d1 36. ♖xd1 ♗ xd1 37. c5 ♔e8 0-1

Giorgi Kacheishvili 0:1 Peter Svidler World Juniors U18Szeged, Hungary, 1994 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. ♗d2 In 1994 in round 7 of the U18s I played the then Georgian and now American chess player Giorgi Kacheishvili. The sideline he employed here is very important today, and a lot of very strong players have played it. Dreev has been its major proponent over the course of the last decade and a half. Vishy Anand also won a very nice game against me, so this is a very serious subline of the Grünfeld. I'm not even going to attempt to cover the theory here, but I think the game I played against Giorgi is very instructive and also covers one subline quite sufficiently. I also want to point out that although in those years there was

obviously some theory on this line I knew none of it. My reaction was, "ok, this is some move, I have no idea what I'm supposed to do here, so let's just make natural moves. This can't be the refutation of my favourite opening, so let's just play chess." In many cases over the course of my career that kind of approach has brought me better results than the times when I've sat at the board tortuously trying to remember what my notebooks and then later files said about something. An uncluttered mind is sometimes a very powerful thing. 5... ♗g7 6. e4 ♘b6 7. ♗e3 O-O The main way of replying to 5.Bd2, and in this position White has a variety of good moves. Perhaps the most critical and most popular here is the somewhat counter-intuitive 8.Bb5. 8.h3 is another very important alternative which Vishy used to great effect in the game he won against me at the Tal Memorial. Of course there's also 8.Nf3, or the move played in the game. 8. ♗e2 Black once again has more than one idea here, but I believe what I did effectively puts an end to this subline (though it doesn't stop White returning to the lines with Nf3). 8... ♘c6 9. d5 9. ♘f3 is probably the best move in this position, transposing to the normal Nf3 theory. 9... ♗g4 10. d5 ♗xf3 11. gxf3 ♘a5 ( 11... ♘e5 It's important to know that this move is imprecise here, since 12. ♕b3actually gives White a decent shot at an opening advantage.) 12. ♗d4♕d6! Theory shows that Black has absolutely nothing to fear and is fine in all lines here. 13. ♗xg7 ♔xg7 a) 14. ♖c1 ♕f6! ( 14... c6 is much less precise, as after 15. ♕d2! the knight on a5 is in trouble.) a1) 15. ♘b1⁈ ♕xb2 16. ♖c2 ♕b4+ 17. ♘d2 f6⁉=/+ ( 17... c618. ♕a1+ ♔g8 19. a3 ♕a4 20. ♗d1 ♕b5 21. ♗e2 ♕a4 22.♗d1 ♕b5 1/2 Ivanchuk-Brkic, Kusadasi 2006) a2) 15. ♘a4 15... ♘xa4 16. ♕xa4 ♕xb2 17. ♖xc7 ♖fc8 18.♕xa5 b6 19. ♖xc8 bxa5 20. ♖xa8 ♕b1+ 21. ♗d1 ♕b4+ 22.♔f1 ♕b7 23. ♖d8 ♕b6 24. ♖a8 ♕b7= Sieciechowicz-Matlakov, Chotowa 2010 b) 14. ♕d2 14... ♕f6 15. b3 ( 15. f4 ♘ac4 16. ♗xc4 ♘xc4 17.♕e2 ♘b6 18. ♕e3 ♘c4= I.Sokolov-Shirov, Las Vegas FIDE WCh 1999)15... ♖fd8! /\ Nc6( 15... c6 16. ♖d1 cxd5 17. exd5 ♖ad8 18. ♘e4♖xd5 19. ♘xf6 ♖xd2 20. ♘h5+!± Zhao XueKosintseva, Sochi 2009)16. ♖c1 e6 17. ♘b5 ♘c6 18. ♘xc7 ♖ac8 19. ♘e8+ ♖xe8 20.dxc6 ♖xc6 21. ♖xc6 bxc6 Arlandi-Konopka, Baden 1999 9... ♘e5 10. ♗d4 This looks very sensible because White is aiming to play f4 next move, trade off the darksquared bishops and then develop naturally. His pieces will be more harmonious because the knight on b6 is somewhat misplaced and White is dominant in the centre. This is actually a very principled way of dealing with Black's setup, but unfortunately it runs into quite serious turbulence. 10... c5 I'm very proud to say I was the first to play this move, which I believe is now the definitive response to this line. 11. ♗xc5 11. ♗xe5 is also possible, and you could argue it's no worse than 11.Bxc5, but giving up the dark-squared bishop like that is no-one's idea of fun. 11... ♘ec4 It transpires that defending the b2-pawn won't be at all easy.

11... ♘bc4 12. ♕b3 b6 13. ♗d4± 12. ♗xc4 This move looks very strong at first glance. 12. ♗xb6 ♕xb6 13. ♗xc4 ♕xb2∓

12. ♕b3 is what White should perhaps play, and after 12... ♘xb2 13.♖c1 White at least attempts to hold his queenside together without giving up either of his bishops. ( 13. ♕xb2 ♘a4−+ ) a) 13... ♗xc3+ The move I suggested in my notes to this game for the Informant. 14. ♕xc3 ♘2a4 I went on to say that Black is slightly better here, which goes to show that in 94 my understanding of the static structures in the Grünfeld was somewhat deficient, because after 15. ♕d4 Black is definitely not better and there have even been some games starting from this position which White won convincingly. I'm not saying White has a definite advantage here, but it's very playable and would pose him no problems. 15... ♘xc5 16. ♖xc5 ♗d717. ♘f3 ♘a4 18. ♖c1 ( 18. ♖c4⁉ b5 19. ♖c2∞ Palo-Jerez Perez, Andorra 2004) 18... ♕a5+ 19. ♕d2 ♕xd2+ 20. ♔xd2 ♖fc8 21. ♘e5♗e8 22. ♗g4 ♖d8 23. ♔e3+/= HuebnerSolozhenki, FIN-tch 2012 b) 13... ♗d7! Ignoring the fact that the knight is hanging on b2. 14.♘f3 1/2 Gleizerov-Ganguly, Abu Dhabi 2005 ( 14. ♕xb2 The point is that after the forcing sequence 14... ♘a4 15. ♕a3 ♘xc5 16. ♕xc5♖c8 17. ♕a3 ♕c7 the knight on c3 is lost. Black is once again very much ahead.) 14... ♖c8 15. ♗d4 ♗xd4 16. ♘xd4 ♘2c4⇆

12... ♘xc4 13. ♕b3

13. ♕e2 ♘xb2 transposes to 13.Qb3. 13... ♘xb2! The point of my entire play. 14. ♕xb2 ♕c7 It transpires that White is unable to hang on to the material. The piece will be won back with a lot of compensation. 15. ♗b4 The correct move. Trying to hold on to more material with 15. ♕a3 b6 16. ♗b4 a5 17. ♖c1 runs into 17... axb4! 18. ♕xa8 and taking on c3 with pretty much anything. Black will have an advantage because White is so undeveloped and uncoordinated. 15... a5 16. ♘ge2 axb4 17. ♕xb4 And in this position "the past me" made a very good move: 17... ♗g4 17... ♕d6⁉ 18. ♕xd6 exd6 19. O-O f5 20. ♖ae1 ♗d7 21. ♖b1 fxe422. ♘xe4 ♖xa2= ChernuschevichJansa, France 2003 18. f3 Pretty much forced. 18... ♗d7 The point is simply to weaken an additional diagonal and create more potential entry points into White's position. 19. O-O b5=/∞ I don't think Black is better here - at least not yet - but it's definitely much easier and more enjoyable for him to play. The game bore out that assessment.

20. ♔h1? A very natural and human move by Giorgi, trying to get the king away from potential checks. 20. ♕xe7 ♖fe8 21. ♕b4 ♗f8 22. d6 ( 22. ♕b3 b4 23. ♘d1 ♖a3 24.♕b1 ♗b5−+ ) 22... ♗xd6 23. ♕b3 ♗xh2+ 24. ♔h1 ♗e5=/+ 20. ♖ab1! a) 20... ♗e5⁉ 21. ♔h1 ♖fb8 22. ♘d4 ( 22. ♕xe7 ♗d6 23. ♕h4 b424. e5 bxc3 1/2 DoettlingJaracz, Bad Wiessee 2000 25. exd6 ♖xb126. ♖xb1 c2 27. ♖c1 ♕xd6 28. ♖xc2 ♕xd5 29. ♕d4 ♗e6 30.♕xd5 ♗xd5 31. ♘c1 h5= ) 22... ♕a5 23. ♕xa5 ♖xa5 24. ♘ce2♖xa2 25. f4 ♗xd4 26. ♘xd4 ♖a4 27. ♖fd1 ♖c4= b) ♖fc8 21. ♘xb5 ♗xb5 22. ♕xb5 ♖xa2=/∞ 20... ♕d6=/+ Once again nice play by the young me, because I'm sure in those years it was difficult for me to trade queens in a position where I'm still a pawn down and could perhaps be tempted by some more Romantic continuation. This is the correct approach, however, and the endgame is very, very good for Black. 21. ♕xd6 21. ♖ab1 ♕xb4 22. ♖xb4 ♖fc8 23. ♘xb5 ♗xb5 24. ♖xb5 ♖xa2 25.♘g1 ♗h6! 26. f4 ♖cc2→ 21... exd6 The following sequence is now almost forced. 22. ♖ab1 ♖fb8 23. ♖b4 You can't allow b4. 23... ♖a3 24.♖c1 24. ♖b3 b4 25. ♖xa3 ( 25. ♖fb1 ♖xb3 26. ♖xb3 ♖c8∓ ) 25... bxa326. ♖b1 ♖b2−+

24... ♖c8 Shifting the rook from b4 and opening the a-file. 25. ♖b3 25. ♘xb5 ♖xc1+ 26. ♘xc1 ♖e3!−+ 25... ♖xb3 26. axb3 ♖a8! 27. ♖b1

27. b4!= This is the move I think White should have played. Strictly speaking, he probably isn't worse yet, but he isn't better either and Black has more options. At some point he could try to open more diagonals by playing f5, and in general this position remains much easier and more comfortable for Black to play. Giorgi instead started to go quite badly wrong from here onwards and allowed b4 at a very inopportune moment. Soon all my pieces were roaming free, so to speak, in White's position. The game didn't last until the first time control and was instrumental in my winning the 1994 World Championship in Szeged. So of course I had a lot of reasons to remember it fondly regardless, but recently I also had a very peculiar and pleasant experience.

I was playing in the Gibraltar Masters and after one of the rounds a man came up to me and said, "Hi, you don't know me, but do you remember a game you won against the Bd2 Grünfeld where your bishops were so strong that you won an endgame a pawn down?" I said, "Yeah, I remember it well, it was a nice game," and he replied, "You know that prompted me to pick up the Grünfeld!" I'm sure people like Kasparov hear that all the time. For instance, I clearly picked up the Grünfeld because Kasparov was the major Grünfeld player of his age and I was basing my opening repertoire on his games. But for me it was a very rare occurrence to hear somebody say that my games and my choices somehow influenced his choices in life. I remember that conversation and it was a very pleasant compliment to hear. Anyway, those are the two games I wanted to show. I hope they're a) somewhat instructive chess-wise, and b) convey a sense of how I view the Grünfeld as an opening, what made it so attractive to me and why I've been a Grünfeld player for the past 20 years or so. 27... ♖a3 28. ♘d1 ♖a2 29. ♘f4 b4 30. ♘d3 ♖e2 31. g3 ♗b5 32.♘xb4 ♖e1+ 33. ♔g2 ♗d4 34. ♘c2 ♖e2+ 0-1

1.d4 Nf6 2.c4 g6 3.Nf3 d5 4.cxd5 Nxd5 5.g3 Bg7 6. Bg2 Nxc3 7.bxc3 c5 8.e3 8.Nf3 3.g3 c6 4.Bg2

4.cxd5 Nxd5 5.g3 Bg7 6. Bg2 Nxc3 7.bxc3 c5 8.e3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. g3 The point of this move order is that White manages to prevent Black from playing the symmetrical systems against g3 that start with the move c7-c6 and then d5. In those systems Black recaptures on d5 with the pawn, keeping the structure very solid. That's always been the most fireproof way for Black to deal with the g3 systems and is currently doing very well, so White players are understandably trying to avoid it. However, the position of the knight on c3 gives Black additional opportunities. 5... ♗g7 6. ♗g2 ♘xc3 7. bxc3 c5 In this position White has a choice between 8.Nf3 and 8.e3. 8. e3 ♕a5 This is more ambitious than 8...0-0 and I think a stronger option. Black aims for immediate equality, and White now once again has a choice. 8... O-O has no particular drawbacks. 9. ♘e2 ♘c6 10. O-O ♕a5 In this structure Black aims to play Rd8, the queen is normally very safe on a5, or later a6, and Black tries to play on the light squares or go for an e5-break at an opportune moment. The position is very playable. a) 11. ♕b3 ♖d8 ( 11... ♗g4 12. ♘f4 e5 13. h3 ♗f5 14. g4 ♗c815. dxe5 ♘xe5 16. ♘d5 ♗e6∞ ; 11... ♕a6 12. ♘f4 e5 13. ♘d5♗e6∞ ) 12. a4 ( 12. ♘f4 e5! ; 12. ♗d5 e6 13. ♗g2 ♕a6 14. ♘f4 e5⇆ ; 12. ♖e1⁈ cxd4 13. exd4 ♘xd4! ) 12... ♗g4 13. f3 ( 13. ♘f4e5⇆ ; 13. ♖a2 ♗e6 14. d5 c4 15. ♕b5 ♗xd5 16. ♗xd5 ♖xd5 17.♕xb7 ♕b6!=/+ ) 13... ♗d7 14. ♗a3 b6 ( 14... cxd4 15. cxd4 ♖ab8 16. ♖fd1 ♗e6∞ ) 15. f4 ♖ac8 16. d5 Lieb-Malisauskas, 1996 16... ♗g4! 17. ♖f2 ♗xe2 18. ♖xe2 ♕xc3=/+ b) 11. a4 11... ♖d8 12. ♖b1 ♕a6 and Black is fine: 13. ♘f4 cxd414. cxd4 ♗f5 15. ♖b5 ♖ac8 Marszalek-Jaracz, Polanica Zdroj 2005 9. ♘e2 This has the disadvantage of allowing 9...cxd4. 9. ♗d2 keeps the cxd4 option open for White, but after 9... O-O 10.♘e2 ♘c6 11. O-O ♕a6 Black's idea becomes apparent. The bishop on d2 is completely misplaced - it really has no business being there - and in general it's quite difficult for White to find suitable posts for his pieces. A possible plan for White is now 12. ♘c1 but after 12... ♗d7 ( 12... cxd413. cxd4 e5 14. d5 e4 15. ♘b3+/= ) 13. ♘b3 b6 it becomes clear that the knight on b3 is only a slight improvement over the knight on e2. 14.a4 ( 14. dxc5 will be met by 14... ♖ac8 and Black gets very typical Grünfeld compensation for the pawn. He's perfectly fine here, as the whole structure on the queenside is now a target for attack and White's pieces are still misplaced.) 14... ♖ac8 15. a5 ♘xa5 , and after massive simplifications on the queenside a draw is the most likely outcome. 16.♖e1 ♕b5 17. ♘xa5 bxa5= 9... cxd4 10. exd4 O-O 11. O-O ♘c6 Black has managed to force White to take on d4 with the e-pawn. I think that should be very beneficial for Black because in general White wants to take with the c3-pawn, opening up the c-file and the c3-square for the knight on e2. Black is very comfortable. Two high profile games continued 12. ♕b3 ♖d8 13. d5 ♘e5 Black is ready to play e.g. Bd7 next move and is just fine here. 14. ♗g5

14. ♘d4 ♗d7 15. ♗g5 ♕c5 16. ♖fe1 h6 17. ♗e3 ♕c7 18. ♗f4 g519. ♗xe5 ♗xe5 20. ♖ac1 ♖ac8 21. ♘f3 ♗f6⇆ Mamedyarov-Grischuk, Almaty IV World Blitz 2008

14... ♖e8

14... ♕c5 is another safe way to meet 14.Bg5. 15. ♘d4 ♗d7 16. ♖ab1 b6 17. ♗f4 ♖ac8 18. ♗xe5 ♗xe5 19. ♘c6♕xc3=/+ Georgiev-Kramnik, Belgrade 1997

4.cxd5 Nxd5 5.g3 Bg7 6. Bg2 Nxc3 7.bxc3 c5 8.Nf3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. g3 ♗g7 6. ♗g2♘xc3 7. bxc3 c5 8. ♘f3 cxd4 This is my proposed solution in this line. 8... ♘c6 is what most people play here, which is a fine move in itself, but after 9.0-0 it allows White a position I would recommend trying to avoid. a) 9. e3 is not what White plays 8.Nf3 for. 9... O-O 10. O-O ♕a5 a1) 11. ♗b2 ♖d8 ( 11... ♗e6 12. d5 ♖fd8∞ Miton-Volokitin, Bermuda 2003) 12. ♕e2 ( 12. ♕b3 ♗e6 13. ♕a3 ♕c7 14. ♘g5♗c4 15. ♖fd1 ♘a5 16. ♘e4 b6 17. ♘d2 Pasalic-Mikhalevski, Chicago 2004 17... ♖ac8= ) a11) 12... ♗e6 13. ♖fd1 cxd4 14. ♘xd4 ( 14. cxd4 ♖ac8 15.♘d2 ♕a4∞ ) 14... ♗d7= Arnett-Yermolinsky, New York 1998 a12) 12... e5 13. ♘xe5 ♘xe5 14. dxe5 ♗xe5 15. c4 ♗xb2 (15... ♖d2 16. ♕xd2 ♕xd2 17. ♗xe5=/∞ Kotsur-Bologan, Moscow 2004) 16. ♕xb2 ♗e6 17. ♕e5 ♖d2= a2) 11. ♗d2 ♕c7 12. ♕c2 ♘a5 13. ♖fc1 ♖d8 14. ♗e1 ♗e6=/+Doss-Shulman, Dallas 2000 a3) 11. ♕b3 11... ♖b8 ( 11... ♕c7 12. ♗a3 b6 13. dxc5♖b8=/∞ Ljubojevic-Topalov, Monaco 1997; 11... b6 12. ♘d2 ♗d713. ♗b2 ♖ac8= Arutinian-Kacheishvili, Tbilisi 66th GEO-ch 2007) a31) 12. d5 ♕xc3 ( 12... ♘e5 13. ♘xe5 ♗xe5 14. ♗d2 b5 15.f4 ♗g7 16. e4 e6=/+ Navara-Bologan, Mainz 14th Ordix op 2007) a32) 12. ♘d2 12... ♗d7 13. ♗a3 b6 14. ♖fd1 ♖fc8 15. ♗f1cxd4 16. cxd4 ♗g4⇆ Filippov-Bologan, Vladivostok 1995

b) 9. O-O 9... cxd4 ( 9... O-O⁈ is not the best: 10. d5 ♘a5 11.♗f4 ♗f5 12. ♖c1 ♗e4 13. c4 ♗b2 14. ♗h3 ♗xc1 15. ♕xc1 ♗xf316. exf3 b5 17. cxb5 ♕xd5 18. ♗h6 Potkin-Belov, Dagomys RUS-tch 2010) 10. ♘xd4⁉ , and now Black has a choice. 10... ♘a5⁉ I played this against Vladimir Potkin as I decided I didn't want to defend the endgame after 10... Nxd4, but I think this position is quite dangerous for Black. White has a number of options here which are potentially very, very unpleasant and make it hard to recommend.

( 10... ♘xd411. cxd4 ♕xd4 , accepting the pawn sac, is now considered more or less a forced draw, but it's one that Black, not White, will clearly have to make, and in general after 12. ♖b1 White will enjoy a slight pull with absolutely zero risk. Although this is of course a very viable option, it's not my recommendation.) b1) 11. ♗g5 OO ( 11... h6 12. ♗e3 OO 13. ♕c1 ♔h7 14.♖d1± ) 12. ♕a4 ♗d7 13. ♕b4 ♖e8 14. ♖ad1 ♖c8∞ b2) 11. ♕a4+ 11... ♗d7 12. ♕a3 ♖c8 13. ♖d1 ♕b6 and here I was already fine in PotkinSvidler, Bundesliga 2011 14. ♕b4 ♕xb415. cxb4 ♗a4 16. ♖d3 ♘c6 17. ♗e3 ♘xd4 18. ♗xd4 ♗xd4 19.♖xd4 ♖c7 , but the whole approach seems a bit risky to me. 9. cxd4 The normal move. 9. ♘xd4 O-O 10. O-O ♕c7! a) 11. ♖b1 ♕xc3 12. ♘b5 ♕a5 13. ♗d2 ♕xa2 The simplest move. (13... ♕d8 leads to problems. 14. ♗e3 ♕a5 15. ♕e1 ♘c6 16. ♕xa5♘xa5 17. ♘c7 ♗f5 18. ♘xa8 ♗xb1 19. ♖xb1 ♖xa8 20. ♗xb7±Aronian-Grischuk, Noyon 2005) 14. ♗b4 ( 14. ♘c7 illustrates the point of 13... Qxa2, as it's met by 14... ♖d8 and after the bishop on d2 is lost only Black can be better.) 14... ♗f5 This results in the total annihilation of everything on the board and a quick draw. 15. ♗xb7♗xb1 16. ♕xb1 ♕xb1 17. ♖xb1 ♘d7= b) 11. ♕b3 11... ♘c6 The following almost forced line once again leaves the board almost empty 12. ♘xc6 bxc6 13. ♗f4 ♕a5 14.♗xc6 ♗e6= 15. ♕b4 ♕xb4 16. cxb4 ♗xa1 17. ♗xa8 ♗c3=Bol ogan-Svidler, Astana Blitz 2012 9... ♘c6 10. O-O Once again inviting Black to take on d4. 10. e3 O-O 11. O-O ♗e6 is very nice for Black: 12. ♕e2 ( 12. ♕a4 a613. ♗d2 ♗d5= ; 12. ♘d2 ♗d5= ) 12... ♖c8∞ 10... O-O Black once again declines the invitation. 10... ♘xd4 11. ♘xd4 ♕xd4 12. ♖b1 O-O 13. ♗e3 Aronian-Radjabov, 2009 11. ♖b1 White continues to insist. 11. ♗f4 ♘xd4 12. ♘xd4 ♕xd4 13. ♖c1 ♕xd1 14. ♖fxd1 ♗e6= 11. ♗g5 ♘xd4 12. ♘xd4 ♕xd4= 11. e3 here, abandoning the idea of sacrificing the d4-pawn, leaves Black perfectly fine after 11... ♗e6 His plan is quite transparent. He wants to put the bishop on d5, the rook on c8 and the queen will go to either d7 or a5. In general, if Black manages to neutralise the potentially very strong bishop on g2 he should be fine in this structure. 11... b6⁉ This is the move I propose here. I played it against Ruslan Ponomariov in the European Club Championship in 2011, and I think it gives Black an interesting position with definite potential for counterplay. 11... ♘xd4 transposes to the endgame I described above.

12. d5 The main move and the one Ruslan played against me. 12. ♘e5 ♘xe5 13. ♗xa8 shows the tactical justification for 11...b6, as Black now has the incredibly strong move 13... ♗f5=/+ hitting the rook on b1 and threatening Qxa8 and then Bh3. Black regains material with interest and is already better. 12. ♘g5 ♗b7 13. d5 ♘a5 14. e4 h6 15. ♘f3 then either 15... e6 or the less concrete ( 15... ♖c8 lead to very double-edged positions which should be very playable for Black. 16. ♗b2 ♗xb2 17. ♖xb2 ♕d6∞ ) 16.♗a3 ♖e8 17. d6 ♕d7∞ ( 17... ♘c4 18. d7 ♘xa3 19. dxe8=Q+ ♕xe8 20. ♖b3 ♘c4=/∞ ) 12. ♘d2 ♗b7 13. e3 ♘a5 14. ♗xb7 ♘xb7 15. ♗a3 ♕d7= 12. ♗a3 is a quiet move Black can meet with 12... ♗e6 because a pawn is no longer hanging on b7. He again wants to play something like Rc8 and Qd7, when Black should be very comfortable. 13. ♖c1 ♖c8 ( 13... ♗d514. e3 ♕d7 15. ♖e1+/= ) 14. e3 ♗xa2 ( 14... ♕d7 15. ♘g5+/= ) 15.♘g5 ( 15. ♕e2 ♗e6 16. ♘g5 ♗d5 17. ♕b5 ♗xg2 18. ♔xg2 ♕e8= )15... ♘a5 16. ♖xc8 ♕xc8 17. ♗x e7 ♖e8 18. ♗b4 ♗c4 19. ♖e1 ♕f520. ♘f3 ♕b5∞ 12. e3 gives Black additional possibilities, for instance playing 12... ♗f5 first. 12... ♗f5 This is Black's point. You don't want to allow e2-e4 without putting up some resistance. The rook is now hanging so White has a choice. 13. ♖b5 13. ♖b2 This isn't an exchange sacrifice Black should even seriously contemplate. I personally prefer a) 13... ♘e5 14. ♘xe5 ( 14. ♘d4 ♘d3 ) 14... ♗xe5 15. e4 ♗d7The machine insists Black is very much alive and well in the position after this additional trade, but I'm not sure Black needs to trade off the knights. 16. ♖b1 ♖c8∞ b) 13... ♘a5⁉ is what Ruslan played here. 14. ♘d4 ♗d7 15. ♕d3 (15. ♖b1 may be the best White has, which would transpose to the game between myself and Ruslan.) 15... ♖c8 16. ♖b1 ( 16. ♖c2♖xc2 17. ♘xc2 ♕c7 18. ♗a3 ♖c8 ) 16... ♕c7 17. ♘b5 ♗xb5 18.♕xb5 ♕c4= 13. ♖b3 ♘a5 14. ♖b4 White's plan here is to transfer the knight to c6. If he succeeds it would give him a big advantage, but I don't think it's very likely to happen. 14... ♖c8 15. ♘d4 ♕d6 Entering a forcing line. 16.♗a3 ♗xd4 ( 16... ♗d7 17. ♕d3 ♕c7 18. ♘c6 ♗xc6 19. dxc6+/= ) 17.♕xd4 ♘c4 Black successfully trades off a lot of material. 18. ♖xc4♖xc4 19. ♕xc4 ♕xa3 This position is equal. Black hangs on with forced lines, making this very playable for Black. 20. e4 ♗d7 ( 20... ♖c8 21.♕b5 ♗g4 22. h3 a6 23. ♕xb6 ♗e2 24. ♖e1 ♕xa2 25. ♕b4+/= ) 21.♕c 7 ♕a4 22. e5 ♖c8 23. ♕b7 ♖c5= 13... ♘a5⁉ 13... a6 is of course possible here, but I was worried White would play14. ♖b2⁉ and after 14... ♘e5 or ( 14... ♘a5 the pawn on b6 might be hanging in some lines. I didn't want to weaken my queenside for no particular reason.) 14. ♘d4 ♗d7 15. ♖b1

15. ♘c6 doesn't work here as the pawn will be surrounded and eaten. 15... ♖c8 Black has rich counterplay because it's difficult for White to come up with any concrete plans, while Black has a lot of squares along the c-file he can use as shuttle points for his pieces - c5, c4 and even c3 in some cases. 16. ♗b2 16. ♗f4 is natural, but I believe Black's play is somewhat easier after16... ♖c4 17. e3 ♕c8⇆ He has very clear and well-defined plans. Next move he'll probably continue with Qc5, the second rook will come to c8 and Black's occupation of the only open file gives him very comfortable play. 16... ♕c7 Planning to put the queen on c4 and apply some pressure on the queenside. 17. ♗a3 , and here I missed a very nice shot at an advantage. 17. ♖c1 is of course met by 17... ♕d6 and the queen is ready to come to b4, which is also unappealing for White. 17... ♕c3! would have given Black an excellent game. 17... ♕e5 is what I played, and after 18. ♘f3 the game was soon drawn. Ponomariov-Svidler, Rogaska Slatina 27th ECC 2011 18. ♘b3 I couldn't quite make this work for Black because White now wants to play either Rc1 or Bb2, leading to beneficial trades e.g. Black might even have to recapture on a5 with the pawn because the queen will be driven away from c3. 18. ♗b2 is of course met by 18... ♕c4 and it's just a waste of a tempo, which wasn't what Ruslan planned. 18... ♘c4 19. ♗xe7 ♘e3! This is the very beautiful shot I missed. Black wins back a lot of material with a fantastic position. I looked at the position after 19... ♖fe8 20. d6 for a while but couldn't find any direct counterplay. It's somewhat risky to give away the e7-pawn for absolutely nothing.

4.Bg2

1. d4 ♘f6 2. c4 g6 3. g3 c6 4. ♗g2 4. ♘f3 d5 5. cxd5 cxd5 6. ♘c3 ♗g7 7. ♘e5 , trying to save a tempo on Bg2, doesn't get White anywhere: 7... ♘fd7 8. f4 ♘xe5 9. fxe5 OO 10. ♗g2 ♗e6 11. OO ( 11. e4 dxe4 12. ♗xe4 ♕b6=/+ ) 11...♘c6 12. ♗e3 f6 13. exf6 exf6 , and Black is perfectly fine. 4... d5 5. b3 This is just a mistake, first of all because in this position with Bg7 and 0-0 Black can transpose to a line covered later where Black is doing quite well, but also because White is somewhat underdeveloped while Black still has all his options open.

5. ♕a4 A serious option, which Gabi Sargissian played against me in the same European Club Championship where the game against Ruslan Ponomariov was played. a) 5... ♘fd7 is the best move in the position, which has been known about for quite a while, but slipped my mind. In this position Nb6 is a huge threat. a1) 6. ♕c2 The main move, trying at least somehow to justify playing Qa4 on the previous move. Here the most precise way of playing is 6... dxc4 ( 6... ♗g7 could be played first, but then after7. ♘f3 dxc4 8. ♕xc4 ♘b6 White can play 9. ♕c2 because you really don't want to take on d4 and give up your pride and joy, the dark-squared bishop.) 7. ♕xc4 ♘b6 Now White pretty much has to play 8. ♕d3 as after 8. Qc2 Black will of course take on d4 and simply be a pawn up. But after 8... ♗g7 9. ♘f3 OO 10. O-O♘a6 the queen is completely misplaced on d3. Black is ready to start counterplay with c5 here and, in general, I think Black stands very well. a2) 6. cxd5 Black may even be able to claim a slight advantage here. After 6... ♘b6 7. ♕b3 cxd5 8. ♘f3 ♗g7 9. O-O ♘c6 10.♖d1 O-O it's clear that Black is already better developed. The bishop will come to f5 and Black has a lot of very obvious strengthening moves like Qd7 and Rc8, whereas White is already somewhat pinned down to the defence of the d4-pawn. The position is probably still equal, but if anyone is better here it's Black. b) 5... dxc4 is what actually happened in my game against Sargissian.6. ♕xc4 ♗e6 7. ♕a4 And only now 7... ♘bd7 8. ♘f3 ♘b6 9. ♕d1Unfortunately I really did go on to lose the game. Sargissian-Svidler, Rogaska Slatina 27th ECC 2011. c) 5... ♘bd7 is the move I showed immediately in the video, and after6. cxd5 ♘b6 7. ♕b3 I said that "the b8-knight is somewhat misplaced on b6 and in general belongs on c6, if at all possible. I couldn't achieve full equality here and went on to lose the game." 5. ♘f3 ♗g7 6. O-O O-O Apart from the three main moves - 7. b3, 7.Qb3 and 7.cxd5 - White has the option of playing 7. ♘bd2 This has the advantage of not allowing any forced theory, but the knight is quite badly misplaced on d2 and has no particular future there. A very harmonious way for Black to develop was pointed out, no doubt correctly, by Boris Avrukh in his book. 7... ♗f5 ( 7... a5 is how many people immediately react to the knight coming to d2, with a4 to follow, but that's not really necessary.) 8. b3 ♘e4 9. ♗b2 If you give White an additional move he'll play Nh4, forcing Nxd2 Qxd2, and White will at least have a harmonious position with some plans, although I wouldn't really term that an opening advantage. 9... ♕a5 A strong and important move, pinning White down. The knight on d2 is stuck. You don't want to play Nxd4 because after Bxd4 Black is very comfortable. The bishop shouldn't be allowed to come to e4 so cheaply in this kind of structure. One attempt by White to solve his problems is 10. a3 ♘d7 11. b4 but after 11... ♕d8 it becomes clear that although Black has seemingly wasted two tempi on Qa5 and Qd8 White's expansion on the queenside has only created potential problems. For instance, he almost never wants to take on d5 because the c4-square will be weak for the rest of the game. If White plays something like 12.♕b3 , which looks very natural, Black has a very nice idea. He can play12... ♘b6 creating the threat of dxc4 and Be6, winning material. White will feel very tempted to play 13. c5 and then just 13... ♘d7 Once again, Black has seemingly lost a couple of tempi but now that the centre has been stabilised he'll have very easy play connected with the e7-e5 break, whereas White's initiative on the queenside is still in its very early stages. I think Black is already very comfortable here. 5... dxc4 6. bxc4 e5 putting White's centre under immediate pressure. This really isn't advisable for White.

1.d4 Nf6 2.c4 g6 3.Nf3 Bg7 4.g3 c6 5.Bg2 d5 6.cxd5 6.Qb3 6.b3

6.cxd5 1. d4 ♘f6 2. c4 g6 3. ♘f3 ♗g7 4. g3 c6 5. ♗g2 d5 6. cxd5 6. ♘e5 ♘e4 7. O-O O-O 8. ♕b3 is too fancy to be good: 8... dxc49. ♘xc4 ♘d6 10. ♘e5 ( 10. ♘xd6 exd6 11. d5 ♘d7 12. ♘c3c5⇆ ) 10... ♗e6 11. ♕c2 ♘d7 12. f4 ♘xe5 13. dxe5 ♕b6+ 14.♔h1 ♘f5=/+ Inarkiev-Bosiocic, Rijeka 2010 6... cxd5 7.♘e5 7. ♘c3 ♘e4 A recurrent theme in these lines. Black aims to trade off the c3-knight as early as possible, preventing White from establishing knights on c3 and e5. The moment one of the knights moves you play Ne4 - which makes it very easy to memorise. 8. ♕b3 is a very important subline here. It featured in a number of interesting games recently, with Hikaru Nakamura using it to beat Vladimir Kramnik at the 2012 Olympiad in Istanbul. ( 8. ♘xe4 dxe4 9. ♘e5 should be met with 9... ♘d7which is followed by very typical annihilation in the centre: 10. ♘xd7♕xd7 11. ♗xe4 ♕xd4 12. ♕xd4 ♗xd4 leading to a position that Black will quite comfortably hold. For instance, after 13. O-O he might even consider playing 13... ♗h3 and castling long in order to safeguard the only target for White's initiative, his weak b7-pawn.; 8. ♘e5♘xc3 9. bxc3 transposes.) 8... ♘xc3 9. bxc3 O-O 10. ♘d2 e6 11.e4 Attempting to force some favourable exchanges in the centre, but Black replied 11... ♘c6 a) 12. ♗a3⁉ was played by Erwin L'ami in a recent game played in the final rounds of the Rilton Cup in Stockholm against Ilya Smirin. 12...♖e8 13. O-O ♘a5 14. ♕d1 dxe4⁉ Under favourable circumstances this exchange is something Black should actively encourage, and here it seems to refute L'Ami's idea.( 14... b6 allowed L'Ami to go on to win a fine attacking game: 15. e5 ♗a6 16. ♖e1 ♖c8 17. ♖e3 ♕d718. h4⇆ L'Ami-Smirin, Stockholm XLII Rilton Cup 2013) 15. ♘xe4♘c4 It's very important the bishop is on a3 so Black gains this important tempo. 16. ♗c1 ♗d7 I think this is very comfortable for Black and he may easily go on to be better if White allows Bc6. 17.♕b3 ( 17. ♗g5 f6 18. ♗f4 ♗c6 ; 17. ♘c5 allows White to and try win the pawn on b7, but Black will have fantastic positional compensation based on the horrible bishop on c1.) 17... ♘a5 The easiest response. 18. ♕b4 Other moves are even weaker. 18... ♗f819. ♕b1 ♗c6 20. ♗g5 ♗e7= Black is very comfortable here. b) 12. exd5 ♘a5 It transpires that White can't keep the pawn. c) 12. O-O 12... ♘a5 13. ♕d1 ♕c7 14. ♕f3 , and now I suggest14... ♗d7⁉ One point of this move is that Black aims in many cases to take on e4 and play Bc6. If Black manages to get his bishop to c6 while White has failed to accomplish anything in the meantime Black will be doing very well. The main point, however, is to play Rfc8 in one tempo. ( 14... b6 was played by Kramnik, and after 15. ♗a3 ♖d816. e5 Hikaru managed to win the game, though to claim he was better out of the opening would be a slight overstatement. He wasn't, but he still got a playable position. 16... ♗a6 17. ♖fe1 ♖ac8 18.♗b4 ♗h6 Nakamura-Kramnik, Istanbul 2012) c1) 15. exd5 exd5 16. ♘b3 ( 16. ♕xd5 ♕xc3 is clearly unattractive) 16... ♘xb3 17. axb3 ♖fc8 18. ♗d2 ♗e6= and Black is just fine.

c2) 15. ♗a3 15... ♖fc8 and it's important the rook attacks the c3-pawn in one tempo. White already has to be slightly careful in this position and should probably capture. 16. exd5 ( 16. ♕e3dxe4 17. ♗xe4 ♗c6=/+ ) 16... exd5 17. ♕xd5 Black has a choice. 17... ♕xc3 is perhaps simpler. ( 17... ♗c6 is the forcing option: 18. ♕d6 ♕xd6 19. ♗xd6 ♗xg2 20. ♔xg2 ♖xc3 21.♗b4 ♖d3 22. ♗xa5 b6 , winning the piece back with an endgame that Black should hold. 23. ♘f3 bxa5 24. ♖fd1 ♖xd125. ♖xd1 ) 18. ♕xd7 ♕xa3 19. ♗d5 ♕f8 Black's next move is Rd8, driving the pieces away, and I think it will very soon end in complete equality. 20. ♖ac1 ♖d8 21. ♕b5 ♘c6= 7. O-O O-O a) 8. ♘e5 a1) 8... ♘e4 After 9. ♘d2 the one thing you shouldn't do is play9... ♘xd2 ( 9... ♗f5 is fine for Black: 10. ♕b3 ♘c6 11.♘xc6 bxc6 12. ♘xe4 ♗xe4 13. ♗xe4 dxe4 and although the mass exchanges have left Black with a somewhat strange pawn structure his far superior bishop on g7 ensures that isn't really felt.; 9... ♘d7 is another option.) 10. ♗xd2 ♘d7 because that allows 11. ♗c3 and this is the one position where White actually has some pull. a2) ♘g4 is an additional option for Black in this move order. 9. f4( 9. ♘xg4 ♗xg4 10. ♘c3 ♘c6 ) 9... ♘xe5 10. fxe5 ♘c6 11.♘c3 ♗e6 and Black is very comfortable. a21) 12. ♔h1 ♕d7 13. e4 ♗g4 14. e6 ♗xd1 15. exd7♗g4 16. ♗g5 f6 17. ♗e3 e5⁉∞ ( 17... dxe4 18. ♘xe4♗xd7 19. ♘c5 ♗c8= ) a22) 12. ♗e3 12... f6= b) 8. ♘c3 8... ♘e4 Once again very comfortable for Black. If White plays 9. ♘xe4 dxe4 10. ♘e5 Black has a choice.( 10. ♘g5 ♕xd411. ♘xe4 ♘c6= ) 10... ♘d7 The simplest: ( 10... ♕d5 is a well-known and decent move here.) 11. ♘xd7 ( 11. ♗xe4 ♘xe5 12.dxe5 ♗h3 13. ♖e1 ♕xd1 14. ♖xd1 ♗xe5= ) 11... ♕xd7 ( 11...♗xd7 12. ♗xe4 ♗h3 13. ♗xb7 ♖b8 14. ♗g2 ♗xg2 15. ♔xg2♗xd4 ) 12. ♗xe4 ♕xd4= B lack is fine. 7... ♘e4 7... O-O is somewhat naive. After 8. ♘c3 e6 9. O-O White gets the tabiya of the g3 cxd5 cxd5 variation. This has been discussed in very high level games over a long period of time, including in a number of Kasparov-Karpov games. Generally speaking, Black is doing quite well here but the character of play isn't to everybody's taste. White gets quite a bit of pressure after 9... ♘fd7 10. f4 ( 10. ♘f3 is the latest trend in this position. It looks somewhat counterintuitive, but has a point. Black needs to be precise here.) 8. ♘c3

8. OO OO 9. ♘d2 ♗f5⁉ ( 9... ♘xd2 10. ♗xd2 ♘d7 11. ♗c3!+/=; 9... ♘d7⁉ ) 10. ♕b3 ♘c6 11. ♘xc6 bxc6 12. ♘xe4 ♗xe4 13.♗xe4 dxe4 14. ♖d1 ♖b8 15. ♕c2 ( 15. ♕a4 ♖b5 16. ♕xa7♖d5→ ) 15... ♖b5! 16. ♗e3 ♕d5= 8. ♘d2 ♘d7 9. ♘xe4 dxe4 10. ♘xd7 ♕xd7 11. ♗xe4 ♕xd4 12.♕xd4 ♗xd4 13. O-O ♗h3= 8... ♘xc3 9. bxc3 O-O 10. O-O We've come to the central position of this whole line. 10... ♘c6 This is a move I've played on a number of occasions with very decent results, but I just thought it was kind of boring. I wanted to show you something else... but nothing else works! 10... ♘d7⁉ This is the move I wanted to recommend to you, but instead I'd like to share a slight disappointment. This move isn't very well-known in this variation and I thought it was an improvement on all the theory starting with 10...Nc6, but with great regret I had to shelve the idea. 11.♘xd7 White should take, as allowing Nf6 just isn't a good idea.( 11.♘d3 ♘f6= ) 11... ♕xd7 If Black is allowed to play b6 and follow it up with Ba6 he may even be slightly better, so White plays 12. ♕b3 ♖d813. e4 dxe4 ( 13... b6 14. ♕xd5 ♕xd5 15. exd5 ♗b7 16. ♖e1♖d7 17. a4+/= ) 14. ♗xe4 a) 14... ♕e6 15. ♗g2+/= b) 14... e5 15. d5 ( 15. ♗g5 ♖e8 ; 15. ♖e1 exd4 16. ♗g5♖f8 17. ♖ad1 ♕e6 18. cxd4 ♕xb3 19. axb3 ) 15... ♕c7 ( 15...f5 16. ♗d3→ ; 15... b6 16. ♗g5 ♖e8 17. d6 ♖b8 18. ♗e7♗b7 19. ♗xb7 ♖xb7 20. f4 ; 15... ♗f8⁉ ) 16. ♗a3 ♗f8 17.♗xf8 ♔xf8 18. c4 ♔g7 19. ♖ab1 ♖b8 20. ♕c3+/= c) ♕c7 I thought this just equalised, but then I discovered White has the very strong 15. ♗d5! This looks odd, but is quite unpleasant to meet. ( 15. ♗f4 e5= ) c1) 15... ♖xd5 16. ♕xd5 ♗h3 17. ♖e1 ♕xc3 18. ♗g5± c2) 15... e6 16. ♗f4 Only now, and e5 is impossible as the pawn on f7 will be hanging after everything is exchanged. 16... ♕a5 17.♗g2 e5 18. ♗g5 I couldn't find equality for Black in this position. c21) 18... ♖d6 19. ♗e7 ( 19. dxe5 ♗xe5 20. ♖ae1+/= ) c22) 18... ♖e8 c221) 19. ♗d5 ♗e6 20. ♗xe6 fxe6 21. ♗e3 b6 22.dxe5 ( 22. ♖ac1 exd4 23. cxd4 ♕d5 24. ♕xd5 exd525. ♖c7+/= ) 22... ♕xe5 23. ♖ac1+/= c222) 19. ♖fe1 19... h6 20. ♗e3 exd4 21. ♗xd4 ♗e622. ♕xb7 ♖ad8 23. ♗xg7 ♔xg7 24. ♕b4+/= c3) 15... ♗h3 White even has a choice here. c31) 16. ♖e1 e5 ( 16... e6 17. ♗f4± ) 17. c4 ( 17. dxe5♗xe5 18. ♗b2 ♖e8 19. c4 ♗xb2 20. ♕xb2 ♖ac8 21.♖xe8+ ♖xe8 22. ♕xb7 ♕e5 23. ♕xf7+ ♔h8 24. ♖f1♗xf1= ) 17... ♖e8 18. ♖b1 exd4 ( 18... ♖ab8 19. dxe5♗xe5 20. ♕e3!+/= ) 19. ♕d1! b6 20. ♗xa8 ♖xa8 21. ♕f3♕c8 22. ♖e7 ♗f5 23. ♖b5+/=

c32) 16. ♗xf7+ 16... ♔h8 17. ♗e6 ( 17. ♖e1 ♕c6 18. d5♕f6 19. ♗e6 ♗xe6 20. ♖xe6 ♕xc3= ) 17... ♗xe6 ( 17...♗xf1 18. ♗f4!± and there's no equality for Black.) 18. ♕xe6♕xc3 19. ♖b1 b6 20. ♗b2 ♕c7 21. ♖fe1+/= 11. ♘xc6 bxc6 12. ♕a4 To my great surprise this was played against me in successive rounds of the 2011 World Cup, first by a young Vietnamese player and then by Fabiano Caruana, as a serious attempt to fight for an advantage. 12. ♗a3 ♗a6 13. ♖e1 ♖e8 14. e4 e6 is how the old theory starts. In most cases the bishops will summarily be exchanged in this position on the f8- and f1-squares. In general, only slight precision is required of Black to hold. It's somewhat unspectacular and less than adventurous, but very sound. a) 15. h4 ♗c4 16. e5 ♖b8 17. ♕d2 ♗f8 18. ♗xf8 ♖xf8 19.♗f1 ♗xf1 20. ♔xf1 h5 ( 20... c5 21. h5→ Tregubov-Yandemirov, Maikop 1998) b) 15. e5 15... ♗f8 16. ♗xf8 ♔xf8 17. h4 ♔g7 18. ♖b1 ♕e719. ♗f1 ♗xf1 20. ♖xf1 ♖eb8 21. ♕d2 ♖b6= Ragger-Miroshnichenko, Austria 2010 12... ♕b6 13. ♗f4 13. e4 a) 13... ♕a6 14. ♕b4 dxe4 15. ♗xe4 ♗e6 This isn't very precise because in the position after 16. ♗f4 ♖fe8 17. ♖fe1 ♖ac8 18.♗g2 ♕c4 19. ♕a3 c5 Fabiano was short of time and repeated moves. 20. ♗f1 ( 20. ♗b7⁉ If he's played this I'm not entirely sure Black has full equality.) 20... ♕d5 21. ♗g2 ♕c4 22. ♗f1 ♕d523. ♗g2 Caruana-Svidler, Khanty-Mansiysk 2011 b) ♗a6 14. ♖e1 e6 15. ♗f4 ♗d3!= An important move, preventing the rook from getting to b1. Black is just fine here. 13... ♕a6 14. ♕b4 ♗f5 15. ♖fe1 ♖fe8 16. e4 ♗xe4 17. ♗xe4dxe4 18. ♖xe4 e6 19. a4 ♖ad8 19... ♕b6 After this move I was slightly worse, although the game was eventually drawn. 20. ♖b1+/= Nguyen Ngoc Truong Son-Svidler, Khanty-Mansiysk FIDE World Cup 2011 20. ♖b1 ♗f8 20... ♖d5 is apparently sensible, but shows the slight danger in the position, since it runs into 21. ♕e7 and Black has serious problems to solve. 21. ♕b3 c5 22. dxc5 ♗xc5= Black should equalise without too much trouble as it's very difficult for me to believe that Black can be seriously worse in this position.

6.Qb3 1. d4 ♘f6 2. c4 g6 3. ♘f3 ♗g7 4. g3 c6 5. ♗g2 d5 6. ♕b3O-O 7. ♘c3 This is best met by 7... ♕b6 8. O-O ♖d8 , and now White has a very wide choice (at least seven reasonably good moves), but I believe Black is fine in every line: 9. c5 9. ♕xb6 isn't really a problem for Black, as after 9... axb6 10. cxd5♘xd5 11. ♘xd5 ♖xd5 12. ♖d1 his position is comfortable almost regardless of what he does, although I think ( 12. e3 ♖d8 13. ♗d2 c514. ♗c3 ♘c6 15. ♘e5 ♘xd4 16. exd4 cxd4 17. ♗b4 ♗xe5 18.♗xe7 ♖e8 19. ♗b4 ♗e6 20. ♖fe1 ♗f6 1/2 Kavalek-Ogaard, Manila 1975) 12... ♗f5⁉ is a very sensible move here, trying to stop e4. If I was Black in this position I'd already be wondering whether I was better. ( 12... ♖b5 13. e4 ♗e6 14. a3 ♗b3 15. ♖e1 ♖a4 16.♗f1 ♖ba5 17. ♗d2 ♖a8 18. ♗c3 c5 19. ♗b5 ♖4a7 20.♖ac1+/= Rogozenko-Donche) 9. cxd5 ♕xb3 10. axb3 ♘xd5 11. ♗g5 ♔f8! Covering the only possible weakness - e7. Ever since the game between Peter Heine Nielsen and Teimour Radjabov it's been known that this is very nice for Black. 12.♖fc1 ♗e6 13. ♗d2 h6 14. e3 ♘d7 15. ♘e1 ♘xc3 16. ♖xc3 g517. ♘d3 ♗f5 18. ♘c5 ♘xc5 19. ♖xc5 ♗g6 20. ♗c3 e6 21. b4a6 22. ♖ca5 with a draw in Nielsen-Radjabov, Tripoli FIDE WCh KO 2004 9. ♗f4 dxc4 10. ♕xc4 a) 10... ♕xb2 is a possible but very dangerous move that I played in a blitz game against Maxime Vachier-Lagrave. White gets a lot of compensation. 11. ♖ab1 ♕a3 12. ♘g5 e6 13. ♗c7! ( 13. e4 is bad, and after 13... b5 14. ♕d3 b4 15. ♖b3 ♕a6 I had a great game in Vachier Lagrave-Svidler, Moscow 5' 2010) 13... ♖e8 14.♘ce4 , and White has compensation for the pawn. b) ♗e6 Black has simple equality after 11. ♕d3 ♘a6 12. ♕c2 c513. ♘a4 ♕b5 14. ♘xc5 ♘xc5 15. dxc5 ♖ac8 , AlekseevRadjabov, Astrakhan 2010 9. ♕a3 This would be a very interesting idea if Black allowed c4-c5, with a big spatial advantage for White, but he of course just plays 9... dxc4 10. ♕xe7 a) 10... ♖e8⁉ is a perfectly sensible move, and 11. ♕d6 ( 11. ♕a3♘a6 ) 11... ♗f5 12. ♕f4 ♘a6 13. h3 ♘b4 14. g4 ♘bd5 15.♕d2 ♘e4 16. ♘xd5 cxd5 17. ♕f4 ♗e6 led to an edge for Black in Burmakin-Midoux, 2006 b) 10... ♗f5? 11. ♘a4 c) ♘a6 This is my idea here, and I'm keeping to the tradition of letting you know what I have written in my notes. c1) 11. ♘e5 ♗e6 12. e3 ♖e8 13. ♕a3 ♕b4 is fine: 14. ♕a4( 14. ♗d2 ♕xa3 15. bxa3 ♘d5 16. ♘e4 ♘b6 ) 14... ♕xa415. ♘xa4 ♖ad8 16. ♗d2 ♗d5

c2) 11. ♘a4 ♕c7 12. ♕xc7 ♘xc7 13. ♗f4 ♘e6 14. ♗e5 b515. ♘c3 b4 , and Black gets a lot of counterplay on the queenside: the bishop will come to a6, the rook to c8 and Black will play c5. I'm not worried about Black's prospects at all. 16. ♘b1♗b7 17. a3 c5 18. dxc5 b3 19. ♘bd2 ♖ac8 c3) 11. ♖d1 11... ♖e8 12. ♕a3 ♗f5 13. ♘e5 ♕b4 This is very strong. After 14. ♕xb4 ♘xb4 the pawn on c4 will probably be lost, but at the very least Black will have a repetition with Nc2-b4 on every move. 15. e3 ♘c2 16. ♖b1 ♖ad8 17. ♘xc4 , and Black has a choice between the immediate draw and the more ambitious 17... b5 ( 17... ♘b4= ) 18. ♘a5 b4 19. ♘e2 ♘xd420. ♘xd4 ♗xb1 21. ♘axc6 ♖d7 22. ♘xb4 a5=/+ 9. h3⁉ This is a useful pass move in this position, but Black has a very concrete way of replying. 9... dxc4 ( 9... ♕xb3 10. axb3 ♘a6 is also possible) 10. ♕xc4 ♗e6 11. ♕d3 a) 11... ♘a6⁉ I couldn't find complete equality here. 12. e4 ♘e8 (12... ♘b4 13. ♕e2 ) 13. ♖d1 ♘d6 14. b3 c5 15. ♗e3 ♖ac8∞ b) 11... ♘d5⁈ looks sensible, but after 12. ♗g5! there are some questions to answer. 12... ♘a6 ( 12... ♕a6 13. ♕d2 ♘d7 14. e4♘xc3 15. bxc3 f6 16. ♗h6 ♘b6∞ ) 13. e4+/= c) ♕a6 is an idea I very much like here. 12. ♕c2 White needs to keep the queens on the board. 12... c5! 13. dxc5 ♘c6 I think you'll recognise this as a structure that's seen in a lot of positions in the Qb3 Grünfeld. Here, as there, Black has fantastic compensation. At the cost of a single pawn he's opened up the whole board for his bishops, while the c5-pawn will most likely drop off in the immediate future. 14.♗e3 ( 14. ♗f4 ♘d5 15. ♘xd5 ♗xd5 16. e4 ♕c4 17. ♕xc4♗xc4 18. ♖fe1 ♘b4=/∞ ) 14... ♘d5 15. ♗d2 ♘xc3 16. ♗xc3♕c4 17. ♖ac1 ♕xc5 18. ♗xg7 ♕xc2 19. ♖xc2 ♔xg7 20. a3 ♗d5= 9. ♖d1 a) 9... ♕xb3 10. axb3 ♗f5 11. ♘e1 ♘a6 ( 11... e6? 12. h3!± )12. ♖a4 a1) 12... h6 13. ♗f4 ♘e4 14. cxd5 cxd5 15. f3! ( 15. ♘b5?♗d7−+ ) 15... ♘xc3 16. bxc3 g5 17. ♗d2 e5 18. e3+/=Portisch-Hort, Tilburg 1979 a2) ♘e4⁉ 13. ♗e3+/= b) 9... ♘a6 is a move it's important not to play. I once did against Ernesto Inarkiev, and White replies 10. ♕xb6 axb6 11. ♘a4 and suddenly Black has tactical problems. c) ♗f5 is Black's normal response, threatening to take on b3 and play Bc2, winning a pawn. c1) 10. ♕xb6 axb6 11. cxd5 ♘xd5 12. ♘xd5 ♖xd5 ( 12...cxd5 13. ♗f4 ♖c8 14. ♘e5 e6 15. f3 ♘c6 16. ♘xc6 bxc617. ♖d2 f6 18. e4 e5 19. ♗e3 Jankovic-Muslic, CRO 2001) 13.♘h4 ♖xd4 14. ♗e3 ♖d6∞

c2) 10. ♘e1 In a FIDE Grand Prix event, once more against Ernesto, I played 10... dxc4 which I think is a decent move.( 10...♗e6 is very solid and fine for Black. 11. c5 ♕xb3 12. axb3♘bd7 13. b4 a6= Anic-Nataf, Vichy FRA-ch Pool A 2000) 11.♕xc4 ♘a6 12. e4 White's only logical continuation. 12... ♗e6! (12... ♗g4? was a serious mistake. I thought it was interesting to draw the bishop to f3 and only then play Be6. 13. ♗f3 ♗e6 14.d5! White was better in Inarkiev-Svidler, Astrakhan FIDE GP 2010, as it transpires that after 14... cxd5 15. exd5 Black seriously misses the tempo he would win by playing Bg4 and hitting the rook on d1.) 13. ♕a4 ( 13. d5 cxd5 14. exd5 ♗g4 , and as I mentioned, the fact that White needs to take a tempo to deal with the threat to his rook leaves Black very comfortable.) 13... ♗g4There's a subtle difference between this position and the position with the queen on c4, but Black is fine here, as the following analysis shows: c21) 14. f3 e5! c22) 14. ♗f3 ♘c7 ( 14... ♗h3 15. e5 ♘g4 ) c221) 15. e5 ♗xf3 16. ♘xf3 ♘fd5 17. ♕c4 ♘e6 18.♘e4 ♕b4 19. ♕c2 ♖ac8 20. ♗e3 f5 21. ♘c3 ( 21.♘eg5 ♘xe3 22. fxe3 ♘xg5 23. ♘xg5 ♗h6 ) 21... f4⇆

c222) 15. ♗xg4 ♘xg4 16. ♘f3 ( 16. h3 ♖xd4 17.♖xd4 ♗xd4 18. hxg4 ♗xf2+ 19. ♔f1 ♗xe1 20. ♔xe1♕g1+−+ ) 16... e5 17. ♗g5 ♕xb2!∓ c223) 15. ♕c4 ♘e6 16. ♗e3 ( 16. ♗xg4 ♘xg4 17.♕e2 ♘f6 18. e5 ♘d5= ) 16... ♗xf3 17. ♘xf3 ♘g4 18.♕e2 ♘xe3 19. fxe3 c5∞

c224) 15. ♕a3 15... ♗xf3 ( 15... ♖xd4 16. ♗e3 e5 17.♗xd4 exd4 18. ♘a4 ♕a6 19. e5 ♗h3 20. ♗g2 ♗xg221. ♔xg2 ♘g4 22. ♘f3 ♗f8 23. ♕b3 c5⇆ ) 16. ♘xf3♘b5 17. ♘xb5 ♕xb5 18. ♖e1 e5! 19. ♗g5 ( 19. dxe5♖d3 ) 19... h6 20. ♗xf6 ♗xf6= c23) 14. ♘f3 c231) 14... ♕b4 15. h3 ( 15. ♕c2 ♗xf3 16. ♗xf3♖xd4 17. ♗e3 ♖xd1+ 18. ♖xd1 ♕c4 19. ♔g2 ♕e6= )15... ♕xa4 16. ♘xa4 ♗e6 17. ♘c3 ( 17. ♘g5 ♗c8 18.♘c3 ♘b4 19. ♘f3 ♘e8 20. ♖b1 ♗e6= ) 17... ♘b4 18.♘e1+/= c232) 14... e5! 15. dxe5 c2321) 15... ♘d7 16. ♗g5 ♖e8 17. ♖d2 ♘xe5 (17... ♗xf3 18. ♗xf3 ♘xe5 19. ♗e2 ♘c5 20. ♕c2± )18. ♘xe5 ♗xe5 19. ♗e3 ♕b4 20. ♕xb4 ♘xb4 21.a3+/= c2322) ♖xd1+ 16. ♕xd1 ♘d7 c23221) 17. h3 ♗xf3 18. ♕xf3 ( 18. ♗xf3 ♘xe519. ♗e2 ♖d8 20. ♕b3 ♘c5 21. ♕xb6 axb6= )18... ♘xe5 19. ♕e2 ♘c7 20. ♗e3 ♕b4 21. ♗f1♘e6=

c23222) 17. ♕c2 17... ♘xe5 18. ♗e3 ( 18. ♘xe5 ♗xe5 19. ♗e3 ♗d4= ) 18... ♕a5 19. ♘xe5♗xe5 20. h3 ( 20. f4 ♗c7 ) 20... ♗e6 21. f4♗c7 22. a3 ♗b6 23. ♕e2 ♗xe3+ 24. ♕xe3 ♕b6 25. ♕xb6 axb6 26. e5 ♘c5 27. ♖d1 f6= 9... ♕xb3 10. axb3 ♘a6 11. ♗f4 In this position Black has a lot of counterplay based around playing e5. 11. h3 ♘d7 12. ♗f4 ♘f8 13. ♖a4 ♗f5 14. ♖fa1 ♘e6 15. e3♗e4 16. g4 ♘xf4 17. exf4 ♗xf3 18. ♗xf3 ♘c7 19. f5=/+ 1/2 Tregubov-Vorobiov, Moscow Aeroflot op 2004 11... ♘d7 12. ♖fd1 12. ♖a3 ♖e8 13. ♗h3 f5 14. ♖c1 e5=/+ A.Petrosian-Lamprecht, Germany 1994 12... ♖e8 Making the threat of e5 very real. 12... e5⁈ is important not to play immediately, as I did against Aronian in the World Team Championship. 13. ♘xe5 ♘xe5 14. ♗xe5 ♗xe5 15.dxe5 In this position I realised that Nxc5 isn't quite right, so I played15... ♗f5 first, but after 16. ♖a5 ♗c2 17. ♖d4 ♗xb3 18. f4+/= I barely managed to draw. It's not very comfortable for Black. 13. e4 dxe4 14. ♘xe4 ♘b4 This position has seen serious discussion in grandmaster practice. I played it myself against Konstantin Sakaev and was fine, while it's also occurred in some of Gata Kamsky's games. Black is generally doing ok, as can be seen from the following analysis. 15. ♗d2 15. ♘c3 ♘f6 16. ♗e5 ♗e6 17. ♘d2 ♖ed8 18. ♘c4 ♘bd5 19.♖a3 h5 20. ♖da1 a6= Bocharov-Kamsky, Khanty Mansiysk 2005 15. ♗g5 h6 16. ♗d2 ♘d5∞ ( 16... ♘c2 17. ♖a4 f5 18. ♘c3 e519. dxe5 ♘xe5 20. ♘xe5 ♗xe5 21. b4∞ Tregubov-Vachier-Lagrave, Paris 2007) 15... ♘d5 16. ♘c3 ♘7f6 17. ♘e5 ♗e6 18. ♘c4 ♖ed8 19. ♘e2 19. ♗g5⁉ ♘b4 ( 19... h6 20. ♗xf6 ♘xf6 21. ♘a5 ♖ab8 22.♗xc6∞ ) 20. ♖a4 ♘fd5 21. ♗xe7 ♘xe7 22. ♖xb4 ♗xd4 23.♖xb7 ♗xc5= 19... ♗f5⁉ would be even better. 19... ♖d7 20. ♘a5 ♘e8 ( 20... ♗f5⁉ ) 21. ♗f3 ♖ad8 ( 21... h5 )22. ♗e1 ♖b8 23. ♘c4 b6 24. b4 ( 24. ♖a3 ♖db7 25. cxb6 axb626. ♘e5 ♖c7= was equal in Sakaev-Svidler, Moscow RUS-ch Super Final 2007) 24... ♘xb4 ( 24... ♖c7 25. ♘e5 bxc5 26. bxc5 ♖xb2 27.♖a6 ) 25. ♘xb6 axb6 26. ♗xb4 20. ♘c3 ♘b4 21. ♖a4 ♘c2 22. ♘a5 ♖xd4 23. ♘xb7 ♗g4=/+

6.b3 1. d4 ♘f6 2. c4 g6 3. ♘f3 ♗g7 4. g3 c6 5. ♗g2 d5 6. b3 O-O7. O-O There are of course many slow possibilities available to Black in this position like a7-a5, Bf5, Nbd7 or Ne4, but I feel it's better to treat this position forcefully. 7. ♗b2 doesn't change anything, since after 7... dxc4 8. bxc4 c5White can't play 9. d5 e6 10. ♘c3 exd5 11. ♘xd5 due to 11...♘xd5 12. ♗xg7 and the beautiful shot 12... ♘e3! , and Black is just better. 7... dxc4 A very forcing line. 8. bxc4 c5 Immediately attacking the white centre. 9. ♗b2 The main move here. 9. e3 ♘c6 10. ♗b2 ♕b6 gives Black a lot of play against the white centre. This is very similar to lines we'll discuss later, but if anything I think it's more comfortable for Black. 11. ♕c1 ( 11. ♕e2 cxd4 12.exd4 ♗g4 13. d5 ♗xf3 14. ♗xf3 ♘d4= ) 11... cxd4 12. ♘xd4♗d7= 9... ♕b6 9... cxd4 10. ♘xd4 ♕b6 is also possible immediately, but I think this is somewhat less precise. 10. ♕b3 This position contains quite a few pitfalls for Black. 10. ♕c1 is met by the simple 10... cxd4 ( 10... ♘e4 11. e3 ♘c6 12.♘e5 ♘d6 13. ♘xc6 bxc6 14. ♘d2 ♘f5 15. ♘b3 cxd4 16. c5 ♕c7 17. e4 ♘h6 18. ♗xd4 e5 19. ♗c3 ♗e6 20. h3 f6= Drasko-N.Nikolic) 11. ♘xd4 ♘c6 and the c1square isn't ideal for the queen and Black can safely simplify in the centre. 12. ♘xc6 bxc6 13. ♘d2♗e6 14. ♖b1 ♕a6 15. ♗a3 ♖fe8 16. ♖b3 ♘d7= Goldin-Yandemirov, Elista 1995 10... ♘e4 10... ♘fd7 looks like a decent move, but it runs into the very strong 11.♘c3! In general, you shouldn't allow any position where White gets the knight to d5, from where it can completely dominate the position. 11. e3 It's time to cement the centre somewhat. 11. ♘c3 ♘xc3 12. ♗xc3 cxd4 13. ♕xb6 axb6 14. ♗xd4 ♘c6=/+ 11... ♘c6 Threatening Na5 with problems for White, so he has little choice. 11... ♘d6 12. ♘c3 ♗e6 13. ♘a4 ♕d8 14. ♖fc1 b5 15. ♘xc5♗xc4 16. ♕c2± Burmakin-Anisimov, Pardubice Skanska op 2004 11... cxd4 12. ♗xd4 ♕a6 ( 12... ♕xb3 13. axb3 ♘c6 14. ♗xg7♔xg7 15. ♘d4+/= ) 13. ♗xg7 ♔xg7 14. ♘d4 ♘c5 15. ♕b2 ♕f616. ♘c3 e6 17. ♕a3 ♘bd7 18. ♖ab1 a6 19. ♖fd1+/= Nielsen-El Kher, Aarhus 1999 12. ♘e5

12. ♖d1⁈ ♘a5 13. ♕xb6 axb6 14. ♘a3 ♗e6 15. ♖ac1 ♘d6=/+ 12. ♘c3 is a possible try, but Black has a variety of options. I quite like the move 12... ♗f5 here. ( 12... ♕xb3 13. axb3 ♗f5∞ ) 13. ♕xb6 (13. ♘d5 ♕xb3 14. axb3 e6 15. ♘f4 cxd4 16. exd4 ♖fd8 17.♖fd1 ♘c5!=/+ ) 13... axb6 14. ♘d5 e6 15. ♘xb6 ♖a6 16. ♘d7♖d8 17. ♘de5 ♘xe5 18. ♘xe5 ♖b6 19. ♗a3 cxd4 20. ♗xe4♗xe4 21. ♗e7 ♖e8 22. ♗c5 dxe3 23. ♗xb6 e2 24. ♖fe1 ♗xe525. ♖xe2 ♗xa1 26. ♖xe4 ♖a8= 12. ♕xb6 axb6 13. ♘a3 ♗g4∓

12... ♘d6 A very good square for the knight, attacking the pawn on c4 and cementing Black's whole structure. 12... ♘xe5⁉ 13. dxe5! ( 13. ♕xb6 axb6 14. dxe5 ♘d6N 15. ♘d2♖a4⇆ ) 13... ♘d6 14. ♘d2+/=

13. ♘xc6 The main move here.

13. ♘d2 It's important to start with 13... cxd4 ( 13... ♗e6? allows White the forced variation 14. ♗xc6 bxc6 15. dxc5 ♕xc5 16. ♗d4and White wins some material. Something to avoid. 16... ♕a5 17.♘xc6 ♕c7 18. ♗xg7 ♔xg7 19. ♕c3+ ♔g8 20. ♘a5 ♖ab8 21.♖fc1 ♖fc8 22. ♕e5 ♗ h3 23. ♘ab3 ♖b4 24. c5 ♕b7 25. f3±Giri-Nielsen, Amsterdam 5th NH 2010) 14. exd4 ♗e6 and Black is fine. This was seen, for instance, in a game between Dmitry Jakovenko and myself some years ago. ( 14... ♘f5⁉ ) a) 15. ♖fe1 ♘a5 16. ♕xb6 axb6 17. d5 ♗f5 18. ♖ac1 ♖fc819. ♗f1 e6 20. ♗d4 exd5 21. cxd5 ♖xc1 22. ♖xc1 b5= b) 15. ♗xc6 bxc6 16. ♘xc6 ( 16. ♕xb6 axb6 17. ♘xc6 ♖fe8 )16... ♕xc6 17. d5 ♖fb8!=/+ c) 15. ♖fc1 15... ♖fc8 16. ♕xb6 ( 16. c5 ♗xb3 17. cxb6 ♗a418. bxa7 ♖xa7 19. ♘dc4 ♘xc4 20. ♖xc4 ♖d8 21. ♘xc6 bxc622. ♗xc6 ♗xc6 23. ♖xc6 ♗xd4 24. ♗xd4 ♖xd4 25. ♖c3 g5=Zhou Weiqi-Ding Liren, Danzhou 1st GM 2010) 16... axb6 17. f4 c1) 17... ♖a4 was a way to play on: 18. d5 ♘xe5 19. fxe5♗d7 20. exd6 ( 20. ♗f1 ♘f5 ) 20... ♗xb2 21. dxe7 ♗a3=/+ c2) ♘xe5 , and a draw was agreed in Jakovenko-Svidler, Sochi 17th RUS-tch 2010 13... bxc6 14. dxc5 14. ♕xb6 axb6 15. ♗xc6 ♖a7=/∞ ( 15... ♘xc4⁉=/∞ ) 14... ♕xc5 15. ♗xg7 ♔xg7 16. ♘d2 This is Boris Avrukh's recommendation in his book on the white repertoire against the Grünfeld, and as with everything else Boris does this has gained quite a cult following. I myself faced this in a game against Viktor Laznicka at the Khanty-Mansiysk Olympiad, where I had serious problems. In general I had to analyse this position carefully, as I realised that people do intend to play it with a view to getting an opening advantage. But after 16... ♗e6 Black is fine. 17. ♖fc1 The main move.

17. ♕a4 can just be met by 17... ♖fc8 18. ♖fc1 ( 18. ♖ab1 ♘xc419. ♖fc1 ♘b6 20. ♕a6 ♕d6 21. ♘e4 ♕e5= ) 18... ♖ab8 a) 19. a3 ♖b2 b) 19. ♖ab1 ♖xb1 20. ♖xb1 a5 ( 20... ♘xc4 21. ♕c2 ) 21. a3♕e5 c) 19. ♘b3 19... ♕e5 and Black is fine due to counterplay against the c4pawn. 20. ♕xa7 ♗xc4 21. ♘a5 ♖a8 22. ♘xc6 ♖xa7 23.♘xe5 ♖xa2= 17... ♖ab8 18. ♕d3 The main line, but since Avrukh's books on the Grünfeld for Black were published White has realised he has little here. 18. ♕c3+ is the move attention has shifted to. 18... ♔g8 a) 19. ♘f3 ♖fc8 ( 19... ♕xc4 20. ♕xc4 ♘xc4 21. ♘d4 ) 20.♕e5 ♕xe5 21. ♘xe5 c5 22. ♗d5 ♗h3 23. ♘c6 ( 23. ♖ab1♖xb1 24. ♖xb1 ♔g7 ) 23... ♖b7 24. ♖ab1 ♖cc7 25. ♖xb7♖xb7 26. e4 ♖b2= b) 19. ♖ab1 was played by Laurent Fressinet in a recent rapid game against Loek Van Wely. Black has a variety of responses, but I quite like the simple 19... ♖xb1⁉ ( 19... ♖fc8 20. ♖b3 a5 21. ♖cb1♖a8 , as in Fressinet-Van Wely, Bastia 2012, is not entirely satisfactory, but is also very solid.) 20. ♖xb1 ♖c8 It seems a bit strange just to give up the b-file, but there aren't actually any real entry points, and in order to start any kind of play at all White has to play something like 21. ♗f1 but after 21... f6 Black has a very safe position. There are no weaknesses and it's hard for White to come up with a plan. The only way White can shift Black's pieces from their very convenient posts is to play 22. ♘b3 but Black can simply respond 22...♕a3 23. c5 ♘b5 24. ♕d2 ♔g7 , and should hold quite comfortably here: 25. ♖c1 ( 25. ♘d4 ♘xd4 26. exd4 ♕xa2= )25... ♕a4 ( 25... ♖c7 26. ♗xb5 cxb5 27. ♘d4 ; 25... ♗xb3 26.axb3 ♕xb3 27. ♗xb5 ♕xb5 28. ♕d7 ♖f8 29. ♕xe7+ ♖f7 30.♕d6 a5= ) 26. ♘a5 ♖c7 27. ♗g2 ♘a3 and it's unclear whose pieces are more misplaced. c) 19. ♘b3 19... ♕xc4 20. ♕a5 ( 20. ♕e5 ♕b5 21. ♖c5 ♕b422. ♖xc6 ♖b5 23. ♕c3 ♗xb3= PertGriffiths, Sheffield 2011)20... ♕b4= 18... ♖fc8 This is my recommendation - a convenient solution that doesn't require the memorisation of the long, computer-like lines that follow 18...Qe5! 18... ♖fd8⁈ 19. ♕c3+ ♔g8 20. ♘b3 ♕xc4 21. ♕xc4 ♘xc4 ( 21...♗xc4 22. ♘a5 ) 22. ♘d4 ♘e5 23. f4 ♘d3 24. ♘xe6 fxe6 25.♖xc6 ♖b2 26. ♖xe6 ♔f8 gives Black good drawing chances, but there's no need for it. 18... ♕e5! is a very forcing improvement that I came up with after analysing my game vs Laznicka deeply. Play is very sharp and in some cases Black gives up the a7-pawn or even both the a7- and c6pawns. He gets a lot of piece activity in return, however, and I believe Black holds in the ensuing complications: a) 19. f4 ♕f5 a1) 20. ♕c3+ ♕f6 21. c5 ( 21. ♕xf6+ exf6 22. c5 ♘f5 23.♗xc6 ♘xe3 24. ♘b3 ♗f5 25. ♘d4 ♖b2= ) 21... ♕xc3 22.♖xc3 ♘b5 23. ♖cc1 ♘a3 24. ♗xc6 ♖b2 25. ♘f3 ♘c2 26.♖ab1 ♖xb1 27. ♖xb1 ♖c8= a2) 20. ♕a3 ♕f6 21. c5 ♕b2 ( 21... ♘b5= ) 22. ♕xb2+♖xb2 23. cxd6 ♖xd2 24. dxe7 ♖e8 25. ♗xc6 ♖xe7 26. a4♗d7=

a3) 20. ♕d4+ 20... ♕f6 21. ♕xf6+ exf6 22. c5 ♘f5 23.♗xc6 ♘xe3 24. ♘b3 ♗f5 25. ♗h1 ♘c2 26. ♖ab1 ♘b4 27.♖a1 ♘c2= b) 19. ♗xc6 ♖fc8 20. ♗d5 ♗xd5 21. cxd5 ♖b5 22. ♖xc8♖xd5 23. ♕c3 ♘xc8 24. ♕xe5+ ♖xe5= c) 19. c5 19... ♘b5 20. ♘b3 ( 20. ♗xc6 ♖fd8 ) 20... ♖fd8 21.♕e4 ♕b2! c1) 22. h4 ♗d5 23. ♕xe7 ♖e8 24. ♕d7 ♖bd8 25. ♕b7♗xg2 26. ♔xg2 ♖xe3=/+ c2) 22. ♕h4 ♘c3 23. ♖e1 ♘e2+ ( 23... ♗d5⁉ ) 24. ♔f1♘c3 25. ♔g1= c3) 22. h3 22... a5 ( 22... ♗d5 23. ♕xe7 ♖e8 24. ♕h4 ) 23.♕c2 ♕f6 24. ♘xa5 ♘a3 25. ♕a4 ♖b2 26. ♕f4 ♖dd2 27.♕xf6+ ♔xf6 28. ♘xc6 ♗f5! , preparing to cut off the rook on a1. 29. g4 ( 29. ♖c3 ♘b1 ) 29... ♗b1! c31) 30. ♘d4 e5 31. ♘f3 ( 31. c6 ♖xf2 ) 31... ♖dc2 32.♖d1 ♖xf2= c32) 30. ♗h1 30... ♖dc2 31. ♖d1 ♖xc5= 19. ♘b3 ♕b4! As correctly recommended by Boris Avrukh. 19... ♕a3⁈ is the move I immediately went wrong with against Laznicka, and after 20. c5 ♘b5 I thought Black was just fine, but instead White has the incredibly strong 21. ♕e4! and once the queen gets to e5 Black actually runs into quite serious problems.The queen on e5 is a powerful piece that controls the entire board. White gets play everywhere, both on the kingside and the queenside. Laznicka-Svidler, Khanty-Mansiysk 39th Olympiad 2010. 20. ♕d4+ 20. c5 ♘b5 is now no longer a problem, as the queen on b4 covers all the squares by which the white queen could try to get to e5. Black has more than enough counterplay. 20... ♔g8 20... f6? 21. ♕xa7 ♖b7 22. ♕c5 ♘xc4 23. ♘d4± 21. ♕xa7 ♘xc4 22. ♕c5 f6 Optically White seems to be better because of the outside passer and the apparent pressure on the queenside, but Black is in fact comfortable here. White has no way of improving his position without giving up some trumps. Black is ready, for instance, to play Kf7 next move and carefully prepare to simplify on the queenside. If White plays the logical 23. ♖c2 , aiming to double on the c-file, Black can force equality with 23... ♕xc5! 24. ♘xc5 ♘a3 25. ♖c3 ♖b1+ 26. ♖xb1 ♘xb1 27.♖c2 (otherwise the pawn on a2 will fall) 27... ♘a3 28. ♖d2 ♗c4 and the a2-pawn is safely blockaded while Black is ready to improve his position further. Although White's position once again looks very nice there's no way to pose Black any particular problems. Black is fine here.

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Bf4 Bg7 5.Rc1 5.Nf3

5.Rc1 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗f4 ♗g7 5. ♖c1 To be honest, you're unlikely to face this move very often, but it's a fun and instructive line to look at. It's a rare occurrence in modern theory that as early as move five Black has an option that leads to unclear and double-edged positions. I'm talking, of course, about 5... ♘h5 , which should be mandatory for anyone playing Black here, and something to welcome with open arms. 6. ♗d2 The main move, giving Black a choice of whether to start with 6...c5 or 6...dxc4. Both will likely lead to the same position, although c5 at least gives White a choice. White has two and a half options. The half is 6. ♗e3 , a really odd-looking move once played by Ivan Sokolov, but after 6... dxc4 7. ♕a4+Black's simplest response would have been 7... ♘c6 8. ♘f3 (of course 8.d5 ♗xc3+ isn't an option) 8... O-O 9. ♕xc4 ♗e6 10. ♕c5 ♕d6=/+Black is clearly fine. His pieces are better and more harmoniously developed while the bishop on f1 will still need quite some time to join the fray. 6. ♗g5 h6 a) 7. ♗d2 is better than 7.Bh4, but Black, as well as being able to play in the same way as after 6.Bd2, has an additional option. After7... dxc4 8. e3 apart from the move c7-c5, which should be fine, Black can also just hang on to the pawn with 8... ♗e6 9. ♘f3 c6 As Ng5 is no longer possible with the pawn on h6 White will have to allow the black pawn to remain on c4. 10. ♘e4 ♗f5 Perhaps even stronger than the natural-looking ( 10... ♗d5 11. ♕c2 b5 PinterPopovic, Thess ol 1988 12. ♘c5 ♗xf3 13. gxf3 ♘d7 14. ♘xd7 ♕xd7 15.b3+/= ) 11. ♘c5 (If 11. ♘g3 Black manages to trade off the knight on h5, which must count as an achievement given the knight was doing little there. 11... ♘xg3 12. hxg3 b5 Of course the position after the possible 13. b3 cxb3 14. ♕xb3 isn't completely clear. White has decent compensation for the pawn due to Black's weakened queenside and the undeveloped knight on b8. The position is still full of life, but in general I think it would be wrong for Black to avoid lines like this in the Grünfeld - it's not often you get such unclear and lively positions.) 11... b6 Playing this move first gains Black tempi. Only after 12. ♘a4 does Black follow up with 12... b5 b) 7. ♗h4 is the most consistent and popular move here. 7... c5 8.e3 ( 8. dxc5 d4 is very dangerous without a bishop on d2 to control the e1-a5 diagonal.) 8... cxd4 b1) 9. ♘xd5 ♘c6 10. exd4 ♗e6 A saner and better choice for Black, who is probably already slightly better as the white centre is in a lot of trouble here. (I played 10... ♗xd4 and I want to show you what happened, not because of its theoretical importance but because of the position after Black's 19th move. 11. ♕d2 g5 A long sequence follows where the players attack the opponent's pieces rather than defending their own. 12. ♗e2 ♘f6 13. ♘f3♗b6 And now a series of captures: 14. ♘xf6+ exf6 15. ♕xd8+♗xd8 16. ♗g3 ♗a5+ 17. ♔f1 f5 18. ♗d6 ♗e6 19. ♘e5O-O-O=/+ It's rare that Black manages to castle queenside on move 19 in a Grünfeld. A fun position to look at. Black is slightly better here although I eventually almost lost the game. It was drawn after a lot of adventures. I. Sokolov-Svidler, Wijk aan Zee 2004) 11.♘f3 O-O

b2) 9. exd4 b21) 9... dxc4 A slight mistake. 10. ♘f3 (After 10. d5! O-O11. ♗xc4∞ the position is quite unclear and White is at least fine.) 10... O-O 11. ♗xc4 ♗g4 12. O-O ♘c6 13. d5♘d4=/+ Dzagnidze-Svidler, Gibraltar Masters 2012 b22) 9... ♘c6 10. ♘f3 ( 10. cxd5 ♘xd4 ; 10. ♘xd5 ♗e6transposes to 9.Nxd5.) 10... ♗g4 11. cxd5 ♘xd4 12. ♗e2clearly favours Black, (so White plays 12. ♕a4+ and after 12...♗d7 13. ♕d1 Black can take an immediate repetition or play13... ♘f5 , when I believe he's better. The bishop on h4 is about to be traded off for one of the knights, giving Black the two bishops, a better structure and the better chances. Paunovic-Krnic, Brezovica 1988) 6... dxc4⁉ 6... c5⁉ a) 7. cxd5= cxd4 8. ♘b5 ♘a6 b) 7. e3 cxd4 8. exd4 dxc4 - 6...dxc4 ( 8... ♘c6 9. cxd5 ♘xd410. ♘ge2∞ ) c) 7. dxc5 7... d4 c1) 8. ♘e4→ a5! ( 8... O-O∞ ) c2) 8. ♘d5→ e6 9. ♘b4 OO 10. ♘f3 ( 10. g3 a5⁉ 11. ♘d3e5 12. ♗g2 ♕e7 ) 10... a5 ( 10... e5⁉∞ ) 11. ♘d3 ♘c6=/∞ /\ e5-e4 c3) 8. ♘a4 The best square for the knight, and the move White should play if he's in an adventurous mood. This leads to very, very interesting positions in which Black has ample compensation for the pawn. 8... ♘f6 (After let's say 8... OO 9. g3 e5 10. ♗g2♘f6=/∞ the position is very double-edged and interesting to look at. I encourage you to do so simply because I think you'll have fun, although you can avoid these lines by starting with 6...dxc4.) 9.g3 ♗d7 10. b4 ♘e4 11. ♘f3 ♘xd2 12. ♘xd2 ♗c6 13. ♘f3d3 14. exd3 ♗h 6 15. ♖b1 ♕d7 16. ♘c3 ♕e6+ 17. ♘e4♗xe4 18. dxe4 ♕xe4+ 19. ♗e2 ♘c6 20. ♕d3 ♕xd3 21.♗xd3 O-O-O 22. ♗e4+/= Komarov-Mikhalevski, Kiev 1995 7. e3 c5 8. ♗xc4 The only serious move in this position. 8. d5⁉ a) 8... e6⁈ A pitfall that needs to be avoided, because after 9.♕a4+! you pretty much have to play 9... ♗d7 and after 10. ♕xc4 I was no longer able to find equality for Black. 10... b6 11. b4⁉ O-O (11... cxb4 12. ♘b5± ) 12. bxc5 bxc5 13. ♕xc5± b) 8... O-O+/= 9. ♗xc4 c) 8... ♘d7∞ 9. ♗xc4 ♘b6 10. ♘f3

d) a6= A simple approach. 9. a4 ( 9. ♗xc4 b5 leads to a very unclear position that Black should welcome with open arms. 10. ♗e2♘f6 11. e4 OO ; 9. ♕a4+? b5 10. ♘xb5 ♗d7 11. ♗xc4♕b6∓ ) 9... e6 10. ♗xc4 ( 10. e4⁈ exd5 11. exd5 OO ) 10...exd5 11. ♘xd5 ( 11. ♗xd5 OO 12. ♘f3 ♘f6 13. ♗c4 ♘c6∞ Black is obviously doing excellently here.) 11... OO 12. ♗c3 ( 12.♘f3 ♗e6 13. e4 b5 ) 12... ♗e6 Black is ready to undermine White's entire setup with b7-b5 or Nf6 in some positions. At the very least Black is no worse here. 8. dxc5 is well met by 8... ♘d7 and after the possible 9. ♕a4 O-O 10.♕xc4 ♘e5 Black gets a classical pawn sacrifice which is very typical of many lines in the Grünfeld where Black sacrifices the c-pawn for open diagonals and a lead in development. As usual in those positions, Black is fine here and shouldn't be at all worried about the prospect of being a pawn down for a while. 11. ♕d5 ♗e6 Schlosser-Groszpeter, Munich 1989 8... cxd4 9. exd4 The main position of the whole Rc1 line. 9... ♗xd4=/∞ is an additional possibility I want to propose for Black. After all, if you're offered a whole pawn why not at least consider taking it? 9... O-O The reasonably sane choice. a) 10. ♘f3 ♗xd4= ( 10... ♗g4 11. d5 ♘d7 12. O-O ) b) 10. ♗e3⁈ ♘c6 11. ♘f3 ♗g4 c) 10. d5 10... ♘d7 11. ♘f3 a6! The thematic break. c1) 12. a3⁈ b5 13. ♗a2 ♘b6 14. O-O ♗b7 15. ♗g5 ♘f6=/+ c2) 12. ♗e3⁉ The main move, in response to which it's better not to immediately commit to b5 but to play 12... ♘hf6! waiting for13. O-O⁉ ( 13. a4= once again runs into 13... b5! 14. axb5♘b6 Black has fantastic compensation for the pawn and is likely to win it back in the immediate future. ) 13... b5 14. ♗b3 ♗b715. ♖e1 (If White could get his knight from f3 to c6 he'd have a large advantage, but it's impossible here. If 15. ♘d4 Black has15... b4 and the pawn on d5 is lost. 16. ♘c6 doesn't help. 16...♗xc6 17. dxc6 bxc3 18. cxd7 cxb2∓ ) 15... ♖c8∞ We've reached a very double-edged unclear position which should be welcomed by any Grünfeld player. Black's ideas include Nc5 and b4.(15... ♖e8∞ ) c3) 12. O-O⁉ b5 13. ♗b3 ♘c5 c4) 12. a4= is well met by 12... b5 13. axb5 ♘b6 and the entire queenside is about to be liquidated, with complete equality. 14. b3axb5 9... ♕xd4⁉ 10. ♘b5 The move White will probably play, creating the threat of Nc7+. You can castle here. 10... ♗b6⁉ The most adventurous move in this position.

10... O-O⁉ 11. ♘c7 (Given the problems in the 11.Nc7 line White is likely to play 11. ♘xd4 and after 11... ♕xd4 White will of course have ample compensation for the sacrificed pawn, but on the other hand Black is currently better developed and by the time White has finished his development I think Black should have been able to coordinate his position - developing the knight to c6 and the bishop to e6 or g4. Black should be fine here.) 11... ♗xf2+! A very nice geometrical tactic. 12.♔xf2 ( 12. ♔f1 ♕xc7 13. ♗xf7+ ♖xf7 14. ♖xc7 ♗b6+ The rook on c7 drops off with discovered check and Black is just completely winning.) 12... ♕xc7 13. ♗xf7+ ♖xf7+ and it's check so the queen can't be taken. 11. ♕b3 If White does something slow like 11. ♘f3 ♘c6 12. O-O O-O I believe Black may well already be better. He'll drive the knight back by playing a6 and start to get his pieces into good positions - plus he is, after all, a pawn up. So White needs to hurry. 13. ♕e2 a6 14. ♘c3 ♗g4 15.♘d5 ♗d4 16. ♗h6 e6 17. ♘e3 ♗xf3 18. ♕xf3 ♘g7=/+ 11... O-O 12. ♗h6 ♘c6 Giving up the exchange on the spot is preferable. Black should avoid playing 12... ♘g7 because that allows White to demonstrate his main idea with 13. ♕c3 forcing Black to play the undesirable 13... e5 14. ♘f3 ♘c6 15. O-O The white rooks are now ready to join the battle on the d- and e-files, with Rd1 in particular threatening to plant the knight on d6. White has tremendous compensation for the pawn. 13. ♗xf8 ♕xf8 14. ♘f3 ♘f4 15. O-O ♗e6 Black has excellent compensation for the exchange and is doing very well here. I sincerely recommend this position to anyone who wants slightly more exciting play after 8...cxd4 9.exd4.

5.Nf3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗f4 ♗g7 5. ♘f3 O-O 6. ♖c1 dxc4 The main line. 7. e4 7. e3 a) My initial proposal was to play 7... ♘bd7⁈ 8. ♗xc4 c5 This position could have been reached in a game between Nakamura and Ivanchuk in the 2011 Tal Memorial, so I'm clearly not the only one who thought this might be playable for Black. Unfortunately the fact checkers found out that after 9. O-O cxd4 White has the somewhat alien-looking 10.Nb5. a1) 10. exd4 ♘b6 11. ♗b3 a11) 11... ♗g4 12. h3 ♗xf3 13. ♕xf3 ♕xd4 ( 13... ♕d714. d5+/= ) 14. ♗e3 ♕h4 15. ♕xb7± a12) ♘bd5 a121) 12. ♕d2 a1211) 12... ♘xf4 13. ♕xf4 b6 14. d5 ( 14. ♘e5♗b7 15. d5 ♘xd5 16. ♘xd5 ♗xd5 17. ♖fd1 e6 )14... ♗b7 15. ♖fd1 ♖c8 16. ♘e5 ♘d7 17. ♘xf7♖xf7 18. d6 ♘f6

a1212) ♗e6 a12121) 13. ♗h6 ♘xc3 ( 13... ♖c8 14. ♗xg7♔xg7 15. ♖fe1 ♘xc3 16. bxc3 ♗c4 17. ♗xc4♖xc4 18. ♘e5 ♖c8 19. c4 ) 14. bxc3 ♗d5 15.♗xg7 ♔xg7 16. ♗xd5 ♕xd5 17. c4 ♕d6 18.♖fe1 e6= a12122) 13. ♗e5 13... h6 14. ♖fe1 1/21/2 Portisch, F-Adorjan,A, Budapest 1970 a122) 12. ♗g5 12... ♘xc3 13. bxc3 b6 14. ♖e1 ♗b715. ♘e5 ( 15. c4 ♘e4 16. ♗f4 e6 17. ♗c2 ♘f6 ) 15...♖c8 16. ♕d3 ♕d6 17. ♗h4 e6 18. ♗g3 ♕e7 a2) 10. ♘b5! And after 10... dxe3 there's the even more alien-looking 11. ♘g5‼± and here the machine suddenly starts going red and claiming White is winning. It's not immediately obvious why, but generally speaking when it starts to say it's winning it probably is, so this line isn't particularly playable. b) 7... ♗e6 This move is recommended by most sources, including Avrukh's excellent book on the Grünfeld, and after the forcing sequence 8. ♘g5 ♗d5 9. e4 h6 10. exd5 hxg5 11. ♗xg5 ♘xd512. ♗xc4 ♘b6 13. ♗b3 ♘c6 Black is currently doing very well in this well-known theoretical position. I by no means want to say that the conclusions reached by Avrukh are wrong, but I think your opponents will expect this of you so it's worth being aware of other options. c) c5! is my recommendation. 8. ♗xc4 cxd4 9. ♘xd4 This line has a somewhat mixed reputation for Black, as on the one hand it's extremely solid, but on the other hand, due mainly to the efforts of Grandmaster Alexey Dreev, Black has been suffering a little in the positions after 9... ♗d7 ( 9... ♘bd7 10. ♗g3 ♘b6 11. ♗b3♗g4 12. f3 ♗d7 13. OO ♘h5 14. ♗h4 ♕e8 15. a4 ♖c8 16.a5 ♘c4 17. ♗xc4 ♖xc4 18. ♕b3 ♗xd4 19. exd4 ♖xd4 20.♗f2 ♗e6∞ Dreev-Azmaiparashvili, Pula 1999) c1) 10. O-O If 10... ♘c6 then after 11. ♘b3 ♖c8 ( 11... ♘a512. ♗e2 ♖c8 13. ♗f3 b5 14. ♕e2 b4 15. ♘d5 ♘xb3 16.axb3 ♘x d5 17. ♗xd5 ♕b6 Dreev-Krasenkow, Dos Hermanas 2001) 12. h3 a6 13. ♗e2 it's very important that Black is just in time to play 13... ♘a5 trading off a pair of knights. The moment pieces start coming off the board in this line it's quite clear that White can't hope for a lot. 14. ♗e5 1/2 Malakhov-Caruana, Rogaska Slatina 27th ECC 2011 c2) 10. ♘b3 It's important to start with this very precise move. c21) 10... ♕b6 11. O-O ♘a6 I played this in an exciting game against Joel Lautier in 2001. I was attempting to avoid the normal positions with the knight on c5. This line can't be recommended, however. c211) 12. ♗e5 ♖fd8 13. ♕f3 ♗c6 14. e4 ♖d7 15. h4♖ad8 16. h5 ♕b4! ( 16... ♘b4 17. hxg6 hxg6 18. ♕f4a5 19. ♕g5 ♗xe4 20. ♘a4 ♕a7 21. ♘ac5 ♖d5⇆Lautier-Svidler, Biel 2001) c212) 12. e4!+/= and Black doesn't equalise.

c22) 10... ♘c6 11. ♘c5 c221) 11... ♗c8 The normal move, but after 12. ♕xd8 (12. OO ♘d7 13. ♘5e4 ♘de5 14. ♗e2 1/2 Dreev-Smirin, Sarajevo 2001) 12... ♖xd8 13. ♗c7 this led to a small but very stable advantage for White in a recent game between Dreev and Kurnosov. 13... ♖f8 14. OO ♘d7 15. ♘b3♘de5 16. ♗b5 ♗f5⁉ ( 16... ♗e6 17. ♘c5 ♖ac8 18.♗xe5 ♘xe5 19. ♘xe6 fxe6 20. ♘e4± Dreev-Kurnosov, Apatity Russian Rapid GP Final 2011) 17. ♘c5 ♖ac8 18.♗xe5 ♗xe5 19. ♘d5 ( 19. ♘xb7 ♘b4= ) 19... e6 20.♘c3 ( 20. ♗xc6 exd5= ) 20... ♘b8 21. ♘xb7 ♖c7 22.♘a5 ♖fc8 23. f3 g5 24. ♘b3 ♗xc3 25. bxc3 ♖xc3 26.♖xc3 ♖xc3 27. e4 ♗g6 28. ♖d1 ♘c6= c222) ♗f5⁉ An important improvement that I think changes the current assessment of this line as one giving White a stable advantage. The machine insists Black is doing well after both White's options. c2221) 12. ♕xd8 ♖fxd8 (not, of course 12... ♘xd8which is very passive and hardly the point of the previous play) 13. ♘xb7 ♖dc8 and Black has wonderful compensation for the pawn because White's position is somewhat unstable. He'll have to look for posts for all his minor pieces, while the black knights can jump to h5, b4 or e4. 14. O-O ( 14. b3 ♘h5 15. OO ♗xc3 16. ♖xc3♘b4 17. ♖a1 ♘xf4 18. exf4 ♔g7⇆ ) 14... ♘b4 15.b3 ♘e4 16. ♘xe4 ♗xe4 17. ♘a5 ♘xa2 18. ♖cd1♗c3= c2222) 12. OO 12... ♖c8 13. ♕e2 ♕b6 14. ♘5a4♕a5 15. ♖fd1 a6 16. ♗b3 ♖fd8 17. h3 ♖xd1+ 18.♖xd1 ♖d8= 7... c5 Official theory recommends going for 7... ♗g4 8. ♗xc4 ♘h5 9. ♗e3♗xf3 10. gxf3 e5 and once again I fully believe that Black is fine here. The positions starting with 11. dxe5 usually lead to quite unexciting endgames in which White will be pretending for a while that he has some pull. Generally speaking, those endgames are objectively equal, but why give White the satisfaction when there's a different choice available? 8. dxc5 ♕a5 9. e5 ♘h5 The old theory here is 9... ♖d8 10. ♗d2 ♘g4 and for many years this was supposed to be quite good for Black, but then people realised that after 11. ♗xc4 ♕xc5 12. ♘e4 something like 12... ♕b6 would be met by 13. ♗xf7+ ♔xf7 14. ♖xc8 ♖xc8 15. ♘fg5+ and Black is close to completely lost. 10. ♗e3 The main move. 10. ♗d2 ♕xc5 ( 10... ♗e6⁉ is also an option. 11. ♘e4 ♕a6 12.♕c2 ♘c6 13. ♘eg5 ♘xe5 14. ♘xe5 ♗xe5 15. ♘xe6 ♕xe6 16.♗xc4 ♕c6 17. OO ♘f4= ) 11. ♗e3 ♕a5 Black needs to display a degree of precision after both of White's possible moves here, but he should be fine in both cases. 12. ♗xc4 ( 12. ♕d5 ♕xd5 13. ♘xd5♘c6 14. ♗xc4 ♗g4= ) 12... ♘c6 13. OO ♘xe5 14. ♘xe5 ♕xe515. ♖e1 ♘f6 16. ♗d2 ♕b8! 17. g3 ( 17. ♖xe7 ♕d6!=/+ /\ Ng4)17... ♗g4⇆

10... ♖d8 11. ♘d2 I think this line is even more attractive for Black, because in many cases he gets a lot of counterplay and real chances of an advantage.

11... ♗xe5 11... ♘c6 12. g4 ♘xe5 13. gxh5 ♗g4 14. ♗e2 ♘d3+ 15. ♔f1♗xe2+ 16. ♘xe2 ♘xc1 17. ♕xc1 ♕xa2⇆

12. ♗xc4 ♘c6 13. ♕b3

If White plays passively with 13. O-O I already prefer Black's position. a) 13... ♗f4⁉ b) 13... ♕b4⁈ 14. ♘d5 ♕xb2 15. ♖b1 ( 15. ♗b3⁉ ) 15... ♕a316. ♖b3 ♕a5 17. ♕c1 c) 13... ♕c7 14. ♘d5 ( 14. h3 ♘f6 The simplest response, and Black has achieved a perfect Grünfeld setup. The pawn on c5 means the knight is very comfortable on c6. The bishop will come to f5 and, generally speaking, I would much rather be Black in this position. 15.♕b3 ♘a5 ) 14... ♗xh2+ ( 14... ♕b8⁉ White's activity is just optical while Black is a pawn up) 15. ♔h1 ♕b8 13... ♘d4 A very interesting continuation, forcing White to take. 13... ♘f4⁉ 13... e6 14. ♘f3 ♗f4 Dreev-Zvjaginsev 14. ♗xd4 14. ♗xf7+ ♔g7 is completely inadvisable, and 15. ♕c4 even loses after15... b5 16. cxb6 ♗a6 14... ♖xd4 15. ♘f3 Almost forced, to which Black can reply with the very interesting and ambitious 15... ♖e4+ 16. ♔f1! ♖xc4! 17. ♕xc4 17. ♘xe5 ♖f4 17... ♗e6 Here I believe Black is at least no worse. 18. ♕e4 The best square for the queen. 18... ♗xc3 And now Black is fine after either recapture. 19. bxc3 19. ♖xc3 ♕xa2 , with a lot of play and the queenside about to collapse. White is completely uncoordinated here. 19... ♕xc5=/∞ Once again Black quite clearly has fantastic compensation for the exchange, which should be at least enough for equality and possibly even more. Black has nothing to fear in a position like this.

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Bf4 Bg7 5.e3 0-0 6.cxd5 6.Rc1 6.Nf3

6.cxd5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗f4 ♗g7 5. e3 O-O 6. cxd5♘xd5 7. ♘xd5 ♕xd5 8. ♗xc7 Old theory that's long since been known to be harmless for Black. 8... ♘a6 It's inadvisable to keep the queens on by retreating the bishop here, while the following move leads to an endgame almost by force. 9. ♗xa6 ♕xg2 9... bxa6 10. ♘f3 ♕b7 11. ♗e5 ♗xe5 12. ♘xe5 ♕xg2 13. ♕f3♕xf3 14. ♘xf3 ♖b8 15. b3 ♗b7 16. ♔e2 a5 17. ♖hc1 ♖fc8 18.♘e5 ♗a6+ 19. ♔d2 a4 20. ♘d7 Dreev-Svidler, playchess.com 2004 10. ♕f3 ♕xf3 11. ♘xf3 bxa6 Here White has half an extra pawn because the pawns on a7 and a6 aren't ideal, but on the other hand Black has the bishops and potential activity. In general, if the position opens up Black shouldn't suffer at all here - and according to the current state of theory the position is going to be opened. 12. ♖g1 12. ♖c1 a) 12... ♗b7 13. ♔e2 f6 ( 13... ♖ac8 14. ♖hd1 f6 15. d5 f516. b4 ♗b2 17. ♖c5 a5 18. ♗xa5 ♗c3!∓ Vallejo Pons-Mikhalevski, Calvia 2007) 14. ♖c5 e6 15. ♖hc1 ♖f7 16. ♗g3♖d8 17. ♖c7 ♖dd7 18. ♖xd7 ♖xd7 19. ♖c7 ♖xc7 20. ♗xc7g5 21. ♘d2 ♗f8 22. f3 h5 Gonda-Navara b) f6 13. ♖g1 transposes to 12.Rg1. ( 13. ♔e2 runs into 13...♗g4 14. ♖hg1 ♗h5 This is surprisingly awkward to deal with and shouldn't be allowed by White.; 13. ♖c5 ♖f7 14. d5 ♗f8 15.♘d2 ♗b7 16. OO ♖c8 17. e4 e5 18. d6 ♖d7 19. ♘c4 ♗xe4 Sambuev-Kurnosov, Voronezh 2004) 12... f6 13. ♖c1 ♖f7 13... ♗b7 14. ♔e2 ♖ac8 15. ♖gd1 ♖f7 16. ♗a5 ♖ff8 17. ♖c5e5 18. ♖dc1 exd4 19. ♘xd4 ♖xc5 20. ♖xc5 ♖c8 1/2 Wang Yue-Jakovenko, Nanjing 2009 14. ♔e2 ♗d7 Now Black has the additional idea of Bb5+ in many lines, while his main idea is of course ...e5, which he'll play against almost any move in this position. That opens the position and when Black's bishops start playing to their full potential it becomes obvious that Black has no problems whatsoever. 15. ♖c3 15. ♘d2 e5 16. d5 ♗b5+ 17. ♔f3 f5 18. b3 e4+ 19. ♔g2 ♗b220. ♖c2 ♗d3 21. ♖c6 ♗b5 22. ♖c2 ♗d3 23. ♖c6 1/2 Radjabov-Ivanchuk, Morelia/Linares 2006

15... e5 16. ♗d6 ♖d8 17. ♗c5 ♗b5+ 18. ♔d2 f5 19. ♔c1 exd420. ♗xd4 f4 21. ♘g5 fxe3 22. fxe3 ♖e7 Postny-Tseitlin, Haifa 2010

6.Rc1 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗f4 ♗g7 5. e3 O-O 6. ♖c1 A serious alternative to 6.Nf3. 6... ♗e6 is the move I'm recommending here. This also isn't a novelty and has been tried in a number of recent high-level games, most notably in a game between Magnus Carlsen and the Chinese grandmaster Wang Yue. 6... dxc4 7. ♗xc4 ♘bd7 8. ♘f3! This is the problem with this setup for Black, and transposes to a line we gave earlier. (After Nakamura's 8.♘b5⁈ here against Ivanchuk the game immediately assumed very strange proportions. Black was fine in a very interesting fight and eventually even won. 8... c6 9. ♗c7 ♕e8 10. ♘c3 e5 Nakamura-Ivanchuk, Moscow 6th Tal Mem 2011 11. ♗d6 exd4 12. ♗xf8 dxc3 13. ♗xg7cxb2 14. ♖b1 ♔xg7 15. ♘f3 b5 16. ♗e2 ♘c5 17. ♖xb2 ♘a4 18.♖d2 ) 8... c5 9. O-O cxd4 10. ♘b5! dxe3 11. ♘g5‼ and White wins. The main line here starts 6... c5 7. dxc5 ♗e6 and Black is currently doing reasonably well, but as usual I decided to look for something else which is playable but doesn't involve a lot of forced theory and/or slightly passive positions, although that isn't always avoidable. 7. ♕b3 7. c5 c6 a) 8. ♗d3 ♗g4⁉ Magnus immediately grabs the opportunity to put the bishop somewhere other than e6.( 8... b6 9. b4 a5 is another plan.) 9. ♕c2 ( 9. ♘ge2 ♘bd7 ) 9... ♘fd7⁉ Strong and original.10. ♗xb8⁈ (If 10. h3 then Black can already play 10... e5! ) 10...♘xb8! It's very rare to see a knight on b8 that didn't start there. The point of this move only becomes obvious a couple of moves later. (10... ♖xb8 11. h3 ♗e6= ) 11. h3 ♗c8 12. f4 Trying to stop the break in the centre. 12... b6 13. ♘a4 e5!⇆ But the break comes anyway. Magnus had fantastic compensation for the pawn and was somewhat unlucky not to win the game. b) 8. h3⁉ is a move I was mildly worried about, trying to restrict the bishop and get a favourable version of some kind of Slav, but I finally spotted improvements for Black in the play that follows. 8... b6 9.b4 b1) 9... ♘e4 10. ♘f3 ( 10. ♗d3 a5 11. ♘a4 ♘d7 12. f3 b513. fxe4 bxa4 14. b5 dxe4 15. ♗xe4 cxb5 16. ♘f3 ♗xa2 17.♕e2 ♗c4 18. ♖xc4 bxc4 19. ♕xc4 ♖a7∞ Jumabayev -So, Istanbul 40th Olympiad 2012) 10... a5 11. ♘a4 ♘d7? ( 11... b512. ♘b6 ♖a7∞ 13. a4 axb4 14. axb5 ♘c3 15. ♕b3 cxb5 )12. b5! bxc5? ( 12... cxb5 13. ♗xb5 bxc5 14. dxc5± ) 13.bxc6 g5 14. ♘xg5 ♘xg5 15. ♗xg5 ♘f6 16. ♘xc5 ♕d6 17.♗b5+− 1-0 Timoscenko-Michalik, Banska Stiavnica 2006 b2) a5 Initially I thought Black's position was dubious here, but after b21) 10. a3 Black just swaps everything with 10... axb4 11.axb4 bxc5 12. bxc5 and then has the important resource 12...♖a3 Black is first to seize the initiative on the queenside. Qa5 will follow, then perhaps Ne4 or Na6-

b4.( 12... ♘e4 13. ♗d3♗f5 14. ♘ge2 ♘d7 15. OO e5 16. ♗h2 ♘xc3 17. ♖xc3exd4 18. ♘xd4 ♗xd3 19. ♕xd3 Wohl-Deepan Chakkravarthy, Nagpur 2008 19... ♗e5!= ) 13. ♘b1 ( 13. ♗d3 ♕a5 14.♘e2 ♘a6 15. ♗xa6 ♕xa6 16. OO ♖a8=/+ ) 13... ♕a5+14. ♕d2 ♕xd2+ 15. ♔xd2 ♖a2+ 16. ♔e1 ♘bd7

⇆

b22) It's also important to know that after 10. b5 bxc5 11.dxc5 b221) Black has the strong 11... a4 and White already needs to tread very carefully, as can be seen in the line b2211) 12. b6⁈ ♘bd7 13. ♘ce2 ( 13. ♘xa4 White is seemingly just a pawn up for nothing very much here, but after 13... ♖a5!∓ he no longer has equality as there's no adequate defence against the very strong threat of Qa8. The whole edifice is about to collapse completely.) 13...♘e4 14. ♘d4 ♕c8 b2212) 12. ♘f3 12... cxb5 13. ♘xb5 ♘e4 14. ♗c7♕d7⇆

b222) ♘h5 12. ♗h2 ( 12. ♗xb8 ♕xb8 13. ♘f3 ♗xc3+14. ♖xc3 cxb5 Retera-Langheinrich, Netherlands 2006) b23) 10. cxb6 The main move. 10... ♕xb6 ( 10... axb4 11.♘a4 is unsatisfactory for Black as the pawn on b6 will be protected by Bc7, leaving White with a huge trump on the queenside.) 11. b5 There have already been games that reached this position. Black failed to equalise comfortably, but I'm offering an improvement. 11... a4 12. ♗d3 White generally wants to develop his knight not to f3, but e2, connecting the knights. 12... cxb5 13. ♘xb5 ♘a6! An important move which poses White some concrete problems. ( 13... ♗f5 14. ♘e2♕a5+ 15. ♘ec3 ♗xd3 16. ♕xd3 ♖c8 17. OO ♘bd7 18.♖c2+/= Ortiz Suarez-Krysa, Mar del Plata 7th Americas Conti 2012) b231) 14. ♘e2 runs into 14... ♕a5+ forcing 15. ♘bc3 (15. ♕d2? ♘b4 ) 15... ♘b4 16. ♗b1 a3 A very important milestone for Black, as it means the knight on b4 can't be challenged and driven away. This is a very unclear position in which I believe Black has nothing to fear. 17. OO ♖fc8⇆

b232) 14. a3 14... ♘e4 15. ♘e2 ♕a5+ 16. ♔f1 This huge concession is almost forced, and after the immediate16... ♗d7∞ the position is once again very unclear and I think Black should be very happy playing it. 7. ♘f3! will be covered separately. 7... c5 Black goes for the central break.

7... b6 8. ♘f3 c5⁉ 9. dxc5 ( 9. ♖d1 dxc4 10. ♗xc4 ♗xc4 11.♕xc4 cxd4 12. ♘xd4 ♕c8 13. ♕e2 ♘c6= Bhat-Kudrin, Tulsa 2008;9. cxd5 ♘xd5 10. ♘xd5 ♗xd5 11. ♗c4 ♗xf3! 12. gxf3 ♘c6⇆ )9... ♘bd7⁉ 10. c6 ( 10. cxb6 ♘c5 11. ♕b4 axb6=/∞ ) 10... ♘c511. ♕d1 ( 11. c7 ♕c8 12. ♕b4 dxc4 13. ♗xc4 ♘h5⇆ ; 11. ♕b4♖c8⇆ ) 11... dxc4 12. ♕xd8 ♖fxd8 13. ♘d4 ♘d5 14. ♘xe6♘xe6 15. ♘xd5 ♖xd5 16. ♖xc4 ♗xb2 VainermanSerebro, USSR 1987 8. ♕xb7 is the main move here. It's important to know that in a number of recent games White tried 8.♘f3 ♘c6 9. dxc5 but Black is very comfortable here if he makes one correct move 9... ♘e4! ( 9... ♖c8 10. ♕xb7 dxc4 11. ♘g5 ) 10.♕xb7 ( 10. ♗e2 ♘xc3 11. bxc3 dxc4 12. ♕xb7 ♗d5 13. ♕a6♕a5 14. ♗xc4 ♗xc3+ 15. ♔e2 e5 16. e4 ♗xe4 17. ♕xa5 ♗xa5Borges Mateos-Herrera, Matanzas 1997) 10... ♗xc3+ 11. bxc3 ♖c8⇆White's queenside pawns are about to fall one after another, so Black is fine. 8... ♕b6 8... cxd4 9. exd4 ♕b6 was a famous accident in a game between Aronian and Nepomniachtchi, because after 10. ♕xa8 Black suddenly realised that after 10... ♘c6 11. c5 Qxb2 doesn't threaten Qxc1, so Black is just lost. 9. ♕xb6 axb6 10. ♘f3 dxc4 Here I played 10... ♘c6 against Malakhov, which was imprecise because after the very forcing line 11. dxc5 bxc5 12. cxd5 ♘xd5 13. ♘xd5♗xd5 14. ♖xc5 e6 15. ♗c4 ♗xc4 16. ♖xc4 ♖xa2 17. OO ♘a518. ♖c5 ♘b7 19. ♖c7 ♖xb2 20. g4 I realised that although nothing remains on the queenside White is actually quite a long way from equality because the knight on b7 is such a horrible piece. I did manage to draw the game, but I wouldn't recommend the experience to anyone. Malakhov-Svidler, 2011 11. dxc5 bxc5 11... ♘d5 is also possible in this position, but I believe it's unnecessary. 12. ♘g5 ♗d5 Black is perfectly fine here. For instance, 13. ♘xd5 13. ♖d1 ♗c6 ( 13... e6⁉ Kohlweyer-Bobras, BL 2011) 14. ♗xc4♘bd7! 15. OO ♘b6 16. ♗e2 ♘fd5 Matlakov-Kovchan, St Petersburg 2011 13... ♘xd5 14. ♗xc4 ♘xf4 15. exf4 ♗xb2 16. ♖c2 ♗f6 17.O-O ♘c6= Ding Liren-Le Quang Liem, Manila 2010

6.Nf3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗f4 ♗g7 5. e3 O-O 6. ♘f3c5 7. dxc5 ♕a5 The normal move here, and the one I'll concentrate on. Another serious alternative is the recently introduced 7... ♘e4 which might save you some time and trouble studying the main line. 8. ♖c1♘d7 9. cxd5 ♕a5 10. ♘d4 ( 10. ♘d2 ♘dxc5 11. ♘cxe4 ♘xe412. f3 ♘xd2 13. ♕xd2 ♕xa2

Shahade-Shankland, Fremont NCal Intl. 2012) 10... ♘xc3 11. bxc3 ♘xc5 12. ♗c4 ♗d7⁉ ( 12... e5 is imprecise, because after 13. dxe6 ♗xe6 14. ♗xe6 ♘xe6 15. ♘xe6♗xc3+ 16. ♔f1 fxe6 Black is slightly worse. White's king will be much safer the moment it reaches the h2-square, and he has a more stable pawn structure. 17. ♕b3 ♕a6+ 18. ♔g1 ♗f6 19. h3+/= Gustafsson-Safarli, Porto Carras 2011) 13. O-O a) 13... ♖ac8 was the theoretical move for a while. 14. ♗b3 ( 14.♕e2 ♘a4 15. ♘b3 ♕b6 16. ♘d4 ♕c5 17. ♗b3 ♘xc3 18.♕d3 a5 PotapovKurnosov, Rijeka 2010) b) ♘a4 is better and it seems to me that Black equalises on the spot because of the strong threat of Rc8. 14. ♕b3 ♘b6 15. ♗b5 ♗xd416. ♗xd7 ♗g7 , and this is just a draw. ( 16... ♗f6 17. ♗c7♕xd5= ) 17. ♗b5 ( 17. ♗c7 ♕xd5= ) 17... a6 18. ♗e2 ♘xd5=SanikidzeGupta, Istanbul 40th Olympiad 2012. 8. ♖c1 dxc4 9. ♗xc4 The reason I recommended 5...0-0 and not 5...c5 is that in this position we've managed to force the white knight to f3, thereby completely excluding the huge theoretical line in which White develops the knight to e2. That line, just as most of the Grünfeld nowadays, stands very well for Black, but your memory comes under a real strain and I felt that if we could somehow triangulate White away from that it would be to our benefit. So this is the move order I'm proposing. Here Black should start with 9... ♕xc5 It's important to note that 9... ♘c6 10. O-O ♕xc5 runs into 11. ♘b5and Black isn't entirely comfortable here. 10. ♗b3 It's been known for a long time that 10. ♘b5 is very well met by 10...♗e6 and Black has no problems. a) If 11. ♘c7 ♗xc4 White has to be careful not to run into difficulties. 12. ♘xa8 ( 12. ♘d2 b5 13. b3 e5 ) 12... ♕a5+ b) 11. ♗xe6 11... ♕xb5 12. ♗b3 ♘c6 is also perfectly fine for Black. 10. ♕b3 ♘c6 11. ♘b5 ♗e6 12. ♘c7 ♗xc4 13. ♖xc4 ♕a5+ 14.♔e2 ♖ac8 10... ♘c6 11. O-O ♕a5 This is the main position of the whole Bf4 variation. It's been very heavily discussed in serious practice and was, for instance, one of the important theoretical points of discussion in the Karpov Kasparov matches. In this position the main move for White is 12. h3 If White starts with 12. ♕e2 Black has 12... ♘h5 13. ♗g5 ♗g4 which equalises on the spot. After 14. ♗h4 Black has a choice between 14...♕b4= , pretty much forcing Qc4 and immediate equality, and the more ambitious ( 14... g5 , which after 15. ♗g3 ♘xg3 16. hxg3 and the very precise 16... h6 may even be slightly more pleasant for Black. He has the bishop pair in an open position and no real weaknesses to speak of.) 12... ♗f5 13. ♕e2 13. ♘g5 e5 14. ♗g3 ♖ad8 15. ♗xf7+ ♖xf7 16. ♕b3 ♖f8

13... ♘e4 14. g4 This move, introduced by Rustam Kasimdzhanov in a game against me, is now the absolute main line. Black needs to be careful. 14. ♘xe4 is a minor move, but one I still managed to lose to in a very painful game against Levon Aronian. 14... ♗xe4 15. ♖fd1 ♕h5 16.♗c2 ( 16. ♖d7 doesn't accomplish much as Rxb7 threatens nothing due to Nd4, so Black should just play something like 16... ♖ac8 here, and he's fine.) 16... ♗xf3 is better. (Here my dislike of trading bishops for non-bishops played a bad trick on me, and I played 16... ♗xc2 which is still okish for Black but not the most precise way of continuing. 17.♕xc2 ♕b5 18. a4 ♕b4 19. ♘e1 e5 20. ♗g3 ♖fd8 21. ♘d3♕e7 22. ♕c5 ♖d6 23. b4 ♖ad8 24. b5 ♖d5 25. ♕c2 e4 26.♘f4 1-0 Aronian-Svidler, Moscow 6th Tal 2011) 17. ♕xf3 ♕xf3 18.gxf3 and Black should play 18... ♖fd8 with zero problems. 14. ♘d5 For many years this was the only serious move in this position.14... e5 Here there are three moves for White. a) Very old theory starts with the move 15. ♗h2 and now the following strange sequence is almost forced (it becomes more outlandish by the move) 15... ♗e6 16. ♖fd1 ♖fd8 17. ♕c4♘f6 18. e4 ♖ac8 19. ♘g5 ♘d4 20. ♘e7+ ♔f8 21. ♘xe6+ ♔xe7 22. ♘xd8 ♖xc4 23. ♗xc4 ♘e8!= This precise move has been known about for quite a while now. After the near-forced (In this position the obvious-looking 23... ♔xd8 just loses to 24. b4 and the whole position collapses because the pawn on e5 can't be protected.24... ♕c7 obviously loses to 25. ♖xd4+ ) 24. ♘xf7 b5 25. ♗d5♘e2+ 26. ♔h1 ♘xc1 27. ♖xc1 ♕d2 Black isn't particularly active so his extra queen isn't really felt and the game should end in a draw. Black is running no risk of losing here. 28. ♖c8 ♕xf2 29. ♗g1♕f1= Kramnik-Kamsky, Moscow 2008 b) Another old line is 15. ♖xc6 , which was played by Karpov against Kasparov all those years ago. Nowadays it's only used as a surprise weapon as it doesn't really offer any chances of an advantage against a prepared player. 15... bxc6 16. ♘e7+ ♔h8 17. ♘xc6 The main move. (It's important to know that after 17. ♗xe5 ♗xe5 18. ♘xc6Black has the very important shot 18... ♗h2+ The king can't take because of Qc7+, so 19. ♘xh2 ♕d2 and White has a choice between playing an endgame where he's clearly not better or giving up the b2-pawn when he also can't really hope for any advantage. 20. ♕b5♕xb2 21. ♘d4 ♘d2 22. ♖d1 ♘xb3 23. axb3 ♗e6= ; You should also be aware that 17. ♘xe5 ♗xe5 18. ♘xc6 runs into 18... ♕d2and after 19. ♗xe5+ f6 White loses additional material.) 17...♕b6 18. ♘cxe5 ( 18. ♘fxe5 f6 ) 18... ♗e6 Immediately neutralising the very important bishop on b3. This has also been played at the top level recently and is known to be fine for Black. 19. ♗xe6♕xe6 20. ♕c2 f5 21. ♕a4 g5 ( 21... a6⁈ 22. ♖d1 ♖ad8 23.♖xd8 ♖xd8 24. ♘d4 ♕e8 25. ♕a5 ♖c8 26. ♕d5 ♖c5=Brunello-Svidler, Khanty Mansiysk 2010) 22. ♗h2 b1) 22... ♕e8 23. ♘c6 ( 23. ♕a5 ♖b8 24. ♘c4 ♖b5 25.♕xa7 g4 26. ♘d4 ♖d5 27. ♕b7 ♖d7 28. ♕b5 f4 29. ♘f5g3 30. fxg3 ♖xf5 31. ♕xf5 fxg3 Iljushin-Belov, Sochi 2006)23... ♖c8 24. ♘fd4 f4 25. exf4 gxf4 Kosyrev-Sakaev, chessassistantclub.com 2004 b2) g4 23. hxg4 fxg4 24. ♕xe4 gxf3 This position is reached by force and is known to be fine for Black. 25. ♕c4 ♖ae8 26.♕xe6 ♖xe6 27. ♘xf3 ♖a6 Riazantsev-Eljanov, Biel 2004

c) 15. ♗g5 A curious line from the Peter Heine Nielsen laboratory. For a number of years this was considered a bad move because of 15...♘xg5 16. ♘xg5 ♕d8 but in this position White has an interesting attacking possibility with 17. h4 (after 17. ♘f3 e4 Black is indeed already better) 17... h6 18. g4 ♗d7 19. ♘e4 ♕xh4 20. f3 ♔h821. ♖f2 Peter Heine's new idea. (The old way of playing was 21.♔g2 f5 22. ♖h1 ♕d8 when White may have enough compensation for equality, but no more.) 21... f5 22. ♖h2 ♕d8 The difference now is that the king is on g1 so White can play 23. ♖xh6+ ♗xh6 24.♕h2 This was first played by Viktorija Cmilyte against Kateryna Lagno. 24... ♔g7 25. ♖xc6 Another rook is thrown onto the fire.25... fxe4 (In that game Black reacted in a poor fashion with 25...♗f4 , was soon in a lot of trouble and only saved the game by a minor miracle.) 26. ♕xe5+ ♔h7 27. ♘f6+ ♖xf6 28. ♖xf6 ♗e8This position is fine for Black as he has play of his own. White has to respond 29. g5 ♕d2 This move leads to a very spectacular forced draw. ( 29... ♖c8⁈ Nielsen-Nyback, BL 2012 30. fxe4! ; Black can play the very prosaic 29... ♗xg5 30. ♕xg5 ♔g7 , forcing 31. ♖f5and an ending that is maybe even slightly better for Black, although objectively it should be a draw.) 30. gxh6 ♕xe3+ 31. ♔f1 ♔xh632. ♕h2+ ♔g7 33. ♖xg6+! ♔xg6 34. ♕d6+ with a perpetual. Jobava-Areshchenko, Burgas Black Sea 2012 14... ♘xc3 15. bxc3 ♗d7 16. ♖fd1 ♖ad8 17. ♖d5 ♕a3 In the stem game against Kasimdzhanov I played 17... ♕b6 here, which could have been a bit problematic for me if Rustam had remembered his analysis, because after 18. ♖b1 a) 18... ♗c8 the very weird but efficient move 19. ♖g5 just wins on the spot, as there's no defence against both Bxf7 and Bc2. The queen will have to go to a6 and Black will have a completely lost endgame. (19. ♖b5 ♕a6 20. ♗c4 ♕a3 21. ♖5b3 ♕c5 was a draw by repetition in Kasimdzhanov-Svidler, Donostia 2009.) b) e6 19. ♖dd1 18. ♖cd1 ♗e6 18... ♗c8 19. e4 19. ♖xd8 ♘xd8 20. ♕b5 This position has been subjected to fairly heavy theoretical discussion recently and I think that by now people have figured out what to do. 20... ♗xb3 21. axb3 a6 The more forceful path to a draw here. In some games Black even played 21... ♗xc3⁉ showing a complete disregard for White's initiative. 22. ♖d7 ( 22. ♗h6 ♗g7 23. ♗xg7♔xg7 24. ♖d7 ♘e6 25. ♖xb7 ♖c8 26. ♔g2 Gundavaa-Le Quang Liem, Zaozhuang Asian Nations Cup 2012) 22... ♕b4 23. ♕d5 ♘c6 24.♘g5 ♕a5 25. ♕xa5 ♘xa5 26. ♖xe7+/= Matlakov-Nilsson, Rogaska Slatina 27th ECC 2011 22. ♕b6 ♘e6 Here White has a choice. 23. c4 The main move in this position, trying to win the b7-pawn for nothing. If 23. ♕xb7 Black simply captures on f4 and c3 - 23... ♘xf4 24. exf4♗xc3 and experiences no problems whatsoever. 25. ♔g2 ♕b4 26.♖d7 ♕xb7 ( 26... ♖b8 27. ♕c7 ♖e8 28. ♕c6+/= ) 27. ♖xb7 e628. ♖c7 ( 28. h4 h6 29. ♘e5 ♗xe5 30. fxe5 g5= ) 28... ♗b2 29.♘g5 h6 30. ♘e4 ♖d8

If 23. ♗e5 it's important not to play Qc5 immediately but instead start with 23... ♗xe5 24. ♘xe5 ♕c5 25. ♕xc5 Evgeny Tomashevsky offered me a draw here. If you look at the position on your own for a minute it'll become obvious why - the white knight's position on e5 gives Black additional tempi in a lot of lines and the queenside is likely to get exchanged off in a matter of seconds. 25... ♘xc5 26. b4 ♘e4 27.♖d7 1/2 Tomashevsky-Svidler, Moscow RUS-ch 2010 ( 27. c4 ♖c8 28.c5 a5 29. ♖d4 ♘c3 30. ♖d7 axb4 31. ♘d3 e5 32. ♖xb7 e4= ) 23... a5 A very forcing reply. 24. ♕xb7 ♕b2 The best and cleanest solution, creating the incredibly potent threat of a5-a4, when the pawn would become very strong. Emil Sutovsky played 24... ♘c5 here against Mamedyarov, which might also be enough for equality, and that game did end in a draw. 25. ♕xe7♕xb3 26. ♖d5 ♘d3 27. ♗g3 a4 28. ♖d8 h6 29. ♘e5 ♕b1+ 30.♔g2 ♘e1+ MamedyarovSutovsky, Turkey WTCh 2010 31. ♔h2 ♖xd832. ♕xd8+ ♔h7 33. ♕d5 a3 34. ♘d7= a2 ( 34... ♔g8= ) 35. ♗e5♗xe5+ 36. ♘xe5 ♕ b8 37. ♕xf7+ ♔h8 38. ♕f6+ ♔g8 39. ♕xg6+♔h8 40. ♕xh6+ ♔g8 41. ♕g6+ ♔h8 42. ♕f6+ ♔g8 43. f4 ♕b744. ♘c6 ♘f3+ 45. ♔g3 ♕g7 25. ♕b5 Forced. 25. ♕e4 ♕xb3 26. ♖b1 ♕c3 27. ♗e5 ♗xe5 28. ♘xe5 a4 25. c5 a4 25... ♕e2 26. ♕d5 ♕c2 It turns out White is unable to hang on to the b3-pawn. 27. ♖d3 If 27. ♕d3 then 27... ♕xd3 28. ♖xd3 ♘c5 27... ♕b1+ 28. ♔g2 ♘c5 29. ♖d1 ♕xb3 In this position the passer on the a-file is so potentially dangerous that White can't utilise his somewhat more active pieces. In most lines the game ends in a draw by force. 30. ♗e5 ♗xe5 31. ♘xe5 a4=

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Nf3 Bg7 5.Qb3 dxc4 6.Qxc4 0-0 7.e4 a6 8.Bf4 8.Qa4

5.Qb3 dxc4 6.Qxc4 0-0 7.e4 a6 8.Bf4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 The Qb3 line has been a constant source of worry and excitement over the course of my career. I've had a lot of fun facing it against some very strong opposition, and I hope that here I can convey some of the enjoyment I get from analysing and playing it. The emphasis will be on getting fresh and interesting positions, and in general on aiming to have some fun. These days it isn't always possible to get positions that can adequately be described as fun when you're playing Black against competent opposition, but you can at least try. 4. ♘f3 The normal starting move. 4. ♕b3 dxc4 5. ♕xc4 ♗e6 (If Black plays the standard 5... ♗g7 6.e4 O-O White has the option of replying 7. ♗e2 a6 8. ♗f4 and while this position is holdable for Black and not at all bad I honestly don't believe any fun is to be had, so I'm not going to recommend it.) 6.♕b5+ Almost forced. 6... ♗d7 a) White can take the pawn with 7. ♕xb7 but the position that arises almost by force after 7... ♘c6 ( 7... ♗c6 8. ♕b3 ♕xd4 9. ♗e3♕e5 10. ♘f3 ♕a5 11. ♘d4± ) 8. e3 ♖b8 9. ♕a6 ♘b4 10.♕e2 c5 is not something you should be worried about if you're a budding Grünfeld player. For a single pawn Black gets fantastic compensation. Whenever possible White should be encouraged to take pawns like this against the Grünfeld. 11. ♘f3 cxd4 12. ♘xd4 ♗g713. a3 ♘c6 14. ♕d1 ♘xd4 15. exd4 ♗c6=/∞ b) 7. ♕b3 7... c5 b1) 8. d5 b5 Switching to Volga/Benko Gambit-like play. 9.♘xb5 ♘a6 This is quite a historical line, as it occurred in a game between two very strong players, Smyslov and Ragozin, back in 1943. After a 70-year break it was seen in a game between Khairullin and Kovalenko. 10. ♘c3 ♖b8 11. ♕d1 ♕a5 Black clearly has good compensation for the pawn. 12. f3 I believe this is White's best move in the position, which is illustrative of just how precarious his situation is. This isn't usually something you'd think of doing when you're so far behind in development. ( 12. ♘f3Khairullin-Kovalenko, Moscow Aeroflot op A 2011 12... ♗g7=/+ )12... ♗g7 13. e4 O-O=/∞ White seems to be doing quite well in this very unclear position - in the sense that he isn't worse and has an extra pawn, which must count for something - but I'm not sure a lot of people will be attracted to the idea of giving Black this much freedom this early in a game. b2) 8. dxc5 was played by Magnus Carlsen against Vassily Ivanchuk, so this is a move to expect. I wanted to try and find something new and fresh to do in this line. b21) 8... ♗g7 9. e4 ♘c6 ( 9... OO 10. e5 ♘g4 11. f4♕a5 12. h3 ♘h6 13. ♗e3 ♘f5 14. ♗f2+/= ; 9... ♗c6might be a decent idea here. As so often in the Grünfeld, it's good to attack the e4-pawn as early as possible. 10. ♕c4♘bd7 11. ♘f3 OO 12. ♗e2 b5 13. ♘xb5 ♗xe4=/∞ ) 10.♘f3 ♕a5 11. ♗c4 O-O 12. O-O ♕xc5 13. ♗e3 ♕h5 14.♗e2 White retains some pressure, but it's obviously not the end of the world for Black. 14... ♕a5 15. ♖fd1+/= Carlsen-Ivanchuk, Linares 2009

b22) 8... ♘a6 This is the move I came up with. 9. ♕a3 White has to try and hang on to the pawn. ( 9. ♕xb7? ♘xc5 10.♕b4 e5∓ This position speaks for itself.) 9... ♖c8 ( 9...♕c8 10. b4 ♘c7 11. e4 ♗g7 12. f3 OO 13. ♗e3 a5 14.b5 ♘h5 15. ♖b1± ) 10. b4 It seems White has managed to hold on to the pawn and it's unclear where Black's compensation lies, as the knight on a6 is very awkwardly placed, but it can simply retreat. 10... ♘b8 When I came up with this I decided I had to include it as it's such a quirky little idea. The point is obviously to play a5 next move against pretty much anything, undermining White's structure on the queenside and getting some compensation. White has three major options. b221) White can attempt to get a lot by playing 11. ♖b1 a512. ♕xa5 b6 ( 12... ♕xa5 13. bxa5 isn't good for Black as the pawn is hanging on b7.) 13. ♕a3 bxc5 If Black manages to open files and diagonals on the queenside, as he will after 14.bxc5 and 14.Nf3, it's clear he'll have at least enough compensation for equality. b2211) So White will probably stick to his guns with 14.b5 , trying to keep the position closed, but now after14... ♗f5 White has to make another awkward move like15. ♖b2 and after 15... ♘bd7 White's pieces are so obviously misplaced that Black should be happy to play this position any day of the week. 16. ♘f3 c4 17. ♕a7e6=/+ b2212) 14. bxc5 e5 15. ♗e3 ( 15. e3 ♗xc5 16. ♕b3 OO ) 15... ♘g4 16. ♘e4 ♘xe3 17. ♕xe3 ♗xc5 18.♘xc5 ♕a5+ 19. ♕d2 ♕xc5=/∞ b2213) 14. ♘f3 14... cxb4 15. ♕xb4 ♗g7 16. e4O-O b222) 11. ♕xa7 is inadvisable, because after the almost forced 11... ♘c6 12. ♕a3 ♖a8 13. ♕b2 ♗g7 Black gets fantastic play for a couple of pawns. He has an edge in development and threats like Nd5 and Ne4 are very difficult to meet. b223) 11. e3 a5 12. ♕xa5 ( 12. bxa5 ♘a6→ ) 12...♕xa5 13. bxa5 ♖xc5 14. ♗d2 ♖xa5 15. ♖b1 ♖a7 16.e4 e6 17. ♘f3 ♗c6 18. ♗b5 ♗g7 19. ♗xc6+ ♘xc620. OO OO 21. ♖b3 ♘g4 22. ♖fb1 ♗xc3 23. ♗xc3♖xa2= b224) 11. ♕b3 ♗g7 12. ♗b2 ♗e6 13. ♕c2 a5 14. a3axb4 15. axb4 ♘c6 Once again Black has fantastic compensation for the pawn. 16. ♖d1 ♕c7=/∞ b225) 11. ♘f3 is the move most sane people will play as White. 11... a5 12. ♕xa5 ( 12. b5 e5 13. ♘a4 ♗xb5→ )12... ♕xa5 13. bxa5 ♖xc5 14. ♗d2 ♖xa5 15. ♖b1♗c6 (If White had played 11.e3 instead of 11.Nf3 Black could just play 15... ♖a7 here, but in this position White has 16. ♗e3 winning the pawn, with a much better position.) 16. ♘d4 ♘fd7 Connecting the knights and preparing to deploy the bishop on g7.( 16... ♗g7 17. ♘xc6♘xc6 18. ♖xb7 OO 19. e3+/= ) 17. ♘xc6

bxc6 18. g3♗g7 19. ♗g2 O-O= The most likely scenario in this position is that White will have to give up the pawn on a2 in exchange for the bishop on g7, i.e. Black will take on c3 and a2 in the near future, after which White will have some kind of "moral" advantage as Black will have to spend some time getting his pieces out of the corner. In general, however, Black will probably give up the c6-pawn in exchange for completing his development and/or exchanging off one of White's bishops. A draw will then be by far the most likely result. I know I promised fun, and this isn't exactly fun, but as I said, fun isn't always possible against competent opposition. I still think this line is original enough, interesting enough and, frankly, good enough to be featured as a possible approach for Black. 4... ♗g7 5. ♕b3 dxc4 6. ♕xc4 O-O 7. e4 This is the starting point of the classical Qb3 variation. Black has a variety of options. A few recent publications, including Avrukh's books on the Grünfeld, suggest Nc6 for Black here. That's a very interesting and worthy move, which I by no means consider wrong in this position, but my personal preference has always been for 7... a6 This is how I was taught to play. I briefly dabbled in 7...Na6 at some point, with reasonable but nonspectacular results, and then went back to 7...a6, which is the move I feel most comfortable with. Obviously it hasn't been all sweetness and light, and at some point I'm planning to do a short video entitled something like "four months in the life of a top Grünfeld professional", featuring me losing three games in a row over the course of four months to World Champions back in 1999. First I lost to Kasparov, then Anand and then Karpov. So that will be a fun video to make! Those games were in many ways instructive, starting with my perhaps picking a wrong line to reply to Qb3 on each separate occasion, considering the opposition. In this position White has two main choices, 8.Be2 and 8.e5. The sidelines here aren't particularly threatening, but it still pays to know what's what in them. One is 8. ♗f4 which has by now completely disappeared from serious grandmaster practice, although there was a slight recurrence in 2009. Black responds very naturally, giving up the c7-pawn immediately. 8... b5 9. ♕xc7 ♕xc7 10. ♗xc7 ♗b7 Attacking the white centre. In order not to end up worse White should play 11. ♗d3 , when Black wins the pawn back immediately. 11. e5 ♘d5 12. ♘xd5 ( 12. ♗a5 ♘f4 13. O-O-O ♘h3 ) 12...♗xd5 White will have to be very careful here if he wants to equalise, and I'm not even sure how exactly he should proceed. I suspect Black may already be slightly better. 13. ♗e2 (Some "chess-related trivia" is that in a fairly recent game White played 13. a3 and after 13... ♘c6 14.♖d1 ♖fc8 15. ♗b6 ♖ab8 16. ♗c5 ♗xf3 17. gxf3 ♘xe5∓ Black was already better in Yanayt-Yermolinsky, Agoura Hills 2007, although Alex Yermolinsky managed to lose the game by blundering something huge in time trouble. This little tactic occurs a lot in this type of position and is somewhat instructive, but the game wouldn't really have attracted my attention all that much if not for the fact that the white player in this game was Evgeny Yanayt, who is better known as "oogee" at PokerStars and "FishosaurusREX" at Full Tilt Poker. He's perhaps one of the Top 5 at 2-7 Triple Draw. Apparently he also plays Qb3 in the Grünfeld.) 13... ♖c8 14. ♗a5 ( 14. ♗b6 ♖c2 15. b3 ♘c6 16.♗d1 ♖c3 17. OO ♖b8 18. ♗c5 ♖d8 19. ♗b6 ♖b8 20. ♗c5 b421. ♖e1 Dreev-Neverov, Uzhgorod 1987 21... e6=/∞ ) 14... ♘c6 15.♗c3 ♖ab8 , preparing additional advances on the queenside, is very decent for Black. 16. a3 a5=/+ Antoshin-Kalinichev, Moscow 1982. 11... b4 12. ♘a4 ♘xe4 13. O-O ♘f6 The knight retreats to avoid coming under attack on e4.

14. ♖ac1 was Leko's move in a rapid match against Anand. 14. ♘e5 even allowed me to play iIn an old game of mine White played14... ♘c6 In an old game of mine White played. After 15. ♘c5 ♖fc8 Black was very comfortable and I went on to win the game. 16. ♘xc6♗xc6 17. ♗e5 ♗b5= Dyachkov-Svidler, Maikop RUS-tch 1998 14... ♘bd7 15. ♘e5 ♖fc8 16. ♖fe1 e6 Once again Black is very comfortable and has no problems whatsoever. 17. ♘b6 ♘xb6 18. ♗xb6 ♘d5 19. ♗a5 ♗h6= Leko-Anand, Miskolc (m) 2009

5.Qb3 dxc4 6.Qxc4 0-0 7.e4 a6 8.Qa4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6.♕xc4 O-O 7. e4 a6 8. ♕a4 is a more serious attempt by White than 8.Bf4, and has been played in a number of reasonably high-level games. The move I am recommending is 8... c5⁉ This is a very thematic break which will occur in a lot of these positions. Here, as in most cases, it's very good. In the 2009 European Team Championship I had this position against Teimour Radjabov 8... ♘c6 Teimour took the bait and played 9. d5 and after (Unfortunately White doesn't have to play 9.d5 here and can just reply 9. ♗e2 , after which I couldn't find any direct equality for Black, so I'm not recommending this.) 9... ♘a7 Black is actually doing very well because the pawn on d5 is a target and the knight is by no means badly-placed on a7. In many cases it will go to b5, and c6 is an immediate threat. 9. dxc5 ♗d7 10. ♕a3 The only serious move in this position. 10. ♕c2⁈ is met by 10... ♕a5 and Black is already doing very well. It's hard to parry the threat of Nxe4. At the very least the pawn on c5 is about to fall. 11. ♗e3? ♘xe4∓ 10. ♕b3 ♗c6 This is also very standard, targetting the pawn on e4.

a) 11. ♕b4 ♘bd7 12. ♗e2 a5 13. ♕c4 a4 14. O-O ♕a5 15.♗e3 ( 15. e5 ♗xf3= ) 15... ♘g4= b) 11. e5⁈ ♘fd7 12. e6 ♘xc5 13. exf7+ ♖xf7 14. ♕c4 ♗xf315. gxf3 ♘bd7→ c) 11. ♕c4 is pretty much the only move that makes any sense here.11... ♘bd7⁉ ( 11... ♗b5 is simple and good, transposing to 10.Qc4) c1) 12. e5⁈ ♘h5 13. e6 ♗xf3 ( 13... fxe6 14. ♗e2 ♗xf3 15.♗xf3 ♘e5 16. ♕xe6+ ♔h8 17. OO ♘xf3+ 18. gxf3 ♖xf319. ♗e3∞ ) 14. exd7 ( 14. gxf3 ♘e5 ) 14... ♗c6=/+ c2) 12. ♗e2 c21) 12... ♕c7 c211) 13. ♗e3 b5 14. cxb6 ♘xb6 15. ♕c5 ( 15. ♕b4♖ab8 ) 15... ♘ bd7 16. ♕c4 ♘b6=

c212) 13. OO 13... ♗b5 14. ♘xb5 axb5 15. ♕xb5♘xe4 16. ♗e3 ♘dxc5 17. ♖fc1 b6 18. a4+/= c22) 12... ♕a5 13. b4 ♕c7 14. ♖b1 b5 15. ♕d3 ♕b7 (15... ♖fd8 16. ♕c2 ♕b7 17. ♗d3+/= ) 16. ♘d2 ♖fd8 17.♕e3+/= c23) 12... b5! gives Black decent play. 13. cxb6 ♕xb6 14.♗e3 ♕xb2 15. ♖b1 ♕a3 16. ♖b3 ♕d6 17. e5 ♘xe5 18.♘xe5 ♕xe5 19. ♕xc6 ♖ac8 20. ♕xa6 ♖xc3= 10. ♕c4 is met by the necessary but strong 10... ♗b5 ( 10... ♕a5 11.♗d2 ) 11. ♘xb5 (After 11. ♕b3 Black has a choice between taking on f1 and 11... ♗c6 , which are both fine.) 11... axb5 12. ♕xb5♘xe4 This is obviously fantastic for Black, because by the time White has finished his development something will have fallen on the queenside and Black will be very comfortable. 13. ♗e3 ♘c6=/+ 10. ♕b4 ♘c6 11. ♕a3 Neither of Black's alternatives to this move is very good. ( 11. ♕b6 ♕xb6 12. cxb6 ♘b4→ ; 11. ♕xb7 especially leads to a very picturesque position for Black after 11... ♖b8 12.♕xa6 ♘b4 13. ♕e2 ♕a5∓ Black is a whole queenside down, so to speak, but if I got this position in a game I'd feel I was already close to winning. Even if not entirely correct this is the stuff dreams are made of for a Black player.) 11... a5 A very strong move, securing the b4-square for the knight. 12. ♗c4 (If White plays 12. ♗e3 ♘b4 13. O-O-O you might conclude he has a slight death wish. After 13... ♕c7 I tried to find a way not to lose with White, but it isn't easy and it's difficult to imagine anyone playing like this. It's risky and obviously unprofitable to be aiming for positions like this with White in the Grünfeld. 14. ♕b3♘g4 15. a3 ♘xe3 16. fxe3 ♗xc3 17. bxc3 ♘a6∓ ) 12... ♘b4 13.♕b3 Forced, and after 13... ♗c6 Black is doing extremely well. ( 13...b5 14. cxb6 ♕xb6 15. OO a4 16. ♕d1 ♕c7 17. ♗e2 a3=/∞ ) 14.♘e5 ( 14. OO ♘xe4 15. ♘xe4 ♗xe4 16. ♘g5 ♗d5 17. a3 ♗xc418. ♕xc4 ♘c6=/+ ) 14... ♘xe4 15. ♘xc6 (After 15. ♘xf7 Black has the strong 15... ♕d4 and 16. ♗e3 loses to the very nice geometrical motif 16... ♘c2+!−+ 17. ♕xc2 ♕xc4 ) 15... ♗xc3+ 16. bxc3 ♘xc617. ♕xb7 ( 17. OO ♕c7 18. ♗h6 ♘xc5 19. ♕b5 ♘a7=/+ ) 17...♕c8 18. ♕xc8 ♖fxc8 19. ♗e3 ♖ab8= 10... ♗c6 Attacking the central pawn. Gyimesi-Sedlak 11. ♘d2 ♘bd7⁉ This makes sense. 11... a5 Grischuk played this against Wang Hao in the 2012 London Grand Prix, somewhat missing, I believe, White's reply. 12. f3 A very important idea, freeing up the d2-knight to go to c4. 12... ♘fd7 13. ♘c4! ♗d414. ♗e3 ♗xe3 15. ♘xe3 e6 16. h4 The move Wang Hao perhaps should have made. ( 16. ♔f2 ♕e7 17. ♗b5 Wang Hao-Grischuk, London GP 2012) 16... ♕e7 17. O-O-O This is very unclear, but White may enjoy a slight pull. 17... ♕xc5 ( 17... ♘xc5 18. ♔b1+/= ) 18.♕xc5 ♘xc5 19. h5 ♘bd7∞ 12. ♗e2 The point is that 12. f3 can now be met by 12... ♘e5 , preventing White from settling a knight on c4. 13. ♘c4 ♘xc4 14. ♗xc4 ♕d4 15.♗e2 ♘d7 The pawn on c5 is about to fall and the position should be equal. 16. ♗g5 ♖fe8 17. ♖d1 ♕xc5 18. ♕xc5 ♘xc5 19. ♔f2 (19. ♘d5 ♗xd5 20. ♖xd5 ♘e6 ) 12... ♕c7!

12... e6 13. O-O ♕e7 was a sensible approach in Gordon-Hawkins, North Shields 99th GBR-ch 2012, but here White missed an opportunity to sacrifice the exchange with 14. b4! , and if Black accepts the sacrifice then after 14... ♘xe4 15. ♘dxe4 ♗xe4 16. ♘xe4 ♗xa1 17. ♖d1!White has a lot of play for the exchange. I'm not claiming he's much better, but it's at least an interesting improvement over White's play in the game. 17... ♗e5 18. ♗g5 f6 19. ♗h6 ♖fd8 20. f4 ♗c7 21.♕c3 ♔h8 22. a3=/∞ 13. O-O ♖fd8 Now Black wants to play Bb5, attacking White's whole structure and perhaps winning the pawn on c5, whereas White has very few constructive ideas as he's tied down to the defence of the e4-pawn. The one idea he does have, 14. f4 , with e5 to follow, runs into a very interesting tactic. 14... ♘xc5! 14... ♗b5 15. e5 ♗xe2 16. ♘xe2 ♘d5 17. ♘e4+/= 15. e5 Now the machine very calmly proceeds to play After 15. ♕xc5 ♘xe4∓ Black is already better due to the huge threat of Bd4+ against any move.

15... ♘d3 16. ♗xd3

16. exf6 ♕b6+ 17. ♔h1 ♘f2+−+ 16... ♖xd3 17. exf6 ♗xf6 and to my surprise it claims Black is better despite being a whole piece down and not having any direct threats for a while. You can check for yourself why that is, but I'm convinced the assessment is correct. 18. ♘f3 ♗xf3 19. gxf3 ♕d7!↑

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Nf3 Bg7 5.Qb3 dxc4 6.Qxc4 0-0 7.e4 a6 8.Be2 b5 9.Qb3 c5 10.dxc5 Be6 11.Qc2 Nbd7 12.b4/e5/Ng5/c6 12.Be3

8.Be2 b5 9.Qb3 c5 10.dxc5 Be6 11.Qc2 Nbd7 12.b4/e5/Ng5/c6 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6.♕xc4 O-O 7. e4 a6 8. ♗e2 This has been the subject of a number of very high-level theoretical discussions. 8... b5 9. ♕b3 Black has a number of options in this position. Nc6 and Bb7 are playable, but the main and by far the most popular move is 9... c5 A thematic break that will feature in almost all the positions starting with a6. 10. dxc5 An only move. 10... ♗e6 For a number of years Bb7 was almost the only serious move here for Black, and as far as I'm aware it's still very playable, but for at least partly selfish reasons I want to recommend something else. Over a number of years I had an ongoing theoretical discussion with the Ukrainian/American grandmaster Alexander Onischuk. I kept playing him in various tournaments, mainly with Black, and he kept on playing the Qb3 Grünfeld against me. In general I tended to do well, but I was usually the one chasing the game because he'd surprise me with something in the opening. I'd have to solve problems at the board, and usually did that sufficiently well, but the whole experience started to grow somewhat tiresome. Whilst preparing for yet another game with Black against him in Foros I thought I should maybe try and find something new, which is when I spotted the idea behind this move (only my next move was actually a novelty). 11. ♕c2 ♘bd7⁉ People would previously automatically play Nc6 or Qc7 first, when White can expect to apply some real pressure, which is why I came up with the idea of this move. In the stem game against Onischuk I was almost lost, but that didn't stop me from developing the line further and it's since featured in a lot of very high profile games, including some in Candidates Matches. This has now become the central line of the whole Grünfeld with Qb3 and Be2. 12. b4 was played against me by Shakhriyar Mamedyarov. Despite spending a lot of time, by my standards, analysing this position over the course of a number of years I never really looked seriously at b4 because I thought it was too weird a move to devote any time to. When Shakh played it at the board I regretted that, but I quite like my solution. 12. e5⁈ has to be considered in almost any position where it's possible, but in this particular case it's really not a very good move, because after12... ♘d5 everything is hanging for White and it's clear Black is doing fine. 12. ♘g5 isn't very good either, because after the fairly forced sequence12... ♘xc5 13. ♗e3 ♖c8 14. ♗xc5 ( 14. ♖d1 ♕a5 15. ♘xe6♘xe6 16. OO b4 17. ♕a4 ♕xa4 18. ♘xa4 ♘xe4 19. ♗xa6 ♖a820. ♗b5 ♘d6 21. ♗c6 ♖a5= ) 14... ♖xc5 15. ♘xe6 fxe6 Black is already slightly better because of the huge threat of b4, and after 16.♕b3 Black has the powerful 16... ♖e5 The pawn on e4 will most likely drop as there's no convenient way to defend it. 12. c6 is a more serious move. I was surprised that it took a while for this move to be played in serious practice, as any move which forces Black to take back his previous move must be an interesting possibility. 12...♘b8

a) 13. e5 ♘g4 A much more adventurous solution ( 13... ♘d5⁉ is also fine) 14. ♕e4 ( 14. ♘e4 ♕d5 15. c7 ♘c6 ) 14... b4Attacking White's position from all sides. a1) 15. ♘a4 is adequately met by 15... ♕a5 ( 15... ♗f5 16. c7♗xe4 17. cxd8=Q ♖xd8 18. ♘g5 ♗f5 19. f4 ♘c6 20. ♘f3f6 21. h3 ♘h6 22. exf6 exf6 23. ♗c4+ ♘f7 ) 16. ♘d4♕xa4 17. ♘xe6 ♘xf2 a2) After 15. c7 ♕xc7 16. ♕xa8 bxc3 Black is doing very well.17. OO ♘c6 18. ♕xa6 ♘cxe5 a3) 15. ♘d4 The main line. 15... bxc3 16. ♘xe6 White would be very happy if not for the tactical detail that in the position after16... cxb2 17. ♗xb2 ♕a5+ 18. ♔f1 Black has the shot 18...♘xf2 19. ♔xf2 fxe6+ 20. ♗f3 ♘xc6 21. ♕xc6 ♕d2+ and Black wins back the bishop and is at least fine. b) 13. O-O ♘xc6 14. ♘xb5 ♘xe4! A beautiful geometrical solution that made me very happy when I first spotted it during preparation for my game against Onischuk. Black already had one knight hanging and now hangs another, but the point is that no matter which knight White captures Black will play Bd5 and recapture on b5, with excellent compensation. ( 14... ♘b4 There's also nothing wrong with this move. After 15. ♕a4 Black would continue 15... ♘xa2 16. ♘bd4♘xc1 17. ♖fxc1 ♗d7 18. ♕c2 ♕b6 19. e5 ♖fc8 20. ♗c4♘d5 21. ♕e4+/= Hernandez Carmenates-Mekhitarian, Sao Paulo 2009) 15. ♕xe4 ( 15. ♕xc6 ♗d5 16. ♕c2 axb5 17. ♗xb5 ♘d618. ♗e2 ♕b8 At the very least Black will win back something on the queenside, and I think most people would prefer to be Black here.; 15.♖d1 ♗d5 ) 15... ♗d5 16. ♕e3 axb5 17. ♗xb5 ♕b8=/∞ c) 13. ♘xb5 axb5 14. c7 ♕c8 15. cxb8=Q ♕xc2 16. ♕xb5♘xe4 17. O-O ♖fd8 d) 13. ♘g5 The main move, after which Black has to be precise. 13...♘xc6 14. ♘xe6 fxe6 For a while this position only existed in my notes - and in everybody else's notes as well, I expect and nobody would play this with either colour, but by now there have been a number of reasonably high-level games. d1) 15. ♗e3 ♘d4 16. ♕d1 ♘xe2 17. ♕xe2 ♕a5! It's much stronger to start with this move, when Black is doing very well. (17... b4 18. ♘a4 ♕a5 19. ♘c5 b3+ 20. ♕d2 ♕xd2+ 21.♔xd2 ♖ac8 22. f3 ♖fd8+ 23. ♔e1 ♘d7 24. ♘xb3 ♗xb225. ♖d1= Buhmann-Le Quang Liem, Khanty Mansiysk 2010) 18.OO ( 18. ♗d2 b4 19. ♘d1 ♖ad8→ ) 18... b4 19. ♘d1♘xe4 20. ♕c4 ♕d5 21. ♕xb4 ♘d6= d2) 15. ♘xb5 More critical. 15... ♖c8 16. ♘c3 d21) People automatically play 16... ♘d4 here, because it seems to be where the knight should go. 17. ♕d3 ♘d7 18.OO ♘c5 19. ♕d1 ( 19. ♕h3 ♘xe2+ 20. ♘xe2 ♕d3 21.♕xd3 ♘xd3 22. ♗e3 ♗xb2 23. ♖ad1 ♖c2= Volkov-Areshchenko, Plovdiv 13th EICC 2012) 19... ♘xe2+ 20. ♕xe2♕d3 21. ♖e1 ♕xe2 22. ♖xe2 ♘d3 23. ♔f1 ♘xb2 24.♗xb2 ♗xc3= Meier-Le Quang Liem, Dortmund 39th GM 2011

d22) ♘b4⁉ is also possible here, almost forcing White to play17. ♕b1 (otherwise Nxe4 comes into serious consideration)17... ♘g4! Black is fine in the lines that follow. 18. OO ( 18.♗xg4 loses to 18... ♘d3+ ) 18... ♘xf2 19. ♗e3 ♘fd3 (19... ♗d4 20. ♗xf2 ♖xf2 21. ♖xf2 ♗xf2+ 22. ♔xf2♕d4+ 23. ♔f1 ♖f8+ 24. ♗f3 ♕c4+= 25. ♘e2 ♘c2 ) 20.♖xf8+ ♗xf8 21. ♔h1 ( 21. ♗xd3 ♕xd3 ; 21. a3 ♗h6! )21... ♗g7 22. a3 ( 22. ♗g4 ♖c6 23. a3 ♘xb2 24. axb4♗xc3 25. ♕c2 ♕d3 26. ♕f2 ♕d8 27. ♖f1 ♗f6 28. e5♘d3 29. ♕d2 ♗xe5 30. ♗c5 ♗c7 31. ♗f3 ♖xc5 32.bxc5 ♘xc5= ) 22... ♘xb2 23. axb4 ♖xc3 24. ♗g1 ♘d3= 12... a5 13. ♖b1 13. c6? axb4 14. ♘xb5 b3−+ 13... axb4 14. ♖xb4 ♘xc5 15. O-O All Black's moves so far have been very natural, but if he doesn't do something energetic here the b5-pawn is going to fall and he'll have serious problems. After pausing to think I came up with the very neat tactical solution – 15... ♕c8⇆ This creates a very unpleasant X-ray along the c-file. 16. ♘d4 ♘cxe4 17. ♘xe4

17. ♘xe6 ♕xc3 is clearly very good for Black. 17... ♕xc2 After taking some time to think Shakh played 18. ♘xc2 It transpires that after 18. ♘xf6+ ♗xf6 19. ♘xc2 ♖xa2 White's minor pieces are incredibly misplaced. After 20. ♗d3 (or 20. ♗d1 ♗c4 )20... ♖d8 he's unable to keep all of his material. 18... ♘xe4 19. ♖xe4 ♗f5 20. ♖xe7 ♗xc2 21. ♗xb5♖xa2 and there was nothing left on the board and the game was soon drawn. 22. ♖e2 ♗b3 23. ♖xa2 ♗xa2 1/2 Mamedyarov-Svidler, Astrakhan GP 2010

8.Be2 b5 9.Qb3 c5 10.dxc5 Be6 11.Qc2 Nbd7 12.Be3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6.♕xc4 OO 7. e4 a6 8. ♗e2 b5 9. ♕b3 c5 10. dxc5 ♗e6 11.♕c2 ♘bd7⁉ 12. ♗e3 White's only option that poses Black any real problems. 12... ♖c8 12... ♘g4 13. ♗d4 ♘de5 14. ♘xe5 ♘xe5 15. ♖d1 ♕a5 16. OO♖fd8 17. ♘d5± ( 17. b3 ♖xd4 18. ♖xd4 ♘c6 Ezat-Sutovsky, Bursa 7th WTCh 2010) 12... b4 13. ♘a4 ♕a5 14. O-O ♘xe4 15. c6±

12... ♕c8 13. ♖c1 ♘xc5 14. ♕b1 ♗c4 ( 14... ♘g4 15. ♘d5♗xd5 16. ♗xc5 ) 15. ♘d2 ( 15. b3 ♗xe2 1 6. ♔xe2 ♘e6 17.♘xb5 Zuger-Komljenovic, Zagreb 1986 17... ♕b7 18. ♘c3 ♖ac8=/∞ )15... ♘cd7 16. ♘xc4 bxc4 17. OO ♘e5 18. h3 ♕b7 19. b3 cxb3 20. f4 ♘c6 21. ♕xb3 ♕xb3 22. axb3± 13. ♖d1 The critical line. 13. b4 ♘xe4 13. c6 is another option for White. 13... ♖xc6 14. ♘d4 ♖d6 15.♘xe6 ( 15. O-O ♘g4 16. ♗xg4 ♗xg4 ; It's possible to postpone 15.Nxe6 for a while, but in a very instructive game from the US Championship a few years ago Nakamura played 15. ♖d1 against Gata Kamsky. Gata's play was, I think, exemplary. 15... ♘g4 16. ♗xg4♗xg4 17. f3 ♘e5 Creating the huge threat of Nc4 and showing that the bishop doesn't yet have to go anywhere. 18. ♗f2 ♘c4 19. ♘c6♗xc3+ A nice finesse, spoiling White's pawn structure. 20. bxc3 ♕c7 I would go as far as to say that Black is slightly better here, although the game was drawn. Nakamura-Kamsky, Saint Louis USA-ch 2009) 15...♖xe6 a) 16. f3 I dislike this move. Although I understand the desire to protect the e4-pawn, f3 is a very weakening move and loses a tempo in a position where tempi are very important. After 16... ♕a5 17. O-O♖c8 Black is already doing very well due to his piece activity and plans that include Nb6-c4 and Nb6-d5. Now that the pawn is on f3 the bishop on e3 is hanging in a lot of lines. ( 17... b4 18. ♘d1 ♘b6 19. ♘f2♘fd7 20. ♘d3 ♖b8 21. ♔h1 ♘a4 22. ♗f4 ♘e5 23. ♗d1♕b5 24. ♕xa4 ♕xd3 25. ♗b3 ♖d6 26. ♕a5+/= Wojtaszek-Piorun, Warsaw POL-ch 2010) 18. ♖fd1 ( 18. ♕b1 ♘b6 19. a3♘bd5 ) 18... ♘b6 19. ♕b3 ♘fd7 20. ♘d5 ♘xd5 21. exd5 ♖xe3 22. ♕xe3 ♗xb2 23. ♖ab1 ♗a3 24. ♔f1 ♘b6 25. ♖d4♖c2 26. ♗d1 ♘c4∓ Braun-Bobras, BL 08-09 2008 b) 16. O-O 16... ♕b8! Much stronger than 16...Qc8. Black will respond to almost any next two moves by White with Rc8 and Nb6, and if the knight manages to land on c4 Black will have excellent play while White will have his hands full just keeping everything under control. (Tigran Petrosian took the bait here in the 2009 European Team Championship with 16... ♕c8 , winning the e4-pawn. This is why people might choose to play 16.f3, but after 17. ♖ac1 ♘xe4 18.♘xe4 ♕xc2 19. ♖xc2 ♖xe4 it's clear it would have been better not to accept the sacrifice. Only White is playing for a win in this endgame. The black queenside is potentially very weak and the bishops on e3 and f3 will be a huge threat. White will also have control of both the c- and dfiles. 20. ♖d1 ♘f6 21. ♗f3 ♖e6 22.g3 h5 23. h3 ♘e4 24. ♔g2 f5 25. ♖d7 ♗f6 26. b3 ♔g7+/= Tomashevsky-Petrosian, Novi Sad 17th EU-tch 2009) 13... b4 It would be very nice to be able to play 13... ♕a5 but unfortunately after 14. a3⁉ no matter how hard I tried I couldn't make this position work for Black. 14... b4 ( 14... ♘g4 15. ♗d4 ♘xc5 16. b4 ) 15.axb4 ♕xb4 16. c6 ♖xc6 ( 16... ♗b3 17. ♖d4!+− ) 17. ♘d4 ♖d618. ♘xe6 ( 18. O-O ♘g4 ) 18... ♖xe6 ( 18... ♖xd1+ 19. ♗xd1fxe6 20. O-O ) 19. O-O+/= 13... ♘g4 14. b4 ♘xe3 15. fxe3± 14. ♘d5⁉ The main line.

Onischuk played 14. ♘a4 against me, although this isn't currently considered to be a huge problem for Black. 14... ♕a5 15. O-O ( 15. b3♘g4 equalises comfortably for Black. 16. ♗d4 ♗xd4 17. ♘xd4♘xc5 18. ♘xc5 ♖xc5= ) 15... ♘xe4 Black wins back the pawn. a) 16. ♕xe4 ♕xa4 17. ♘g5 ♘xc5 a1) It's important to note that 18. ♗xc5 is no problem because after 18... ♖xc5 19. ♘xe6 Black has ( 19. ♕h4 ♖xg5 20.♕xg5 ♕xa2 ) 19... ♖e5∓ when he's at least fine.

a2) 18. ♕h4 18... h6 19. ♘xe6 ♘xe6 20. ♗c4 Here Morozevich clearly had very nice compensation for the pawn in a game against Kasimdzhanov, but after the correct 20... ♗f6 21.♕e4 ♔g7 22. ♗xe6 fxe6 23. ♕xe6 ♕c6 although White has some pull it's only temporary. Black should be able to equalise with careful play. Kasimdzhanov did indeed hold the draw. 24. ♕h3 h525. b3 ♖fd8 26. ♖c1 ♕b7 27. ♕e6 ♖xc1 28. ♖xc1 ♖d629. ♕c4 a5 30. h3 ♕d7= Morozevich-Kasimdzhanov, Eilat 17th ECC 2012 b) 16. c6 16... ♘d6 17. b3! b1) 17... ♗f5⁉ b11) 18. ♕c1 ♗e4 19. ♘d4 ( 19. cxd7 ♖xc1 20. ♖xc1♕d8 21. ♘b6 ♗f5 ) 19... ♗xd4 20. ♖xd4 ♖xc6 21.♕e1 b12) 18. ♗d3 18... ♗g4 ( 18... ♗xd3 19. ♖xd3 ♕b5 20.♖c1 ) 19. ♗e2= b2) Here I committed a less than obvious blunder with 17... ♗d5?thinking I had a very nice position, but he replied with the very strong 18. ♖xd5 ♕xd5 19. ♖c1 and I was in a lot of trouble and objectively lost at some point in the game, although ultimately I won. Onischuk-Svidler, Foros 2008. b3) ♖c7! 18. ♘d4 ( 18. ♗d4 ) 18... ♗xd4 19. ♗xd4 ( 19.♖xd4 ♖fc8 20. ♕d2 ♖xc6 ) 19... ♗f5 20. ♕c1 ♗e4= 14... ♗xd5 15. exd5 ♘xc5 16. O-O This position has been a hot topic of discussion over the past four years or so. Black has tried many different moves. 16. ♗xc5 ♘d7 is clearly good for Black. 16... ♘ce4 My recommendation. 17. ♕d3! The main line. 17. ♕a4 By far the worst square for the queen, as Black can just play17... ♘xd5! 18. ♗xa6 ♘c5 In general, if Black manages to trade off the d5-pawn for the b4-pawn, or especially for the a6-pawn, he has a very good position, because the bishop on g7 starts to come into play and White has almost no active plans. 19. ♕b5 ♘xa6 20. ♖xd5 ♘c7 21.♖xd8 ♘xb5= 17. ♕b3 ♕d6 It's important to play this move first, when it's very difficult for White to achieve anything. (If you start with 17... ♘c5White has 18. ♕c4+/= and Black already has to be very careful.)

a) 18. ♖d4 ♘c5 ( 18... a5 19. ♖fd1 ♘c5 20. ♕c2 ♘e6 21.♖c4 ♖xc4 22. ♗xc4 ♘c7 23. ♕d2 ♖d8 24. ♗f4 ♕c5 25.♕d4 ♕xd4 26. ♘xd4± ) 19. ♕c4 ♖fd8 20. ♖fd1 a5= 21. g3♘a4 22. ♕b3 ♘c5 23. ♕c4 ♘a4= b) 18. g3 At least attempting to play Bf4. 18... ♘c5 ( 18... a5 19.♗f4 ) 19. ♕xb4 Almost forced.( 19. ♕c4 ♘a4 ) 19... ♘xd5 20.♕h4 ( 20. ♕a3 e6 ) 20... ♕c6 Aiming for the important a4-square. ( 20... e6 21. ♗c4 ♕d8 22. ♗g5 ♕b6 23. ♗xd5 exd5 24.♖xd5 ♕e6 25. ♖d2 ♘e4 26. ♖e2 f5 27. b3 ) 21. ♗d4 ♖fd8Structurally you could argue that White is slightly better, but after careful checking I think this position is very holdable for Black and he should be able to draw reasonably comfortably. 17. ♕b1 was played against me by Karjakin in the 2012 Russian Team Championship. That was a somewhat bizarre experience, first of all because Sergey normally plays 1.e4 against me, so this was just about the first game in a number of years where I played the Grünfeld without revising this line. I half expect people to play Qb3 against me all the time, but generally not people who don't play 1.d4... So I got to this position with a less than 100% fresh memory of what's what. That's not the most bizarre thing, however, which was that not three metres away from our board, on the first board of a different match, Morozevich and Caruana reached the same position at precisely the same time we did. That's always a bit awkward because it becomes something of a game of chicken. There's a huge temptation for people to wait for the other game to move one ply further and see what's going on there. It was slightly better for me, as I clearly remembered the first move of the variation, which was 17... ♘d6! After some delay this move was also made by Caruana against Morozevich. a) 18. ♘e5 ♘f5 b) 18. ♖fe1 A strange thing happened here. As I couldn't remember what my notes said I started to think what they might say, and I had a complete visual hallucination. b1) I thought I'd "remembered" what my notes said and played 18...♕a5 , in the absolute certainty that after 19. ♘d4 ♕xd5 20.♗f3 Black could play 20...Nfe4. I had a "memory" in my notes of a position where Black had knights on d6 and e4 and a heavy piece on d5, with the same pins, and the evaluation was "Black is fine". I probably looked so convincing that Sergey believed me and didn't go for the line, whereas Morozevich did. ( 20. ♘b5 ♕e4 21. ♘xd6♕xb1 22. ♖xb1 exd6 23. ♗xa6 ♖c7+/= Karjakin-Svidler, Sochi 13th RUS-tch 2012) 20... ♕e5 ( 20... ♘fe4 is met by 21. ♗g4!±and the huge threat of Bxc8 and Ne6 means that Black is close to lost.) 21. ♘c6 ♖xc6 22. ♗xc6 ♘g4 This position is quite bad for Black but Caruana is very good at defending and eventually drew the game, although it took him about eight hours, by my reckoning.23. g3 ♕h5 24. h4 ♘xe3 25. ♖xe3+/= Morozevich-Caruana, Sochi 13th RUS-tch 2012 b2) 18... ♘f5! Karjakin and Caruana both correctly surmised that this is much better, and is what my notes actually said. Black is fine in the play that follows, as the pawn on d5 is doomed. 19. ♗g5 (19. ♗d2 ♘xd5 20. ♗xa6 ♖c7= ; 19. ♗xa6 ♘xe3 20. ♖xe3♖c5= ) 19... ♖c5=

c) 18. ♗xa6 is the move Black is most prepared for. After 18...♖a8 19. ♗e2 ♕a5 this position is much easier for Black to play, because if e.g. 20. ♘d4 ♕xa2 21. ♗f3 ♘d7 22. ♘c6 ♖fe8=Black should be able to equalise reasonably comfortably because it's very likely the b-pawns will be exchanged and the beautiful knight on c6 will have little to do. 17... ♕d6 This pawn sacrifice is the main line of the whole variation. The first time I got this position on the board I played the very imprecise17... a5⁈ and after 18. ♘e5 I only managed to save the game due to some trickery and imprecise play by White. 18... ♕c7 19. f4+/=Sandipan-Svidler, BL 2010 17... ♘c5 18. ♕c4+/= 17... ♘d6 18. ♘e5 ♘f5 a) 19. ♗g5 ♕d6 20. ♗f4 ♘h5 ( 20... ♘d7 ) b) 19. ♗f4 19... ♖c5 20. ♘c6 ( 20. ♗f3 ♘h4 21. ♘c6 ♕d7 22.♕e3 ♘xf3+ 23. gxf3+/= ) 20... ♕xd5 21. ♘xb4 ♕xd3 22.♗xd3 a5 23. ♘a6+/= 18. ♕xa6 The only sensible move. 18... ♖c2 The most direct approach. 19. ♗d4 After 19. ♗c1 Black replies 19... ♕c5! with a lot of play. He has the threat of Nxf2 and is already doing quite well here. 20. ♕b5 ♖c8 21.♕xc5 ♖2xc5 22. ♘e5 ♘xd5 23. ♗a6 ♖e8= 19... ♘c5 Forcing White to trade queens. 20. ♕xd6 exd6 21. ♗b5 21. ♖fe1 ♖a8 21... ♘fe4 For a while this position only featured in my notes, but then during the 2012 Russian Championship Superfinal Dmitry Jakovenko played it against me. Probably to the slight surprise of the people watching on the internet and live in the playing hall we reached this position about five minutes after the start of the round. 22. ♘e1 ♖d2 23. ♗xg7 ♔xg7! Surprisingly this is a better way to play. I say surprisingly, as in this line the pawn isn't won back at all. Somewhat unfortunately I did manage to remember what my notes said in this position and played 23... ♖xd1 and after 24. ♗xf8 ♔xf8 25.♘f3 ♖xd5 instead of Bc4 Rf5=, as I clearly remembered my notes said, Dmitry played 26. ♗c6! ♖d3 27. ♖a1! and although the machine insists this is just a dead equal position I'm somewhat less convinced, because it says 0.00 and then simplifies into endgames where White has an outside passer on the a-file and a bishop against a knight. If that's a draw then chess really is a draw everywhere. It might be, but it's no fun to check at the board. Jakovenko-Svidler, Moscow 65th RUS-ch Superfinal 2012 24. ♘d3 The point of White's previous play.

24... ♖c2! 25. ♖c1 25. f3 ♖b8! 26. ♗a6 ( 26. fxe4 ♖xb5 27. ♖f2 ♖xf2 28. ♘xf2♖a5= ) 26... ♘d2 27. ♖f2 ♘xf3+ 28. gxf3 ♖xf2 29. ♔xf2♘xa6= 25... ♖xc1 26. ♖xc1 26. ♘xc1 ♖b8 27. ♗d3 ( 27. ♗c6 b3! ) 27... ♘xd3 28. ♘xd3♖b5 29. f3 ( 29. ♖a1 ♘c5 30. ♘xc5 ♖xc5 31. a4 bxa3 32. bxa3♖xd5 33. a4 ♖a5= ) 29... ♘c5 30. ♖d1 ♘xd3 31. ♖xd3 ♔f6 32.♖b3 ♖a5 33. ♖xb4 ♖xa2= 34. ♔f1 ♔e5 ( 34... ♖a1+ 35. ♔e2♖g1 36. ♔f2 ♖d1 37. ♔e3 ♖xd5 38. ♖d4 ♖h5= ) 35. ♖b7 h536. ♖xf7 ♖xb2= 26... ♘xd3 27. ♗xd3 ♘c5 This solution came as a surprise to me, as Black just exchanges everything and liquidates into an endgame a pawn down. It does seem to be a draw, however. The main idea is 28. ♗c4 28. ♗f1 ♖a8 29. ♗c4 ♖c8 transposes 28... ♖c8 29. ♔f1 29. ♖c2 ♘a4 30. ♗b3 ♖xc2 31. ♗xc2 ♘xb2 32. ♔f1 transposes to 29.Kf1 29. ♖b1 ♘a4 30. ♗b5 ♖a8 31. ♗c6 ♖a5 32. a3 bxa3 33. bxa3♘c3 34. ♖a1 ♘xd5 35. ♗xd5 ♖xd5 36. a4 ♖a5 37. ♔f1 ♔f6= 29. ♗b3 ♖a8 30. f4 ♘xb3 31. axb3 ♖a5 32. ♖d1 ♔f6 33. ♔f2h5 34. ♔f3 ♔f5 35. g3 ♖a2 36. ♖d2 ♖a1= 29... ♘d3 30. ♖c2 ♘xb2 31. ♗b3 ♖xc2 32. ♗xc2 ♘c4 It's very important to start with this move. 33. ♔e2 ♔f6 34. f4 ♘b6 35. ♗b3 Once the bishop has been forced on to b3 it's pretty much impossible for White to attack Black's only weakness, the pawn on b4. This position should be a draw. I must apologise again. I promised you fun, but instead of fun I keep on showing you slightly worse endgames. Unfortunately in main line theory you sometimes have to go for positions you know are drawn even if they're somewhat less than exciting.

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Nf3 Bg7 5.Qb3 dxc4 6.Qxc4 0-0 7.e4 a6 8.e5 b5 9.Qb3 Nfd7 10.Be3 10.h4 10.Ng5 10.Be2 10.e6

8.e5 b5 9.Qb3 Nfd7 10.Be3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6.♕xc4 O-O 7. e4 a6 8. e5 The most principled and sharpest move. 8... b5 9. ♕b3 ♘fd7 In this position White has five moves that are worth mentioning. 10. ♗e3 This is an old move that was the main line for a while until Black figured out that after 10... c5 11. e6 you can play 11... cxd4 12. exf7+ If 12. exd7⁈ then after 12... ♕xd7 13. ♖d1 e5 Black has tremendous compensation for the piece with his dominant central pawns. In the 1997 World Championship in Groningen Vladimir Epishin played 12.♘xd4⁈ against me. 12... ♘c5 13. exf7+ ♖xf7 14. ♕d1 (I think objectively White has to play 14. ♕d5 , but Vladimir had lost the first game of a knockout match and was obliged to play for a win, so he tried to keep pieces on the board.) a) 14... ♖a7⁉ may be even stronger, with the huge threat of Rd7, and Black is already fighting for an advantage. 15. ♘f3 ( 15. ♘c6♘xc6 16. ♕xd8+ ♘xd8 17. ♗xc5 ♖c7∓ KozulAzmaiparashvili, Elista 1998) 15... ♖d7 16. ♕c2 ♘e6 , and White will have to be careful. b) 14... ♗b7 is the move I chose when I was first faced with this position. 15. ♖c1 ( 15. ♘dxb5 ♘bd7! with excellent play for the pawn) 15... ♕d6 , and Black was very comfortable in Epishin-Svidler, Groningen 1997 12... ♖xf7 13. ♗xd4 13. ♘xd4 ♘c5 14. ♕d5 ♕xd5 15. ♘xd5 ♘bd7 16. ♖d1 ♗b7 17.♘c7 ♖c8 18. ♘de6 ( 18. ♘ce6 ♗e5 ) 18... ♗xb2=/+ 13... ♗xd4 14. ♘xd4 Black has a very pleasant choice. 14... ♗b7! Perhaps the best move, keeping all the options open. 14... ♘c5 15. ♕d5 ♕xd5 16. ♘xd5 ♘bd7 17. b4 ♗b7 18. ♘c3e5= Piket-Timman, Rotterdam 1997 15. ♘e6 ♕b6=/+ Nc5 will follow, exchanging off White's remaining active piece. Black is doing very well.

8.e5 b5 9.Qb3 Nfd7 10.h4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6.♕xc4 OO 7. e4 a6 8. e5 b5 9. ♕b3 ♘fd7 10. h4 This is a more critical move that has also now disappeared from practice. 10... c5 11. e6 c4 It's not such a good idea to play 11... cxd4 as after 12. exf7+ ♔h8 13.h5 you might get mated on the spot. 12. ♕d1 It's important to start with this move to force Black to waste a tempo on 12... ♘b6 If Black plays 12... fxe6 White has the very strong 13. a4 13. exf7+ ♖xf7 The following sequence is pretty much forced as White needs to justify pushing his h-pawn. 14. h5 ♘c6 15. hxg6 hxg6 16. ♗e3 There's a lot of theory here. 16... ♘d5⁉ This is my proposed solution - a relatively fresh move that has been played a number of times in grandmaster practice and has a pretty bad reputation, but I think that's entirely undeserved. 16... ♗f5 I'm very proud of this move, which I came up with over the board during a game against Vishy. 17. ♘g5 ♖f6 18. g4 ♗e6 19.♘ce4 ♗d5! This is the idea behind 16...Bf5 - Black doesn't intend to fight for the exchange, and now his bishop occupies the wonderful d5square. 20. ♕d2 ( 20. f3⁈ e5 21. dxe5 ♘xe5 22. ♗e2 ♗xe4 23.♘xe4 ♖e6 24. g5 ♕c7 25. ♕c2 ♖d8 26. ♖d1 ♘d5 0-1 Kasimdzhanov-Khalifman, Wijk aan Zee 2002) a) 20... e5 21. dxe5 ♘xe5 22. OOO ♖c6 ( 22... ♖c8 23.♗xb6 ♘d3+ 24. ♗xd3 cxd3+ 25. ♔b1 ♖xb6 26. ♕xd3 ♗xe427. ♘xe4 ♕xd3+ 28. ♖xd3± Gmuer-Kund, ICCF 2005) 23. ♗xb6♘d3+ 24. ♗xd3 cxd3+ 25. ♔b1 ♕xb6 26. ♕xd3 ♖d8 27. ♕f3♗d4 28. b3± Hauf e-Mukherjee, corr 2003 b) I got a bit carried away here: 20... ♖d6⁈ 21. f3! , and with the queen about to get to h2 White's setup is suddenly very potent. 21...♗xe4 22. fxe4 ♘d7 23. ♕h2 ♘f8 24. e5 ♖d7 25. ♘e6!+−Anand-Svidler, Linares 1999 c) 20... ♕d7 c1) Perhaps White needs to bite the bullet and play 21. OOO⁉ b4 22. ♘c5 ( 22. ♖h4 c3 23. bxc3 bxc3 24. ♕xc3 ♘a4 25.♕a3 ♖c8=/∞ ) 22... ♕c7 23. ♖h3 c3 24. bxc3 bxc3 25.♕xc3 , and the knight on c5 shields the king while White is ready to start his own attack on the kingside. c2) 21. ♗g2 21... a5 With an incredibly messy position. 17. a4 It makes sense for White to get the a1-rook into play with this move.

White can start with 17. ♗h6 immediately, but I don't think that changes much. 17... ♗g4 ( 17... ♗xh6 18. ♖xh6 ♘f4 19. g3 ♕f8 20. ♖h4♘h3 21. ♕c2 ♔g7 22. ♕d2 ♖xf3 23. ♕h6+ ♔f7 24. ♕xf8+♔xf8 25. ♗xh3 ♗xh3 26. ♖h8+ ♔g7 27. ♖xa8 ♘b4=/∞ ; 17...♗f6⁉ ) 18. ♗xg7 ( 18. ♗e2 ♘f4∓ ) 18... ♔xg7 19. ♕d2 ♘f4 20.♘g5 ♖f5∞

17... ♖b8 18. axb5 axb5 19. ♗h6 Black has a number of very attractive new ideas: the move I like most is another novelty,

19... ♗f6⁉N This creates the very unpleasant threat of Rh7, after which White's whole idea will be on the point of collapse. Black has a better situation in the centre and if he also takes control of the h-file it's not at all clear what White is doing. 19... ♗g4 is the move Ftacnik played against Babula, and after 20.♗xg7 he replied 20... ♗xf3 and entered an endgame he failed to hold. (Black could instead simply have played 20... ♖xg7 with a very unclear position.) 21. ♖h8+ ♔xg7 22. ♖xd8 ♗xd1 23. ♖xb8 Babula-Ftacnik, Moravia Team Ch 1997. 19... ♗f5⁉N is a novelty the machine comes up with after some thought, showing that 20. ♗xg7 ( 20. ♘g5 ♘db4 ) 20... ♖xg7 a) 21. ♗e2 ♘f4 ( 21... ♘db4 22. ♔f1 ♘c2 23. ♕d2 ) b) 21. ♕d2 ♘db4 22. ♕h6 ♘c2+ 23. ♔d1 ♘xa1 c) 21. ♘e5 21... ♘cb4 22. ♖c1 ♘f4 23. ♕d2 g5 24. f3 ♘fd3+25. ♗xd3 cxd3∓

20. ♘e5 This is quite bad news for White.

20. ♗e2 ♖h7! 21. ♕d2 ♗f5 22. ♔f1 ♘db4 23. ♘xb5 ♗d3 20. ♘g5 ♗xg5 21. ♗xg5 ♘db4∓

20... ♗xe5 21. dxe5 ♘db4 If he has nothing better than this position the whole line is pretty much unplayable. Black already has direct threats, so White should probably play 22. ♕xd8+ but after 22... ♘xd8 it's very clear where Black's pieces belong (the knight on e6, the bishop on f5), while White will still have to prove that the position of any of his pieces makes sense. I like Black here and in general I think 16... Nd5 is a very convincing solution to the 10.h4 problem. 23. O-O-O 23. ♖d1 ♗f5 24. ♖h4 ♘e6 , and Black has an edge. 23... ♘dc6 24. ♗e3 ♗f5 25. b3 ♗d3 26. e6 ♖f6=/+

8.e5 b5 9.Qb3 Nfd7 10.Ng5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6.♕xc4 O-O 7. e4 a6 8. e5 b5 9. ♕b3 ♘fd7 Certainly the most unexplored move and the one resulting in the most exciting play is 10. ♘g5⁉ This was introduced into grandmaster practice a couple of years ago by Magnus Carlsen, which is always a sign that a move is very interesting. 10... ♘c6⁉ strikes me as the most interesting move here. Black aims to restrict White's options. Black has generally played 10... ♘b6 11. ♗e3 ♘c6 12. ♖d1 ♘a5This position is very unclear and I don't see any reason for Black to go for this. ( 12... h6 13. ♘ge4 b4 14. ♘a4 ; 12... ♗f5 13. ♗e2 ♘a5 14.♕b4 ♘ac4 15. OO f6 16. ♘f3 ♘xe3 17. fxe3 fxe5 18. dxe5♕e8 19. ♕c5+/= Carlsen-Dominguez Perez, Wijk aan Zee 2010; 12...♗g4 13. f3 ♗f5 14. ♘ge4 ♘a5 15. ♕c2 b4 16. ♘e2 ♘d5 17.♗f2 c5 18. g4 b3 19. axb3 ♘b4 20. ♕xc5 ♗xe4 21. fxe4 ♖b822. ♘c1+/= Moiseenko-Volokitin, Kiev 81st UKR-ch 2012) 13. ♕c2⁉ (13. ♕b4 ♘bc4 14. ♗c1 ♗b7 15. ♘ce4 h6 16. ♘c5 hxg5 17.♘xb7 ♘xb7 18. ♗xc4 bxc4 19. ♕xb7 c5= Korobov-Gupta, New Delhi AICF-AAI Cup 2012) 11. ♗e3 The only good move. After 11. ♘xf7 ♖xf7 12. e6 Black sacrifices the exchange with 12...♘xd4 13. exf7+ ♔f8 14. ♕d1 and after the possible 14... ♘c5 15.♗e3 ♗b7 16. ♖c1 ♕d6 I think it's quite clear to anyone that Black is dominating the board and the exchange is a very small price to pay for compensation like this. If 11. ♘e6⁈ Black has the very nice reply 11... ♘a5! and after 12.♘xd8 ♘xb3 13. axb3 ♖xd8 Black is ready to break in the centre. 14.♘xb5 ( White is just worse after 14. ♗xb5 ♗b7! as both the pawn on g2 and the bishop on b5 are hanging. White's centre will collapse after e.g. 15. ♗f1 c5 ) 14... c6 Driving the knight back. 15. ♘c3 c5 For a single pawn Black has a position with a huge lead in development and a very strong initiative against the white centre. 16. ♘d5 ( 16. ♗g5cxd4 17. ♘d5 ♗b7 ) 16... ♗b7∓

After 11. e6 ♘xd4 12. exf7+ ♔h8 Black hasn't even had to sacrifice anything to get a brilliant position. 11. ♕d5 looks interesting at first glance, but it runs into the very strong11... ♘db8!∓ Suddenly the pawn on d4 is in a lot of trouble and I think Black is already better. 11... ♘a5 12. ♕d1 The white queen being forced to this square is a big advantage over the 10...Nb6 lines. In a recent game White failed to appreciate the difference between this and the lines starting with 10...Nb6 and went wrong with 12. ♕c2 when Black was able to play 12... c5! It's now impossible to play 13. e6 as after ( 13. dxc5 ♘xe5 14. ♖d1 ♗f5 15. ♖xd8 ♗xc2 16. ♖d2♗f5=/+ Bai-Negi, CHN-tch 2012 2012) 13... cxd4 14. exf7+ ♖xf7 it turns out that the fork is inescapable and White loses material. 15.♘xf7 ♔xf7∓ 12... ♘b6 In this position Black's play is clear. He wants to play Nc4, and most importantly f6. If the centre is blown apart Black's lead in development may become a decisive factor so to justify the previous play I think White has to play his following move. 12... h6 13. ♘ge4 ♗b7 ( 13... b4 14. ♘a4+/= ) 14. b4 ( 14. e6fxe6 15. ♕g4 ♕e8 16. ♕xe6+ ♔h8 17. a3 ♖d8 18. b4 ♘c6 19.♖d1 ♘f6 20. d5 ♘xe4 21. ♘xe4 ♘e5+/= ) 14... ♘c6 ( 14... ♘c415. ♗xc4 bxc4 16. ♕e2+/= ) 15. ♕b3 ( 15. a3 f5! ) 15... e6 ( 15...a5 16. ♗xb5 axb4 17. ♗xc6 ♗xc6 18. ♕xb4+/= ) a) 16. ♗e2 a5 17. ♗xb5 axb4 18. ♗xc6 ♗xc6 19. ♕xb4♕h4!∓

b) 16. a3 f5 ( 16... a5 17. ♖b1 axb4 18. axb4 ) 17. exf6 ♘xf618. OOO ( 18. ♘c5 ♘xd4 ) 18... ♘xe4 19. ♕xe6+ ♔h7 20.♕xe4 ♕d6⇆ c) 16. ♖b1⁉ ♘e7 ( 16... f5 17. exf6 ♘xf6 18. ♘c5 ) 17. ♗e2♘f5 18. O-O+/=

d) 16. ♖d1 16... ♕e7 17. a3 ♖fd8 ( 17... f5 18. exf6 ♘xf6 19.♘c5± ) 18. ♗e2 a5∞ 13. h4 Here I can offer you a choice. 13. b3 f6=/+ 13... h6 This is my alternative. For those who like material and don't mind defending positions that look scary for a while, there's 13... f6⁉ and now once again, to justify the preceding play, White almost has to go for 14. ♘xh7 ♔xh7 15. ♗d3(if 15. h5 ♖h8 16. hxg6+ ♔xg6 17. ♗d3+ ♔f7−+ White runs out of checks and this really isn't very attractive for him.) 15... ♖h8 16. h5♔g8∞ The machine says this is perfectly fine for Black, and it may well be so, but for someone like me who is, generally speaking, much more content giving away material than accepting sacrifices, this isn't ideal. 14. ♘ge4 ♘ac4 Black is ready to play f5, and in general White has very little time to waste since he's somewhat behind in development and his centre is about to become vulnerable. 15. h5 Forced, after which Black again has a choice. For sane people there's 15. ♗c1 b4 16. ♘a4 ♕d5∓ 15. b3 ♘xe3 16. fxe3 f6∓

15... g5 , and the huge threat of f5 forces White's next move. 15... ♘xe3 16. fxe3 f5 17. ♘c5+/= For people who don't particularly like sanity and would like a little crazy in their lives there's 15... f5 16. exf6 ( 16. ♘c5 ♘xb2 17. ♕c2 ♘2c418. ♖d1 f4 19. ♗c1 ♗f5 20. ♕b3 g5=/+ ) 16... exf6 White can choose between two positions. 17. ♗xc4+ is perhaps the stronger move, getting rid of the unproductive bishop on f1. (after 17. hxg6 ♘xe3 18.fxe3 ♕e7⇆ Black is planning to play f5-f4 against pretty much anything.) 17... ♘xc4 18. hxg6 ♘xe3 19. fxe3 f5 20. ♘c5 f4 This is the position Black is aiming for in this line. It looks truly weird and isn't a pawn structure you see every day. For now Black is a pawn down, but White's central pawn structure has come under direct attack and if Black manages to open up the diagonals and get his bishops into play I think he has very real chances of an advantage here. In general, this is a very fresh and interesting way of dealing with 10.Ng5. It should have excellent shock value, as your opponent won't be expecting you to be so free with your kingside pawns. I was very pleased when I came up with this idea. At the time it was almost a novelty, while now there have already been some games, which makes it somewhat less exciting, but I still think Ng5 Nc6 is a very decent way to play and I heartily recommend it to anyone. 21.♕b3+ ♔h8⇆

16. ♘xg5 After the pretty much forced

16... ♘xe3 17. fxe3 hxg5 18. h6 ♗xe5 19. dxe5 ♕xd1+ 20.♖xd1 and 20... ♗f5 , preventing the white bishop from getting to the ideal d3-square unchallenged, this endgame looks slightly scary but I checked it and think the most likely outcome is a draw.

21. ♗d3 ♗xd3 22. ♖xd3 ♖ad8 22... ♔h7 23. ♔e2 ♖ad8 24. ♖xd8 ♖xd8 25. ♘e4 ♔g6 26.♘c5+/= 22... ♘c4 23. ♖d7 ♘xe3 24. ♔e2 ♘xg2∞ 23. ♔e2 23. ♖xd8 ♖xd8 24. ♘e4 ♘c4 23... ♘c4=

8.e5 b5 9.Qb3 Nfd7 10.Be2 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6.♕xc4 OO 7. e4 a6 8. e5 b5 9. ♕b3 ♘fd7 10. ♗e2 A new-fangled way of handling this position, which was revived if not invented by Alexander Morozevich in his game against me in the last round of the 2011 Russian Championship Superfinal. 10... c5 11. e6 fxe6 This was my own reaction, and I think it's perfectly adequate. It's possible to play 11... cxd4 12. exf7+ ( 12. exd7 ♘xd7 13. ♘e4♗b7 14. ♗d3 ♖c8 leads to a position where Black has a lot of compensation for the piece: 15. OO ♗xe4! 16. ♗xe4 ♘c5 17. ♕d5e6 18. ♕xd8 ♖fxd8 19. ♗c2 d3 20. ♗b3 ♘e4=/∞ ) 12... ♔h8 13.♘e4 Black does run the risk of getting mated here, but on the other hand, if he doesn't get mated he has a potentially very nice central majority. 13... ♘c6 I think this is the best move, although the position is extremely unclear (and very interesting to analyse). (One pitfall is 13...♘e5 14. ♘xe5 ♗xe5 15. h4 ♗b7 which Ray Robson played in the 2012 US Championship, thinking that the knight on e4 was hanging and Bd5 would follow next move. He was sadly mistaken on both counts. After 16. h5 ♗xe4 17. ♗f3!± Lenderman-Robson, St. Louis 2012, he realised that 17... ♗xf3 18. hxg6 is mate and the only other choice is to lose material, so it was a bit of an opening catastrophe.) 14. h4 h6 (14... ♘de5 15. h5 ♗f5 16. ♘eg5± ) 15. h5 g5 16. ♗xg5 ♘de5 17.♗f4 ♗f5 18. ♘xe5 ♗xe5∞ Shabalov-Gupta, Philadelphia 40th World op 2012 12. ♕xe6+ ♔h8 13. dxc5 Something like 13. ♕d5 cxd4 14. ♕xa8 dxc3 is clearly acceptable for Black. 13... ♘e5 was my choice and is still my recommendation. I want to base this series around my own games where possible. 13... ♘xc5⁉ 14. ♕e3 b4 is a line that gives Black very interesting counterplay and has already been pointed out in a number of sources. 15.♕xc5 bxc3 16. bxc3 ♘d7 17. ♕a3 ♘f6=/∞ 14. ♕d5

14. ♕b3 ♘d3+ ( 14... ♗b7 15. ♘xe5 ♗xe5 16. OO+/= ; 14...♘xf3+ 15. ♗xf3 ♖xf3 16. gxf3 ♘c6 17. ♗e3 ♘e5 18. ♕d5♕xd5 19. ♘xd5 ♘xf3+ 20. ♔d1 ♗b7 21. ♘xe7 ♗e4 22. c6♗xb2 23. c7 ♗xa1 24. ♔e2 ♗e5 25. c8=Q+ ♖xc8 26. ♘xc8♘xh2= ) 15. ♗xd3 ♕xd3 16. ♗e3 ♘c6 17. ♖d1 ♕f5 Black has the huge threat of Be6, and in general in positions like this one, where Black gets free play and both of his bishops are very active, his compensation for the single pawn is clearly more than sufficient. 18. ♕d5 ( 18. ♖d5♕f6 19. O-O ♗e6∞ ) 18... ♕xd5 19. ♘xd5 ♗xb2= 14... ♕xd5 15. ♘xd5 ♘xf3+! is what I should have played. I was one move from safety but started to go completely crazy - five moves later I could pretty much resign. 15... ♗b7 16. ♘c7 ♘xf3+ 17.gxf3!± Taking with the pawn makes a huge difference. 16. ♗xf3 16. gxf3 can simply be met by 16... ♘c6=/∞ and the knight is about to land on d4, the c1-bishop is stuck defending the b2-pawn, the other black bishop might come to f5 or even h3 in some lines, and in general it's obvious this position is fine for Black. 16... ♗b7 Only now, and in this position there's really nothing better for White than 17. ♘c7 ♗xf3 18. gxf3 and it makes a huge difference that there are no light-squared bishops on the board, because after 18... ♖a7 19. ♘e6 Black can just play 19... ♖xf3= and a number of grandmaster games in which this position occurred were all drawn. Black is perfectly safe here. The knight will come to c6, cementing the black queenside, the entrance points along the d-file are completely covered and in general I don't see any reason why Black should be worse.

8.e5 b5 9.Qb3 Nfd7 10.e6 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6.♕xc4 OO 7. e4 a6 8. e5 b5 9. ♕b3 ♘fd7 10. e6 The classical move in this position. 10... fxe6 This line was considered completely void until 1999 when people started playing 11. ♗e3 against me. First Kasparov played it in Wijk aan Zee, then Karpov in Dos Hermanas. Official theory pre-1999 used to be 11. ♕xe6+ ♔h8 12. ♘g5 ( 12.♕e4 ♘b6 13. ♕h4 ♘c6 It's been known for a number of years that Black is fine here. For instance, Kamsky demonstrated a very nice double exchange sacrifice against Karpov: 14. ♗d3 ♖xf3 15. gxf3 ♘xd4 16.♗e4 ♗f5 with wonderful compensation in the World Championship match in Elista. 17. ♗e3 c5 Karpov-Kamsky, Elista 1996) 12... ♘f613. ♘f7+ ♖xf7 14. ♕xf7 ♘c6 The queen is stuck on f7 and Black has wonderful compensation for the exchange. 11... ♘f6 It's curious that I chose 11... ♘b6 against Kasparov, when White replies12. h4 with a kingside attack. Kasparov is reasonably well-respected as an attacking player...

12. a4 My 11...Nf6 against Karpov encouraged him to respond with this move, playing for small advantages on the queenside. Karpov is a reasonably well-known positional player... I lost both games, but I don't want to go into too much detail. As I said, I'm planning to make a small video out of the three games I lost to World Champions in quick succession (the other loss was against Anand). 12... b4⁉ This is my recommendation in this line rather than the 12... bxa4 I played against Karpov. 13. ♕xb4 Forced. 13... ♘c6 When I first started looking at this position I was under the misguided impression that Black might even be better here. I think that's a huge overstatement, as Black isn't really fighting for an advantage against precise play, but I still believe Black is perfectly fine. The best move here is 14. ♕a3 14. ♕c5 ♕d6 15. ♗c4 ♖b8 In this position White has to negotiate some pitfalls. 16. O-O (Onischuk played 16. ♘g5? against me, and after16... ♕xc5 17. dxc5 ♘g4!=/+ it turned out the hanging pawn on e6 isn't an issue and Black is already much better. I was very disappointed not to win that game.) 16... ♖xb2 Black is a pawn up here, and although he's not yet better I still managed to win in a Bundesliga game against Zoltan Gyimesi some years back. 17. ♖ab1 ♕xc5 18. dxc5♖c2 19. ♖fc1 ♖xc1+ 20. ♗xc1 ♖d8 21. ♘g5 ♘d5=/+ Gyimesi-Svidler, BL 2002 14... ♕d6 15. ♗c4 ♘b4! is much better. It's important not to play 15... ♖b8 on autopilot, as 16. O-O leads to positions that are better for White, as can be seen, for instance, from a game of mine against Kramnik at one of the Tal Memorial blitz tournaments. 16... ♕xa3 17. bxa3 ♖d8 ( 17... ♔h8 18. h3 ♘a5 19.♗a2 c5 20. dxc5 ♘d5 21. ♘xd5 exd5 22. ♖ab1 ♖xb1 23. ♖xb1♘c4 24. ♗d4 ♗f5 Miton-Swiercz, Warsaw 9th Amplico Life EU-ch Rapid 2009 25. ♖b7± ) 18. ♖ac1 ♘a5 19. ♗a2 c6 20. ♗f4 ♖b721. ♖fe1 ♘d5 22. ♗d2+/= Kramnik-Svidler, Moscow Tal 5' 2009 16. ♕b3 Compared to White's other option of Rc1 this at least wins the e6-pawn. 16. ♖c1 ♗b7→ 16. O-O White could ignore Nc2 and play 16... ♘c2 White could ignore Nc2 and play 17. ♕xd6 exd6 after which Black's structure is actually better than White's - the pawn on a4 is ugly. Black has no problems whatsoever here. 16... ♗b7! 17. ♗xe6+ 17. ♘g5 can lead to a very beautiful variation that I want to show you for purely aesthetic reasons: 17... ♗xg2 18. ♘xe6 ♗xh1 19. ♗f4 It seems as though there's no good square for the queen and White's about to start collecting material with checks, but here Black has a sequence of beautiful intermediate moves. 19... ♘d3+! 20. ♗xd3 ♗d5 Suddenly all of Black's problems along the a2-g8 diagonal have been solved and he may even be better. 21. ♕xd5 ♕xd5 22. ♘xd5 ♘xd5 23. ♗g3 ♖fc8 24.♗c4 c6=/+

17... ♔h8 18. O-O ♗xf3 19. gxf3 ♖ab8 Black is perfectly comfortable in this position as he's eliminated his bad pieces. The pawn on e6 was unimportant and he's rid himself of the potentially bad bishop on c8. White's structure on the kingside has been spoiled and Black is ready for some forcing compensation. Play might continue 19... ♘c6 20. ♗c4 Otherwise the bishop might get stuck on e6. 20... ♕c6 21. d5 21. ♗e2 ♘fd5 22. ♘xd5 ♘xd5 23. ♕c4 ♕d7 24. ♖fc1 ♖b4=/+ 21... ♕a8 Black will win back the pawn on d5 in a matter of moves, by which time White will have been able to regroup somewhat and the position will be very close to complete equality. 22. ♕d1 ♖fd8 23. ♕e2 ♘fxd5 24. ♖ac1 ♘xe3 25. fxe3 c6=

As I mentioned when recording my videos about the 5.Qb3 Grünfeld I thought it might be interesting to make a kind of "infotainment" video. What follows is somewhat less chess-heavy than the other material in this eBook, but I'd like to tell you about the three games I played in 1999 in pretty much the same line of the Grünfeld against three World Champions - the reigning World Champion, the ex-World Champion and the future World Champion and in general to tell you a little bit about how my 1999 went. I know it was a long time ago and has no particular relevance but, as they say, it's my series and why not! I think the chess content is quite interesting as all three games are remarkable in their own way. 1998 went quite well for me - I did ok. I finished fine in my first ever Linares, tied for first in the Russian Championship (as was my norm in those years, if not now - I was very consistent in the Russian Championships) and I got quite a lot of top-level invitations because I was the new kid on the scene and people were interested in seeing me play in the top events.

Garry Kasparov 1:0 Peter Svidler Wijk aan Zee30.01.1999 The first top event I had to prepare for was Wijk aan Zee in January. I went there with very high hopes and thought this was my chance to shine! I should play well because I could remember playing well pretty much throughout 98. Wijk was the first tournament to establish a pattern which was something I was completely unused to in those years but came to recognise and accept as my own in future years i.e. out of the 13 rounds in Wijk I won one, lost one and made 11 draws. Towards the second half of the tournament I was playing pretty much kamikaze chess with Black because I was going more than slightly mad with all the drawing I was doing. Yet even when I played lines I knew were almost losing by force I'd still somehow make a draw. And that drove me even more mad. The one game I won was against my current chess24 colleague Rustam Kasimdzhanov, who was playing perhaps his first major tournament and felt that he also needed to play for a win in every single game. In the game against me he overpressed in a position where the normal result would have been yet another draw... And I lost to Garry Kasparov, which is of course perfectly fine and normal - there's no shame in losing to Kasparov and it was quite a remarkable game. I remember not feeling even remotely disappointed afterwards because I felt I was beaten by someone at the top of his game and someone who had shown a very interesting opening idea in a line I felt I knew quite well. So all in all, this particular game gave me no great cause for alarm. 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 As a small sidenote... it's perhaps not the most obvious choice to play the Grünfeld as Black against Kasparov because he has some ideas about the opening, to put it mildly, but first of all, in those years - it's still true even now - but in those years in particular I didn't have that wide an opening repertoire, and secondly, I thought, "wouldn't it be interesting to find out what will happen to me if I play the Grünfeld against Kasparov?" A feeling not of invincibility, as such, but one of wondering - I knew this wasn't going to be a walk in the park but I was very interested to see what he would show if I did it. And he did show something... 4. ♘f3 ♗g7 5. ♕b3 dxc4 6. ♕xc4 O-O 7. e4 a6 8. e5 b5 9.♕b3 ♘fd7 This was a very topical line in those years as well. I had some experience with it and felt I was reasonably well-prepared. 10. e6 I knew this move, which had always been considered completely harmless, so what was his plan?

10... fxe6 11. ♗e3 At this point in time this was a novelty. I sank into thought and could immediately see that my choice was pretty clear cut. I could play 11...Nf6, to which White would reply 12.a4, completely demolishing my queenside structure and leaving me to defend a position in which for a meagre pawn I'd have a very passive position with unclear prospects. Or I could play 11...Nb6, to which he'd more than likely reply 12.h4 and I'd have to work very hard not to get mated. Remember I was playing Kasparov, who is universally accepted to be one of the better dynamic attacking players in the history of chess. So from the practical point of view I think 11...Nf6 was a somewhat better idea, but once again I was very curious to see what exactly would befall me if I played 11...Nb6 against Kasparov... So I did! 11... ♘b6 12. h4 ♘c6 Black obviously needs to create some kind of counterplay or he'll get mated for absolutely nothing. 13. h5 ♖xf3 Other options for Black really aren't worth mentioning. 13... ♘xd4 14. ♘xd4 ♗xd4 15. ♖d1 ( 15. hxg6 ♗xe3 16. gxh7+♔h8 17. fxe3 ♕d6 18. ♘e4 ♕e5 ) 15... c5 16. hxg6 hxg6 17.♕c2+− Black is already almost lost here. 14. gxf3 ♘xd4 15. ♖d1! At this point I realised - not that it surprised me greatly - just how difficult my task was, because I thought my reaction was the most principled one and that this was how Black should react to a new idea: tackling it head-on. But after 15.Rd1 I realised that this was perhaps something my opponent was expecting and prepared for and that my position was very, very risky. 15... c5 Pretty much forced. 15... ♘xf3+? 16. ♔e2 ♘d4+ 17. ♗xd4 ♗xd4 18. hxg6+− 16. ♗xd4 16. hxg6 ♘xf3+ 17. ♔e2 ♘d4+∞ 16... cxd4 17. hxg6 h6 I thought this was a reasonable idea as I don't want to be opening files on the kingside. For a second here I felt mildly optimistic because I thought that on my next move I'd play ...Nd5 and he'd have to either take the knight - when I'd have some kind of a pawn fist in the center I could at least play behind - or if he didn't take I'd be blocking off some of the avenues for his attack and I'd again have something to base my position on. The knight on d5 would be a very strong piece. 17... hxg6 has actually been played since, but White has two incredibly strong moves in reply: a) 18. ♕c2 ♕e8 ( 18... ♕f8 is almost forced. 19. ♕xg6 ♕f5 20.♕e8+ ♕f8 21. ♕xf8+ ♔xf8 22. ♘e2+/= If this endgame isn't lost it's pretty close to it.) 19. ♗d3 White will probably give mate. 19...♗b7 ( 19... dxc3 20. ♗xg6 ♕f8 21. ♗h7+ ♔h8 22. ♗e4+♔g8 23. ♗xa8 ♘xa8 24. ♕h7+ ♔f7 25. bxc3 ♘b6 26. ♖g1♘d7 27. ♖d4 ♕h8 28. ♖h4+− ) 20. ♗xg6 ♕b8 21. ♘e2! ( 21.♘e4 ♕f4 22. ♕e2 ♗d5 Lima-Guimaraes, Araruama 2010) 21...♗xf3 22. ♖h4± b) 18. ♖g1! 18... ♘d5 ( 18... ♕e8 19. ♘e2 ♕f7 20. ♘xd4+− )19. ♘xd5 ( 19. ♘e2 e5 20. f4 ♗f5± ) 19... exd5 20. ♖xg6+/= 18. ♖h5! Kasparov played this after a little thought. This is when I finally figured out that this probably wasn't going to be my day. Now that ...Nd5 is impossible White will be free to develop his initiative if I do nothing and I also don't really have any useful moves apart from ...Qe8. I remember Kasparov being very proud of this move after the game, which should tell you something.

18. ♗g2⁈ ♘c4 19. f4 ♖a7+/= 18. f4 ♕d6 19. ♗g2 ♖a7 20. ♖h4± 18. ♖h4 Shabalov decided to "improve" on Kasparov's play in 2010 and also won quite convincingly, though I think 18.Rh5 is stronger: 18...♘d5 19. ♖hxd4 ♗xd4 20. ♖xd4 ♗b7 21. ♗h3 ♕d6 22. ♖e4+/=Shabalov-Kaufman, Irvine 2010 18... ♕e8 18... ♕c7 19. ♘e2 ♕d6 20. ♘xd4! ♗xd4 21. ♖h4 ♕e5+ 22.♖e4 ♗xf2+ 23. ♔xf2+− 18... ♔h8 19. ♘e2 e5 20. ♕f7 ♕g8 21. ♕xe7+− 18... ♖a7 19. ♘e2 ♘c4 20. ♘xd4 ! 20... ♗xd4 ( 20... ♖d7 21.♘c6 ) 21. ♗xc4 bxc4 22. ♕e3± 19. ♘e2 ♕xg6 20. ♖h1 Here I realised that I was in a lot of trouble because I was more or less forced to take the pawn on g6, which I was very much counting on using as a kind of human shield for the remainder of the game. It was blockading one of the avenues along which I'm getting mated later on and the pawn on d4 is also now falling which makes White's attacking task much easier. I'll go through the remainder of the game reasonably fast because I think by this point Black's position is beyond repair. 20... ♔h8 20... ♕f7⁉ 21. ♘xd4 e5 22. ♘c6 ♕xb3 23. axb3± Taleb-Khachiyan, Abu Dhabi 2000 21. ♖g1 ♕f7 21... ♕f6 22. ♖xd4 ♗b7 23. ♖dg4 ♗d5 24. ♕d3± 21... ♕h5 22. f4 ♘d5 23. ♕g3 ♕f7 24. ♗g2± 22. ♘xd4 ♘d5 Once again, if Black could just play ...Bd7, ...Rf8 and maybe one more move he'd be doing ok-ish, but that isn't going to happen. 23. ♕d3 ♗d7 23... e5 24. ♘xb5 ( 24. ♘c6 ) 24... ♗b7 25. ♘a3 ♘f4 26. ♕e3+− 24. ♕e4 White is just in time to install his pieces in the proper positions. 24... ♖c8 24... ♖g8 25. ♗d3 ♘f6 26. ♕g6± 25. ♗d3 Threatening mate-in-one. 25... ♘f6 25... ♕g8 26. ♖g6 ♘f6 27. ♕h4

26. ♕h4 You can see how all White's pieces are just "clicking" into place. I think Kasparov's play throughout most if not all of this game is exemplary. 26... ♖c5 Trying to establish some sort of defence along the fifth rank. 26... ♘d5 27. ♖g6 ♕f4 28. ♖g4 ♕f6 29. ♕g3 27. ♘e2! Another excellent move because now the knight is sometimes going to f4 27... ♘d5 Not really a particularly good move, but I think nothing really helps Black here. 27... e5 would be met by 28. ♘c3!+− with the knight perhaps then going to e4. Black is just stuck watching White improve his position and give mate. 28. ♖g6 ♕f8 You can't allow Rxh6+. 28... ♗e8 29. ♖xh6+! 29. ♕e4 Once again threatening Rxh6+ and Qh7#. 29... ♕f7 30. ♔d2! A very nice finishing touch. Now Rdg1 or Rdh1 are coming with devastating effect. 30... ♘f6 31. ♕e3 Here I resigned. I've been criticised for this particular resignation because people felt it was a bit premature, but in this position I think I was perfectly justified in stopping the clocks. You'll see the same decision in the second game and there I really could and maybe should have continued. Anyway, I was quite happy with this game because I felt that I'd been deservedly beaten by a legend and it wasn't really through my own bad play. I was playing someone who was better prepared and in general a better player than I am. 1-0

Viswanathan Anand 1:0 Peter Svidler Linares27.02.1999 I went back home and thought I repaired the line, so when I went to Linares in February I was more than happy to repeat it because I now knew what to do. 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6.♕xc4 O-O 7. e4 a6 8. e5 b5 9. ♕b3 ♘fd7 I got the same position as against Kasparov in Round 6 against Vishy Anand. I have to say I started quite well in Linares. In Round 1 I almost beat Kasparov with Black. I felt confident that the tournament was going well and by Round 5 I was on +1 and, if anything, I felt I could have had more points. Then I lost against Vishy and also lost the three games that followed, which didn't really improve my mood at all. This game was the beginning of what was by far the longest series of losses in my career. We reached this position reasonably fast and here Vishy went for 10. h4 , giving me a bit of a break from the 10.e6 line. 10... c5 11. e6 c4 12. ♕d1 The correct move, but in his notes to the game Vishy said he hadn't concentrated on this particular line very much, so he wasn't as booked up as he might have been in some other positions.

12... ♘b6 13. exf7+ ♖xf7 14. h5 ♘c6 15. hxg6 hxg6 16. ♗e3This is the "tabiya" of the 10.h4 line. I'd like to remind you that I covered this quite extensively in my 5.Qb3 chapters and in this position, apart from the move I made in this game and the most obvious move in the position - 16...Bg4, Black has a very interesting idea with 16...Nd5. I think that's very sensible and it's highly recommended that you look at it. In this game I started to think and came up with 16... ♗f5⁉ Maybe this isn't the best move in the position but it's certainly very playable. 17. ♘g5 Otherwise it's not clear what he's doing. 17... ♖f6 17... ♘xd4⁉ is also very playable, but I was consumed by the idea of the 17...Rf6 line. 18. ♘xf7 ♔xf7 19. ♖c1 e5 20. g4 ♗e6 21. ♗g2♖a7=/∞ 18. g4 ♗e6! 19. ♘ce4 19. ♗g2 ♗d5 20. ♘xd5 ♘xd5=/+ 19... ♗d5 It very much appealed to my aesthetic sense just to leave the rook on f6 en prise and occupy the allimportant d5-square. 20. ♕d2 I think Vishy's move is correct. 20. ♘xf6+ is of course unattractive - 20... exf6 21. ♘f3 ♕d7 and Black is in total control. ( 21... ♘a4 ) 20. f3⁈ was tried immediately here by Rustam Kasimdzhanov in a later game, but after 20... e5! 21. dxe5 ♘xe5 he was perhaps already worse against Khalifman in a game he didn't manage to draw. 22. ♗e2 ♗xe423. ♘xe4 ♖e6 24. g5 ♕c7 25. ♕c2 ♖d8 26. ♖d1 ♘d5∓KasimdzhanovKhalifman, Wijk aan Zee 2002 20... ♖d6? I got very carried away here. Sometimes during a game I have a feeling that I normally treasure and generally trust - it rarely lets me down - of being completely in control, of knowing what's what and what the correct move is. In my particular case when I feel like that I think it's likely that some kind of left-field solution will come to mind. Here the move I made on the board appealed to that sense so much because I felt it was a rare case of a move where the rook goes from one square where it's attacked to another square on which it's also attacked. I also thought it was a decent move in general because I'm trying to capture on d4, which is a very sensible idea. Unfortunately it has tactical problems, as you'll soon see. 20... ♕d7! is the move Black should play instead. I've tried many times in my life to analyse the position after 21. ♗g2 but always stopped short. I don't really see how you can analyse such a complete mess. ( 21. f3♖xf3! ) 21... a5 The fact that this is the first line of the machine (or at least of my machine), to which White replies 22. a3 , is just too much for me in my current state. Since this position really isn't very critical to our story I'll leave it at that. 21. f3! This is what I'd missed. The threat of Qh2 is just incredibly powerful. Now Black is in a lot of trouble, and I was unable to save the game. It's very important that the rook is no longer on f6 as otherwise Black would be able to play ...Rxf3 in this position and he'd be doing well. Now that's impossible and Black is under a severe attack. The remainder of the game is a very instructive attack by Vishy. 21... ♗xe4 I could see no other choice.

21... ♘xd4? 22. ♗xd4 ♗xe4 23. ♗xg7 ♖xd2 24. ♗e5+− 22. fxe4 ♘d7 22... ♘xd4 loses pretty much by force. A very nice variation, for instance, is: 23. ♗xd4! ♖xd4 ( 23... ♗xd4 24. ♕h2 ) 24. ♕f2! ( 24.♕h2 ♕d6 ) 24... ♕f8 ( 24... ♗f6 loses on the spot to 25. ♖h8+!+−Quite beautiful!) 25. ♕h4 ♕f6 26. ♗e2!+− 23. ♕h2 ♘f8 23... ♕b6 24. ♗xc4+ bxc4 25. ♕h7+ ♔f8 26. O-O ♘f6 27. ♕xg6♔g8 28. ♕f7+ ♔h8 29. ♖f3 24. e5! The attack develops very naturally. 24... ♖d7 24... ♖xd4 25. ♕f2 ♕a5+ 26. ♔e2+− 25. ♘e6! ♕a5+ I was very proud to have found this geometrical idea. 25... ♘xe6 is met by 26. ♗xc4! bxc4 ( 26... ♕a5+ 27. ♗d2 ♕xd2+28. ♔xd2 ♖xd4+ 29. ♔e3 bxc4 30. ♕h7+ ♔f7 31. ♖af1++− ) 27.♕h7+ Wherever the king goes the check on the f-file just wins on the spot e.g. 27... ♔f7 28. O-O+− 26. ♗d2 ♘xe5! A very beautiful move which unfortunately doesn't help because White doesn't take anything. 26... ♕b6 27. ♘xg7+− 27. ♗e2! c3 Forced. 27... ♕b6 28. ♘xf8+− 28. ♗xc3⁈ This isn't very good. 28. bxc3! was instead almost completely winning. 28... ♘f7 29. ♘xg7♔xg7 30. O-O+− 28... b4 29. ♘xg7 When looking at this position I realised I had my chance to enter the history books here and now and thought I had to grab it with both hands - a decision which will probably get me laughed out of the room in most polite company, but I just thought I had to do it because when would my next chance come? 29... bxc3? Vishy looked at me somewhat quizzically because this isn't a particularly good move. 29... ♘f7! I saw perfectly well that I could play this move, which almost forces 30. ♗d2 ♔xg7± but I honestly didn't believe I would save the game. 30. ♕h8+ ♔f7 Before he could react I looked up and said, "Vishy, please castle". He was somewhat surprised but then he smiled and did castle. The point of this dialogue was that I was very worried he might choose not to do that and instead play 31.Rf1+.

31. O-O Both moves win on the spot, but the whole point of my playing 29...bxc3 was that I wanted at least one of my games - and probably this will end up being that one game in my entire life - to end with castles and check on move 31. In general castling on move 31 is a very infrequent occurrence and castling with check and winning the game with the move is going to be even rarer, so I thought this was something I absolutely should have on my resume. So instead of continuing the game by playing 29...Nf7 I did that. Perhaps it was somewhat muddled thinking if you consider yourself a chess professional, but such was my decision. I finished Linares tied for dead last, although I almost rescued the tournament. After losing four in a row I won the next game and then had a completely winning position against Veselin Topalov which I eventually lost to finish on -3.

1-0

Anatoly Karpov 1:0 Peter Svidler Dos Hermanas09.04.1999 In that already somewhat depressed state - I think Linares was my first ever last place in a tournament and was kind of a new experience for me - I went to play the Dos Hermanas tournament in April. That also didn't go according to plan and one of the reasons it didn't go according to plan was this game against Karpov. 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♘f3 ♗g7 5. ♕b3 dxc4 6.♕xc4 OO 7. e4 a6 8. e5 b5 9. ♕b3 ♘fd7 10. e6 Karpov returned to this move, which I was very ready for. 10... fxe6 11. ♗e3 ♘f6 I blitzed out this move. 11... ♘b6 I think this move, which I played against Kasparov, is problematic for Black and there's no equality, but from a purely practical point of view getting the position I was about to get against Karpov in particular also wasn't a very brilliant idea. It's notable how I chose exactly the wrong positions against perhaps all three players. I chose an incredibly messy position by design against Vishy, who I have to say is very, very good in messy positions. I chose to try and withstand a direct attack against Kasparov and I chose a slightly worse endgame against Karpov. Well played me! 12. a4 I was prepared for this. My coach Andrey Lukin and I had looked at the position and I thought I knew everything I needed to know. 12... bxa4 12... b4 is a perfectly legitimate move in this position, as I pointed out in the 5.Qb3 chapters, and perhaps better in terms of simplicity of play 13. ♖xa4 ♘c6 14. ♗c4 14. ♗xa6? ♗xa6 15. ♕xe6+ ♔h8 16. ♕xc6 ♕c8! is very good for Black. 14... ♖b8 15. ♗xe6+ ♔h8 16. ♕c4 ♗xe6 16... ♘b4 17. ♘g5 ♘c2+ 18. ♔d1 ♘xe3+ 19. fxe3→ 16... ♕d6 17. ♘g5± 17. ♕xe6 ♕d6 This whole line was blitzed out by both myself and Karpov. Of course White isn't going to take on d6.

18. ♘g5 18. ♕xd6 cxd6 19. ♖xa6 ♘b4 18... ♖xb2 19. O-O! 19. ♘f7+ is premature as White needs to finish development first. 19...♖xf7 20. ♕xf7 e5! ( 20... ♘b4 21. OO± ) 21. ♗c1 ♘d8 ( 21...♖c2 22. ♗a3 ♕d7 23. ♕xd7 ♘xd7∞ ) 22. ♕c4 ♖c2↑ 19... ♘d8 I'm not quite sure why I played this, expertly heading for a slightly worse endgame against perhaps the most dangerous player of positions where he has an advantage and no risk. It's too long ago for me to remember whether I forgot something or I'd just missed 19.0-0 in my analysis at home, though that's kind of hard to believe as it's the most obvious move in the position. 19... ♖c2 is the best move. I want to show you a particular line which I found when I was preparing to make my video for chess24. It has no particular relevance but it ends in a position which I want to show you since this is, as I said, meant to be more infotainment than just purely informative. 20. ♖c4 ♘a5 21. ♖c5 These moves are quite forced.21... ♘b7! A more critical move. ( 21... ♕xe6 22. ♘xe6 ♖c8 23.♖b1 ♘d7 24. ♖xa5 ♖xc3 25. ♖xa6 ♘f8+/= , and Black should hold but couldn't claim to have equalised completely in Sasikiran-Bacrot, Lausanne 2001) 22. ♘f7+ ♖xf7 23. ♕xf7 ♘xc5 24. dxc5 ♕e5 (24... ♕d8 allows White to catch the rook with 25. ♕b3! as 25... ♕d3loses to 26. ♖d1± ; 24... ♕d3 straightaway also loses to 25. ♖d1 )25. ♖d1 h5 26. ♗d4 ( 26. ♘e4⁉ ♕xe4 followed by 27. ♗h6 is an attempt to give immediate mate and seems to be succeeding for a second, but then you find 27... ♘e8! and realise that Black is maybe even a little better. There's no mate and White's back rank is also quite weak.) 26... ♕f4 In this position I had very long analysis trying to prove this position is a draw, which may not be the case, but I want to show you a particular line which I think is very interesting geometrically. ( 26...♕g5 27. ♕xe7 ♖d2 28. h3+/= ) 27. ♕xe7 a) 27... c6 28. h3 ♖c1 ( 28... ♖d2 29. ♖xd2 ♕xd2 30. ♗xf6♕c1+ 31. ♘d1 ♕xd1+ 32. ♔h2 ♗xf6 33. ♕xf6+ ♔h7 34.♕xc6+− ) 29. ♗e5 ♖xd1+ 30. ♘xd1 ♕g5 31. ♘e3± b) ♖d2 28. ♕d8+ ♔h7 29. g3 ♕e4⁉ This is the move I wanted to show you as it's very appealing from an aesthetic point of view.( 29...♖xd1+ is probably objectively the strongest move for Black. 30.♘xd1 ♕f3 31. ♗xf6 Forced. 31... ♕xf6 32. ♕xc7 ♕e5 Black does have decent drawing chances but it's quite dificult to prove it's an absolute draw.) 30. ♘xe4⁈ This is the reason I started this whole thing. ( 30. ♖xd2 leads to a bishop endgame that White is probably winning after 30... ♕e1+ 31. ♔g2 ♕xd2 32. ♗xf6 ♕xd8 33.♗xd8 ♗xc3 34. ♗xc7 I'm not recommending you play this way with Black.) 30... ♖xd1+ 31. ♔g2 ♘xe4 32. ♕xc7 ♖xd4 33. c6 It seems as though White is winning here, but... 33... ♖d2 34. ♔f3♘xf2 35. ♕b7 ♘g4 36. c7 ♘xh2+ 37. ♔e4 (all of this is actually quite forced) 37... ♖c2 38. c8=Q ♖xc8 39. ♕xc8 ♘g4= This is the position I wanted to show you because I think not everybody knows that even if White somehow manages to exchange the pawn on g3 for Black's three remaining pawns - which I don't think he will - if the black knight gets to e5 then the setup with the king on h7, the bishop on g7 and the knight on e5 is a draw against the queen. This is a little-known nugget of a fact which I quite like and the whole reason I went through this very long and not particularly theoretically interesting line. I thought it might be interesting to some of you to learn that this is actually a fortress which in some rare cases you can actually aim for. 20. ♕h3 ♕d7⁈ 21. ♕xd7 Karpov very happily took.

21... ♘xd7 22. ♖xa6+/= This endgame is very difficult to annotate so I'll just give you the next 15 moves or so. You can marvel at how without doing very much Karpov just constantly improved his position and waited for my inaccuracies. By the first time control I was almost, or perhaps completely, lost. 22... ♘b6 22... c5⁈ 23. ♘a4 23. h4 Advancing on the kingside and gaining some space. 23... ♖c2 24. ♘ce4 h6 25. ♘f3 ♘c6 To begin with it seems as though Black isn't really doing all that badly because the weaknesses are somewhat covered and the rook on c2 is active. That's the first thing that White does something about. 26. ♖aa1 ♘b4 26... ♖d8 27. ♖ac1 ♖xc1 28. ♖xc1 ♘xd4 29. ♘xd4 ♗xd4 30.♗xh6+/= 27. ♖fc1 Exchanging Black's one active piece. 27... ♖xc1+ 27... ♘6d5 28. ♗d2 ♔h7+/= 28. ♖xc1 ♘4d5 29. ♘c5 ♖f6 30. ♗d2 Very quietly keeping the pressure on. White just improves his position step by step. 30... ♖d6 Perhaps aiming to play ...Nd7. 31. ♘b7 White doesn't allow that. 31... ♖f6 31... ♖d7⁈ 32. ♘e5 ♗xe5 33. dxe5± 31... ♖e6 would allow the attractive idea of 32. ♘d8 and Nc6. 32. ♘e5 The second knight comes in. 32... ♔h7 33. g3! Bit by bit White is improving his position and it becomes very, very difficult to breathe. 33... ♖f8 Trying to activate at least one piece. 34. ♘c5 ♖a8 35. ♖c2 Very quietly White prevents the rook from getting to the second rank. 35... ♖d8 35... ♖a1+ is tempting but I realised that after 36. ♔g2 ♖d1 37.♔f3+/= the king would just come over to e2 and drive the rook away. I would just be wasting another two tempi achieving absolutely nothing. 36. ♔g2 ♗xe5⁈ Perhaps in some mild time trouble I could tolerate it no longer and started exchanging pieces.

37. dxe5 ♘d7 38. ♘xd7! 38. ♘e6 also looks quite dangerous for Black, but Karpov's approach is even more classical. 38... ♖c8 39. ♔f3 c5 40. ♔e4 ♘c7 41. ♘xc7♖xc7+/= 38... ♖xd7 39. ♖c6± Fixing the entire black structure. The black knight now has absolutely no constructive moves, the rook on d7 is stuck protecting c7 and e7 and there's a constant threat of h4-h5, destroying the black kingside. I think here is as good a time as any to stop because the position is perhaps already lost and obviously I wasn't able to save it. This concludes the overview of the three 5.Qb3 games I lost to exceptional players in 1999. As you can perhaps glean from this chapter 1999 wasn't a particularly good year for me. By the end of it I was out of the Top 20 and out of the 2700 club and I think it took me the best part of two years to recover from that half-a-year stretch. But I did eventually recover and I didn't stop playing the Grünfeld. I never really blamed the Grünfeld for any of this and, in general, I don't think I even played particularly badly. I just encountered very strong players who played very well. I wanted to show you these games as an amusing anecdotal story about half a year in the life of a Grünfeld player to give you an idea of the opening when you're playing against strong players and what can happen to you if you're not very precise. I'm not sure what you'll think of this format but I thought I'd try it once and see what the reaction is. 1-0

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 7.Nf3 c5 8.Rb1 0-0 9.Be2 cxd4 10.cxd4 Qa5+ 11.Bd2 Qxa2 12.0-0 Bg4 13.d5 13.Be3

8.Rb1 and 13.d5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♘f3 c5 8. ♖b1 This line was immensely popular in the mid to late-90s and the start of the new millennium, due mainly to the efforts of Alexander Khalifman and Boris Gelfand, who worked on it both together and separately and really moved the theory forward. Alexander Chernin perhaps also deserves a mention as a player who did a lot for the popularity of this line. 8... O-O By far the most important move. 9. ♗e2 This line has been a source of worry for me over a number of years because, as you may already have surmised from my videos, I tend to look for replies to White's ideas in the Grünfeld which aren't very theory-heavy and don't involve memorising long lines just to make a draw. That's not always possible, however, and in particular in the 8.Rb1 line if Black wants to get a position which is "interesting" or non-standard he's very likely to end up playing something substandard where he runs the risk of being worse after the opening. Of course that didn't stop me on many occasions playing both the very playable 9...Nc6 and also 9...b6. 9... cxd4 Objectively Black should bite the bullet and follow the main line with this move. It could also be added that in the past 20 years the deeper the analysis went the clearer the situation became, and by now playing this line with Black doesn't require memorising quite as much as it once did. Black is actually reasonably comfortable in the main lines. 9... b6 10. O-O ♗b7 leads to unclear positions after both 11. ♕d3(and the interesting recent pawn sacrifice 11. d5 Those positions are playable but potentially dangerous for Black.) 10. cxd4 ♕a5+ 11. ♗d2 The main move. 11. ♕d2 ♕xd2+ 12. ♗xd2 b6 has been known to be fine for Black for a number of years. You can separate this line into two different structures - those in which White tries to refrain from playing d4-d5, and those where White plays d5 as early as possible. a) If White does something slow like 13. ♗e3 you could consider starting with 13... e6 , because in general fixing the pawns on e4 and d4 is a decent idea, and after the possible 14. ♗d3 ♗b7 15. ♔e2♘c6 16. ♖hc1 ♖fd8 it's quite clear that this is a very favourable version of the classical Grünfeld ending for Black. He's already attacking the white centre and, in general, White is much better off in positions like this if his knight is on e2 not f3. This line is simply comfortable for Black. ( 16... ♖fc8 17. ♖c4 1/2 Dao Thien Hai-Sasikiran, Esfahan 2005) 17. ♖c4 f5⇆

b) 13. d5 In general the moment the white d-pawn crosses this threshold Black should try to put the knight on c5. 13... ♘a6 14.♗e3 f5 Black is already fighting for the initiative and this sequence of only two moves shows Black's main ideas very succintly. In order to stop the knight coming to c5 White responds with Be3, and then the bishop immediately comes under fire from Black's f5-f4. Kasparov was soon even better here as Black, although that had a lot to do with the play of his opponent, Gerald Hertneck. 15. ♗c4 ( 15. e5⁈ f4 16.♗d4 ♗f5 17. ♖c1 ♘b4 18. ♗c4 ♘d3+ 19. ♔d2 ♘xc1 20.♖xc1 ♖fc8 21. d6+ ♔f8−+ Hertneck-Kasparov, Munich 1994) 15...♔h8 16. e5 f4⇆ c) 13. ♖c1 ♗b7 14. d5 ( 14. ♗d3 ♖d8 15. ♖c7 ♖d7 16.♖xd7 ♘xd7 17. ♔e2 ♖c8= is clearly fine for Black.) 14... ♘a615. ♗g5 ♖fc8 16. OO ♔f8 17. e5 h6 18. ♗h4 g5 19. ♗g3♖xc1 ( 19... ♘c5⁉ ) 20. ♖xc1 ♘c5∞ Pavlovic-Mikhalchishin, Trnava II 1988 d) 13. O-O The main line. 13... ♗b7 14. d5 ( 14. ♗d3 e6 15.♖fc1 ♘c6 16. ♗e3 ♖fd8 is once again very nice for Black.) 14...♗a6 White has a minor choice here. d1) 15. ♖fe1 ♗xe2 16. ♖xe2 d11) Black might consider starting with 16... ♖c8⁉ , as White's general idea is to attack the e-pawn next with 17. ♗g5 , trying to make Black's life at least slightly uncomfortable. After the rook move Black can now simply reply ( 17. ♗b4 ♖c7 18. e5♘a6 ) 17... ♔f8 and his position is fine, at the very least. 18.e5 h6 19. ♗f4 Koch-Bakalarz, Germany 1991 19... ♖c4!⇆ d12) ♘a6 is also a reasonable move.

d121) 17. ♗e3 ♖fe8 ( 17... f5⁉ 18. ♗g5 fxe4 19.♖xe4 ♖f5 20. ♖xe7 ♖xd5 21. ♗e3 ♗f8 22. ♖b7♘c5 23. ♗xc5 ♗xc5 24. h4 ♖ad8= LputianTseshkovsky, Kiev SU-ch 1986) 18. ♘d4 ♗xd4 19. ♗xd4 e6 20. dxe61/2 Khalifman-Tseshkovsky, Minsk 1985 d122) 17. ♗g5 17... ♖fe8 18. ♖d1 e6 19. dxe6 1/2 Krasenkow-Gavrikov, Budapest 1989 d2) 15. ♗xa6 15... ♘xa6 The following sequence is almost forced, as allowing Nc5 isn't appealing for White 16. ♗e3 f5 17.e5 f4 18. ♗c1 ♖ad8 19. ♖d1 ♘c7 20. d6 exd6 21. exd6♘e6! This endgame is very safe for Black and he has nothing to worry about. In the Grünfeld in general the passer on d6 either wins pretty much straight away, especially when White has bishops against bishop and knight, or it's safely blockaded and not a major concern for Black. (The old approach here was 21... ♖f5 , aiming to play Nd5 next move, but that runs into 22. g4!+/= and Black can't play 22... fxg3 because of 23. ♗g5 with serious tactical problems.) 22. ♗b2 ♘c5 ( 22... ♖f5 23. ♗xg7 ♔xg7 24.♖d2 ♘c5 25. ♖e1 ♔f6 26. h4 Ernst-Nijboer, Dieren 2002) 23.♗xg7 ♔xg7 24. ♖b4 ♖f5= 11... ♕xa2 12. O-O Black has a variety of possible moves here. I played 12...b6 against Vassily Ivanchuk in one of the Linares tournaments many, many years ago. Black also has the option of 12...Qe6, or there's the weird move that I believe Suat Atalik once played, 12...Na6. There's also 12...a5, which leads to crazy positions, but the main and safest move has been known for a while now to be 12... ♗g4 , when White has two major options, 13.d5 or 13.Bg5.

13. d5 In most cases when this move is made without gaining a tempo - i.e. without a knight on c6 - it benefits Black, firstly because it gives Black a target for breaks in the centre and secondly because the queen on a2 immediately feels completely secure. It can no longer be attacked as the g7-bishop controls all the squares on the long diagonal. Here in particular, after 13... ♘d7! 14. ♖xb7 ♖fb8 in order to keep control of the b-file, at least for a while, White sometimes plays 15. ♖c7 15. ♕b1 ♕xb1 16. ♖fxb1 ♖xb7 17. ♖xb7 ♘c5= was equal in a game played by my coach: S. IvanovLukin, Russia 1992 15... ♖c8 16. ♕b1 16. ♖xc8+ ♖xc8 17. h3 ♗xf3 18. ♗xf3 ♘e5= 16... ♕xb1 17. ♖xc8+ ♖xc8 18. ♖xb1 but Black experiences no problems whatsoever after either move here. 18... ♘c5 18... ♖c2= 19. h3 ♗d7 20. e5 ♗f5=

8.Rb1 and 13.Be3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♘f3 c5 8. ♖b1 OO 9. ♗e2 cxd4 10. cxd4 ♕a5+11. ♗d2 ♕xa2 12. O-O ♗g4 13. ♗e3 White could try 13. ♖xb7 , but after 13... ♗xf3 14. ♗xf3 ♗xd4Black is fine. 15. ♗b4 ( 15. e5 is simply met by 15... ♘a6 and Black is ready to attack the pawn on e5. 16. ♖xe7 ♖ad8= ) 15... ♘c6 ( 15...♖d8⁉ is also a very viable alternative ) 16. ♗xe7 ♖fe8 ( 16...♖ab8 17. ♖c7 ♘xe7 18. ♕xd4 ♕e6 19. ♕xa7 ♘c6 20. ♕c5♖fc8+/= Wojtaszek-Krasenkow, Lubniewice 2005 ) 17. ♗g5 ♗b6! A very strong move, creating the positional idea of Nd4. I checked this and think Black is very comfortable here. 18. e5 ( 18. ♕d5 ♕xd5 19.exd5 ♘d4 20. d6 ♖e5 21. ♗e7 ♘xf3+ 22. gxf3 ♗d8 23. ♗xd8 FiguraSchmidt, Police 2007) 18... ♘d4 19. ♖d7 ♘xf3+ 20. ♕xf3♕e6 21. ♖fd1 ♕f5 22. ♕xf5 gxf5 23. ♗f6 ♖ac8 24. ♖7d3 f4 25. g4= 13... ♘c6 Developing while also attacking the white centre. 14. d5 is almost forced. There was a lot of old theory starting with 14...Ne5, but there's no need to go into that here. 14. ♖xb7 ♖ab8 15. ♖xb8 ♖xb8 16. h3 ♗xf3 17. ♗xf3 ♖d8 Black wants a target to attack in the centre so he uses this move to force18. d5 ♘e5 Black already wants to play e6. 19. ♕c1 ( 19. ♗g5♗f6= ) 19... ♘c4 Correctly keeping the queen boxed in on c1. 20.♗g5 ♖e8 21. ♖e1 e6 An old game between Khalifman and Stohl was completely equal here and

soon finished in a draw. ( 21... a5⁉ ) 22.♖e2 ♕b3 23. dxe6 ♖xe6 24. ♖c2 ♘d6 25. ♖c6 h5 26. ♗e3 1/2 Khalifman-Stohl, BL 1997 14... ♘a5 15. ♗g5 Black tried a variety of moves in this position in the 90s. 15. ♖b4 makes no sense whatsoever without the inclusion of Bg5 and h6 first. 15... b6 Adequately protecting against Ra4. 16. ♗g5 ( 16. e5 is simply met by 16... ♗xf3 17. ♗xf3 ♗xe5 and White really needs to be taking a pawn on h6. Without that this is completely pointless. ) 16...♕a3 17. ♗xe7 ♖fe8 18. ♗d6 ♘b7 19. e5 ♗xf3 20. ♗xf3♗xe5!−+ ( 20... ♘xd6 21. ♖a4 ♕c5 22. exd6 ♕xd6 23. ♖a6∓ ) 15. ♖e1 b6

15. ♗c5 will of course be met by 15... b6 , transposing to the main line. 15... b6! was a major breakthrough. Black realised he can simply ignore the threat of Bxe7 and get a very safe position. 15... ♕a3 is very strongly met by 16. ♗d2 I believe I played 15... ♗xf3 16. ♗xf3 ♕a3 here against Kramnik once, but that's also less than ideal. 16. ♗xe7 This is what White should play. 16. ♖e1⁈ ♗c3 Correctly played by Ian Nepomniachtchi against Li Chao in the World Team Championship in Ningbo. ( 16... h6⁉ ) 17. ♘d2 a) 17... ♗xe2 is a comfortable way for Black to equalise. 18. ♖xe2♗f6 ( 18... f6 19. ♗h6 ) 19. ♗xf6 exf6 20. ♘b3 ♕a4= Li Chao2Nepomniachtchi, Ningbo 8th WChT 2011 b) ♗d7! is more amitious, after which it's hard for White to untangle himself. If Black assumes control of the c-file he'll simply be better, so White should probably play 18. ♗a6 A sample line showing Black's ideas very clearly is: 18... e6 19. ♕c1 Trying to drive back the bishop. ( 19. d6 f6 20. ♗h6 ♗a4 21. ♕c1 ♗c2 22. ♗xf8 ♖xf8) 19... ♗g7⁉ 20. d6 h6 21. ♗e7 ♖fc8! 22. ♗xc8 ♖xc8 23.♕d1 ♖c2 , and despite being a pawn down I believe it's Black who's playing for a win. White is completely discoordinated, the bishop on e7 is far from the action and the moment the black pawns on the queenside start to roll it will be very dangerous for White. I'd very much like to get this position as Black. 24. ♖e2 ♘c6 25. ♖c1♖xc1 26. ♕xc1 b5=/+ 16. h3 ♗xf3 17. ♗xf3 a) 17... ♖fe8⁈ 18. d6 exd6 19. e5 dxe5 ( 19... ♖xe5 20. ♗xa8♖xg5 21. ♕xd6+/= ; 19... ♗xe5 20. ♗d5 ♕a3 21. ♗xa8 ♖xa822. ♕d5± ) 20. ♗d5 ♕a3 21. ♗xa8 ♖xa8 22. ♕d5 ♖f8 23.♖fc1+/= b) ♕a3 This is very comfortable for Black with the pawn on b6. b1) 18. ♕e2 ♕d6 19. ♖fc1 ♖fc8 20. ♕a6 ♖xc1+ 21. ♖xc1h6 22. ♗d2 ( 22. ♗e3 ♘b3 23. ♖c6 ♕b4 ) 22... ♘b3 23.♖c6 ♕d8 ( 23... ♕b8 24. ♗e3 ♘d4 25. ♖c4 ♘xf3+ 26.gxf3 ) 24. ♗e3 ♘d4 25. ♖c4 ♘xf3+ 26. gxf3 ♕d7 27. ♔g2e6 28. dxe6 ♕xe6 29. ♖c7=

b2) 18. ♗d2 18... ♕d6 The most straightforward way to proceed. ( 18... ♗c3 19. ♗f4 ) 19. ♗b4 ♕e5 A very simple plan for Black here is to play Nb7-c5 against almost any two moves by White, and the moment the knight reaches c5 and closes the c-file Black will clearly be very comfortable. 20. ♕d3 ( 20. ♗g4 ♘b7=/+ ;20. ♖c1 ♘b7 21. ♖c6 ♘c5=/+ ; 20. d6? ♖fd8∓ ; 20. ♕a4♘b7 21. ♖fc1 ♘c5 22. ♗xc5 bxc5 23. ♖xc5 ♖fb8= ) 20...♘b7 21. ♖fc1 ♘c5= ( 21... ♘d6 22. ♖c6 f5 23. exf5 ♕xf524. ♕xf5 ♘xf5 25. d6 ♘d4 26. dxe7 ♘xf3+ 27. gxf3 ♖fe828. ♖c7 a5 29. ♗a3 ♗d4= ) 22. ♕a3 ♕g5= ( 22... a5 23.♗c3 ♕d6 24. ♗xg7 ♔xg7 25. e5 ; 22... ♘xe4 23. ♗xe4♕xe4 24. ♗xe7 ♕xd5= ) 16... ♖fe8 17. d6 Almost forced. 17... ♘c6 Here it transpires that Black is prepared to give up the exchange in order to swap off all of White's bishops, and Black seems to be very comfortable in the ensuing positions. When White realised that he started to look for ways to avoid playing 18.Bb5. 18. ♗b5 Almost forced. Etienne Bacrot played 18. ♕c1 against me in the 2010 French Team Championship. 18... ♕xe2 (That very clever move is based on the idea that 18... ♘xe7 runs into the strong 19. ♗c4 ♕a4 20. ♘g5 and Black has tactical problems. ) 19. ♕xc6 ♖ec8 (It's important to note that19... ♖ac8 runs into 20. d7 and Black has to resign, which isn't ideal. ) a) In the game Etienne played 20. ♕d5 ♗e6 21. ♕b7 ♖cb8 22.♕c6 ♖c8 23. ♕a4⁈ He felt obliged to play for a win, as his board was important for the result of the match. (I believe if this had been a game played in a normal tournament and not a team tournament Etienne would have played 23. ♕b7= because objectively White simply isn't better.) 23... ♕c2 24. ♕xc2 ♖xc2 25. e5 ♗d7 A very precise move, removing the bishop from all kinds of Nd4 jumps, after which I was already better and was ultimately very disappointed not to win that interesting game. It involved crazy complications towards the first time control, but generally speaking Black's pawns are simply faster and more dangerous than White's safelyblockaded pawns. 26.♘e1 ♖c4 27. f4 b5 28. ♖b2 a5 Bacrot-Svidler, Marseilles 2010 29. ♖c2 ♖xc2 30. ♘xc2 b4 31. ♘d4 a4∓ b) 20. ♕b7 This is what White has to start with if he wants to play on, but after 20... ♖cb8 21. ♕d5 ♗e6 22. ♕d1 ♕xd1 23.♖fxd1 ♖b7 24. ♖bc1 ♗d7 25. ♖c7 ♖ab8 26. e5 a5 Bl ack clearly has his own trumps in this position, while his blockade of the d-pawn is very convincing and will take a while for White to dismantle. The black pawns on the a and b-files are very dangerous, so at the very least Black is no worse here - I checked this position in some detail. 27.♗f6 a4 28. ♘d4 ♗xf6 ( 28... b5 29. ♗xg7 ♔xg7 30. e6 fxe631. ♘c6 ) 29. exf6 b5 30. ♘f3 ( 30. ♘c2 b4 31. ♘xb4 a3 32.♘a2 ♔f8 33. h3 ♔e8= ) b1) 30... b4 31. ♘e5 ♗e8 32. ♖xb7 ( 32. ♖e7 a3 ) 32...♖xb7 33. d7 ♗xd7 34. ♘xd7 h5 35. ♘c5 ♖a7 36. ♖a1 a337. ♔f1 a2 38. h3 ♖c7 39. ♘b3 ♖c2 40. ♔e1 ♖b2 41.♘d2 b3 42. ♘xb3 ♖xb3 43. ♖xa2 ♖b6= b2) 30... a3 31. ♘e5 ♗e8 32. ♖xb7 ♖xb7 33. d7 ♗xd7 34.♘xd7 h6 35. ♘c5 ♖c7= 18... ♘xe7 19. h3 The main move. Here Gelfand played 19. ♗xe8 ♖xe8 20. dxe7 against Loek Van Wely in Wijk aan Zee, to which Loek replied with the very precise 20... ♕a321. ♖e1 (If White simply allows Black to take on e7 and keep both bishops then I believe Black will be playing for a win, so Boris replied 21.h3 after which Loek exchanged everything off with 21... ♗xf3 22.♕xf3 ♕xf3 23. gxf3 ♖xe7 With such a kingside structure White

obviously has no hope of an advantage and the game was soon agreed a draw. 24. ♖fd1 ♗e5 25. ♔f1 ♔g7 26. ♔e2 1/2 Gelfand-Van Wely, WaZ 2006 ) 21... ♗xf3 22. ♕xf3 ♕xf3 23. gxf3 ♖xe7 24. f4 ♗d4= 19... ♗xf3 20. ♕xf3 ♕e6 20... ♖ec8 21. dxe7 ♕e6 22. e8=Q+ ♖xe8 23. ♗xe8 ♖xe8 24.♖fd1 21. ♗xe8 ♖xe8 22. dxe7 I reached this position against Yury Shulman in the same World Team Championship in Ningbo 2011. Here I began to think. I remembered there was a game starting in this position where the bishop somehow reached c5. I looked for alternatives and came up with an idea I'm quite proud of – 22... ♕xe7 It seemed to me that after 22... ♖xe7 the best move was 23. ♖fd1 and after the almost forced ( 23. ♖fe1 was, as I discovered when I got back to my room, the way the game between Gelfand and Shirov progressed, and after 23... ♗d4 the position was incredibly unclear and Shirov eventually went on to win a brilliant game. 24. ♖bd1 ♕e5 25. ♖d3a5 26. ♕d1 ♗c5 27. ♖e2 ♖e6 28. g3 ♖d6 29. ♔g2 ♖xd3 30.♕xd3 a4⇆ Gelfand-Shirov, Odessa Pivdenny Bank Cup 2007 ) 23...♕xe4 24. ♕xe4 ( 24. ♖d8+ ♗f8 25. ♕xe4 ♖xe4 26. ♖d7 a5 27.♖xb6 a4 28. ♔f1 a3= ) 24... ♖xe4 25. ♖d7 it's very likely that White will win back one of the pawns on the queenside. By that time Black will probably have put the bishop on c5, so he won't be completely helpless, but I felt that defending a position like that for seven hours wouldn't be an ideal use of my time. 25... ♗d4 26. ♖xa7 ♖e2 27.♖f1 ♔g7+/= 23. ♖fd1 Played immediately, which confirmed my assessment of what Yury's plan was. 23... ♗f8! After this move Black is absolutely ready to take on e4 if White does nothing. After sinking into quite deep thought Yury ended up playing 24. ♖d4 , resigning himself to the fact that he had no advantage. By nothing here I mean something like 24. ♖bc1 ♕xe4 25. ♕xe4♖xe4 26. ♖c7 ( 26. ♖c6 ♗c5 27. ♖d7 ♖f4 ) 26... a5=/+ 24... ♕e6 25. ♕b3 ♕xb3 26. ♖xb3 ♗c5 Shulman switched to defence with 27. ♖a4 a5= and in this position it's quite obvious that Black can be no worse. If he could somehow rapidly swing the king over to c6 he'd even be much better, but White is in time to stop that plan. At some point I allowed Yury to return the exchange and the game was drawn. Shulman-Svidler, World Team Ch, 2011.

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 7.Nf3 c5 8.Rb1 0-0 9.Be2 cxd4 10.cxd4 Qa5+ 11.Bd2 Qxa2 12.0-0 Bg4 13.Bg5 h6 14.Bh4 14.Be3

8.Rb1 and 13.Bg5 h6 14.Bh4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♘f3 c5 8. ♖b1 OO 9. ♗e2 cxd4 10. cxd4 ♕a5+11. ♗d2 ♕xa2 12. O-O ♗g4 13. ♗g5 This attack on the e7-pawn isn't yet a genuine attack. 13... h6 14. ♗h4⁉ This used to give White very interesting and non-standard positions, but in the main line Black has found a very concrete solution to his problems. 14. ♗xe7 isn't a serious option. After 14... ♖e8 15. ♖xb7 Black has a choice between the comfortable 15... ♘d7 (or 15... ♘c6 which was fine for Black in Kramnik-Kasparov many years back. This line has completely disappeared from grandmaster practice. ) 14... g5! 15. ♗g3 ♘c6 16. d5 ♖ad8 Immediately starting counterplay against the white centre. In practice White has only played 17.Qe1 and 17.Rxb7, but I want to cover the other moves as well. 17. ♖xb7 The main move. For a number of years it was considered problematic for Black until a game between Kramnik and Anand at the Tal Memorial in 2009. 17. ♕d3?∓ f5

17. ♕e1= ♗xf3 18. ♗xf3 a) After 18... ♘e5 Black should draw with relative ease. 19. ♖xb7 (19. ♗e2⁉ ♕c2 20. f4 gxf4 21. ♖xf4 ♘g6 22. ♖c1 ♕b2∞ ; 19.♗h5⁉=/∞ ) 19... e6 20. ♗xe5 ♗xe5 21. dxe6 ♕xe6 22. ♖xa7♖a8= b) ♘d4 19. ♖xb7 Here both 19...e6 and 19... f5 have been shown to be quite adequate for Black. ( 19. ♗g4 ♕a6 20. ♗c7 ♖a8 21.f4 ) 19... f5 ( 19... e6 20. dxe6 ♕xe6 21. ♖xa7 f5 22. exf5 ♘xf3+ 23. gxf3 ♕xf5 24. ♕e7 ♕f7 25. ♖e1+/= ) 20. exf5♘xf3+ 21. gxf3 ♕xd5 22. ♖xa7 ♖de8 23. ♖a5 ♕xf3 24.♕e6+ ♔h8 25. ♖e1 ♗f6= 17. h3= is lukewarm, to say the least. After 17... ♗xf3 18. ♗xf3 e619. ♖xb7 ( 19. ♗c7 ♖d7 20. d6 ♗e5 ) 19... ♘e5= we get a very typical position in which Black doesn't even attempt to hang on to the b7-pawn. Instead he attacks the white centre and tries to force White to trade one of the bishops for the knight on e5. Black once again has no problems. 17. ♖e1= involves a very beautiful idea. a) After 17... ♗c3 White will play 18. ♘xg5 and 18... ♗xe1 is met by ( 18... ♗xe2 is impossible because of 19. ♖xe2 , hitting the queen on a2.) 19. ♗xg4↑ giving White a very serious initiative on the kingside.

b) 17... ♗xf3 is therefore what Black should play, when after 18.♗xf3 e6 White has the additional option of 19. ♖e2 ( 19. ♖xb7♘e5= ; 19. d6 ♗e5 20. ♖e2 ♕a3 21. ♖d2 ♖d7=/+ ) 19...♕a5 ( 19... ♕a6⁉ , trying to keep everything under control, runs into 20. ♖d2 ♘e5 21. ♗e2 and it turns out that holding on to the b7-pawn is more difficult than Black originally thought. ) 20. ♖d2Taking control of the d4square. 20... ♘e5 21. ♖xb7 It initially seems as though White has made some progress here. The rook is obviously much better on d2 than f1, the d5-pawn is very well-protected and White is ready to develop some sort of initiative, but Black has a very concrete solution that starts with 21... ♘c4 22.♖c2 (If 22. ♖e2 the knight simply returns: 22... ♘e5= Note this would be impossible if Black had started with 21...exd5. ) 22...exd5 23. exd5 ♖fe8 Creating the huge threat of Re1+. Black has a lot of counterplay and White just isn't better at all. b1) 24. ♖c1 g4 b11) 25. ♗h4 ♘b2 ( 25... gxf3 26. ♗xd8 ♕xd8 27. ♕xf3♘d6 ) 26. ♗xd8 ♖xd8 27. ♕e2 gxf3 28. ♕xf3 ♕xd5 29.♕xd5 ♖xd5 30. ♖xa7= b12) 25. ♗xg4 ♖xd5 26. ♕f1 ♘d2∓ b13) 25. ♗c7 ♕a6!=/+

b14) 25. ♗e2 25... ♖xd5 26. ♕c2 ♘d6= b2) 24. ♖e2 24... ♖xe2 25. ♕xe2 ♘d2 26. ♗h5 ♕a1+ 27.♕d1 ♕xd1+ 28. ♗xd1 ♖xd5 29. ♗h5 ♖a5= 17... e6 17... f5 18. h3! 18. ♗c7 After 18. d6 the simplest way to proceed is 18... ♗xf3 19. ♗xf3♕a6= , preparing to play Ne5 next move, and once again the pawn on d6 will be unsustainable and the game will soon end in complete equality. 20.♖c7 ♘e5 21. d7 ♕b6= 18... ♗xf3! Vishy's huge improvement, which solves all Black's problems. 18... ♖d7 19. ♗b5± is just bad for Black, which is why the whole line was considered dubious. 18... exd5 19. ♗xd8 dxe4 20. ♗e7 ♖e8 21. ♗c5 ♖d8 22. ♕b3♕xb3 23. ♖xb3 ♖d5 24. ♗a3 exf3 25. ♗xf3 ♗xf3 26. ♖xf3±Chernin-Horvath, Hungary 1992 19. ♗xf3 19. gxf3 ♘d4 is clearly unattractive for White. 19... ♖d7 This is played only now, and the absence of the Bb5 move means that Black gets additional time to attack White's somewhat entangled pieces. Kramnik played 20. dxc6 The queen sacrifice looks very promising, but Black is just in time.

20. ♕b1 ♕xb1 21. ♖fxb1 ♖c8! This move equalises on the spot. (It's important not to make the mistake Bykhovsky made with 21... exd5when he got into some trouble. 22. ♗g4 f5 23. ♗xf5 ♖e7 24.exd5+− Gareev-Bykhovsky, Berkeley Intl. op 2011 ) 22. dxc6 ( 22. d6♗f8= ; or 22. ♗g3 ♖xb7 23. ♖xb7 ♘d4 when Black's activity clearly compensates for the passed pawn on d5. ) 22... ♖dxc7= is just a draw. 20... ♖xd1 21. ♖xd1 ♕c2 22. ♗d6 g4 Very precise - Black manages to take on c6 and liquidate into a drawn position. 23. ♗xg4 ♕xc6 24. ♖c7 ♕b6 25. ♖d7 25. e5 ♖d8 26. ♗h5 ♗f8 27. ♗xf7+ ♔h8=/+ 25... ♖d8 26. ♗c7 ♖xd7 27. ♗xb6 ♖xd1+ 28. ♗xd1 axb6 29.♔f1 1/ 2 Kramnik-Anand, Moscow Tal Mem 2009

8.Rb1 and 13.Bg5 h6 14.Be3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♘f3 c5 8. ♖b1 OO 9. ♗e2 cxd4 10. cxd4 ♕a5+11. ♗d2 ♕xa2 12. O-O ♗g4 13. ♗g5 h6 14. ♗e3 ♘c6 15. d5 15. ♖xb7 ♖ab8 16. ♖xb8 ♖xb8 17. h3 ♗xf3 (Black might consider playing 17... ♗d7⁉ 18. d5 ♘e5 19. ♗f4 ♕b2 and once again the fact that White can win the pawn on h6 is less important than the potentially huge passed pawn on the afile. ) 18. ♗xf3 ♖d8 19. d5♘e5 20. ♕c1 It would now be less advisable to play 20...Nc4 (the move previously recommended in a similar position), as the pawn on h6 is hanging, but Black can simply play 20... a5= , ignoring the pawn on h6 as the passer on the a-file is a more important factor. 21. ♕c7 ♖e8 22.♗f4 In the game Arun Prasad-Negi, Nagpur 2008, Black played 22...Qb2 and held, although it's also possible to play 22... ♘xf3+ 23. gxf3 ♕b3=, which perhaps holds even more comfortably. 15... ♘a5 15... ♗xf3 a) 16. ♗xf3 ♘e5 17. ♖xb7 a5! 18. ♖xe7 ( 18. ♗c5 a4 19.♗xe7 ♖fb8 20. ♖xb8+ ♖xb8 21. d6 ♕e6 22. ♕xa4 ♘xf3+ 23.gxf3 ♕h3 24. ♕a5 ♕xf3 25. ♕d5 ♕g4+ Van Wely-Svidler, Biel 2000 26. ♔h1 ♕f4!=/+ ) 18... a4 19. ♗d4 ♘xf3+ 20. gxf3♗xd4 21. ♕xd4 a3 22. d6 ♕e2 Kramnik-Svidler, Zurich 2001 b) 16. gxf3 16... ♘d4 ( 16... ♘a5 17. ♕d2!+/= ) 17. ♗d3 ♕a318. f4 ♕d6 19. ♖xb7 ♖fb8 20. ♕b1 ♘f3+ 21. ♔g2 ♘h4+ 22.♔h1 g5 23. ♖g1 gxf4 24. ♗c5! ♕e5 25. ♗d4 ♖xb7 26.♕xb7 ♕xd4 27. ♕xa8+ ♔h7 28. ♗b1+/= Krasenkow-Svidler, Polanica Zdroj 2000 16. ♗c5 The main move. 16. h3 ♗xf3 17. ♗xf3 ♘c4= 16. ♖b4

a) If Black plays 16... b6⁉ White has 17. e5 ♗xf3 18. ♗xf3 ♗xe5( 18... ♖ac8 19. d6!± ) 19. d6 The fact that the pawn on h6 is hanging makes this position much less comfortable for Black than in the Be3 line, and some precision is required. I spent some time looking for a solution that would be completely satisfactory. a1) It's possible to give up the exchange with 19... ♗xd6 20.♖a4 ♕b2 21. ♗xa8 which isn't too bad for Black. 21... ♖xa822. ♕f3 ♖c8 23. ♗xh6 ♖c4 24. ♖xc4 ♘xc4 25. g3 b5 26.h4 ♔h7 27. ♗c1 ♕f6 28. ♕c6 ♕f5=/+ Giddins-Baquero, Amsterdam 2005 a2) 19... ♕a3 20. ♕b1 ♗xd6 21. ♖g4 ( 21. ♗d5 ♕xb4 22.♕xg6+ ♔h8 23. ♕xh6+ ♔g8= ) 21... ♔g7 22. ♗xa8 ( 22.♗d5 f5 23. ♖h4 h5 24. ♖d1 ♗e5 ) a3) 19... ♕e6 20. ♗xa8 ♕xd6 was the idea I came up with, 21.♕d5 ♗xh2+ Forced. ( 21... ♕xb4 is met by 22. ♕xe5 ♖xa823. ♗xh6 and Black loses material. ) 22. ♔h1 ♕xb4 23. ♔xh2If the pawn was back on g7 I'd be quite happy to recommend this for Black as he does, after all, have four pawns for the piece, but with a pawn on g6 he runs a serious risk of coming under a menacing attack on the kingside. 23... e6 24. ♕e4 b) ♗xf3 is therefore my suggestion. 17. ♗xf3 ( 17. gxf3 b6 ) 17...♘c4 18. ♗d4 b1) 18... ♘d2 is the normal theoretical move. It's supposed to lead to positions that are completely equal, and perhaps they are. 19.♖b2 ♘xf3+ 20. ♕xf3 ♕a6 21. ♗xg7 ♔xg7 22. ♕c3+ ♔g823. ♕d2 ( 23. ♕c7 ♕d6 24. ♖xb7 ♕xc7 25. ♖xc7 a5= )23... ♔h7 ( 23... h5 24. f4 ) 24. ♕b4 ♕f6 25. ♕xb7 ♕e5= b2) ♗xd4⁉N is how I propose fine-tuning that idea. 19. ♕xd4 b5This makes it much harder for White to get a position with any kind of initiative. b21) 20. ♖a1 ♕c2 (Of course it's important not to blunder a piece with 20... ♕d2 21. ♖xc4 ) 21. e5 ♖ab8 Creating the huge threat of a5, with a very double-edged position. b22) 20. e5 ♘d2 b23) If White plays 20. h4 Black is now ready to play 20...♘d2 , exchanging some more pieces. 21. ♖a1 ♘xf3+ 22.gxf3 ♕e2 23. ♔g2 a5 24. ♖b2 ♕c4=/+ b24) 20. ♖xb5 This is White's most natural reaction, but after20... ♖fb8= Black is first with some very serious counterplay. He wants to occupy the b-file and then play Nd2, and perhaps Rb1 if White allows him. In general Black experiences no problems whatsoever and this may be a cleaner solution than the official theory. White is left with an unattractive position with no initiative. 16... b6 17. ♗xe7 ♖fe8 18. d6 ♘c6 19. ♗b5 Forced. Compared to the positions with a black pawn on h7 White has even less scope for play here. 19. ♕c1 isn't an option, since after 19... ♘xe7 20. ♗c4 ♕a5 White doesn't have Ng5 and Black is just better.

19... ♘xe7 20. h3 20. ♗xe8 ♖xe8 21. dxe7 ♕a3 22. ♖e1 (The fact that the pawn is on h6 means that after 22. h3 Black could even consider playing 22...♗e6=/∞ As Ng5 is impossible Black gets to keep both bishops, which could be beneficial. ) 22... ♕xe7 23. h3 ♗e6 24. ♕a4 ( 24. ♘d4♗xh3 25. gxh3 ♕g5+ 26. ♔h1 ♖d8 ) 24... ♕d7 25. ♕a6 ♖d826. ♖e3 ( 26. e5⁉ ) 26... ♕c7 27. ♖d3 ♖c8=/∞ Bunzmann-Van Wely, BL 2001 20... ♗xf3 21. ♕xf3 ♖ec8 21... ♕e6 22. ♗xe8 ♖xe8 23. dxe7 ♕xe7 24. ♖fd1 ♗f8 22. dxe7 ♕e6 23. e8=Q+ ♖xe8 24. ♗xe8 ♖xe8 This position is very similar to those covered previously, with the fact that the pawn is on h6 changing precious little. Black is extremely comfortable and has nothing to fear. 25. ♖fe1 25. ♖fd1 ♗f8! 25... ♗f8! 26. e5 a5 27. ♖e4 ♗c5 28. ♖f4 ♕xe5 29. ♖xf7♕e4 30. ♕xe4 1/2 Sundararajan-Gupta, Visakhapatnam 2006

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Bg5 Bg7 5.e3 5.Nf3

4.Bg5 Bg7 5.e3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗g5 ♗g7 This is the move I'm going to be recommending and for me it's a welcome return to what was perhaps my first serious contribution to opening theory back in the early 90s. I'm recommending it not only for sentimental reasons, but also because while preparing for this series I found a very fresh and interesting new idea in the only line which is currently thought to pose Black any problems. It's very difficult nowadays to find such an idea as early as move 7 in any kind of main line opening, so I was very eager to feature it here. For the past 6-8 years I've almost exclusively replied 4... ♘e4 That move is perfectly fine and objectively Black is comfortable, but the reason I'm not recommending it is because it involves a lot of very complicated transpositions and difficult theory. It can lead to problems unless Black is very precise. For instance, despite 4.Bg5 being the Grünfeld line I've perhaps spent the most hours of my career on I still mix up move orders in practice and keep on getting positions I shouldn't be getting. 5. e3 A rare move that isn't very good at all. It's generally enough to know that you should respond with 5... c5 6. dxc5 ♕a5 , and Black is absolutely equal in every single line here. In my game against Moskalenko he played 7. ♗xf6 7. cxd5 ♘e4−+ 7. ♕a4+ ♕xa4 8. ♘xa4 ♘e4 9. ♗f4 ♗d7∓

7. ♕d2 dxc4 8. ♗xc4 ♕xc5 ( 8... ♘bd7 9. ♖c1 OO 10. ♘f3♘xc5 11. OO ♗e6 12. ♗xe6 ♘xe6= Moskalen ko-Adorjan, Alushta 1994; 8... O-O 9. ♗xf6 exf6 10. ♘e4 ♕xd2+ 11. ♘xd2 ♘d7 12.♘b3 f5 13. O-O-O a5 14. a4 ♘e5=/∞ Jobava-Belov, playchess.com 2005 ) 9. ♗xf6 ♗xf6 a) 10. ♖c1 a1) 10... O-O 11. ♘d5± a2) 10... ♕xc4⁉ 11. ♘d5 ♕g4 12. h3 ♕f5 13. ♘c7+ ( 13. g4♕d7 14. ♘c7+ ♔d8 15. ♘xa8 ♕xd2+ 16. ♔xd2 ♘c6⇆ )13... ♔f8 14. ♘xa8 ♔g7 15. ♘f3 ♗d7 ( 15... ♘c6 16. O-O♕h5 17. ♘c7 ♗xh3 18. ♕d5 ♗f5 19. ♕b5+/= ) 16. O-O♗c6=/∞ a3) ♘c6 11. ♘e4 ♕e5 12. ♘xf6+ ♕xf6= b) 10. ♘d5 10... ♕d6 11. ♘xf6+ ♕xf6 12. ♘e2= Sharif-Rowson Liechtenstein 1996 7. ♖c1 dxc4 8. ♗xc4 Here castling is interesting for Black and 8...Nbd7 is fine, but you can also simply play 8... ♕xc5 and Black has no problems. 7... ♗xf6 8. ♖c1 Instead there's absolutely nothing wrong with the simple

8... dxc4 In those years, and probably still today, I was reluctant to win back material if I could continue playing in an exciting, Romantic style, so I went for 8... ♘a6 9. cxd5 ♘xc5 but after 10. ♗b5+! ( 10. ♗d3♗xc3+! 11. bxc3 ♕xa2 12. ♗b5+ ♗d7 13. ♕d4? ♗xb5 14. ♕xc5♗a6∓ ) 10... ♔f8 11. ♘e2 I was slightly worse in Moskalenko-Svidler, 1993 9. ♗xc4 O-O 10. ♘f3 ♕xc5 10... ♘d7⁉ 11. ♕e2 ♘c6= and Black is fine.

4.Bg5 Bg7 5.Nf3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗g5 ♗g7 5. ♘f3 This is a much more serious option. When I was growing up as a Grünfeld player I was very attached to the idea of playing Bg7 with the intention of never playing Ne4 in any line, but here it's pretty much forced. 5... ♘e4 White has three main options here. In this position I used to play 5... c5⁈ The problem with this move, and the reason it can't be recommended, is that White can play 6. dxc5 (If White does something like 6. ♗xf6 ♗xf6 7. cxd5 then after 7...♕b6 the position is quite unclear. I'm not sure Black gets full equality in every single line, but this was fine for me as I would at least get some kind of fresh non-theoretical battle where Black is doing ok. ) 6... ♕a57. cxd5! and at some point it dawned on me that what we're getting is a transposition to the Bf4 Grünfeld. 7... ♘xd5 ( 7... ♘e4 8. ♗d2♘xc3 9. bxc3 ♗xc3 10. ♖c1+/= ) 8. ♕xd5 ♗xc3+ 9. ♗d2 The bishop returns to d2 from g5 and not f4, but that really isn't much of a difference, and in this position Black is in trouble. 9... ♗e6 (For many years now it's been well-known that 9... ♗xd2+⁈ leads to a horrible position after the forced 10. ♕xd2 ♕xc5 11. ♖c1 ♕f5 12. ♘d4♕d7 13. ♕h6± and Black will find it very hard to finish his development.) 10. ♕xb7 ♗xd2+ 11. ♘xd2 O-O a) 12. ♕xa8 ♖d8 13. b4 ( 13. ♖d1 ♗d5∓ ) 13... ♕xb4 14.♖d1 ♖xd2 15. ♖xd2 ♕b1+ 16. ♖d1 ♕b4+= b) 12. b4 This is quite an exciting position and was considered reasonably ok for Black when I was growing up, but it's since been determined that Black doesn't equalise, or even come close. 12...♕a4 ( 12... ♕a3 13. ♕xa8 ♖d8 14. e4+− ) 13. e3 b1) 13... ♖d8 b11) 14. ♗c4⁉ is not so clear. b111) 14... ♘c6 is wrong: 15. ♗xe6 fxe6 16. b5 ♘a5 17.♕xe7± ♖xd2 ( 17... ♕xb5 18. ♕xe6+ ♔g7 19. ♕e7+♔g8 20. ♘e4 Grigore-Rogozenko, Odorheiu Secuiesc 1992 )18. ♕xe6+ ♔g7 19. ♕e7+ ♔g8 20. ♔xd2 ♖f8 21. f4♘c4+ 22. ♔d3 ♘b2+ 23. ♔e2+− Grigore-Odeev, Romania 1994

b112) 14... ♗xc4! 15. ♘xc4 ♘d7 16. OO ♖ab8 17.♕e4 ♕xb4 18. c6 ♘f6 19. a3 ♘xe4 20. axb4 ♖xb4 , and Black should hold. b12) 14. a3! 14... ♕c2 15. ♖d1 ♗d5 ( 15... ♗b3 16. ♗e2♕c3 17. OO ♗xd1 18. ♖xd1 ♘d7 19. ♘f1 ♘f6 20.♕xe7 ♖xd1 21. ♗xd1 ♖e8 22. ♕xa7± 1-0 Giese, R-Mueller, Porz op 1993 ) 16. ♕xe7 ♘c6 17. ♕f6 ♗b3 18.♗e2 ♗a4 19. b5 ( 19. ♕a1 a5 20. b5 ♕xc5 21. bxc6♗xd1 22. ♗xd1 ♕xc6 23. ♗f3 Van Wely-Hjelmaas, EU-chJ U20 Arnhem 1989; 19. h4!± ) 19... ♘a5 20. ♘b3 ♖xd1+ 21.♗xd1 ♕d3 22. ♕d4± Akhmeto v-Turov, Moscow 2005 b2) ♘d7 14. a3 ♖fd8 ( 14... a5 15. ♗b5 ♕c2 16. ♗xd7♗xd7 17. OO ♕xd2 18. ♖ad1 ♕c3 19. ♖xd7 ♕xa3 20. c6♖fb8 21. ♕xb8+ ♖xb8 22. c7 ♖f8 23. ♖d8 ♕c3 24. c8=Q ♕xc8 25. ♖xc8 ♖xc8 26. bxa5+− Avrukh-Kalod, Menorca 1996;14... ♖ad8 15. ♕a6 ♕c2 16. ♕d3 ♕xd3 17. ♗xd3 ♘xc518. ♗c2± ; 14... ♕c2 15. ♕e4± ) 15. ♕a6 ♕c2 16. ♕d3♕a4 ( 16... ♕xd3 17. ♗xd3 ♘xc5 18. ♗c2+/= ) 17. ♕c3 a5 , and now White has a pleasant choice: b21) 18. b5⁉ ♘xc5 19. ♕xc5 ♖ac8 20. ♕e5 ♕c2 ( 20...♖d5? 21. ♕e4+− ) 21. ♖d1 ♖d5 22. ♕a1 ♖cd8 23. ♗e2♖xd2 24. O-O a4+/= ( 24... f6+/= ) b22) 18. ♖c1 18... axb4 19. axb4 ♘f6 20. ♗c4 ( 20. e4♖ab8 21. ♗d3 ♖xb4 22. O-O ♖bd4 23. ♗c2 ♕c6 24.♘f3 ♖c4 25. ♘e5 ♖xc3 26. ♘xc6 LukacsLiptay, Hungary 2002 ) 20... ♗xc4 21. ♕xc4 ♘d5 Black has some drawing chances, but this definitely can't be recommended: 22. ♖b1 (22. OO Ftacnik 22... ♕xb4 23. ♕xb4 ♘xb4 24. ♘e4♘c6+/= ; 22. c6 ♖ac8 23. OO ♕xb4 24. ♘f3 ♕xc4 25.♖xc4 f6 26. e4 ♘f4 27. g3 ♘e6 28. ♖fc1+/= Simon-Atakisi, ICCF Email 2004 ) 22... ♕a2 ( 22... ♖ab8 23. OO♖xb4 24. ♕d3 ) 23. ♕xa2 ( 23. c6 ♘xe3 Van Wely-Kamsky, 1995 ) 23... ♖xa2 24. ♘f3 ♖b8⁉ ( 24... ♘xb4 25. O-O )25. O-O ♖xb4+/= 6. ♗h4 6. ♗f4 is a well-known line. The following moves (the move order may vary) will probably appear on the board. 6... ♘xc3 7. bxc3 c5 8. e3O-O 9. cxd5 cxd4 10. cxd4 ♕xd5 11. ♗e2 ♘c6 12. O-O ♗f5 This small tabiya of the Bg5 line has been known about for many years and has been tried and tested by very strong players. It's currently considered completely safe for Black. 13. ♕a4 ( 13. ♘h4 , trying to cause Black some discomfort and improve the position of the white pieces, is best met by 13... ♗c8! , and the strong threat of g6-g5 means White will most likely have to bring the knight back to f3. ) 13... ♕a5! I spent a while considering Rac8 here but finally settled on starting with this move. a) 14. ♕xa5 This move is easy to deal with if Black isn't worried by the prospect of trading off all four rooks and playing a somewhat passive but very safe endgame. 14... ♘xa5 15. ♖ac1 ( 15. ♖fc1♖ac8 16. ♘d2 ♖xc1+ 17. ♖xc1 ♖c8 transposes ) 15... ♖ac816. ♘d2 This was seen in two games between myself and Pavel Eljanov.

a1) I played 16... ♘c6 both times. 17. g4 ♗e6 18. a3 ♖fd8 19.♗f3 ♗d5 20. ♔g2! An improvement by Pavel in our later blitz game, after which I had to solve some problems to fully equalise. (In the first game he played 20. ♗xd5 ♖xd5 21. ♘e4? , which looks aesthetically pleasing, but unfortunately runs into 21... ♗xd4when Black wins material. I went on to win the game as well. Eljanov-Svidler, Nalchik FIDE GP 2009 ) 20... ♗xf3+ 21. ♘xf3f6 22. ♗g3 e6 23. ♖c4 ♗f8 24. a4 ♘b4 25. ♖fc1 EljanovSvidler, Moscow VI WCh Blitz 2010. a2) 16... ♖xc1! is the best approach. 17. ♖xc1 ♖c8 18.♖xc8+ ♗xc8 19. ♗c7 ♘c6 and the endgame should be easy for Black to hold. He has no weaknesses and too much material has been exchanged for White to hope for much of an advantage. b) 14. ♕b3 14... ♕b4 15. ♖fc1 b1) It's important to know that after 15... ♖ac8 White has 16.♗c4! ♕xb3 More or less forced. (After something like 16... a6White has the very important tactical trick 17. ♗xf7+ ♖xf7 18.♖xc6 , which wins a pawn. ) 17. axb3 a6 18. ♗d5 This position is probably still defendable, and I held it against V.Milov in Biel 2000, but Black really has no reason to go anywhere near this. b2) ♕xb3⁉ 16. axb3 a6! The latest finesse, which was played by Morozevich against Radjabov and solves all Black's problems. ( 16...♖fc8 17. ♗c4 a6 18. ♗d5+/= ) 17. ♖c5 ( 17. ♗c4 is now met very strongly by 17... ♘b4 and the bishop doesn't get to d5.; The point of this move order is that if 17. ♗xa6 Black has 17...♘xd4! 18. ♘xd4 ♗xd4 19. exd4 ♖xa6= and a draw will very soon be agreed due to the limited material remaining. ) 17...♖ac8 18. ♖ac1 ♘b4 19. ♘e5 ♖xc5 20. ♖xc5 ♖c8 21.♗f3 b6 22. ♖xc8+ ♗xc8 And Black was once again perfectly comfortable in RadjabovMorozevich, Moscow 7th Tal Mem Drawing of Lots 2' 2012 6. cxd5 ♘xg5 7. ♘xg5 e6 8. ♕d2 ( 8. ♕a4+ c6 9. dxc6 ♘xc6 10.♘f3 ♗d7=/∞ ) 8... h6! This is very comfortable for Black. (Some heavy theory starts with 8... exd5 9. ♕e3+ ♔f8 10. ♕f4 ♗f6 11. h4 Black may be fine even here, but I never felt there was any need for this.) a) 9. ♘h3 exd5 a1) 10. ♕e3+ ♔f8 11. ♘f4 c5 12. dxc5 d4 13. ♕d2 a11) Black used to play 13... ♘c6 after which the position is very unclear, although even here Black has good play for the pawn. 14. ♘b5! ( 14. ♘e4 g5 15. ♘d3 ♕d5 16. ♘d6♗e6 17. ♖g1 ♘e5↑ Kuzubov-Eljanov, Moscow 2005 ) 14...b6 15. g3 ♗a6 16. a4 bxc5 17. ♗g2 ♖c8 18. O-O±MarkosTukmakov, Czechia 2009 a12) ♘a6⁉ 14. e3 is an improvement, and after both Black's responses he has a very comfortable position and experiences no difficulties. 14... ♗f5 ( 14... ♘xc5 15. OOO ♗f5 16.exd4 ♘e4 17. ♘xe4 ♗xe4 18. ♗d3 ♕xd4=/+ Volkov-Dvoirys, Omsk 1996) 15. exd4 ♕xd4 16. ♗xa6 bxa6 17.OOO ♕xc5 18. ♕d5 ♕xd5 19. ♘fxd5 ♖c8 20. ♖he1h5=/+ Damaso-Avrukh, Albufeira Algarve 1999

a2) 10. ♘f4 10... O-O It's important to note that Black doesn't waste any time defending the d5-pawn, and if 11. g3 Black has the very strong ( 11. e3 c5 12. dxc5 d4 A recurring theme in this structure - Black should always be happy to sacrifice this pawn, most often temporarily, in order to open up the position for his g7-bishop. Black is fine here. 13. ♖d1 ♘c6 14. ♘b5 dxe3 15.fxe3 ♕g5=/∞ ) 11... ♖e8! and after 12. ♗g2 ♘c6 it transpires that the d5-pawn can't be protected and will be exchanged for the d4-pawn. After that Black is better due to having the bishop pair in a completely open structure. b) 9. ♘f3 is what White should objectively play, with 9... exd5 10.e3 O-O and something like 11. ♗e2 following. This is a Carlsbad structure in which Black should be perfectly fine. He'll follow up with c6 and in some cases a5 to hold back White's advance on the queenside. The b8knight generally develops towards d7 and f6. It often makes sense for Black to play Re8 and then swing the bishop over from g7 to d6, a square from which it can cover the queenside and also support an attack on the kingside if White gets too carried away pushing his queenside minority. 6... ♘xc3 7. bxc3 dxc4 This is where Black gains a lot by having forced White to include Nf3 Bg7. Rather than going for the pawn sacrifice most people play 8. ♕a4+ If White plays 8. e3 Black can respond 8... b5! , and after 9. a4 c6White can't win the rook by axb5, cxb5 and Qf3. That means this is a genuine pawn sacrifice, and hence not to everybody's taste. Play might continue 10. ♗e2 O-O 11. ♘d2 a6 a) 12. O-O ♕c7 This natural move was played by Rustam Kasimdzhanov in a game against Gurevich. (Something to avoid here is12... ♗f5? 13. f4 ♗e6 and it's quite clear that wasting a whole tempo can't be best play for Black, although even this isn't completely horrible. ) 13. f4 ♘d7 14. f5 ( 14. ♗xe7 ♖e8 is clearly bad for White ) 14... ♘f6 and Gurevich very soon ran into some serious problems. In general, if Black manages to get his pieces out of the corner on the queenside he'll be a reasonably healthy pawn up and playing for a win. 15. e4 gxf5 16. ♗xf6 exf6 17. ♕e1 ♖e8∓Gurevich-Kasimdzhanov, Konya TUR-tch 2011 b) 12. ♗f3 The main theoretical move in this position.

b1) It's possible to play 12... ♗b7⁉ , planning to meet 13. ♘e4with 13... ♖e8 and after 14. ♘c5 Black will play ( 14. OO or) or 14... ♕c8 or and follow it with Nd7 next move. White's pieces will gradually be driven back. Black is a pawn up, which isn't something to be sneezed at. b2) 12... ♖a7 The normal response. 13. O-O This is a well-known position with a number of possible moves. 13... ♗f5! The move I like best and one which carries the stamp of approval of having been playing by Garry Kasparov himself (in a simul in 1997).( 13... ♕b614. ♕b1 ♗e6 15. ♕a2 ♖c7 16. ♖fb1 ♖e8 17. ♕a3=/∞ Wang YueGao Rui, China 2010; 13... ♖d7 14. ♕b1 ♕c7 15. ♘e4♖dd8 16. ♕b4 f5 17. ♘c5 ♖de8 18. ♖fd1 ♗f6 19. ♗xf6♖xf6 , and now, in Beliavsky-Kamsky, Belgrade 1991, White should have played 20. e4=/∞ ; Boris Gulko many years ago played13... h6 , in some lines preparing an advance with g5 and f5. 14.♕c1 ♗f5 15. ♕b2 ♗d3 16. ♖fd1 ♕b6 17. ♗e4 ♗xe4 18.♘xe4 g5 19. ♗g3 ♘d7=/+ Lengyel-Gulko, Sombor 1974)

b21) 14. ♖e1 ♗d3 15. ♘b3 ♗f5 16. ♘d2 White has nothing better, and Black is already equal. ( 16. ♘c5 runs into 16...♘d7 ) 16... ♗d3 17. g4 ♖c7 18. ♘b3 cxb3 19. ♕xd3 c5SorinKasparov, Buenos Aires 1997 b22) 14. e4 is an attempt by White to prevent Bd3, but after14... ♗c8 Black will meet pretty much any move by White with c5, when the white centre will come under direct attack. That was demonstrated by Jean-Philippe Karr in a game against Aleksander Delchev. He played very nicely, up to a point, and had a huge advantage with Black. 15. ♕b1 c5 16. axb5 ♖b7 17.♘xc4 cxd4 18. e5 ♖xb5 19. ♕d3 ♘d7=/+ Delchev-Karr, France 2010 8... ♕d7 8... ♘d7 9. e3 ( 9. ♕xc4 c5 10. e3 ♕a5⇆ ) 9... OO 10. ♕a3 c511. ♗xc4 ♕c7 12. ♗b5 ( 12. ♖c1 e5⁉ ; 12. ♗xe7 ♖e8 13. ♗xc5♘xc5 14. dxc5 ♗g4 15. ♘d4 ♖ac8 16. ♕b3 ♖e7 17. h3 ♗d7 18.OO ♕xc5 19. ♗d5 ♕xc3 20. ♕xb7 ♗xd4 21. exd4 ♕xd4=Korobkov-Vorobiov, Serpukhov 2008) 12... ♘f6 13. ♗g3 ♕b6 14.♗d3 ♗f5 15. ♗xf5 gxf5 16. OO ♘e4 17. ♖ab1 ♕c6 18. ♗e5b6 19. c4 ♗xe5 20. ♘xe5 ♕e6 21. ♖bc1 ♖fd8 22. ♕b2 f6 23.♘d3 cxd4 24. exd4 ♖ac8 25. ♘b4+/= Ki.GeorgievMelkumyan, Benasque 2009 9. ♕xc4 White wins back the pawn, but after 9... b6 the queen comes under direct attack in many lines, and for a number of years now this has been known to be very comfortable for Black. 10. e3 The main move. 10. ♗g3 c5 11. ♘e5 ♗xe5 12. ♗xe5 ♗a6 13. ♕b3 O-O , and although the idea of giving up the darksquared bishop is quite traumatic for any Grünfeld player you gain so much time in this particular case that you shouldn't hesitate. 10... ♗a6 11. ♕b3 ♗xf1 12. ♔xf1 The other capture is possible, but in the vast majority of cases that will just transpose to the same line. 12... O-O 13. ♔e2 c5⁉ 14. ♖hd1 is the somewhat more critical attempt here. 14. dxc5 is very well met by 14... ♘a6 15. ♖hd1 ♕c7 ( 15... ♕b7should be avoided, as after 16. c6 ♕xc6 17. ♗xe7 ♖fe8 18. ♗a3♕xc3 19. ♕xc3 ♗xc3 20. ♖ac1 ♗b4 21. ♗b2 ♗f8 22. ♘d4 ♘c5 23. g4+/= Magnus Carlsen got a small but pleasant endgame advantage against Pavel Eljanov and went on to win. ) 16. cxb6 axb6It's well-known that Black has fantastic compensation for the pawn here.17. ♗g3 (If White overreaches and tries too hard to preserve his material advantage he might run into quite serious problems very early on, as shown in a game between Joel Lautier and Vassily Ivanchuk: 17.a4 ♘c5 18. ♕b4 ♖a5 19. ♗g3 e5 20. ♘d2 ♖fa8 21. ♘c4♘xa4∓ Lautier-Ivanchuk, Terrassa 1991) 17... ♕c6 18. ♔f1 ♘c5 19.♕d5 ♕xd5 20. ♖xd5 ♗xc3 21. ♖c1 ♗f6= Bosboom LanchavaGustafsson, Amsterdam Lost Boys op 2001 14... ♘c6 15. dxc5 ♕f5⁉ 15... ♕c8 16. ♖ab1 bxc5 17. ♕c4 It's somewhat problematic for Black to get full equality, although his position isn't that bad. 17... ♘a518. ♕a4+/= ( 18. ♕b5 ♘c6 19. ♕c4 ♖b8 20. ♖xb8 1/2 FressinetVachier Lagrave, Pau 2008 )

16. ♖d5 16. ♕d5 ♕c2+ 17. ♔f1 ♖fd8 18. ♕b3 Forced. 18... ♕xb3 19.axb3 bxc5 and the endgame is completely equal. 16. cxb6 axb6 is a very typical position for this line where Black gets a lot of counterplay against the white queenside and the somewhat misplaced king on e2. Black is fine here. 16... ♕e4 A wonderful square for the queen. 17. cxb6 axb6 Once again Black has terrific compensation for the pawn because White's position is incredibly discoordinated and Black has a lot of play against the targets on the queenside.

imm1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Bg5 Bg7 5.Bxf6 Bxf6 6.cxd5 c6 7.e3/dxc6/Rc1 7.e4 0-0 7...cxd5 7...Qb6 7...Bg7

4.Bg5 Bg7 5.Bxf6 Bxf6 6.cxd5 c6 7.Rc1 and other moves 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗g5 ♗g7 5. ♗xf6 ♗xf6 6.cxd5 This is White's only way of fighting for an advantage, and it was a big challenge to find something to recommend. Official theory, as it currently stands, doesn't promise Black full equality here. That discourages a lot of players from entering this line, and not without reason, but I've found a couple of new ideas I want to share with you. 6. ♘xd5 ♗g7 7. e3 c5 is clearly good for Black: 8. ♕a4+ ♘c6 9.dxc5 ( 9. ♖d1 cxd4 10. exd4 OO 11. ♘f3 ♗g4∓ )

a) 9... e6 10. ♘b6 ♕f6 ( 10... axb6 11. ♕xa8 ♗xb2 12. ♖d1♗c3+ 13. ♔e2+− ; 10... ♖b8 11. ♘xc8 ♖xc8 12. ♖d1 ♕e7 )11. ♖b1+/= b) 9... ♗xb2 10. ♖d1 ♗d7 11. ♖b1 ♗g7 12. ♖xb7 ♖b8 13.♖xb8 ♕xb8 14. ♕b3 OO 15. ♘f3 e6 16. ♘f4 ♕c7=/∞

6... c6 6... c5 is a move people have started to play due to the problems Black is experiencing after 6...c6. It's a very interesting line, sacrificing a second pawn for development and to prevent White from playing e4 as he can after 6... c6. I encourage you to have a look at it, although I'm not sure it equalises for Black and it's not my recommendation. 7. ♖c1 For a while White had reasonable success playing this move. 7. e3 doesn't trouble Black. 7... cxd5 8. ♗b5+ ♔f8 ( 8... ♗d7?blunders away the central pawn after 9. ♗xd7+ ♕xd7 10. ♕f3!± ) 9.♘ge2 ♔g7 This is a very nice Slav-like position for Black. He wants to play e6 and in many cases the bishop will be better off on d6 (via e7). The knight on b8 can choose between d7-f6 and c6-a5. Black is very comfortable here. 10. ♕b3 ( 10. ♘f4 e6 11. OO a6 12. ♗e2 ♕d613. ♘d3 ♘d7 14. ♕b3 b6 15. f4 ♗b7 16. ♘e5+/= Goldin-Svidler, SPb mem Petroff 1994 ) 10... e6 11. OO a6 12. ♗d3 ♘c6 13.♖ac1 ( 13. ♘a4 e5 ) 13... ♘a5 ( 13... ♗e7 14. ♘a4 ♘a5 Nakamura-McShane, playchess.com 2004 ) 14. ♕d1 ♗d7= 7. dxc6 ♗xd4 8. cxb7 ♗xb7 is covered in my Early Games chapter, and is quite frankly something you're quite unlikely ever to meet at the board. White has nothing here. a) 9. ♕a4+ ♘c6 10. e3 ♕b6 11. ♗b5 ♗xc3+ 12. bxc3 a6 13.♗d3 ♕b2 14. ♖b1 ♕xc3+ 15. ♔e2 ♖d8= Lyrberg-Kreiman, Medellin WChJ B20 1996

b) 9. ♘f3 ♗xc3+ 10. bxc3 ♕a5 11. ♕b3 OO 12. e3 ♗xf3 13.gxf3 ♘d7 14. ♕b4 ♕f5 15. ♗e2 ♖ab8 16. ♕e4 ♕xe4 17.fxe4 ♖b2=/+ Orlinkov-Najer, Moscow-ch 1996 c) 9. ♖c1 OO 10. e3 ♗xc3+ 11. ♖xc3 ♕a5 12. ♕b3 ♗d5 13.♕a3 ♕xa3 14. bxa3 ♘d7 15. ♘f3 ♖ab8 16. ♗d3 ♖b2=Ferenc-Sakalauskas, MK Cafe Cup Open 1999 d) 9. e3 ♗xc3+ 10. bxc3 ♕a5 d1) 11. ♕d4 O-O 12. ♕b4 ♕c7 13. ♘f3 ( 13. ♖c1 ♘a6 14.♕a3 ♘c5 ) 13... ♘a6=/∞ d2) 11. ♕b3 ♘d7 ( 11... O-O 12. ♕xb7 ♕xc3+ 13. ♔e2♕c2+ 14. ♔f3 ♕f5+= ) 12. ♘f3 d3) 11. ♕c2 11... O-O 12. ♘e2 ( 12. ♖b1 ♖c8! ) 12...♘d7=/+ Popov-Svidler, 1993 e) 9. ♕b3 e1) Even 9... ♗xc3+⁉ is possible, although I can't see why you would want to give the bishop up so easily: 10. bxc3 ( 10. ♕xc3 OO 11. ♘f3 ♘d7 12. ♖d1 ♖c8 13. ♕b4 ♕b6 14. ♕xb6♘xb6=/∞ ) 10... ♕c7 11. ♘f3 ♘d7 12. e3 ♗xf3 13. gxf3 OO 14. ♗e2 ♖ab8 15. ♕a3 ♖b6 16. OO g5 17. ♕a5 ( 17.♕xe7 ♖h6 18. h4 ♖e6! 19. ♕a3 ♖xe3 20. fxe3 ♕g3+= )17... ♘c5=/∞ Korniushin-Yandemirov, RUS-Cup 1997 e2) 9... ♕b6⁉ 10. e3 ♗f6 ( 10... ♕xb3 11. axb3 ♗f6 12.♘f3 OO 13. ♗c4+/= ) 11. ♗b5+ ♔f8 12. ♘f3 ♘a6 13. OO♘c5 14. ♕a3 ♔g7 15. ♗e2 ( 15. ♖ac1 ♖ab8=/∞ ) 15...♖ab8 16. ♘d4 ♖hd8 17. ♖fd1 ♗a8 18. ♖ab1 ♘e4 e3) 9... ♕c7 10. e3 ♗g7 e31) I won one of the most important games of my junior career in this line: 11. ♗b5+ ♘c6 12. ♘f3 ( 12. ♘ge2 OO 13.OO ♘e5 14. ♖ac1 a6 15. ♗a4 ♖ab8 ) 12... OO 13.OO ♘e5 14. ♘xe5 ♕xe5 15. e4 ♖ab8 16. ♕a4 ♕c5 17.♗e2 ♗a8 18. ♖ab1 ♖fd8→ , and Black was already better in Papaioannou-Svidler, Duisburg WChJ B16 1992 e32) 11. ♘f3 11... O-O 12. ♗e2 ♘d7 13. O-O ♖ab8 14.♕a3 ♘e5 , and once again Black has wonderful play for the pawn: 15. ♖ac1 ♗xf3 16. gxf3 ♕b7 17. f4 ♘f3+ 18.♗xf3 ♕xf3 19. ♕xe7 ♖xb2 20. ♘e4 ♖xa2 21. ♘f6+♔h8 22. ♘d7 ♕g4+ 1/2 Crouch-Svidler, London Lloyds 1991 7... O-O 7... ♕b6 doesn't quite fit the bill: 8. e3 ♕xb2 9. ♗c4 OO ( 9...♕b4 10. ♕b3 ♕xb3 11. ♗xb3 OO 12. ♘f3 ♖d8 13. OO ♔f814. ♘e4 ♗f5 15. ♘eg5 ♗xg5 16. dxc6 ♘xc6 17. ♘xg5 f6 18.♘e6+ ♗xe6 19. ♗xe6 ♖d6+/= Mchedlishvili-Najer, Saint Vincent 2005 ) 10. ♘ge2 ♕b4 11. ♕b3 ♕a5 12. OO ♗g7 13. ♘f4 ♘d714. ♘e4 ♗h6 15. dxc6 bxc6 16. ♘d3 ♖b8 17. ♕c3 ♕h5 18.♗b3 Iljin-Nepomniachtchi, Cheboksary RUS-chj B20 2006 7... cxd5 is of course met by 8. ♘xd5 and the bishop on c8 will be hanging. 8. dxc6 Later Black realised that he can play

8. e4 ♗g7! transposes to 7.e4 8... ♕xd4⁉ , completely solving the problems in this line. 8... ♗xd4 9. c7 ♗xc3+ 10. ♖xc3 ♕xd1+ 11. ♔xd1 ♘c6 should be fine for Black, as the pawn on c7 obviously isn't long for this world, but I think White derives much more pleasure from playing this position than Black. The resulting endgames are slightly better for White, who has some pressure and is taking absolutely no risks. White has even won some games at a reasonably high level. 12. e4 ( 12. g3 ♗e6 13. ♗g2♖ac8 14. ♗xc6 bxc6 15. ♘f3 ♖xc7 16. ♔c1 ♖b8 17. b3 a5 18.♘g5 ♗f5 19. ♖c4 ♖b4= Dreev-Sakaev, Tripoli 2004 ) 12... ♗d7 (12... ♗g4+ 13. ♗e2 ♗xe2+ 14. ♔xe2 ♖fc8 15. ♘f3 ♖xc7 16.♖d1 ♖cc8 17. ♖d7 ♖ab8 18. ♖b3 ♘a5 19. ♖b5 b6 20. ♖xa7±AzmaiparashviliSutovsky) 13. ♘f3 ♖fc8 14. ♗b5 ♖xc7 15. ♔d2 f6( 15... ♖d8 16. ♔e3 ♖cc8 17. ♖hc1 ♔f8 18. ♗xc6 ♖xc6 19.♖xc6 ♗xc6 20. ♘d4 ♔e8 21. ♘xc6 ♖c8 22. g4 bxc6 23. ♖c5+/=Kacheishvili-Atakisi, Turin ol 2006; 15... a6 16. ♗a4 ♖cc8 17. ♖hc1b5 18. ♗b3 e6 19. e5 ♔f8 20. ♗c2 ♘a5 21. ♗d3 ♖xc3 22.♖xc3 ♘c6 23. ♗e4 Jobava-Miton, Skanderborg 2005; 15... ♖cc8 16.♖hc1 ♖d8 17. ♔e3+/= ) 16. ♖hc1 e5 17. ♘e1 ♖d8 18. ♘d3♗e8 19. ♔e3 ♖e7 20. f3 ♖d4 21. a3 ♔g7 22. ♗xc6 ♗xc6 23.♘b4 ♗b5 24. ♖c7 ♖xc7 25. ♖xc7+ ♖d7 26. ♖c5 a6=Kachei shvili-Deng Kong Liang, Las Vegas op 2006 9. ♕xd4 ♗xd4 10. cxb7 After 10. c7 Black may already be better. 10... ♘c6 11. ♘f3 ♗b6 12.e3 (White can't hold on to the pawn with 12. ♘d5 as that runs into12... ♗a5+ 13. ♔d1 e6 14. b4 ♘xb4 15. ♘xb4 ♗xb4 16. e3♗d7=/+ ) 12... ♗xc7 Black has won back the pawn and has bishops against knights. 13. ♗b5 ♘e5 14. ♘d4 ♖d8 15. O-O ♗b6=/+ Wang YueBerkes, Taiyuan GM Schev 2006 10. ♘f3 ♗xc3+ 11. ♖xc3 ♘xc6= Ivanchuk-Dominguez Perez, Havana 2010 10... ♗xb7 11. ♘f3 11. e3 ♗f6 12. ♘f3 transposes ( 12. f3 gives Black fantastic compensation. He has play along the long diagonals and the c- and d-files for his rooks. 12... ♘d7 13. ♗b5 ♘e5 14. ♘h3 ♖fd8 15. ♘f4♖ac8 16. O-O a6 17. ♗e2 ♖d2=/∞ ) 11... ♗f6 12. e3 12. g3 ♘d7 13. ♗g2 ♖ab8 14. ♘a4 ♖fc8⁉ ( 14... ♘b6 15.♘xb6 ♗xb2 16. ♖c2 ♗xf3 17. ♗xf3 ♖xb6= Jumabayev-Ganguly, Guangzhou 16th Asian Games tt 2010 ) 15. ♖xc8+ ♖xc8 16. ♔d2♘b6=/+ 12... ♘d7 12... ♖d8 13. ♗e2 ♘a6 14. OO ♘b4 15. a3 ♘d3 16. ♗xd3♖xd3 17. ♖fd1 ♖xd1+ 18. ♖xd1 ♗xc3 19. bxc3 ♖c8 20. ♖c1♗xf3 21. gxf3 ♖c4=/∞ Vuckovic-Mekhitarian, Moscow 2011 13. ♗b5 The only serious move.

13. ♗e2 ♖ab8 is already an immediate draw as White can't hang on to the material. 14. ♘a4 ( 14. OO ♗xf3 15. ♗xf3 ♖xb2 16. ♘d5♖xa2 17. ♘xe7+ ♔h8= ; 14. ♖c2 ♗xf3 15. ♗xf3 ♖xb2 16.♖xb2 ♗xc3+ 17. ♖d2 ♖b8=/+ ) 14... ♘b6 15. ♘xb6 axb6 16. b3♖a8 17. a4 ♗d5 18. ♗c4 ♖fc8 19. ♔e2 ♗xc4+ 20. ♖xc4 1/2 Brunello-Erdos, Sibenik Mitropa Cup 2012 13... ♖fd8! 13... ♘c5 is strongly met by 14. ♗c4 ♖ac8 15. b3 ♘d7 16. ♘a4and White actually manages to hang on to his extra material. Although Black will always have some drawing chances in a position like this it can't be recommended. 16... ♗c6 17. ♗a6 ♗b7 18. ♗c4 ♗c6 19. ♔e2♗xa4 20. bxa4 ♘c5 21. ♗b3 ♘e4 22. ♘d4 ♖fd8 23. ♖xc8♖xc8 24. f3+/= Moiseenko-Vachier Lagrave, Plovdiv 2010 13... ♘e5 14. ♘xe5 ♗xe5 15. O-O ♖fd8 16. f4+/= 14. O-O 14. ♔e2 ♖ab8 15. ♗xd7 ♖xd7 16. ♖hd1 ♖xd1 17. ♘xd1 ♗xb218. ♘xb2 ♗xf3+ 19. ♔xf3 ♖xb2= 14... ♗xf3 14... ♖ab8 15. ♖fd1 15. gxf3 ♖ab8 This position has by now been shown to be an immediate draw. 15... ♘e5 16. f4+/= 16. a4 The only critical move. 16. ♗xd7 ♖xd7 17. ♖c2 ♖db7 Winning back the pawn. 18. ♖b1♗xc3= Akobian-Le Quang Liem, Gibraltar Masters 2012 16. b3 ♗xc3 17. ♗xd7 ♗d2 18. ♖c7 An only move. 18... ♗a5 19.♖c5 Once again an only move, or else White loses material. 19... ♗b420. ♖b5 ( 20. ♖d5 ♗d6 isn't a particularly good idea. ) 20... ♖xb521. ♗xb5 ♖d2 22. a4= 16... ♘e5 is advisable. 16... ♘c5 17. ♘e4 ( 17. ♖fd1 a6= ) 17... ♘xe4 18. fxe4 a6 19.♗xa6 ♖xb2 20. ♗c4 17. ♔g2 ♘d3 18. ♗xd3 ♖xd3 19. ♖c2 ♖b4= Here the completely compromised structure on the queenside ensures that White has no winning chances whatsoever.

7.e4 0-0 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗g5 ♗g7 5. ♗xf6 ♗xf6 6.cxd5 c6 7. e4⁉ This poses Black serious problems. 7... O-O Currently the main line in this position. 8. e5

8. ♘ge2 simply looks wrong - it's not a very good square for the knight.8... cxd5 9. exd5 ( 9. ♘xd5 ♗g7 10. ♕d2 ♘c6 11. ♖d1 ♗g4 12.f3 ♗e6 13. ♘dc3 ♗c4 14. b3 ♗a6 15. e5 ♖c8 16. a3 ♕b6 17.b4 ♖fd8=/∞ ) 9... ♘d7=/∞ 8. ♘f3 isn't a problem for Black: a) 8... ♗g4⁉ a1) 9. ♕b3 ♗xf3 10. gxf3 ( 10. ♕xb7 ♗xd4 ) 10... ♗xd4 11.♕xb7 ♘d7 12. dxc6 ♘e5 13. O-O-O ♖b8=/+ a2) 9. e5 9... ♗g7 10. ♕b3 b) 8... cxd5 b1) 9. ♘xd5 ♗g7 b11) 10. ♖c1⁈ ♘c6 11. ♗b5 ♕a5+ 12. ♘c3 ( 12. b4♕xb5 13. a4 ♕a6 14. ♘c7 ♕b6 15. ♘xa8 ♕xb4+ 16.♕d2 ♗g4 17. ♘c7 ♕xa4=/+ ) 12... ♘xd4 13. ♘xd4♖d8∓ b12) 10. ♘c3 10... ♗g4 11. d5 ♕a5 ( 11... ♕b6 12. ♕b3♕xb3 13. axb3+/= ) 12. ♖c1 ♖c8 13. ♕d2 ♘d7=/∞ b2) 9. e5 9... ♗g7 b21) 10. ♗d3 ♘c6 11. h3 ( 11. h4 ♗g4 ) 11... ♕b6 12.♘xd5 ♕a5+ 13. ♘c3 ♖d8∓ b22) 10. h3 10... ♘c6 11. ♗e2

b221) 11... ♕b6 12. ♕d2 ♗e6 13. O-O ♖ac8∞ b222) 11... f6 12. exf6 exf6 ( 12... ♗xf6 13. OO ♗f514. ♖c1 e6 15. ♕d2 ♗e4 16. ♕e3 ♗xf3 17. ♕xe6+♔g7 18. ♗xf3 ♘xd4 19. ♕xd5 ♘xf3+ 20. ♕xf3 ♗g521. ♕xb7+ ♖f7∞ ) 13. OO ♗e6 14. ♖e1 ♗f7 15.♕b3 ♕d7= b223) ♗f5 12. O-O ♗e4 13. ♕d2 ♗xf3 14. ♗xf3 e6∞ 8... ♗g7 9. ♗c4 In order to win material back Black has to play 9. dxc6 ♘xc6 10. ♘ge2 ♗f5=/∞ 9... b5 9... cxd5 10. ♗xd5 ♘c6 11. ♘ge2± 9... ♕b6 10. ♕d2 10. ♗b3 10. ♗d3

a) 10... a6⁉ a1) 11. ♘f3 cxd5 12. O-O ♘c6 13. h3 f6 a2) 11. ♕b3 c5 12. dxc5 ♘d7 13. c6 ( 13. ♗e2 ♗xe5 14.♕b4 a5 15. ♕e4 ♕c7 16. ♗xb5 ♘xc5 17. ♕e3 ♖b8∓ )13... ♘xe5 14. ♗e4 e6 15. ♖d1 exd5 16. ♗xd5 ♕c7 17.♘ge2 ♘xc6=/+ a3) 11. h4 11... cxd5 ( 11... c5 12. dxc5 ♘d7 13. f4 ♘xc514. ♗c2 ) 12. h5 ♘c6 13. ♘ge2 ♗g4 14. f3 ( 14. ♕d2 ♖c8) 14... ♗e6 15. ♕d2 ♕b6 16. ♗c2→ b) b4 11. ♘ce2 ♕xd5 12. ♕c2 ♕xg2 13. ♘g3 ♕d5 14. ♘1e2♗a6 15. ♗e4 ♕c4 16. ♕d2 ♖d8 Vovk-Astaneh Lopez, London e2e4 Diamond Jubilee Premier 2012 10... b4 10... a5 11. a3! After this move (or 11. a4 play often becomes very sharp, but in general White is currently very happy to go for this. Practice shows that he has somewhat better chances and I think the character of the ensuing play also very much suits White. 11... b4 12.♘ce2 cxd5 13. h4 ♗a6 14. h5 ♘c6 15. hxg6 fxg6 16. ♕c1 ♕d717. ♕e3 e6 18. ♕h3 h6 19. ♕e3 ♖ac8 20. ♘f3 ♕f7 21. ♖h3♘e7 22. g4 g5∓ Pantsulaia-Gopal, Khanty-Mansiysk 39th olm 2010 )11... ♗a6 12. ♘f3 b4 13. axb4 axb4 14. ♘e4 cxd5 15. ♘c5 e616. ♗c2 ♕b6 17. ♗d3 ♗b7 18. ♖xa8 ♗xa8 19. h4 ♗c6 20. h5♗b5 21. hxg6 hxg6 22. ♘g5 GuptaDas, Kochi IND-tch 2011 11. ♘a4⁉ 11. ♘ce2 cxd5 12. h4 is very ambitious, but quite unclear: a) 12... h6 13. ♘f4 e6 14. ♘f3 ( 14. ♖c1 ♕b6 15. ♘f3 ♘c616. ♗a4 ♘e7 17. h5 ♗a6 18. hxg6 fxg6 9. ♕d2 ♖ad8 20.♖c5 Kazhgaleyev-Laylo, Manila 6th Prospero A Pichay Cup 2010 )14... ♘d7 15. ♗c2 ♗a6 16. ♗d3 ♗xd3 17. ♕xd3 ♕e7 18.OO ♖fc8 19. ♖ac1 ♕d8 20. ♕a6 ♔h7 21. g3 ♘b6 22. b3+/=Rodshtein-Khusnutdinov, Moscow 9th Aeroflot op A 2010 b) 12... ♘c6 13. h5 e6 14. ♘f3 ♗a6 15. ♘f4 ♖c8 16. hxg6hxg6 17. ♗c2 b3 18. axb3 ♕b6 19. ♘e2 ♗xe2 20. ♔xe2♘b4 21. ♗d3 ♖c7=/∞ Bukavshin-Matlakov, Moscow Open E 2011 c) 12... h5⁉ 13. ♘f4 e6 14. ♗c2 ♘c6 15. ♖h3 ♕a5⇆ Pashikian-Edouard, Khanty-Mansiysk 39th olm 2010

11... cxd5 12. ♘e2 ♘c6 13. ♘c5 e6 14. ♖c1 ♘a5 15. OO ♕b616. ♘f4 ♖d8 17. ♗a4 ♗a6 18. ♖e1 ♗b5 19. ♗xb5 ♕xb5 20.♕g4 ♖ac8 21. h4→ Zhou Weiqi-Gao Rui, Shenzhan CHN-tch 2010

7.e4 cxd5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗g5 ♗g7 5. ♗xf6 ♗xf6 6.cxd5 c6 7. e4⁉ cxd5 , as played in a game between Vovk and Tregubov, was my first attempt at reviving this line, but it doesn't quite work: 8. ♗b5+ is much stronger than 8.Nxd5.

After 8. ♘xd5 ♗g7 9. ♗b5+ I believe Black has good compensation. a) It's important not to play 9... ♗d7 because 10. ♖c1! is very strong. a1) 10... ♗c6 11. ♘e2 O-O 12. ♗c4± a2) 10... ♗xb5 11. ♘c7+ ♔f8 12. ♘xb5 a6 ( 12... ♘c6 13.d5± ) 13. ♘c7 ♖a7 14. ♘f3 e6⁈ ( 14... ♘c6 15. d5 ♕xc716. dxc6 bxc6 17. ♕d2± ) 15. ♕b3 ♘c6 16. d5! ♕xc7 17.dxc6 bxc6 18. O-O+/= Gretarsson-Olafsson, Gentofte 1999 a3) 10... ♘c6 11. ♘e2± a4) 10... O-O 11. ♗xd7 ♘xd7 ( 11... ♕xd7 12. ♘c7+− ) 12.♘f3± b) 9... ♘c6 10. ♘e2 O-O b1) 11. ♗xc6 bxc6 12. ♘dc3 ♖b8 b11) 13. ♖b1 ♗a6 14. OO e5 ( 14... c5 15. ♕a4 ♗d3 16.♖bd1 ♗xe2 17. ♘xe2 ♖xb2 18. dxc5 ♕c7 19. ♘c1+/=Gretarsson-Littlewood, Birmingham 2000 ) 15. d5 cxd5 16.exd5 e4!=/∞ b12) 13. ♕d2 13... c5 ( 13... ♕a5 14. OO ♖d8 15. ♖fd1♗g4 16. f3 ♗e6 17. b3 ♖b4 18. ♘a4 ♗c4 Wunnink-Lahaye, Amsterdam 2004 ) 14. d5 ♗a6=/∞ b2) 11. O-O ♖b8! 12. ♖b1 ♗g4 13. f3 ♗e6 14. ♗xc6 bxc615. ♘dc3 ♕b6 b3) 11. ♖c1! b31) 11... ♕a5+ can be met by the strong 12. ♕d2! (A typical tactic for this position is 12. ♘dc3 ♘xd4 13. ♘xd4 ♖d8 and the knight on d4 is lost, with an advantage for Black. ) 12...♕xb5 13. ♘c7 and I couldn't make this work for Black. The knight on a8 escapes and White ends up in positions where he's an exchange up and can at least hope for some advantage. 13...♕h5 14. ♘xa8 ♗g4 b311) 15. ♘c7 ♗xe2 16. ♕xe2 ♕a5+ 17. ♔f1 ( 17.♕d2 ♕xc7 18. d5 ♖d8 ) 17... ♕xc7 18. d5 ♕f4=/+ b312) 15. ♖c5 15... ♕h4 16. ♘c7 ♗xe2 17. ♕xe2♘xd4 Black has an attractive position, but it's not quite enough to compensate for the exchange: b3121) 18. ♕d3 e6 19. OO ♖d8 20. ♘b5 ( 20. b4♘c6 ) 20... a6 21. ♘a3 ♕g4 ( 21... ♘c6 22. ♕e2♘d4 23. ♕e3 ♗h6 24. ♕d3 ) 22. ♔h1 ( 22. ♖c2♕h4!= ) 22... ♘c6 23. ♕f3 ♕xf3 24. gxf3 ♗xb2 25.♘c2 ♖d2 26. ♖b1 ♗d4 27. ♘xd4 ♘xd4 28. ♔g2b5 29. a4+/=

b3122) 18. ♕e3 18... ♘c6 19. ♖c4 ♗xb2 ( 19...♕f6 20. b3 ♕a1+ 21. ♖c1 ♕xa2 22. OO ♕a5 23.♘d5 e6 24. ♘f4 ♗h6 25. ♖a1 ♕b5 26. ♖ad1a6+/= ) 20. OO ♖d8 21. ♘d5 e6 22. ♘c3+/= b32) 11... e6 12. ♗xc6 bxc6 13. ♘b4 ♕b6 14. ♕b3+/= b33) 11... ♗e6 12. ♗xc6 bxc6 13. ♘b4± b34) 11... ♖b8 A very strong move, threatening to play e6. 12.♕d2 Once again restoring the idea of Nb4 in the positions following e6. ( 12. ♕a4 e5 13. dxe5 ♘xe5 14. ♖d1 a6 15.♗d3 ♗e6=/∞ 16. ♘ec3 b5 ; If Black doesn't react to the threat then after e.g. 12. OO e6 13. ♗xc6 bxc6 the point of Rb8 is clear - White doesn't have Nb4, which would otherwise almost be winning.) 12... ♘xd4! ( 12... ♗e6 13. ♗xc6 bxc614. ♘b4± ) 13. ♘xd4 e6 Another point of 11... Rb8 now becomes clear - Nc7 doesn't hit the rook. 14. ♗c4 exd5 15.♗xd5 b341) 15... ♕b6 16. ♘b3 ( 16. ♖d1 ♗g4 17. f3 ♗e6 ) 16... ♗e6 17. OO ♗xd5 18. exd5 ♖bd8 19. ♖fd1♖fe8 20. g3 ♖e7 21. ♘a5 ♕xb2 22. ♕xb2 ♗xb2 23.♖b1± b342) ♕f6! Here Black should hold as White won't be in time to finish his development while Black will be able to create some serious counterplay after, for instance, b3421) 16. ♖d1 ♗e6! ( 16... ♗g4 17. f3 ♗e6 18.♘xe6 fxe6 19. ♗b3 ♖bd8 20. ♕c2 ♖xd1+ 21.♕xd1 ♕xb2 22. O-O with a small edge for White ) 17.OO ( 17. ♘xe6 fxe6 18. ♗b3 ♖bd8 ) 17... ♗xd5 18. exd5 ♖fd8 b3422) 16. ♘f3 b34221) 16... ♕xb2 17. ♕xb2 ♗xb2 18. ♖c7♗e6 ( 18... ♗f6 19. ♔e2 ♗d8 20. ♖xf7 ♖xf721. ♘e5± ) 19. ♗xe6 fxe6 20. O-O ( 20. ♔e2♖fc8 ) 20... ♖f7 ( 20... ♖fc8 21. ♖d7 ) 21. ♖c2♗f6= b34222) ♗g4 17. ♖c3 ( 17. ♘g5 ♖bd8 18. OOh6 19. ♘h3 ♕xb2 20. ♕xb2 ♗xb2 21. ♖b1 ♗e522. ♖xb7 ♖d7= ) 17... ♗xf3 18. ♖xf3 ♕xb2 19.♕xb2 ♗xb2 20. ♖b3 ♖fc8 21. OO ♖c2 22.♗xb7 ♗d4 23. ♗d5 ♖xb3 24. ♗xb3 ♖e2 25.♗d5 ♔g7= b3423) 16. ♘b5 16... ♗d7 17. ♘c3 ♕a6 18. ♕e2♖fc8=/∞

8... ♗d7 9. ♗xd7+ ♕xd7 In this position White has two possibilities, both of which are quite unpleasant for Black. 10. ♘xd5! 10. exd5 a) 10... OO 11. ♘f3 ♘a6 12. ♕b3 ♖ab8 ( 12... ♖fd8 13. OO) 13. OO ♖fd8 14. ♖ac1 ( 14. ♖fe1 ♘c7 15. ♘e5 ♕d6 16.♘c4 ♕d7 17. d6 exd6 18. ♘e4 Vovk-Tregubov, Lviv 8th Vasylyshyn Mem GM 2011 18... ♗xd4 19. ♖ad1 d5 20. ♖xd4♕e7= ) 14... ♘c7 15. ♘e5 ♗xe5 16. dxe5 ♘xd5 17. ♖fd1 e618. ♘e4+/= b) ♘a6 Stronger than Tregubov's 10...0-0. Black needs to be forceful and has no time to waste. 11. ♘f3 b1) 11... ♘c7 12. ♕b3 OOO ( 12... OO 13. ♘e5 ♕d6 14.OO ♖fd8 15. ♕xb7 ♘xd5 16. ♘c6 ♘xc3 17. bxc3 ♖dc818. d5 ♔f8 19. ♖ad1 ♗xc3 20. a4 ♗f6 21. a5+/= ) 13. ♘e5♗xe5 14. dxe5 ♘xd5 15. ♖d1 e6 16. O-O ♔b8 17. ♖d4♕c6 18. ♖fd1± b2) ♘b4 b21) 12. ♘e5 ♗xe5 13. dxe5 ♖d8 14. O-O Black faces some problems as he's behind in development. After 14... ♘xd5 (14... OO 15. ♕f3 ♘xd5 16. ♖ad1 e6 ) 15. ♘xd5 ♕xd5 White has the somewhat unpleasant 16. ♕a4+ b5 17. ♕xa7♕xe5 ( 17... OO 18. ♕xe7 ♖fe8 19. ♕b4 ♖xe5 20.♖fe1 ) 18. ♖fe1 ♕g5 19. b4 b211) 19... OO 20. ♕xe7 ♕xe7 21. ♖xe7 ♖d2 22.♔f1 ♖c8 23. ♖e2 ♖cc2 ( 23... ♖xe2 24. ♔xe2 ♖c2+) 24. ♖xd2 ♖xd2+/= b212) 19... e6 20. a4 bxa4 21. ♕xa4+ ♔f8 ( 21... ♔e722. b5 ♖b8 23. ♕a7+ ) 22. b5 ♔g7 23. b6± b213) ♔f8 20. ♖ac1 e6 ( 20... ♔g7 21. ♕xe7 ♕xe7 22.♖xe7 ♖d2 23. g3 ♖xa2 24. ♖cc7 ♖f8 25. ♖b7 ) 21.♕c5+ ♕xc5 22. ♖xc5 ♖d2 23. ♖a1 ♔e7 24. ♖xb5♖a8 25. ♖a5 ♖xa5 26. bxa5 ♖d5 27. a6 ♖a5 , and Black should hold.

b22) 12. ♕b3⁉ is a more ambitious option. 12... ♘d3+ 13.♔d2 (Not the vegetarian 13. ♔f1 ♘f4 14. ♘e5 ♗xe5 15.dxe5 OO=/∞ ) 13... ♘f4 ( 13... ♘xf2 is just bad because of14. ♖hf1 ♘g4 15. h3 ♘h6 16. g4 and despite material being equal Black is in a lot of trouble due to his horrible knight on h6. 16... ♗g7 17. ♘e5 ♗xe5 18. dxe5 OO 19. ♔c2± )14. g3 ♘h3 15. ♕b5 Very precise, and after this move I was unable to find complete equality for Black. (If White plays 15.♖hf1 and allows 15... ♘g5 I believe Black might have enough compensation here. 16. ♘xg5 ♗xg5+ 17. f4 ♗f6 18. ♔e3 OO 19. ♖fd1 ♖ac8 20. ♔f2 h5=/∞ ) 15... ♕xb5 16.♘xb5 OOO 17. ♘c3 ( 17. ♘xa7+ ♔b8 18. ♘b5 ♖xd519. ♘c3 ♖f5 20. ♔e3 g5 21. ♖hf1 ♖d8 22. ♖ad1 g4 23.♘e1 h5=/∞ ) 17... ♘xf2 18. ♖hf1 ♘g4 ( 18... ♘h3 19.♘e5 ♖hf8 20. ♔e2 h5 21. ♖ad1+/= ) 19. h3 ♘h6 20. g4♗g7 21. ♖fe1 ♖he8 ( 21... f5 22. ♖xe7 ♗f8 23. ♖e2fxg4 24. hxg4 ♘xg4 25. ♘e5 ) 22. ♖ac1 ♔b8 23. ♔d3♖d7+/= ( 23... f6? 24. ♘b5!± ) 10... ♗g7 11. ♕f3! The only way to fight for an advantage. 11. ♘f3 ♘c6= 11... O-O Forced. 11... ♗xd4 12. ♖d1 ♘c6 13. ♘e2+− simply drops material. 11... ♘c6 12. ♘e2+/= 12. ♖d1 12. ♘e2 e6 a) 13. ♘dc3 ♘c6 ( 13... ♗xd4 14. O-O-O ♘c6 15. ♔b1 ) 14.d5 exd5 15. ♘xd5 f5= b) 13. ♘f6+ 13... ♗xf6 14. ♕xf6 ♕b5 15. ♕f3 ♖c8 16. OO♖c2 17. ♘f4 ♘c6 18. ♕e3 ♕xb2 19. d5 exd5 20. ♘xd5 ( 20.exd5 ♘a5 ) 20... ♕d4= 12... e6 13. ♘f6+ 13. ♘c3 ♗xd4= ( 13... ♘c6 ) 13... ♗xf6 14. ♕xf6 ♕b5 15. ♖d2 I had some hope of finding equality here for a while, but in the end I settled on the evaluation that Black doesn't fully equalise, so it can't really be recommended. 15... ♘d7 15... ♕a5 16. ♘e2 ( 16. a3 ♘d7 17. ♕f3 ♘b6=/∞ ) 16... ♘d7 17.♕f4 ( 17. ♕e7 ♘b6 ) 17... ♕xa2 18. O-O+/= 16. ♕f3 16. ♕f4 e5 17. ♕g5 f6 18. ♕e3 exd4 19. ♕xd4 ♘c5→ 16... ♕a5

16... ♖ac8 17. ♘e2 a) 17... ♘b6 18. b3 ♕a5 19. ♕d3 f5 20. h4 ( 20. exf5 exf5 21.O-O ♘d5=/∞ ) b) ♕a5 18. a3 ♖fd8 19. b4 ♕a4 20. h4 h5 21. O-O ♘e5 22.♕g3 ♘c4 23. ♖d3± 17. ♘e2 17. a3 ♘b6 17... ♕xa2 18. O-O ♖ac8 18... ♘b6 19. e5 ♘d5 20. ♘g3 ♖ac8 21. ♘e4 ♖c6 22. h4+/= 19. e5 ♕d5 20. ♕e3 20. ♕xd5 exd5 21. ♘c3 ♘b6 22. f4 f5= 20... ♘b6 21. ♘c3 ♕b3 22. ♕h6 ♘d5 23. ♘e4↑

7.e4 Qb6 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗g5 ♗g7 5. ♗xf6 ♗xf6 6.cxd5 c6 7. e4⁉ ♕b6 8. e5 White can postpone e5, but that's less dangerous for Black: 8. ♘ge2♕xb2 9. ♖b1 ♕a3 10. g3 ( 10. ♕d2 OO 11. g3 ♖d8 12. ♗g2e5 13. OO ♘d7 14. f4 exd4 15. ♘xd4 cxd5 16. ♘xd5 ♗g7 17.♘b5 ♕c5+ 18. ♔h1 ♘b6 Gretarsson-Kalod, Pardubice 1999 19.♖fc1! ♕f8 20. ♖c7+/= ) 10... O-O ( 10... ♗g4 11. ♗g2 ) 11. ♗g2 a) 11... e5⁈ 12. dxe6 ♗xe6 13. e5 ♗e7 14. O-O ( 14. ♖xb7♗b4 15. ♕d3 ♗c8 1/2 Bhat-Shivaji, Berkeley Masters 2005 16.♖xb4 ♕xb4 17. OO ♗a6 18. ♕e3 ♗xe2 19. ♘xe2→ ) 14...♕a6 15. ♘f4 ( 15. d5 cxd5 16. ♘xd5 ♗xd5 17. ♕xd5 ♘c6 18.♘f4+/= ) 15... ♗c4 ( 15... ♗f5 16. ♗e4 ♗xe4 17. ♘xe4± ) 16.♖e1→ b) 11... ♗g7 12. ♕d2 ( 12. OO ) 12... ♕a5 ( 12... cxd5 13.♘xd5 ♘c6 14. OO ) 13. e5 ♕d8 14. f4 cxd5 15. ♘xd5 ♘c6 16. ♘e3 f6 17. ♕c3 fxe5 18. dxe5 g5 19. O-O± MchedlishviliSasikiran, Vlissingen 2004 c) cxd5⁉ 12. ♘xd5 c1) 12... ♘c6 13. ♘xf6+ ( 13. ♕d2 ) 13... exf6 14. ♕d2 ( 14.OO ♕xa2 15. ♖a1 ♕b2 16. ♕d3 ♕b4 17. ♖fb1 ♕d6 )14... ♗g4 15. ♖b3 ♕d6 16. f3+/= c2) ♗g7 13. OO ♘c6 14. ♕d2 e5 ( 14... ♗g4 15. f3 ♗c816. ♖fc1 ♖d8 17. ♖b3 ♕d6 18. f4+/= ) c21) 15. ♖b3 ♕a5 16. ♕xa5 ♘xa5 17. ♖b5 b6 ( 17...♘c6 18. dxe5 ♗xe5 19. f4 ♗g4 20. ♘c1 ♗d4+ 21.♔h1+/= ) 18. ♘c7 ♖b8 19. dxe5 ♗d7=

c22) 15. dxe5 15... ♘xe5 16. ♖fc1 ( 16. f4 ♘c4 17. ♕c2♘e3 18. ♘xe3 ♕xe3+ 19. ♖f2 ) 16... ♘f3+ 17. ♗xf3♕xf3 18. ♘e7+ ♔h8 19. ♘xc8 ♖axc8 20. ♖xc8 ♖xc821. ♖xb7+/= 8... ♕xb2 9. ♘ge2 9. ♕c1⁉ is another interesting idea: 9... ♕xc1+ 10. ♖xc1 ♗g7 11.♗d3 ( 11. ♗c4 OO 12. f4 cxd5 13. ♗xd5 ♘c6 14. ♘f3 ♗h6 15.g3 ♖d8 16. ♔f2 ♗f5 17. ♔e3∞ Kazhgaleyev-Le Quang Liem, Khanty-Mansiysk 39th olm 2010 ) 11... OO 12. ♘ge2 ♖d8 13. OO ♔f8 14.♗e4 cxd5 15. ♘xd5 ♘c6 16. ♖fd1 ♗h6 17. ♖b1 ♖b8∞ Tihonov-Zhigalko, Minsk 2004 9... ♗g7 10. g3 The material balance has been restored and Black has the bishop pair against a bishop and knight. It hardly looks like the end of the world for Black, but the more I looked at this position the less I liked it, because the bishop on g7 is now completely out of play. It's very difficult to challenge the pawn on e5 and, in general, White just develops naturally with Rb1 and Bg2, giving him a pleasant edge and decent pressure. 10. ♖b1 ♕a3 11. ♕d2 ( 11. g3 cxd5 12. ♘xd5? ♕f3 ) 11... O-O12. g3 10... O-O 11. ♖b1 ♕a3 12. ♗g2 12. ♕d2⁉ f6 13. exf6 ♗xf6 14. ♗g2 ♕a6 15. OO ♗f5 16. ♘e4♘d7 17. dxc6 bxc6 18. ♖bc1+/= Chernyshov-Oral, Pardubice 1999 12... cxd5 12... f6 13. e6 cxd5 14. ♗xd5 ♘c6 15. O-O± 13. ♘xd5 13. ♗xd5 ♘c6 14. O-O ♖d8 15. ♕c1 ♕a6 16. ♕e3 ♗f5 17.♖b5 e6 18. ♗f3 h5+/= Tomilin-Vorobiov, St Petersburg 2003 13... ♘c6 14. O-O ♖b8 14... ♕xa2 15. ♖a1 ( 15. ♘c7 ) 15... ♕b2 ( 15... ♕c4 16. ♖a4♕b5 17. ♘c7 ) 16. ♘ec3 e6 17. ♖a2+− 14... ♗g4 15. ♖xb7 ♖ab8 16. ♖c7 ( 16. ♖xb8 ♖xb8 17. ♘dc3♗xe2 18. ♘xe2 ♘b4 19. e6 ♕xa2 20. exf7+ ♕xf7 21. d5 e6 22.dxe6 ♕xe6 23. ♘f4 ♕f7 Eljanov-Kurnosov, AUT-tch 2009 ) 15. ♘e3!+/= 15. ♘ec3 a) 15... ♗e6 16. ♕d2 ♖fd8 17. ♖b3 ♕a6 ( 17... ♕a5 18.♖b5 ♕a3 19. ♖fb1+/= ) 18. ♖d1 ♖d7 19. ♕b2 Sokolov-Ristic, Neum 2002 b) ♖d8 b1) 16. ♖b3 ♕a5 17. ♖b5 ( 17. ♕d2 b5 ) 17... ♕a3 18.♕d2 e6 ( 18... a6 19. ♖b6 e6 20. ♘f6+ ♗xf6 21. exf6♘xd4 22. ♕f4 ♕c5=/+ ) 19. ♘f6+ ♗xf6 20. exf6 ♘xd4 21.♘d5 ♖xd5 22. ♖xd5 exd5 23. ♕xd4 ♗e6 24. h4 b6 25. h5♖c8=/+

b2) 16. ♕d2 b21) 16... e6 17. ♘b5 ( 17. ♘f6+ ♗xf6 18. exf6 ♘xd4 19.♘b5± ) 17... ♕f8 18. ♘f6+ ♗xf6 19. exf6 a6 20. ♘c3 (20. ♗xc6 bxc6 21. ♘c3 ♖xb1 22. ♖xb1 e5∞ ) 20...♘xd4 21. ♕e3 b6 22. ♖fd1 ♘f5 23. ♖xd8 ♕xd8 24.♕e5+/= b22) 16... ♗f5 17. ♖b3 ♕a5 18. ♖c1 a6 19. h4 b520.♕b2+/=

7.e4 Bg7 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. ♗g5 ♗g7 5. ♗xf6 ♗xf6 6.cxd5 c6 7. e4⁉ ♗g7⁉ I was stuck without a proper recommendation for Black until it dawned on me that this move hasn't yet been played in this position and might be very interesting. The point is that Black steps away from a possible tempo White might gain from playing e5 or Nxd5 after Black takes on d5. Black is aiming for a variety of positions where he doesn't even attempt to win the pawn back. He does, however, get a lot of positional compensation due to his very strong bishops and play against the white centre and queenside. If you don't like positions where you're behind in material you perhaps won't be as excited about this as I am, but I honestly believe this is very, very playable for Black, and it also has serious surprise value. White has four main moves in this position. 8. dxc6 The main move. 8. ♗c4 cxd5 ( 8... O-O transposes to the lines we're trying to avoid. 9.♘ge2 cxd5 10. ♗xd5 e6 11. ♗b3 ♘c6 12. d5 exd5 13. ♗xd5± ;8... ♕b6 9. ♘ge2 ♕xb2 10. OO O-O 11. e5 ♕b4 12. ♗d3 ♕a513. ♖b1+/= ) 9. ♗b5+ This is very similar to 7...cxd5, but the bishop on g7 gives Black extra options: (f 9. exd5 can be met by 9... O-O 10.♘ge2 ♘d7 and Black once again gets fantastic compensation for the pawn. After the possible 11. O-O ♘f6 Black wants to play b6 and Bb7. Even if the pawn on d5 doesn't drop in the immediate future White will be completely tied down to its protection. I think Black is just fine here.12. a4 b6 13. a5 bxa5 14. ♕a4 ♘e8 15. b3 ♘d6 16. ♕xa5♗d7=/∞ ) a) 9... ♗d7⁉ isn't completely satisfactory. 10. ♗xd7+ ♕xd7 11.exd5 OO⁉ ( 11... ♘a6 12. ♘f3 ♘b4 13. ♘e5+/= ) 12. ♘f3 e6=/∞ b) 9... ♘d7 10. ♘xd5 a6 b1) 11. ♗e2 ♘b6 b2) 11. ♗a4 b5 12. ♗c2 ♘b6 and Black has fantastic compensation for the pawn. ( 12... ♗b7 13. ♘c3 O-O 14. ♘f3♕b6 ) 13. ♘xb6 ♕xb6 14. ♘e2 O-O=/∞ b3) 11. ♗xd7+ is the main move. 11... ♗xd7 12. ♘e2 This is one of the many positions where Black won't win the pawn back, but after the possible 12... O-O 13. OO ♗c6 14. ♘dc3 b5 he gets a lot of play on the queenside and in the centre. Black of course wants to expand on the queenside by playing b4, and if 15. a3 then15... a5=/∞ before playing b4. If White feels obliged to play 16. d5the bishop on g7 will become an absolute monster, and in both structure and spirit the position will resemble my game against Kacheishvili that you may have seen in my Early Games chapter. Black has no need to worry about this position. ( 16. ♕d3 b4 17.♘b5 ♕d7 18. a4 ♖fd8 19. ♖fd1 ♖ab8= ) 16... ♗d7 17.♕d2 ♕b6 18. ♖fc1 ♗e5 ( 18... b4 19. axb4 axb4 20. ♘d1♖xa1 21. ♖xa1 ♖c8 22. ♘e3 b3=/∞ )

19. ♕e3 ( 19. ♘d1f5 20. exf5 ♖xf5 21. ♘e3 ♖f7 22. ♖c2 ♖af8 23. ♖f1 b424. axb4 axb4 25. f3 ♗b5=/∞ ) 19... ♕xe3 20. fxe3 b4 21.axb4 axb4 22. ♘d1 ♔g7 23. ♘d4 ♖fc8 24. ♘f3 ♗d6= 8. ♘f3 O-O , and here White has a number of options. ( 8... cxd5 9.♗b5+ ♗d7 10. ♗xd7+ ♕xd7 11. exd5± ) a) White can try to keep the pawn on d5 with 9. ♕b3 but after 9...cxd5 10. exd5 ♘d7 there's a lot of concrete play against the white centre. a1) 11. ♗d3 If 11... ♘b6 12. O-O e6 13. dxe6 ♗xe6 Black has a lot of compensation. 14.♕a3 (After 14. ♕b4 Black has the very strong idea of 14...♘d5 and if 15. ♕xb7 ♘xc3 16. bxc3 ♗d5 after taking on f3 Black will have good play against White's weakened kingside. The pawns on c3 and d4 will also come under constant attack. There are very few comfortable squares for the white queen. ) 14... ♗g4 15.♗e4 ♗xf3! 16. ♗xf3 ♘c4=/∞ with Nd2 to follow. This is quite comfortable for Black. a2) 11. a4 11... a5 12. ♗d3 ♘b6 leads to similar play to that in the above line. 13. OO e6 14. ♖ac1 ( 14. dxe6 ♗xe6 15.♕b5 ♗xd4= ) 14... ♘xd5 15. ♘xd5 ♕xd5 16. ♕xd5 exd517. ♖c7 ♗e6 18. ♖xb7 ♖fb8 19. ♖xb8+ ♖xb8 20. ♗b5♖c8= b) 9. ♗c4 ♗g4 10. O-O cxd5 11. exd5 ( 11. ♗xd5 ♘c6 gives Black excellent compensation for the pawn. 12. e5 ♕b6 13. ♗xc6bxc6=/∞ ) 11... ♗xf3 Capturing on f3 and d4 is the simplest approach. 12. ♕xf3 ♗xd4 , and Black has nothing to fear. He'll soon finish development, and with opposite-coloured bishops the structure is quite safe. 13. ♖ad1 ♗e5 14. d6 ♘c6= c) 9. ♗e2 cxd5 10. exd5 ♘d7 11. OO ♘b6 12. ♕b3 e6 13.dxe6 ♗xe6 14. ♕b4 ♘d5 15. ♕xb7 a5! 16. ♕c6 ( 16. ♘xd5♗xd5 17. ♕b5 ♖b8=/+ ; 16. ♖fe1 ♖b8 17. ♕c6 ♖xb2=/∞ )16... ♖c8 17. ♕a4 ♘xc3 18. bxc3 ♖xc3=/∞ d) 9. dxc6 9... ♘xc6 10. d5 ♘e5 This is a recurring structure in this line. White is a pawn up, but the bishop on g7 is extremely strong and Black gets a lot of play on the queenside. It's also not so easy for White to finish development in this particular position. d1) If 11. ♘xe5 ♗xe5 12. ♗e2 then 12... ♕d6⁉ is very strong and Black gets a lot of counterplay after White's replies. ( 12... f513. O-O ♕b6=/∞ ) d11) 13. h3 may invite Black to play f5 in some lines, as well as allowing Black's typical play on the queenside: 13... b5 14. a3♗d7 15. OO ♖fc8 ( 15... ♕f6 ) 16. ♖c1 a6=/∞ d12) 13. g3 13... ♗h3 ( 13... ♕b4 14. ♕d2 ♗h3=/∞ ) 14.♗f1 ♗d7 15. f4 ♗xc3+ 16. bxc3 ♖ac8 17. ♖c1 ♕a3 18.♕d2 e6 19. ♔f2 exd5 20. exd5 b5⇆

d2) 11. ♗e2 is perhaps a better option for White, but after 11...♘xf3+ 12. ♗xf3 the bishop will be completely misplaced on f3, where it really has no future. 12... ♕d6 ( 12... ♕b6 13. ♕d2♗d7 14. OO ♖ac8 15. ♖ac1 ) 13. OO ♗d7 Black gets fantastic counterplay. He's planning Rc8, a6 and b5 and in general he gets a lot of play on the queenside. After the possible 14. ♕e2 (14. ♗e2 a6 15. ♕d2 ♖ac8 16. ♖ac1 b5 17. f4 e5!=/∞ ; 14.♗g4 ♗xg4 15. ♕xg4 ♗xc3 16. bxc3 ♖ac8= )

d21) 14... a6 15. ♗g4 ( 15. ♖ac1 ♖ac8 16. ♕e3 b5 17.♗e2 b4 18. ♘d1 e6 19. ♖xc8 ♖xc8 20. ♕d2 a5 21.♘e3 ♕e5= ) 15... ♗xg4 16. ♕xg4 ♗xc3 17. bxc3 ♖ac818. ♖fb1 b5 19. a4 bxa4 20. ♖xa4 ♖xc3 21. g3 f5 22.♕e2 ♕c5= d22) 14... ♖ac8 15. ♖ac1 ( 15. ♖fc1 ♖c5 16. ♖c2 ♖fc817. ♖ac1 b5=/∞ ) 15... a6 White's only reasonable idea is 16.♗g4 , attempting to exchange one of the bishops, but Black has a very good response with 16... ♗xg4 17. ♕xg4 ♖c4=/∞ After the second rook comes to c8 Black will have tremendous pressure against White's entire position. At the very least Black is no worse. 8. ♖c1 is perhaps White's least critical option. 8... O-O 9. ♘f3 ( 9.dxc6 ♘xc6 10. d5 ♘d4 leads to a position that's very similar to the main line, but I'm not sure the tempo has been well-spent on Rc1, so this isn't particularly threatening for Black. ) 9... cxd5 10. exd5 (if 10.♘xd5 ♘c6 11. ♗b5 ♕a5+ 12. ♘c3 we finally get to play the wonderful tactic I already described - 12... ♘xd4 13. ♘xd4 ♖d8∓ )10... ♘d7 Aiming to play Nb6 or b6. In general Black gets fantastic compensation for the pawn in positions like this. ( 10... ♗g4 is also possible ) 11. ♗d3 (If White wants to hang on to the pawn he should play something like 11. ♗c4 ♘b6 12. ♗b3 but after 12... ♗g4 it turns out that while the pawn on d5 is now protected the d4-pawn will probably fall because White won't, of course, want to take on f3 with the g2pawn. ) 11... ♘b6 12. ♗e4 ♗g4 13. O-O ♖c8 14. h3 ♗xf3 15.♗xf3 ♖c4 16. ♕e2 ♖xd4 17. ♖fe1 ♘c8= 8... ♘xc6 8... ♗xd4 9. cxb7 ♗xb7 10. ♗b5+ ♘c6 11. ♘ge2 is no longer any good for Black. 8... ♕xd4 9. ♕xd4 ♗xd4 10. ♘d5± 9. d5 ♕b6 An important inclusion, as this move poses White additional questions. 10. ♖b1⁉ It's important to know that White can't play 10. dxc6 because 10...♕xb2 11. ♘d5 ♕xa1 12. ♘c7+ ♔f8 13. ♕xa1 is just fine for Black.13... ♗xa1 14. ♘xa8 ♗c3+ 15. ♔d1 ♗a5! The knight on a8 won't return, so Black may already even be slightly better. 16. ♗b5 bxc6 17.♗xc6 ♔g7 18. e5 ♖d8+ 19. ♔e2 ♗a6+∓ 10. ♕d2 ♘d4

a) 11. ♘a4 is simply met by 11... ♕d6 and after 12. ♘e2 ♘xe213. ♗xe2 ♗d7 14. ♘c3 OO 15. O-O ♖ac8 this is again a version of lines we've looked at before. Black is fine and will either launch play connected to a6 and b5 or start play in the centre connected with e6 and exd5 very soon. 16. ♖ac1 a6 , and once again Black has very good compensation: 17. a3 ( 17. f4 e5 ; 17. g3 b5 18. a3 ♖c7 19.♘a2 ♖fc8 20. ♖xc7 ♖xc7 21. ♘b4 a5 22. ♘d3 ♖c4 23. ♗f3♕c7 24. ♖c1 ♗h3=/∞ ) 17... ♖c7 18. ♖c2 ♖fc8 19. ♖fc1 e620. dxe6 ( 20. g3 exd5 21. ♘xd5 ♖xc2 22. ♖xc2 ♖xc2 23.♕xc2 ♕e5 24. b4 ♕b2= ) 20... ♕xe6 ( 20... ♕xd2 21. exf7+♔f8 22. ♖xd2 ♗h6 23. ♖cd1 ♗xd2 24. ♖xd2 ♗e6 25. f4♖xf7 26. g3 ♖d7 27. ♖xd7 ♗xd7 28. ♔f2∞ ) 21. g3 h5 22.♗d3 ♕d6=/∞ b) 11. ♖d1 O-O 12. ♘ge2 e5! ( 12... ♘xe2 13. ♗xe2 e6 14.O-O exd5 15. exd5 ) b1) 13. dxe6 ♗xe6! ( 13... ♘xe6=/∞ ) 14. ♘xd4 ♖fd8 15.♘ce2 ♗c4=/+

b2) 13. ♘c1 b21) 13... ♕d8 14. ♗c4 ( 14. ♘d3 ♕h4 ) 14... ♕h4 15.♕d3 ( 15. OO ♘f3+ 16. gxf3 ♗h6 17. ♕c2 ♗f4 18.♖fe1 ♕h3 19. ♘d3 ♗xh2+= ) 15... ♗g4 16. ♘1e2 b5 17.♗xb5 ♗xe2 18. ♘xe2 ♘xb5 19. ♕xb5 ♕xe4 20. f3♕e3 21. ♕b3 ( 21. ♖d3 ♕g5 ) 21... ♕c5=/∞ b22) 13... h5⁉ 14. ♗d3 ♔h7 15. O-O ♗h6 16. ♕e1♕xb2∞ c) 11. ♘ge2 The more natural move. 11... OO ( 11... e5 12.♘xd4 exd4 13. ♘a4 ♕d6 14. ♗b5+ ; 11... ♘xe2 12. ♗xe2O-O 13. O-O ♗d7 ) c1) 12. ♖d1 is met by the very strong 12... e5 c11) After 13. ♘c1 Black has two ambitious moves. 13... ♕d8(Another geometric possibility is to play 13... h5 14. ♗d3♔h7 and it turns out White has no convenient way of dealing with the threat of Bh6. Once again, Black gets fantastic compensation for the pawn. ) 14. ♗c4 ♕h4 when Black has serious play against the white king. For instance, 15. OO♘f3+ 16. gxf3 ♗h6 and White is lucky he has a draw here. c12) 13. dxe6 would be fine after 13... ♘xe6 (but 13...♗xe6 14. ♘xd4 ♖fd8 is even stronger and Black is already better, as White's whole edifice is on the point of collapse.) c2) 12. ♘xd4 12... ♗xd4 13. ♖d1 Driving the bishop back. 13...♗g7 14. ♗e2 ( 14. ♗d3 ♗d7 15. OO ♖ac8 16. ♔h1 ♕a517. ♖c1 e6 18. ♗e2 exd5 19. exd5 ♖fe8=/∞ ) c21) 14... f5 15. exf5 ( 15. OO fxe4 16. ♘xe4 ♕xb2 17.♗c4 ♕xd2 18. ♖xd2 ♔h8 ) 15... ♗xf5 16. OO ♖ac8 17.♗f3 ♗e5 18. ♖fe1 ♗d6+/= c22) 14... ♗d7 15. O-O ♖ac8 Here we get a position that's already become typical for our analysis, although White is somewhat more harmoniously placed. It's time to discuss plans. Despite White's position looking very nice the only serious plan he has to improve his lot here is somehow to prepare and play f4 and e5. On the other hand, if he does manage to do that he'll have a huge advantage, because Black's whole play centres around the bishop on g7. If that bishop is shut out of the game Black will suffer greatly. 16. ♔h1 is therefore the most critical move, and Black has to respond forcefully to avoid getting into serious trouble. c221) 16... ♖c5 17. f4 ♗xc3 18. bxc3 ♖fc8 19. c4± c222) 16... ♕b4 17. a3+/=

c223) 16... ♕a5 17. ♖c1 b5 18. a3 b4 ( 18... e6 19. d6) 19. ♘a2± c224) 16... e6 17. f4 c225) 16... ♖fd8 17. f4 e5 18. f5 c226) 16... ♕c5 is the move Black should start with, driving the rook from d1 to c1. 17. ♖c1 ♕b4 Creating the threat of Bxc3 and Qxe4, and forcing White to waste a tempo on 18.f3 ( 18. a3 ♗xc3 19. bxc3 ♕xe4 20. ♖fe1 ♕e5 ) c2261) 18... ♕d4 19. ♕xd4 ♗xd4 20. ♖fd1 ♗e5 21.g3 g5 22. ♔g2 a6 23. ♖f1 b5 24. f4 gxf4 25. gxf4♗d4 26. ♔f3+/= c2262) 18... ♖fd8 19. ♖fd1 ( 19. a3 ♕d4 20. ♕xd4♗xd4 21. g3 h5 22. ♔g2 e6 23. ♖fd1 ♗e3 24.♖b1+/= ) c2263) 18... ♖c7 19. a3 ♕d4 20. ♗d3 ♖fc8 21. f4 c2264) 18... e6! This is a very important source of counterplay for Black. If he manages to take on d5 and put the bishop on 19. a3 ( 19. ♖fd1 exd5 20. a3 ♕a5 21.♕xd5 ♕xd5 22. ♘xd5 ♖xc1 23. ♖xc1 ♗e6 24.♖c2 ♗xd5 25. exd5 ♖d8 26. ♗c4 ♖c8 27. d6♖c6= ) 19... ♕d6 20. f4 The critical move. (If White does something quiet like 20. ♖fd1 Black will be fine after forcing simplifications. 20... exd5 21. ♘xd5 ♗a4 22. ♖xc8 ♖xc8 23. ♖c1 ♖xc1+ 24. ♕xc1 ♗c6= ) 20... e5 This is why it was so important to move the white rook to c1, and is the point of all the preceding play. It may be something of an achievement for White to have closed down the bishop on g7, at least for now, but exf4 is a huge threat. 21. g3 Almost forced. ( 21. f5 is impossible because White loses material after 21... gxf522. exf5 ♗h6 ) 21... ♖fe8 Putting additional pressure on the white centre. After, for instance, 22. ♗f3 a6 it's very hard for White to make much progress. It's almost never possible to play f5, while Black has a lot of ideas for improving his position. In many cases he'll play b5, and he can also put the rook on c4, where it exerts additional pressure against the central pawns on the 4th rank. In many positions Black could consider playing f7-f5, which will most likely result in the complete liquidation of all the pawns on the e and f-files. In combination with

Black's wonderful bishops that should give him complete equality. (22... exf4 23. gxf4 ♗xc3 24. bxc3 ♕xa3 25. e5→ )23. ♗g2 b5 ( 23... f5 24. exf5 ♗xf5 25. fxe5 ♗xe526. ♖ce1 ) 24. ♕d3 f5=/∞ 10... ♘d4 11. ♗d3 O-O 12. ♘ge2 This is an attempt by White to get a slightly different setup with the bishop on d3. 12... ♗d7 Consistent with the previous play. 12... ♘xe2 13. ♕xe2 12... f5⁉ is an interesting alternative for Black, giving him concrete play against the white centre, but I think it's a bit premature and I don't like the idea of breaking up the pawn structure. a) 13. exf5 ♘xf5 14. O-O ♗d7 ( 14... ♘e3 15. ♘a4!+/= ) 15.♔h1 ♖ad8 16. ♕d2 a1) 16... ♗h6 17. f4 ( 17. ♕c2 a6 ) a2) e6 17. ♗xf5 ♖xf5 18. ♘g3 ♖ff8 ( 18... ♖f7 19. d6♗c6 20. ♘ge4 ) 19. d6 ♗c6 20. ♖bd1 ( 20. ♘ge4 ♕b4= )20... ♗xc3 21. bxc3 ♗b5 22. f3 ♗xf1 23. ♘xf1 ♖f5 24.♘e3 ♖b5∞ b) 13. OO 13... fxe4 14. ♘xe4 ♗g4 15. ♕d2 ♖ad8=/∞ ( 15...♘xe2+ 16. ♗xe2 ♗f5 17. ♗f3+/= ) 13. ♘xd4 13. O-O ♖ac8 14. ♕d2 ♘xe2+ 15. ♘xe2 e6 16. ♘c3 ♕a5 17.♖fd1 ♖fe8=/∞ 13... ♗xd4 14. O-O ♖ac8 14... e6 15. ♗c2 ♖fd8 16. ♗a4+/= 14... a6 15. ♘e2 ♗e5 16. ♔h1 ♖ac8 17. f4 ♗b8 18. ♕d2+/= 15. ♕d2 15. ♘e2 ♗g7 16. ♔h1 e6 17. ♘c3 ♕a5= 15... ♖c7 In this position the only serious play for White is on the kingside. 15... e6 16. ♖fd1 a6 ( 16... ♕a5 17. ♘b5 ) 17. ♖bc1 ♖fd8 18.♗f1+/= 15... a6 16. ♔h1 ( 16. ♖bc1 ♖c7 ; 16. ♖fd1 ♕f6 ) 16... e6 (16... e5 17. ♖bc1 ♖c7 18. ♖c2 ♖fc8 19. ♖fc1 ; 16... ♖c7 17.f4 ♖fc8 18. e5 ) 17. f4 ♕c5 18. dxe6 ( 18. ♗e2 ) 18... fxe6 19.♕e2+/= 16. ♔h1 Again the critical move.

The machine likes 16. ♕f4 for a while, because it creates the threat of d5-d6 followed by Nd5 in some positions. 16... f6 Securing a different outpost so the bishop can switch to e5 if it's forced to move. Black will just continue with his play - Rfc8, Kg7, perhaps a6, whereas the queen on f4 is really a bit of a sore thumb and does little unless Black immediately blunders some tactics.( 16... ♖fc8 17. d6! ) 17. ♖fc1 ♖fc8 18. ♖c2♔g7 19. ♕d2 a6 20. ♖bc1 ♕d6 21. g3 b5 22. a3 h5 ( 22... e623. ♗e2 ♗b6 24. ♗g4 ) 23. ♗f1 ♗a7=/∞ 16... ♖fc8 17. f4 ♕a5 Necessary precision. 18. f5 Trying to come up with some direct threats on the kingside. Black needs to simplify, and after the following forced sequence – 18... ♗xc3 19. bxc3 ♕xc3 20. ♕xc3 ♖xc3 21. ♖xb7 ♖xd3 22.♖xd7 gxf5 23. exf5 Black ends up with a rook endgame where he's a pawn down. It's not always such a great idea to recommend a rook endgame a pawn down, so I spent some time checking and I'm pretty sure that in most cases this is a forced draw. As in all the previous examples, Black holds reasonably comfortably and I'm happy to recommend this group of positions to you. I think Black gets fantastic compensation for the pawn, and as this is all absolutely fresh there should be a decent element of surprise 23. ♖xa7 fxe4 24. ♖xe7 ♖a8 25. h3 ♖xa2 26. ♖xe4 ( 26.♖fxf7 ♖g3= ) 26... ♖g3 27. ♖g4+ ♖xg4 28. hxg4 ♖d2 29. ♖f5♖d4= 23... ♖c2 23... a5 24. a4 ( 24. ♖xe7 ♖xd5 25. f6 h5 26. h3 ) 24... ♖c2 24. a4 24. ♖xe7 ♖xa2 25. f6 h6= 24... a5 25. ♖xe7 ♖xd5 26. f6 h6 27. h3 ♖g5 28. ♖d1 ♖cxg229. h4 ♖5g4 30. h5 ♔h8 31. ♖d8+ ♖g8 32. ♖xg8+ ♖xg8 33.♖e5 ♖g4 34. ♖xa5 ♖f4 35. ♔g2 35. ♖a6 ♖h4+ 36. ♔g2 ♖xh5= 35... ♖xf6 36. ♖b5 ♖f4 37. a5 v♖a4=

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 7.Be3 c5 8.Qd2 8.Rc1 8.Nf3 Qa5 9.Bd2 9.Nd2

7.Be3 c5 8.Qd2 We're now going to look at 7.Be3 in the Grünfeld, a line which for a long time was living a very quiet life away from the spotlight. That all changed in 2000, when Vladimir Kramnik used it to beat Garry Kasparov in Game 2 of their World Championship match in London. From that moment on its popularity completely exploded, with constant developments. That was mainly due to the efforts of Kramnik himself, but of course you shouldn't forget people like Khenkin and Dautov, who developed the theory of the line for many years before Kramnik gave it "celebrity status". Over the years it's caused a great deal of headaches for me personally, and as a Grünfeld player I've never particularly liked the positions you get with Black. Therefore it's one of the lines I've analysed the most, and it actually has the somewhat strange distinction of being the only line of any opening I've ever written an article about in a chess magazine. That was published in 2002 in Schach magazine, and to this date remains my only involvement with official chess theory in published form. That's all about to change, obviously, with the chess24 project, but up until now that was my one claim to theoretical fame. Anyway, 7.Be3 is a very serious line and it will take us some time to get through all the possibilities White can employ... so let's get started! 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗e3 c5 Black responds in the usual manner, and now White already has a variety of options. 8. ♕d2 Not a particularly threatening line. This move and also 8.Rc1 are attempts to triangulate Black into unfavourable move orders i.e. to get him to choose something he wouldn't normally have chosen against 7.Be3. 8... cxd4 9. cxd4 ♘c6 The most natural recommendation and by now the established norm. 10. ♖d1 O-O 11. ♘f3 The main line. 11. ♗e2 e5 This is a forcing solution which I was very happy to finally get to play on the board, because it was one of the first positions I saw when looking at the Grünfeld with my coach Andrei Lukin all the way back in the mid-90s. 12. d5 ♘d4 13. ♘f3 This is the idea of 11.Be2. ( 13.♗xd4 exd4 , with the rook coming to e8 with great force, isn't particularly favourable for White. ) 13... ♘xe2 14. ♕xe2 ♕a5+ 15.♗d2 ♕a4 A very important and nice move to make. ( 15... ♕xa2 It's important to note that this isn't particularly pleasant for Black, as after16. O-O the white rooks will come to a1 and b1 and the queen will have to waste a lot of time getting back to home base. White will have fantastic compensation for the pawn. ) 16. O-O b6 Black has a lot of pressure on the white queenside and the black queen is extremely well-placed to add some pressure along the fourth rank - the pawn on e4 will be a source of worry for White. The bishop will come out to a6 and one of the rooks will occupy the c-file. Somewhere around this point a draw was agreed in a game between yours truly and Boris Gelfand in one of the Euro Cups in 2009. If anyone's better it's probably Black, but I just felt that in a team tournament a draw with Black on first board is always a nice outcome because it makes the rest of the team play in a more relaxed fashion.

11. d5 White can try this if he wants to avoid 11.Nf3 Bg4. 11... ♕a5!12. ♕xa5 Otherwise ...Bc3 is obviously a huge threat. 12... ♘xa5 13.♗d2 b6⇆ Black gets a very favourable version of the endgame you're about to see a lot of - this type of structure constantly appears on the board in this line. 11... ♗g4 A very natural response. 12. ♗e2 ♖c8 13. O-O ♕a5 In this position Black is fine regardless of which of the two endgames White chooses. 14. ♕xa5 14. h3 was played by Vishy Anand, but after 14... ♗xf3 15. ♗xf3♕xd2 16. ♖xd2 ♘a5 Magnus Carlsen was absolutely fine as Black. 17.e5 ♘c4 18. ♖d3 ♘xe3 19. fxe3 ♖c7 20. ♗e4 ♖fc8= Anand-Carlsen, Moscow 6th Tal Memorial 2011 14... ♘xa5 15. ♖c1 ♗xf3 16. ♗xf3 ♘c4 17. e5 b5⁉ It's better to play this first, completing all the preparations before taking on e3. The bishop really isn't going anywhere. 17... ♘xe3 is a move I played myself in one of the Monaco tournaments against Vassily Ivanchuk, but after 18. fxe3 ♗h6 19. ♔f2 b5 20.♔e2 e6 21. ♗b7 ♖c4 22. ♖xc4 bxc4 White can play 23. ♗a6! and Black will actually have to suffer a little. ( 23. ♗e4 ♖b8 24. ♖b1♖xb1 , and the game was drawn in Ivanchuk-Svidler, Nice Amber Blind 2010 ) 23... ♖b8 24. ♗xc4 ♖b2+ 25. ♔f3 ( 25. ♔d3 is less clear:25... ♖xg2 26. d5 ♖g5 27. d6 ♖xe5 28. d7 ♗g5 29. ♔d4 ♖f530. ♖b1 ♔g7 31. ♖b8 ♖f2 32. ♔d3 ♖xh2⇆ Volkov-Zinchenko, Rethymnon 2010 ) 25... ♖c2 26. ♗b3 ♖c3 27. ♖e1 ♗g5 28. g3h5 29. h4 ♗h6 30. ♔e4+/= , and Black will have to defend for a while, although a drawn outcome is still favourite. 18. ♖fd1 18. ♗d5 ♘xe3 19. fxe3 ♗h6 20. ♔f2 e6 21. ♗b7 ♖c4 wins Black an important tempo. 18... e6 Black can even wait for another move like this. Generally speaking, in this position the knight on e4 is really no worse than any of the bishops, so you don't need to rush to exchange it. This endgame is very safe for Black.

8.Rc1 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗e3 c5 8. ♖c1 This is a very fashionable and serious move and Black has to be precise so as not to land in lines he doesn't want to play. 8... ♕a5 9. ♕d2 cxd4 9... O-O 10. ♘f3 ♖d8 11. d5 Black has ended up stuck in a very topical line with a lot of possibilities for both sides. This isn't something I want to recommend because I think it's potentially very dangerous for Black. 9... ♘c6 is completely wrong because 10. d5 is possible - the main point of 8.Rc1 is that White wants to have the option of playing d5 the moment Black plays ...Nc6. 10. cxd4 ♕xd2+ 11. ♗xd2 The modern and critical way of handling this position. 11. ♔xd2 An important additional option for White. 11... O-O

a) 12. ♘f3 ♘c6 will transpose to the lines arising after 8.Nf3. b) 12. ♗d3 perhaps has some separate value. 12... ♘c6 13. ♘e2♖d8 14. ♖c4 If Black proceeds carefully and doesn't force matters I think he'll be very close to complete equality, but there's also a forcing solution: 14... ♗e6 b1) 15. d5 ♘e5 16. ♖c7 f5! is fine for Black, as he blows up the white centre and starts immediate counterplay against the somewhat over-extended white pawns on d5 and e4. 17. ♘f4 ( 17.f3 ♘xd3 18. ♔xd3 ♗xd5! 19. exd5 ♖xd5+ 20. ♔c2 ♖e5 21.♖c3 ♖a5!=/+ ) 17... ♗f7 18. ♖xe7 ♗f6 19. ♖xb7 g5 20.♖xf7 ♔xf7 21. ♘e6 ♖db8⇆

b2) 15. ♖a4 15... ♗d7 White either has to allow an exchange of the light-squared bishop or repeat moves. 16. ♖b1 ♘e5 17.♗b5 ♘c6! This return is very strong and now the threat of a6 means that after 18. ♗d3 ♘e5 19. ♗b5 ♘c6 White will once again have a choice of either repeating moves or playing something like 20. f3 a6 21. ♗xc6 after which it's quite clear that Black cannot be worse. 21... ♗xc6 22. ♖a5 e6 23. ♖c5 ♖d7 Black was doing well here in MelkumyanNielsen, Plovdiv 2010 c) 12. ♗b5 12... ♘c6 This is once again perfectly comfortable for Black. 13. ♘f3 ( 13. ♘e2 ♖d8 14. ♗xc6 bxc6 , and the white centre is under attack.; 13. ♗xc6 bxc6 14. ♖xc6 is impossible because of 14... ♗b7 ) 13... f5 It's been known for many years that this is perfectly satisfactory for Black. 11... O-O 12. ♘f3 e6 12... ♗g4 13. ♗e3 is also possible, but to my mind these endgames are unsafe and as things stand now I wouldn't recommend them. 13. h4 Black was supposed to be completely safe in this line until Vladimir Kramnik - who else? - came up with this new approach, which sparked another revival of the Be3 variation just when White had been getting a bit stuck in the other main lines. Play becomes very sharp. 13. ♗b5 is another move that's been played by Kramnik. a) 13... ♘c6 was played in a couple of high-profile games: a1) In 2013 Kramnik tried a somewhat different approach: 14.♗e3⁉ ♖d8 15. ♗xc6 bxc6 16. ♘e5 ♗b7 17. f3 , and now after 17... ♖ac8! ( 17... ♗xe5⁈ 18. dxe5 ♖d3 19. ♔f2♖ad8 20. ♗g5 ♖8d7 21. h4 it was highly unpleasant for Black in Kramnik-Areshchenko, Tromso World Cup 2013 ) 18. ♘d3 ( 18.♖c5⁈ f6 19. ♘d3 e5=/+ ) 18... ♗xd4 19. ♗xd4 ♖xd4 20.♘c5 White has obvious compensation for the pawn, but Black should be perfectly fine. a2) 14. ♗xc6 14... bxc6 15. O-O ♗a6 16. ♖fe1 Kramnik enjoyed a safe advantage and was pressing for the entire game, although he didn't manage to win. 16... ♗b5 17. ♗b4 ♖fd8 18.♗c5 a5 19. h4 a4 20. ♖c3 f6 21. e5 fxe5 22. ♘xe5 ♖d523. f4 Kramnik-Mamedyarov, Tal Memorial, Moscow 2010 b) 13... ♗d7⁉ I think this is the best option. 14. ♗xd7 The critical move. ( 14. ♗c4 is met by 14... ♗c6 , when Black starts play against the white centre immediately and is just fine. 15. d5 exd516. exd5 ♗a4 17. O-O ♘d7 ) 14... ♘xd7

b1) 15. ♖c7 ♘f6 16. e5 ♘d5 17. ♖xb7 ♖fb8=/∞ 18. ♖xb8+♖xb8 19. O-O ♖b2 It's quite clear that Black has fantastic compensation for the pawn, and this has actually already been played: 20. a4 ♗f8 21. a5 ♗b4 The monster knight on d5 gives Black ample play, Stocek-Votava, CZE-tch 2012 b2) 15. ♗e3 ♖fc8 ( 15... ♘f6 16. ♘d2 ) 16. ♔e2 b5= b3) 15. ♔e2 White aims for a slight advantage, but here again Black has a forcing option which I think solves all of his problems. b31) 15... f5⁈ 16. ♖c7 fxe4 17. ♘g5± b32) 15... ♖fc8 16. a4 ♘b6 ( 16... e5 17. d5 ♘c5 18.♖c4 ♘b3 19. ♖xc8+ ♖xc8 20. ♖b1 ; 16... f5 17. e5 h618. a5 b5 19. axb6 ♘xb6 20. ♖xc8+ ♖xc8 21. ♖a1 ♖c722. ♖a6 ♗f8 23. d5! exd5 24. ♗e3+/= ) 17. ♖xc8+ ♖xc8 18. ♖b1+/= b33) 15... ♘f6 16. ♔d3 ♘g4! Attacking the f2pawn. 17.♖hf1 ( 17. ♔e2 would of course be met by Black returning with 17... ♘f6 ) 17... ♖ad8 Creating the important threat of ...Ne5+. White's pawn centre comes under a direct attack and if18. ♔e2 Black once again plays 18... ♘f6 and White has a choice between playing Kd3 and accepting the repetition or 19.♗g5 when Black has the simple 19... ♖d7= White's options are again Kd3, with a repetition, or taking on f6, when Black is no worse. 13. ♗c4 was tried by Kramnik in an epic game against Caruana, but I found a very comfortable improvement for Black: 13... ♘c6 14. d5exd5 15. ♗xd5 ♖e8! ( 15... h6 16. OO ♖e8 17. ♖b1 ♖e7 18.♖fc1 ♗e6 19. ♗xe6 fxe6 20. ♔f1 left White with a lasting plus in Kramnik-Caruana, Dortmund 2013 ) 16. OO ( 16. ♘g5 ♖e7 17. O-Oh6 18. ♘f3 ♗e6 is at least fine for Black. ) 16... ♗e6 17. ♖b1 (17. ♘g5 ♗xd5 18. exd5 ♘e5 ) 17... ♗xd5! The easiest option. (17... ♖ad8 18. ♗xe6 ♖xe6 19. ♖xb7 ♖xe4 20. ♖xf7! ) 18. exd5♘d4! 19. ♘xd4 ♗xd4 20. ♖xb7 ♖ed8 21. ♖e1 ♖xd5= 13... ♘c6 14. e5 ♖d8 Hitting the pawn on d4. This rook shuffle solves the immediate problems. 14... ♗d7⁈ 15. ♖b1 Attacking the b7-pawn and forcing Black to make another small concession on his next move. 15... b6 16. h5 ♘e7 17.hxg6 fxg6 ( 17... hxg6 In this position White has all kinds of plans connected with simply giving mate, such as Ng5 and Rb3-h3. ) 18. ♗c4♖ac8 19. ♗b3 Structurally Black's position is very suspect and Anish Giri wasn't able to hold it in Kramnik-Giri, Dortmund 2011. 14... h6 This is the most natural move and the reaction Black employs to h4 in almost any position, but here it runs into 15. h5 g5 16. ♘xg5hxg5 17. h6 This piece sacrifice is incredibly dangerous and Black is in a lot of trouble. 17... ♗h8 18. h7+ ♔g7 19. ♗xg5 ♔g6 20. f4 f5 21.♗e2+− 15. ♗g5 ♖e8 15... ♖d5 is a very intriguing alternative: a) 16. ♗e3 ♖a5 17. ♗c4 ♗d7

a1) 18. ♖b1 b5! 19. ♗b3 ( 19. ♖xb5 ♖xb5 20. ♗xb5 ♘xe521. ♗xd7 ♘xd7= ) 19... ♖a6 20. ♘d2 ♘a5 21. ♗d1 ♘c4⇆ a2) 18. O-O 18... ♖a3 19. ♗b3 a5⇆

b) 16. ♗c4 16... ♖d7 17. ♗e3 ( 17. ♗b5 ♖d5 18. ♗xc6 bxc619. ♖xc6 ♖a5=/∞ ) 17... ♖c7 18. ♗d3 ( 18. h5 ♗d7 19. ♗b3♖ac8 20. ♘g5 ♘a5∞ ) 18... ♗d7 19. ♔e2 ( 19. h5 ♖ac8 20.♖b1 ♘e7 21. ♘g5 ♘f5 22. ♔e2 ♗c6∞ ) 19... ♖ac8 20. ♖b1♘e7∞ Diermair-Grandelius, Reykjavik Open 2013 16. ♗e3 h6 The point of the rook manoeuvre from f8 to e8 becomes clear in this position, as the 17.h5 piece sacrifice no longer works . 17. ♗d3 White is forced to switch to positional play. 17. h5 g5 18. ♘xg5 hxg5 19. h6 ♗f8! A very important additional resource. The king is also able to go to h8, and this setup is much better for Black than the other way around. 20. ♗xg5 ♗e7∓ White doesn't have enough for the sacrificed piece so Black is doing very well. 17... ♗d7 18. ♖b1 Black was already threatening to play ...Nb4. Black has a couple of options in this position. 18... ♖ab8 The move I like best, and something that's already been played at the very highest level by Fabiano Caruana against Teimour Radjabov in Wijk aan Zee. 18... ♘a5 isn't losing, but it's not very safe because White gets an initiative in the following line: 19. ♘d2 ( 19. ♗d2 b6 20. ♗xa5 bxa521. ♔e2 ♖ab8 22. ♘d2 ♖b4 ) 19... ♗c6 20. ♘e4 ♘c4 21. ♔e2♗d5 22. ♖hc1 ♖ec8 23. ♘f6+ ♗xf6 24. exf6 h5⇆ Jankovic-Gupta, Kavala Open A 2012

18... b6 was played against me by the young Indian grandmaster Negi.19. ♔e2 ( 19. ♘d2⁉ is also a decent idea. ) 19... ♗c8⁉ Once again Black aims to transfer the bishop onto the long diagonal. ( 19... ♘e7 20.♘d2 ♘d5 21. ♘e4 ♖ec8 22. ♗d2 ♗a4 23. ♖bc1 ♗f8 24. g4was somewhat better for White in Li Shilong-L'Ami, Golden Sands 1st Grand Europe Open 2012 ) 20. ♖hc1 ♗b7 21. ♘d2 I tried to achieve one of the main aims for White in this position - transfering the knight from f3 to e4. 21... ♖ed8 22. ♖c4 ♖ac8 23. ♖a4 I had to waste this additional tempo to prevent ...Ba6. On the one hand I managed to get the knight all the way to e4, but on the other hand the rook on a4 is constantly under threat of getting caught. Eventually the game was drawn, but in general I think 18...b6 allows White some additional freedoms. 23... ♗f8 24. g4 ♖c7 25. ♘e4 ♘a5 Black's counterplay is just in time. 26. ♗d2 ♘c6 27. ♗e3 ♘a5 28. ♗d2 ♘c6 And a draw was agreed in Svidler-Negi, Gibraltar Masters 2012 19. ♘d2 19. ♔e2 ♖ed8 20. ♖hc1 b6 , and it's hard for White to make much progress since any knight move will be met by 21...Be8, and the d4-pawn is under attack. ( 20... ♗f8⁉ ) 19... ♘e7 The most natural way to proceed for Black. Without b7-b6 the bishop will be very solid on c6. 19... ♖ed8⁉ is also a very decent move: 20. ♘e4 ♘xd4 21. ♗xd4♗c6 22. ♘d6 ♖xd6 23. exd6 ♗xd4=/∞ 20. ♘e4 ♗c6 21. g4 Very logical, cutting the knight off from the f5-square.

21... ♖ed8 22. ♔e2 ♘d5 23. ♖hc1 ♗f8 Black is just in time with his counterplay on the queenside. He wants to play ...Nb4 if circumstances allow. 24. ♗d2 ♘b6 25. ♗e3 ♘d5 26. ♗d2 ♘b6 27. ♗e3 And Radjabov-Caruana, Wijk aan Zee A 2012, was drawn by threefold repetition. In general, these endgames are quite sharp and obviously require some attention because if White manages to get an attack rolling on the kingside it will be quite unpleasant for Black. Nevertheless, with precise play it should be ok for Black to go for this, which is how I propose you handle this particular variation.

8.Nf3 Qa5 9.Bd2 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗e3 c5 8. ♘f3 The main move in this position. 8... ♕a5 8... ♘c6 is a decent back-up choice. The game might continue: 9. ♖c1 a) 9... cxd4 is wrong, though: 10. cxd4 ♕a5+ 11. ♗d2 ♕xa2 12.d5 ♘e5 13. ♘xe5 ♗xe5 14. ♗b5+ ♗d7 15. ♕g4! ( 15. ♕e2a6 16. ♗xd7+ ♔xd7 17. ♕g4+ ♔d8 18. ♕g5 f6 19. ♕e3 ♕a420. OO ♔e8 21. ♗c3± Seirawan-Mecking, Sao Paulo 1992 ) 15...e6 16. ♗xd7+ ♔xd7 17. ♕g5! , and Black is in a lot of trouble.17... ♕b2 ( 17... f6 18. ♕h6 ) 18. OO h6 19. ♕e3 ♕b6 20.♕f3 ♔e8 21. ♗c3 ♕d6 22. dxe6 ♕xe6 23. ♖fd1 ♗xc3 24.♖xc3 ♕e7 25. ♕f4± b) ♕a5 10. d5 ( 10. ♕d2 cxd4 11. cxd4 ♕xd2+ 12. ♔xd2 O-Ois yet another transposition to 8.Nf3. ) 10... ♘e5 11. ♘xe5 ♗xe512. ♕d2 O-O 13. f4 ♗g7 14. c4 ♕xd2+ 15. ♔xd2 b6 , and the endgame after 16. ♗e2 e5! 17. ♖hf1N ( 17. g4 exf4 18. ♗xf4♖e8 19. ♔d3 ♗d7= ) 17... f5 18. fxe5 ♗xe5 seems quite playable for Black, e.g. 19. ♗h6 ♖e8 20. ♗d3 fxe4 21. ♗xe4♗g7 22. ♗xg7 ♔xg7 23. ♖ce1 ♗d7= GrischukGanguly, Dresden Olympiad 2008 8... ♗g4 is a move I've played myself on a number of occasions, but White currently has a very unpleasant reply: 9. ♖c1 I'm not quite sure how to equalise, which is why I'm not recommending 8...Bg4 and one of the reasons Black is stuck playing the main line in the Be3 Grünfeld. 9. ♗d2 O-O 10. ♗e2 This line was made popular some time ago by Viktor Korchnoi, but I think by now Black has worked out how to deal with it. 10... ♗g4 11. O-O 11. ♕b3⁈ is more than adequately met by 11... cxd4 12. cxd4 ♕b6=/+and White is already in some trouble. His pawn centre is about to collapse. Volodin-Nyback, 2012 11... ♖d8 The more critical line.

11... e6⁉ immediately also comes into consideration as 11...Rd8 allows an interesting pawn sacrifice. 12. a4 cxd4 ( 12... ♘c6 13. ♖b1 cxd4 14.cxd4 ♕d8 15. ♗g5 ♕d7 16. d5 exd5 17. exd5 ♗xf3 18. ♗xf3♘d4 19. ♖b4 ♘xf3+ 20. ♕xf3 ♖fe8 21. h4 ♖ac8= Stocek-Borisek, Kallithea 2008 ) 13. cxd4 ♕d8 14. ♗g5 ♕d6 15. ♖b1♗xf3 16. ♗xf3 ♕xd4 17. ♕xd4 ♗xd4 18. ♖xb7 ♘c6 19. ♖d1 (19. ♖c7 ♖ac8 20. ♖xc8 ♖xc8 21. ♖c1 ♘e7= Korchnoi-Ruck, Leipzig 2002) 19... ♖ac8 20. ♖d7 ♗b6 21. ♗f6 ♘a5 22. g3 ♖c4=Eljanov-Macieja, Moscow 2002 12. d5 Pretty much the only critical move. 12. ♖b1 gives Black a very favourable version of the classical Rb1 variation. 12... cxd4 13. cxd4 ♕xa2 14. d5 ♘d7 15. ♖xb7 ♘c516. ♖b4 a5= Roiz-Avrukh, Netanya 2009 12. ♘g5⁈ Since a game between Korchnoi and Smirin it's been known that12... ♗xe2 13. ♕xe2 ♕a6! is very strong. 14. ♕xa6 ( 14. ♕f3 , attacking the f7-pawn, is the move White would very much like to make, but it runs into the strong 14... f6! The queen on a6 is covering the all-important e6-square, forcing the knight all the way back. 15. ♘h3Black now just wins a pawn for not very much with 15... cxd4 16. cxd4♖xd4 ) 14... ♘xa6 15. d5 c4=/+ Black was already better in Korchnoi-Smirin, Bled Olympiad 2002 12. a4 ♗xf3⁉ ( 12... e6 13. h3 ♗xf3 14. ♗xf3 ♘c6 15. ♖b1♕c7 16. d5 ♘e5 17. ♗f4 ♘xf3+ 18. ♕xf3 ♕d7 19. ♗g5 ♖e8+/=Shipov-Ganguly, Guelph 2005 ) 13. ♗xf3 cxd4 14. cxd4 ♕a6 15. d5♗xa1 16. ♕xa1 ♘d7=/∞ 12... ♗xc3 Black has to accept this pawn sacrifice, and I think it's stronger to accept it like this rather than by first taking on f3. 12... ♗xf3 13. ♗xf3 ♗xc3 14. ♗xc3 ♕xc3 15. ♖b1⁉ ( 15. ♖c1♕a5 16. e5 ♘c6 17. ♖e1 ♘d4∓ Ikonniko v-Voekler, Katowice 1991 )15... b6 16. e5 ♘a6 ( 16... ♘c6 17. d6! ♖ac8 18. ♗xc6 ♖xc6 19.dxe7!± ) 17. d6 ♖ab8! ( 17... ♕xe5⁈ 18. ♗xa8 ♖xa8 19. ♕f3+/=Korchnoi-Balinov, Leipzig 2002 ) a) 18. ♕a4 ♘b4 19. dxe7 ♖e8 20. ♕xa7 ( 20. a3 ♘d3 ) 20...♕xe5 21. ♖fe1 ♕f6 22. ♕c7 ♘c2= b) 18. ♕e2⁉ 18... c4 ( 18... ♘b4 19. ♖fd1! exd6 20. ♖b3!♕c2= ) 19. dxe7 ♖e8 20. ♖fc1 ♕d3 21. ♖d1 ♕xe2 22. ♗xe2♖bc8= 13. ♗xc3 ♕xc3 14. ♖c1 ♕a3 White obviously has to play very energetically here. He's a pawn down and needs to prove he has something in return. 14... ♕a5 15. ♘g5 ♗xe2 16. ♕xe2 h6 17. ♘f3 ♘d7 18. e5 ♘f819. ♖fd1 ♖ac8 20. ♖c4 b5 21. ♖h4 c4 22. e6 fxe6 23. ♖xh6♕c3 24. ♘d4± Girya-Khotenashvili, Gaziantep EU-chw 2012 15. ♘g5 The best move in this position, trying to start the f-pawn running up the board. 15. ♘e5 ♗xe2 16. ♕xe2 ♘d7∞ 15... ♗xe2 16. ♕xe2 h6 Played precisely in order to keep the pawn on f2 for as long as possible. If you allow f4 and then e5 or f5 next White will get a big initiative on the kingside. 17. ♘f3 ♘d7 Black is ready to consolidate, so White has to be very fast.

18. e5 ♘b6 19. e6 f5 20. ♘e5 ♔h7 This looks a bit worrying for Black, but on the other hand the pawn on d5 is about to fall, and after the possible 21. h4 ♖xd5 22. h5 g5 23. ♖fe1 , creating the huge threat of Qc2, Black only has to make one more precise move: 23... c4∞ Now that the queen on a3 joins the defence by protecting the pawn on e7 Black shouldn't have much to fear. After e.g. 24. ♕c2 ♖f8 25. ♘g6 ♖f6 Black can perhaps start thinking about playing for a win. In general, this kind of pawn sacrifice needs to be taken very seriously, but in this particular case Black is fine.

8.Nf3 Qa5 9.Nd2 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗e3 c5 8. ♘f3 ♕a5 9. ♘d2⁉ This move has been known about for a while, but it's become more popular in the last three or four years due to the efforts of Levon Aronian. Black has many options. 9... ♗d7⁉ I think this is the strongest move and it's the one I want to concentrate on. 9... ♘d7 is quite playable and something I've tried myself in this position. a) 10. ♖c1⁉ ♕xa2 11. ♗e2 OO 12. OO ♕a5 13. d5 b5 14.c4 b4 15. f4 ♕c7 16. e5 a5 17. ♘e4 ( 17. ♖a1 ♗b7 18. ♘e4f6 19. e6 f5∞ ) 17... a4 18. ♖a1 a3 19. ♖a2=/∞ Lima-Svidler, Khanty-Mansiysk 2011 b) 10. ♕c2 10... OO 11. ♗e2 ♘f6 12. OO ♘g4 13. ♗f4 cxd414. cxd4 e5 15. ♘c4 ♕c7 16. ♗g3 b5 17. h3 ♘h6 ( 17...♘xf2 18. ♖xf2 bxc4 19. ♗xc4 ♗b7 20. ♕d3+/= Laznicka-Kulaots, Gibraltar Masters 2012 ) 18. ♘e3 ♕xc2 19. ♘xc2 exd420. ♖ad1 ( 20. ♗xb5 d3 21. ♗xd3 ♗xa1 22. ♖xa1 ♖d8= )20... ♗b7 21. f3 ♖ac8 22. ♘xd4 a6= 9... cxd4 isn't entirely accurate: 10. ♘c4 a) 10... dxe3 is a very beautiful positional sacrifice, but objectively it's not enough for equality: 11. ♘xa5 ♗xc3+ 12. ♔e2 ♗xa5 13.♔xe3 OO 14. ♗b5 a6 15. ♗a4 b5 16. ♕d5! ♖a7 17. ♗b3 a1) 17... ♗b6+ 18. ♔e2 ♖d7 19. ♕g5 ♘c6 20. h4 ♖fd8 21.♖ad1 ( 21. ♖hd1⁉ ♘d4+ 22. ♔f1± ) 21... ♘d4+ 22. ♖xd4♖xd4 23. f3+/= Berkes-Baramidze, BL 2013 a2) ♗g4 18. ♕g5 ♗b6+ 19. ♔f4 ♗e6 20. ♗xe6 fxe6+ 21.♔g3 ♘c6 22. ♔h3 ♘d4 23. ♖hd1 ♖c7 24. ♖ac1+− Aronian-Sutovsky, Ningbo WChT 2011 b) ♕d8 11. cxd4 OO 12. ♖c1! ( 12. ♖b1 f5 13. exf5 ♗xf5 14.♖xb7 ♘c6 15. d5 ♗c3+ 16. ♗d2 ♘d4 17. ♘e3 ♕a5 18.♗xc3 ♕xc3+ 19. ♕d2 ♕a1+ 20. ♕d1 ♕c3+ 21. ♕d2 ♕a1+ 22.♕d1= Avr ukh-Kovchan, Mulhouse FRA-tch 2011 ) 12... f5 13. ♕b3♔h8 14. ♘e5 ♗xe5 15. dxe5 ♘c6 16. ♕c3 ( 16. ♖d1 ♕a5+17. ♗d2 ♕b6 18. ♗c4 fxe4 19. O-O ♕xb3 20. axb3 ♗g4Babula-Hoffmann, BL 2012 ) 16... ♗e6 17. exf5 gxf5 18. ♗e2♕d5 19. ♗f3 ♕xa2 20. O-O f4 21. ♗c5=/∞

9... O-O 10. ♗e2 ♕xc3 11. ♖c1 ♕a3 12. ♖xc5 ♕xa2 13. O-O is very unclear: 13... ♘c6 ( 13... ♘a6 14. ♖c2 ♕a5 15. ♕b1 ♕d816. ♖fc1 b6 17. ♕a1 ♘b4 18. ♖b2 f5 19. g3 e5 20. dxe5 ♘d321. ♗xd3 ♕xd3 22. ♖xb6 fxe4 23. ♕a2+ ♔h8 24. ♖d6+−Laznicka-Navara, Novy Bor m 2011 ) 14. ♖c4 ♕a5 15. d5 ♘e5 16.♖a4 ♕c7 17. f4 b5 18. ♖b4⁉ ( 18. ♖a3 ♘c4 19. ♕c2 e6 20.♗xc4 bxc4 21. ♕xc4 ♕d8 22. ♗d4 exd5 23. exd5 ♗b7=Kasimdzhanov-Navara, Wijk aan Zee B 2009 ) 18... a5 19. ♖xb5 ♘d720. ♖b3+/= 10. ♖b1 An important move, not only attacking the b7-pawn but also preparing to meet 10...Ba4 by capturing the bishop. 10. ♕b3 cxd4 11. cxd4 ♘c6=/+ 10. ♗e2 ♗a4 11. ♕c1 cxd4 12. ♗xd4 ♗xd4 13. cxd4 ♘c6=/+Black is putting a lot of pressure on the white centre and may already be better here. 10... ♕xc3! This move has already been played, but I'm recommending it in connection to a slightly different idea from the one that's been shown in practice. 10... ♗a4 a) 11. ♕c1 cxd4 12. ♗xd4 ♗xd4 ( 12... OO 13. ♖xb7 ) 13.cxd4 ♘c6 14. ♕b2 e5 15. d5 ♘d4 16. ♗d3 O-O 17. O-O ♗c218. ♘c4 ♕c7 19. ♗xc2 ♕xc4 20. ♗d1 b6= b) 11. ♕xa4+ 11... ♕xa4 12. ♗b5+ ♕xb5 13. ♖xb5 b6 This endgame may be holdable for Black, but it's not much fun and I don't want to recommend yet another boring, slightly worse endgame. I have something else in mind. b1) 14. ♔e2 cxd4 15. cxd4 ♘c6 16. d5 ( 16. ♘f3 e6 17.♖c1 ♖c8 18. a4 OO= ) 16... ♘d4+ 17. ♗xd4 ♗xd4 18.♖c1 ♔d7= b2) 14. dxc5 14... ♗xc3 15. ♔e2 ♗xd2 16. ♔xd2 ♘d7 17.♖c1+/= ( 17. c6 ♘f8 was too hasty in Melkumyan-Khalifman, Moscow 11th Aeroflot op A 2012 ) 10... b6 11. ♗c4 cxd4 12. cxd4 ♘c6 13. O-O O-O 14. ♗b5!± 10... cxd4 11. ♘c4 ( 11. cxd4 ♘c6 ) 11... ♕c7 12. cxd4 b5 13.♘d2 ♘c6 14. ♘b3± 11. dxc5 Almost forced. 11. ♖c1 ♕a3 12. ♖xc5 ♘a6∞ is very unclear because Black is quite well-developed and White will still have to settle on a square for the rook. 11... ♘a6⁉ This is my proposal, and I think a stronger move. Black immediately starts concrete play against the c5-pawn. White can meet this in three different ways. 11... ♗c6 was played by Grandelius against Fressinet in one of the last rounds of the 2013 Bundesliga. Black equalised quite comfortably, but I'm not sure he would be as happy against very precise play. 12. ♗b5 O-O13. O-O a) 13... ♖d8 14. ♕a4!±

b) 13... ♗d4 14. ♖c1 ♕b4 15. ♖c4 ♕xb5 16. a4 ♕a5 ( 16...♕a6 17. ♖xd4 ♘d7 18. ♘b3 ♘f6 19. e5 ♘h5 20. ♗h6 ♘g721. ♗xg7 ♔xg7 22. ♕d2+/= ) 17. ♘b3 ( 17. ♖xd4 ♘d7 18.♘b3 ♕c7 ) 17... ♕c7 ( 17... ♕a6 18. ♕xd4 ♘d7 19. ♗h6 e520. ♕c3 ♖fe8 21. ♘a5 ) 18. ♘xd4 ♘d7 ( 18... ♗xe4 19. ♗h6) 19. f3+/= c) 13... ♕a5 14. a4 ( 14. ♕b3 ♗xb5 15. ♕xb5 ♕xb5 16.♖xb5 b6= ) 14... ♘d7 15. ♗xc6 ( 15. ♘c4 ♕c7 16. ♗xc6bxc6 17. ♕e2 ♖ab8= ) 15... bxc6 16. ♖b7 ♘xc5 17. ♘b3♘xb3 18. ♕xb3 e6 19. g3→ d) ♘d7 14. ♗xc6! ( 14. ♖b3 ♕f6 15. ♘c4 ♗xb5 16. ♖xb5♕c6 17. a4 b6 18. cxb6 ♕xc4 19. ♕xd7 axb6 20. ♕xe7 ♖xa421. ♖xb6 ♕xe4 22. ♕xe4 1/2 Fressinet-Grandelius, BL 2013 )14... bxc6 15. ♖b3 ♕a5 16. ♘c4 ♕xa2 ( 16... ♕c7 17. g3+/= )17. ♖b4 ♖fb8 ( 17... ♖ab8 18. ♖a4 ♕b3 19. ♕xb3 ♖xb3 20.♖xa7+/= ) 18. ♖xb8+ ♖xb8 19. ♕xd7 ♕xc4 20. ♕xc6+/= 12. ♗b5 Black has a few options. 12. ♗c4 is probably the least critical choice for White. 12... ♗c6 13.O-O O-O 14. ♕e2 ♘c7 Black is very harmoniously developed - the bishop on c6 can't really be touched and the knight on c7 covers both the d5 and b5-squares. Black's pawn on b7 is therefore very safe and I think Black has very decent play. 15. ♘f3 ♖ad8 16. ♖fc1 ♕a3= 12. ♗xa6 bxa6 13. ♖c1 Trying to push the c-pawn. ( 13. O-O If White allows something like 13... ♗b5 14. ♖e1 O-O= then Black is clearly fine because the bishops control the entire board. One will eventually be very well-positioned on c6 while the other is a monster on g7. ) 13...♕f6 Stopping the c-pawn. 14. ♘f3 Once again creating the direct threat of Bd4. 14... ♕e6 Black is again ready to play ...Bc6 next move.15. ♕d5 ♕xd5 16. exd5 ♗b5 Despite optically having a very dangerous pawn pair on c5 and d5 White really has nothing at all here. Black is ready to castle queenside yet again and the pawn on d5 will come under attack. The game might continue 17. ♘d4 ( 17. c6 O-O-O= )17... ♗xd4 The cleanest solution. ( 17... ♗d3⁉ is also possible. 18. f3♖d8 ) 18. ♗xd4 ♖g8 After, for instance, 19. ♗e5 OOO 20. d6♖ge8 Black is ready to blockade the white passed pawn on c5, or d6 if White decides to recapture on d6 with the pawn. In this position I think Black is completely safe and has nothing at all to worry about. 21. ♔d2f6 22. ♗g3 ♔d7= 12... ♗xb5 My preferred move. 12... ♖d8 13. O-O O-O 14. ♕e2+/= 13. ♖xb5 ♗d4 Once again, this is a very concrete way of dealing with the situation. Black keeps up the attack on the c5-pawn. 14. ♕b3 Almost forced. 14... O-O-O! The safest way for Black to proceed. I have to admit I quite like lines in the Grünfeld Defence where Black castles long. They're not so frequent and every time I get to do that in the Grünfeld I feel some excitement. It's like a badge of honour - you found a position where you can castle queenside in the Grünfeld! 14... ♕c1+ , accepting the exchange sacrifice, is also possible. 15.♔e2 ♕xh1 16. ♗xd4 O-O 17. ♗b2↑ 14... ♕xb3 15. ♘xb3 ♗xe3 16. fxe3 ♖b8= 15. ♔e2 ♕xb3 16. ♘xb3 ♗c3! One last precise move. This isn't forced, but I think it's the best way to proceed because it captures the rook on b5 and ...Nc7 is a huge threat.

17. ♗f4 e5 18. ♗g5 ♘c7 19. ♗xd8 ♖xd8 20. a4 ♘xb5 21.axb5= Black is completely fine.

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 7.Be3 c5 8.Nf3 Qa5 9.Qd2 Nc6 10.Rc1 10.Rb1 cxd4 and 10...0-0 10...a6 11.Rc1 Bg4 11...f5 12.d5 12.e5 12.Bd3

9.Qd2 Nc6 10.Rc1 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗e3 c5 8. ♘f3 ♕a5 9. ♕d2 We're now going to concentrate on this position, which features a large number of forcing and very important lines. Black has a clear choice. 9... ♘c6 The safer move for Black, after which White has two major options, 10.Rc1 and 10.Rb1. 9... O-O is possible, when play would continue 10. ♖c1 ♖d8 11. d5 e6 12. ♗g5 This is a very well-known position which has been debated at the highest level for a number of years. There have been a lot of recent developments and, as things stand right now, I'm not quite sure Black equalises. 10. ♖c1 This is the somewhat more classical approach, which was seen in a lot of Kramnik games from the early 2000s. It was recently revived when Kramnik beat me by introducing an important novelty in the London Candidates 2013. 10... cxd4 11. cxd4 ♕xd2+ 12. ♔xd2 O-O 13. d5 13. ♗b5 f5! poses no problems at all for Black, as has been known for a long while. For instance: 14. exf5 ♗xf5 15. ♗xc6 ( 15. ♔e2 ♗e616. ♗c4 ♗xc4+ 17. ♖xc4 ♖ad8 18. ♖b1 ♖d7 ) 15... bxc6 16.♖xc6 ♖ab8 Black has a lot of play for the single pawn he had to sacrifice. 17. ♖a1 ( 17. ♔c3 ♗e4 18. ♖c7 ♖fc8= ) 17... ♖b7 (17... ♖b2+ 18. ♔e1 ♗h3 19. ♘h4 ♗g4=/∞ ) 18. ♖c5 ♗e4 19.♘g5 ( 19. ♘e1 ♖d8 20. f3 ♗f5 21. g4 ♖b2+ 22. ♘c2 ♗e6=/+ )19... ♗xg2 20. ♖g1 ♗f3 21. ♖g3 ♗h1= 13... ♖d8 14. ♔c2⁉ This is the more important move, and what Kramnik played against me. 14. ♔e1 This move used to be the gold standard, and Kramnik played it to great effect in a number of games. This was a very important position in the early years of this millennium and in fact remains - until the chess24 project - my one involvement with official chess theory. More than ten years ago I wrote an article for the German chess magazine Schach, which mainly dealt with this particular position. I remember I was very happy when some of my suggestions began cropping up in high-profile games. Theory has moved on since then and what I'm going to recommend for Black is something that wasn't featured prominently in my article, but was popularised by Emil Sutovsky and Michal Krasenkow. 14... ♘a5 15.♗g5 ♗d7 16. ♗d3 f5! I think this completely puts an end to the old 14.Ke1 line, because White has no good options here. ( 16... ♖dc8 used to be the old main line.)

a) 17. ♘d2 ♖dc8 18. ♔e2 e6 Black gets a much improved version of 16...Rdc8 17.Ke2 e6, with the inclusion of Nd2 and ...f5. That's of course greatly in his favour, as no-one in his right mind would have played Nd2 if Black had started with ...Rdc8. 19. ♗e3 ( 19. h4exd5 20. exd5 ♖e8+ 21. ♔f3 ♖e5=/+ Markos-Sutovsky, Bled 2002) 19... fxe4 20. ♗xe4 ♗b5+ 21. ♔f3 exd5 22. ♗xd5+ ♔h8=Ftacnik-Sutovsky, 2002 b) 17. ♗xe7 ♖e8 18. d6 fxe4 19. ♗xe4 ♗c6 20. ♘g5 leaves Black with a good choice. 20... h6 ( 20... ♗f8 is less adventurous, and after 21. ♗xf8 ♔xf8 22. f3 h6 23. ♖c5 hxg5 24. ♖xa5♖ad8= the game is very likely to end in a completely drawn rook endgame.) 21. ♗xg6 hxg5 22. ♗xe8 ♖xe8 This position is very unclear. 23. f3 ♔f7∞ c) 17. e5 The main move. 17... ♗e8 Attacking the d5-pawn. 18. d6The pawn has to advance. 18... ♖dc8! Black is very comfortable in this position whatever White does. c1) 19. ♔d2 exd6 20. exd6 ♗f8 The pawn on d6 is a huge target and practice has shown that Black is at least doing fine here.21. ♗e7 ( 21. ♗f4 ♖d8 22. ♖he1 ♗f7 23. ♘g5 ♗xd6 24.♗xd6 ♖xd6 25. ♘xf7 ♔xf7 26. ♖c7+ ♔f6 27. ♖xh7 ♘c628. f4 ♖ad8 29. ♖e3 ♘b4 30. ♖ee7 ♖xd3+ 31. ♔e2♖d2+ 32. ♔e1 ♘c2+ 33. ♔f1 ♖f8−+ Ki. Georgiev-Krasenkow, Rethymnon 2003 ) 21... ♗xe7 22. dxe7 ♔g7 ( 22... b5⁉ 23.♘e5 ♔g7 24. f4 a6 25. ♖c3 ♖xc3 26. ♔xc3 ♖a7=/+ Wang Jue-Yang Kaiqi, Qinhuangdao 1st op 2011 ) 23. ♘g5 ♖c6 24.♖xc6 ♘xc6 25. ♖e1 ♖c8 26. ♘e6+ ♔f6 27. ♘f8 h6=BanikasBelov, Kavala 2004 c2) 19. dxe7 19... h6 20. ♗d2 ♖xc1+ 21. ♗xc1 ♖c8 22.♔e2 ♘c6 Black gets his material back and is very comfortable.23. ♗e3 ♘xe5 ( 23... ♘xe7 24. ♗xa7 ♘d5= ) 24. ♘xe5♗xe5 25. ♗xh6 ♔f7 26. h4 ♔xe7 27. h5 gxh5 28. ♗xf5♖c5= 14... ♘a5 This standard response is what Black should play. Very sharp play ensues. 14... ♘e5 I tried for half an hour to remember what my notes said, failed miserably, and played this move in London. It's not very critical. Although I still almost equalised, against Vlad "almost" isn't quite enough. In the end I wasn't able to make a draw and this game started my opponent on his way to a fantastic score in the Candidates - one that was almost enough to qualify him for a World Championship match. 15. ♗g5 ♔f8! 15... ♗d7 has already been played in some games. a) 16. ♗xe7 does in fact allow Black to equalise after 16... ♖e8 17.♗b4 ( 17. d6 ♘c6 18. ♗b5 ♘xe7 19. ♗xd7 ♖ed8 ) 17...♖ac8+ 18. ♔b1 ♖xc1+ 19. ♔xc1 ♘c4! A very beautiful move, and if White doesn't take the knight Black gets a lot of play against the king. a1) 20. ♗d3 ♖c8! 21. ♔b1 ♗b5!=/+ 22. a4 ( 22. ♖d1 ♘b2−+) 22... ♘a3+ 23. ♔a2 ♗xd3 24. ♔xa3 ♗xe4 a2) 20. ♘g5 ♗h6 21. f4 ♗xg5 22. fxg5 ♖xe4 23. ♗d3 ♖d424. ♗c2 ♘b6=/+

a3) 20. ♗xc4 20... ♖xe4 21. ♘d2 ♗h6 Black wins all the material back and is just fine. 22. ♔b2 ♗xd2 23. ♗xd2 ♖xc4= b) 16. ♔b1! Black's life isn't so simple after this move. 16. ♗f4 Black needs to be very precise here and either have a lot of knowledge or a lot of courage. The fact that the rook can't go to f8 will make it very vulnerable, and you'll actually have to give it up on many occasions. 16. ♔b1 a) 16... b5⁉ is also possible and a pretty decent idea. 17. ♗f4 ( 17.♖c7 ♖d7 18. ♗f4 a6 19. ♗d3 ♘c4 20. ♗xc4 bxc4 21. ♖xc4♖b7+ 22. ♔c1 ♗d7=/∞ ) 17... ♘c4 18. ♗xc4 bxc4 19. ♖xc4♗a6=/∞ b) b6 This is quite safe for Black. In general, Black wants to transfer the knight from a5 to c5, and if he manages that he'll be fine, so White needs to try and stop it. b1) 17. ♗d3 ♘b7 b2) 17. ♗b5 ♗d7 18. ♗xd7 ♖xd7 19. ♖hd1 b5 20. ♘d2♖b7∞ b3) 17. ♗f4 ♘b7 18. ♗b5 ( 18. ♘e5 g5 19. ♗g3 f5=/+ )18... ♘c5!= 19. ♖xc5 bxc5 20. ♗c6 ♗a6= b4) 17. ♖c7 17... ♖d7 18. ♖c2 ( 18. ♖xd7 ♗xd7 19. ♗a6♗c8= ) 18... ♖d8 I think Black has enough counterplay because White either has to repeat moves or allow some kind of positive restructuring of Black's pieces. 19. ♗b5 ( 19. ♗d3 ♘b7∞ ; 19.♗f4 f5! 20. ♗d3 ♘b7! 21. ♗c7 ♖d7 22. ♗b5 fxe4 23.♗xd7 ♗xd7 24. ♘g5 ♘c5=/∞ ) 19... a6 20. ♗e2 ♘b7 21.♗e3 ♘c5∞ 16... b6! 17. ♗c7 17. ♗d3 ♘b7 18. ♗c7 ♖d7 19. ♗e5 ♘c5 20. ♗b5 ♖d8 21.♗c6 ♗xe5 22. ♘xe5 ( 22. ♗xa8 ♗f4 23. ♖ce1 ♘xe4=/+ ) 22...♖b8 23. f3 ♗a6→ 17... ♖d7 18. ♔d1 Bb5 seems to be a huge threat, but Black can just ignore it altogether. 18. ♔b1? simply can't be played here, because 18... ♖xd5!∓ more or less wins on the spot. 19. exd5 ♗f5+ 20. ♖c2 ♗xc2+ 21. ♔xc2♖c8 Black emerges a pawn up in a better position. 18. ♗g3 f5! 19. ♘g5 ( 19. ♗b5 ♖d8 20. ♗d3 fxe4 21. ♗xe4♗b7=/+ ) 19... fxe4 20. ♘e6+ ♔f7 21. ♘c7 ♖b8 22. ♗e2 ( 22.♔b1 ♖d8!=/+ ) 22... ♖b7 23. ♔b1 b5! 18... ♘b7 18... f5⁉ 19. ♗b5 ♗b2 20. ♖c2 fxe4 21. ♗xd7 ♗xd7 22. ♖xb2exf3 23. gxf3 ♖c8 24. ♗f4 ♘c4 25. ♖b4 a5 26. ♖b1 ♖c5=/∞ 19. ♗b5 ♘c5! It turns out White wins an entire exchange in this position, but that's not enough to give him any advantage at all. Black is fine. 20. ♗g3 Black even has a choice.

20. ♗xd7 is possible, but after 20... ♗xd7 White has to be careful not to be worse. 21. ♔e2 Trying to at least connect the rooks. ( 21. ♗e5♗a4+ 22. ♔e2 runs into the very strong 22... ♗h6 , when White no longer has any good ideas. The pawn on e4 is hanging, the rook on c1 is hanging and ...Bb5+ will drive the king back to unpleasant squares like d1. White may already be worse. 23. ♖c3 ♗b5+∓ ; 21. ♘e5 ♗xe522. ♗xe5 ♘d3= ) 21... ♗h6 Yet again the most forceful way of handling this position. ( 21... ♘xe4 is possible, aiming for positional compensation. 22. ♘e5 ♗b5+ 23. ♔e3 ♘c5=/∞ ) 22. ♖xc5 ...Bb5+ was a huge threat, so objectively I think returning the exchange like this is the best option. ( 22. ♖b1 would be met by 22... ♘xe4∓ and White is in a lot of trouble - all kinds of forks are threatened. ) 22... bxc5 23.♖b1 a6! Securing the important b5-square for the second bishop, and with the bishops controlling the whole board Black simply can't be worse.24. ♖b6 ♗b5+ 25. ♔e1 ♖c8= 20. ♗f4 can be met by 20... ♖d8 and there are now additional forks on d3 in some lines. If 21. ♖xc5 bxc5 22. ♗c6 the position of the bishop on f4 means that Black can even play 22... e5! If White takes on e5 Black has an additional square on b8 for the rook, while ( 22... ♗a6= ) 23.♗xa8 exf4 is just fine for Black. 20. ♗c6 ♖xc7 21. ♗xa8 ♗a6=/+ leaves Black doing very well because of the huge threat of ...Rc8, forcing the bishop to c6, when the pawn on e4 will finally fall. Black has fantastic play and is perhaps already better. 20... ♗h6⁉ A more human move that I'm at least suggesting as an alternative. 20... ♖d8 is a line the machine suggests is perfectly fine. 21. ♗c6 ( 21.♖xc5 bxc5 22. ♗c6 ♗d7 23. ♗xa8 ♖xa8 24. ♔d2 ♗b5= ) 21...♗a6 22. ♗xa8 ♖xa8 23. ♖e1 ( 23. ♘d2 ♖d8 24. f3 f5 25. ♖e1♗b5!= A recurring theme in this position - Black is in no hurry to win the material back and instead calmly improves his pieces. There's not much White can do: 26. ♖e3 ♗a4+ 27. ♔e1 ♗b5=/∞ ; 23. e5 ♖d824. d6 exd6 25. exd6 ♗b5!=/∞ ) 23... ♗b5 The machine's move - apparently Black isn't worse as White is so completely discoordinated. Nevertheless, this looked a bit alien to me when I found it. 24. ♘d2♗h6=/∞ a) 25. f4 ♘d3 26. ♖c7 ♘xf4⁉=/∞ ( 26... ♘xe1 27. ♔xe1 ♔e8= ) b) 25. ♖c3 25... ♗g7 26. ♖a3 ( 26. e5 ♖d8 ) 26... ♘a4! 27.♖ee3 ♗d4 28. ♖eb3 ♗c5= 21. ♖xc5 Once again the most practical approach for White is to start (preemptively!) returning material. 21. ♖c2 ♖d8 22. ♗c6 ♗d7 23. ♗xa8 ♖xa8 Black has the huge threat of ...Ba4 and the pawn on e4 is hanging. I think White already has to pay attention not to be worse here. 24. ♘d4 ♗a4 25. f3 ♖d8 26.♔e2 ♗xc2 27. ♘xc2 f5 28. ♘d4 ♔e8 29. ♘c6 ♖d7= 21. ♗xd7 ♗xd7 22. ♖xc5 bxc5 23. ♔c2 f5= Yet again the simplest way for Black to proceed, hitting the white centre. The draw is the overwhelming favourite here. 21... bxc5 22. ♗c6 ♖b7 23. ♗xb7 ♗xb7 24. ♘e5 e6 Starting to undermine the white centre immediately. 25. ♘d7+ ♔e8 26. ♘xc5 ♖c8= This tactical finesse means Black is just fine.

10.Rb1 cxd4 and 10...0-0 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗e3 c5 8. ♘f3 ♕a5 9. ♕d2 ♘c6 10. ♖b1 I consider this the most problematic move for Black. 10... cxd4 10... O-O is also possible. 11. ♖b5 cxd4 12. ♖xa5 dxe3 13. ♕xe3♘xa5 I once played this myself against Rustem Dautov, one of the godfathers of the Be3 line. I made a reasonably comfortable draw, but I'm no longer sure Black actually equalises here so I can't really recommend this. 11. cxd4 ♕xd2+ 11... O-O doesn't promise full equality either: a) 12. ♕xa5 ♘xa5 a1) 13. ♗e2 b6 14. ♗d2 ♖d8 15. d5? e6 16. ♗xa5 bxa5 17.♗c4 exd5 18. ♗xd5 ♗a6−+ Bosiocic-Volokitin, AUT-tch 2013 a2) 13. ♗d2 b6 14. h3 ( 14. ♗b5 ♗g4 15. ♗b4 a6 16. ♗a4♗xf3 17. gxf3 ♗xd4 18. ♗xe7 Khenkin-Stohl, Kaskady 200218... ♖fc8=/+ ; 14. ♗xa5 bxa5 15. ♗c4 ♗g4 16. ♘e5 ♗xe517. dxe5 ♖ac8 18. ♗d5 ♖c5 19. O-O e6 20. ♗b3 ♗e2 21.♖fe1 ♗b5 22. f4 ♖d8=/+ Sedlak-Sutovsky, Aix-les-Bains 2011)14... ♘b7 15. ♗b4 ♖d8 16. ♗b5 ♘d6 17. ♗c6 ♗b7 18.♗xb7 ♘xb7 19. O-O ♗xd4 20. ♗xe7 ♖e8 21. ♘xd4 ♖xe722. f3 ♘c5= Seirawan-Cseshkovsky, Yugoslavia 1992 a3) 13. ♗d3 13... ♗g4 transposes to 12.Bd3. ( 13... b6 14.♔e2 ♗b7 15. ♖hc1 ♖fc8 16. d5 f5 ; 13... ♗d7 14. ♔e2♖fc8 15. ♖hc1+/= ) b) 12. ♗d3 12... ♗g4 ( 12... ♕xd2+ 13. ♔xd2 e6 14. ♖hc1♖d8 15. e5 f6 16. ♔e2 fxe5 17. dxe5 ♘xe5 18. ♘xe5 ♗xe519. ♗e4 ♖b8 20. ♖c5 ♗d6 21. ♖a5 a6 22. ♗b6 ♖d7 23.h4+/= Kozul-Polajzer, Ptuj 1989 ) 13. ♕xa5 ♘xa5 14. OO ( 14.♘d2⁉ ) 14... ♖fc8 ( 14... ♖fd8 15. d5 e6 16. ♗g5 f6 17. ♗d2b6 18. ♗xa5 bxa5 19. ♘d4± ; 14... ♗xf3 15. gxf3 e6 16.♖fd1 ♖fd8 17. ♗f1 ♖ac8 18. d5+/= ) 15. ♖fc1 ( 15. d5 ♖c3 16. ♖fd1 a6 17. ♖b6 ♖ac8 18. h3 ♗xf3 19. gxf3 ♗e5 ) 15...♖xc1+ 16. ♖xc1 ♖c8 17. ♖xc8+ ♗xc8 18. d5 b6 19. ♘d4+/= 12. ♔xd2 O-O This was played against me by Jon Ludvig Hammer in Stavanger 2013. I was even worse by move 17 because I completely misplayed it. 13. d5 13. ♗d3 ♖d8 14. d5 ♘a5 transposes.

13. ♗b5 f5 14. exf5 ♗xf5 15. ♗d3 ♖ad8 16. ♖xb7 ♘xd4 17.♘xd4 ♗xd4 18. ♗xf5 ♖xf5 19. ♖hb1 ♔f7 20. ♔e2 ♗xe3 21.♔xe3 ♖e5+ 22. ♔f4 ♖e2 23. ♖1b2 ♖d4+ 24. ♔f3 ♖xb2 25.♖xb2 ♖a4 Oral-Babula, CZE 1994 13... ♖d8 14. ♗d3 ♘a5 15. ♔e2! The most natural move in the position. After this I don't think Black completely equalises in any of the lines, so this also isn't something I can recommend. I immediately went wrong: 15. ♗g5 f6 16. ♗f4 b6 a) 17. ♗c7⁉ ♖d7 18. ♖hc1? ♗h6+ 19. ♔e2 ♗xc1 20. ♖xc1 e6!=/+ b) 17. ♗b5 ♗b7! ( 17... e5 18. ♗e3 ♗d7 19. ♖hc1 ♗xb5 20.♖xb5 ♘b7 21. ♔e2 ♘d6= SvidlerHammer, Stavanger 2013 ) 18.♔e2 f5=/+ c) 17. ♖hc1 e6 d) 17. ♔e2⁉ 17... f5 ( 17... ♘b7⁉ ) 18. exf5 gxf5⁉ ( 18... ♗xf519. ♗xf5 gxf5 20. ♖hd1+/= ) 19. ♖hd1 ♖xd5 ( 19... e5⁉ ) 20.♗c4 ♘xc4 21. ♖xd5 ♗a6⇆

15. ♗f4 b6 16. ♖hc1 f5 17. ♖c7 fxe4 18. ♗xe4 ♗b7 19. ♖b5♖dc8 20. ♖d7 ( 20. ♖xe7 ♖e8 21. ♖xe8+ ♖xe8 22. ♗d3 ♖d8=) 20... ♖d8= 15... f5 15... e6⁈ 16. ♗g5! f6 ( 16... ♖e8 17. d6 ) 17. ♗d2± b6 18. ♗xa5bxa5 19. ♘d4⁉ ( 19. ♗c4 exd5 20. ♗xd5+ ♔h8 21. ♗xa8 ♗a6+22. ♔e1 ♖xa8+/= ) 15... b6 a) 16. ♗g5 f5! has been known to be quite good for Black since Damljanovic-Kozul, Yugoslavia 1989 b) 16. ♖hc1 e6 17. ♗g5 f6 18. ♗d2 exd5 b1) 19. ♗xa5 dxe4 20. ♗xe4 ( 20. ♗c4+ ♔h8 ) 20... ♗a6+ b2) 19. exd5 19... ♖xd5 20. ♗e4 ♗a6+ 21. ♔d1 ♖xd2+ 22.♘xd2 ♖d8=/∞ c) 16. ♘d4! 16... f5 ( 16... e6 17. ♘c6 ♘xc6 18. dxc6 ♖xd319. ♔xd3 ♗a6+ 20. ♔c2 ♖c8 21. ♖hc1 ♗c4 22. ♔d2 ♖xc623. ♔e1+/= Sakaev-Hoffmann, Groningen 1991 ) 17. ♘e6 ( 17. f3fxe4 18. fxe4 ♗g4+ 19. ♔d2 e6 20. ♘xe6 ♗xe6 21. dxe6♖d6 22. ♔e2 ♖xe6 23. ♖bc1 ♗e5 24. g3 ♔g7+/= Volzhin-Urban, Katowice 1992; 17. ♘c6 ♘xc6 18. dxc6 ) 17... fxe4 18.♗xe4 ♗a6+ 19. ♔f3 ♖d7 20. ♖hd1 ♗f6 21. ♖bc1+/= 16. ♖hc1 16. ♘g5 fxe4 17. ♗xe4 b6 18. ♖hc1 ♗a6+ 19. ♔e1 ♖ac8 20.♘e6+/= 16... b6

16... fxe4 17. ♗xe4 ♗f5 18. ♗xf5 gxf5 19. ♖b5 b6 a) 20. ♖c7 ♗f6 ( 20... ♖dc8 21. ♖xe7 ♖c2+ 22. ♔f1 ♖xa223. d6± ) 21. ♗d4+/= b) 20. ♘h4 20... ♖f8 Sherbakov-Yandemirov, USSR 1991 ( 20... a621. ♖xb6 ♖xd5 22. ♖c5 ♖xc5 23. ♗xc5 ♖c8 24. ♗xe7 ♖e825. ♘xf5 OvsejevitschAlexeev, Alushta 2000 ) 21. ♗g5!+/= ( 21.♗f4 ♘b7 22. ♖c7 ♘d6 23. ♗xd6 exd6+/= ; 21. d6 e6 22.♗g5 ♗f6 23. ♗xf6 ♖xf6 24. ♖c7 ♖d8 25. ♖xa7 ♖xd6 26.♘f3 ♘c6+/= Sherbakov,R ) 17. ♖c7 17. ♘g5⁉ is, as usual in this line, a very worthy alternative. 17... fxe4 18. ♗xe4 ♗a6+ 18... ♗b7 19. ♖xe7 ♗xd5 ( 19... ♔f8 20. ♖c7 ♗xd5 21. ♗xd5♖xd5 22. ♖b4± ) 20. ♖e8+⁉ ( 20. ♗xd5+ ♖xd5 21. ♖c1+/= ) 20...♔f7 ( 20... ♖xe8 21. ♗xd5+ ♔f8 22. ♗xa8 ♖xa8 23. ♘g5 ♔g824. ♖c1+/= ) 21. ♖xd8 ( 21. ♗xd5+ ♔xe8 22. ♗xa8 ♖xa8 23.♖c1 ♖d8 24. ♘g5± ) 21... ♖xd8 22. ♘g5+ ♔e7 ( 22... ♔g8 23.♗xd5+ ♖xd5 24. ♖c1 ♗e5 25. f4± ) 23. ♗xd5 ♖xd5 24.♘xh7+/= Korchnoi-Brunner, Bern 1996 19. ♔e1 ♗b7 20. ♖bc1 20. ♖xe7 ♗xd5 21. ♖e8+ ♔f7 22. ♖xd8 ♖xd8 23. ♘g5+ ♔e724. ♗xd5 ♖xd5 25. ♘xh7 ♖h5= 20... ♗f6 20... ♗xd5 21. ♗xd5+ ♖xd5 22. ♖xe7 ♗f8 23. ♖e4 ♗d6 ( 23...♖ad8 24. ♗g5 ) 24. ♔e2 ♘b7= 21. ♘g5 ♗xd5 22. h4 b5 23. ♗xa7 ♘c4 24. ♖d1 ♗xe4 1/2 Tukmakov-Romanishin, URS-ch49 Frunze 1981

10.Rb1 a6 11.Rc1 Bg4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗e3 c5 8. ♘f3 ♕a5 9. ♕d2 ♘c6 10. ♖b1 a6Protecting against Rb5. 11. ♖c1 ♗g4⁉ This is recommended by Boris Avrukh in his excellent book on the Grünfeld, when the following line is very likely to take place: 11... cxd4 12. cxd4 ♕xd2+ 13. ♔xd2 e6 ( 13... O-O 14. d5 ♖d815. ♔e1 illustrates the idea behind White's strange rook shuffle with 11.Rc1. 15... ♘a5 is now completely impossible because 16. ♗b6 wins a piece. ) 14. d5 ( 14. ♗d3 O-O 15. ♖c4 ♗d7 16. ♖hc1 ♖fd8 17.♗g5 f6 18. ♗e3 ♗e8 19. ♔e2 ♖ac8 1/2 Karjakin-Nepomniachtchi, Moscow ch-RSCU Final m 2011 ) 14... exd5 15. exd5 ♘e7 16. ♗c4♗g4 ( 16... OO 17. ♗c5 ♗h6+ 18. ♔c3 ♗xc1 19. ♖xc1 ♖e8 20.♖e1 1-0 Holzke-Ftacnik, BL 11-12 2012; 16... ♘f5 17. ♗c5 ♗h6+18. ♔c3 ♗xc1 19. ♖xc1 f6 20. ♖e1+ ♔d7 21. ♖e6 b5 22. ♗b3 a5 23. g4+− Aronian-Nepomniachtchi, Moscow 2010 ) 12. d5 ♖d8 13. ♗e2 The only move that can cause Black any problems.

13. ♗d3 Black can completely ignore the fact that the knight is en prise and play 13... OO 14. OO ♗xf3 15. gxf3 e6 ( 15... ♘e5 16. ♗e2f5 17. f4 ♘g4 18. e5 ♘xe3 19. fxe3+/= ) 16. dxc6 ( 16. c4 ♘b417. ♖b1 ♖c8 18. ♖b3 b5 19. cxb5 axb5 20. a3 ♕a4 21. axb4♕xb3 22. bxc5 exd5 23. ♖b1 ♕a3 24. exd5 ♖xc5= ) 16... c4 This position is known to be good for Black. 17. cxb7 ♖xd3 18. ♕b2 ♖b819. ♖b1 ( 19. ♗a7 ♕g5+ 20. ♔h1 ♗e5 21. ♗xb8 ♗xb8 22. ♖g1 ♕h5= ) 19... ♗xc3 20. ♕b6 ♕xb6 21. ♖xb6 ♗d4 22. ♗xd4♖xd4 23. ♖c1 ♔f8 24. ♔g2 ♔e7 25. ♔g3 ♔d7= 13. ♘h4 O-O 14. h3 ♗c8 15. ♗e2 e6 16. O-O exd5 17. exd5 c418. ♖fd1 ♘e5 19. ♘f3 ♘d3=/+ KhenkinFernandez Garcia, Las Palmas 1994 13... O-O 14. O-O 14. ♕b2 ♗xf3 15. ♗xf3 ♘d4 16. ♗d1 ♘b5 17. ♗d2 ♘d6 18.♕c2 ♘c4∞ Ehlvest-Rytshagov, Tallinn 1998 14... ♗xf3 15. ♗xf3 e6 This is what Avrukh and others have been suggesting, and currently this position is assessed as completely fine for Black, but as a matter of fact after the very precise 16. ♗e2! , stopping Black's main objective of playing ...Ne5 and putting the pawn on c4, I think this position is far from completely equal. Black's life could become quite difficult. 16. ♖fd1 exd5 17. exd5 ♘e5 18. d6 c4 16... ♖d7 16... exd5 17. exd5 ♘e7 18. c4 ♕xd2 ( 18... ♕c7 19. ♗f4 ♗e520. ♗xe5 ♕xe5 21. ♖fe1 ♕c7 22. ♗d3 ♘c8 23. h4 ♖de8 24.h5 ) 19. ♗xd2 ♘f5 ( 19... ♖fe8 20. ♖fe1 ♘f5 21. ♗d3 ♖xe1+22. ♖xe1 ♘d6 23. a4 ) 20. ♗d3 ♘d6 21. ♗e3 b6 22. a4 a5 23.♖b1 ♖b8 24. ♖b3+/= 16... ♘e7 17. c4 ♕xd2 18. ♗xd2 ♘c6 19. ♗e3 ♘d4 20. ♗d3+/= 17. ♖fe1 exd5 18. exd5 ♖fd8 19. ♗g5!+/= 19. c4 ♕xd2 20. ♗xd2 ♘d4 21. ♗d1 ♖c8 22. a4 b5 23. cxb5axb5 24. axb5 ♖xd5 25. b6 ♗f8 26. ♗c3 ♖b8 27. ♖b1 ♖d6 28.b7 ♖d7 29. ♗xd4 cxd4 30. ♗f3+/= Bosiocic-Erdos, Sibenik Mitropa Cup 2012

10.Rb1 a6 11.Rc1 f5 12.d5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗e3 c5 8. ♘f3 ♕a5 9. ♕d2 ♘c6 10. ♖b1 a6 11.♖c1 f5⁉ This move leads to very unclear and incredibly unbalanced positions where Black requires a lot of precision, but nonetheless I think they're very playable and I'll try to provide a complete overview of this position so you can feel comfortable playing it. White has a lot of options, the main two of which are 12.e5 and 12.Bd3, but let's first look at other moves. 12. d5

12. exf5 is another option for White. 12... ♗xf5 13. dxc5⁉ Black has a choice between two decent moves. ( 13. d5 O-O-O 14. ♗e2 e6 15.d6 ♖d7 16. O-O ♗f8 17. ♗f4 ♗xd6 18. ♗xd6 ♖hd8=/+ ) a) 13... ♗e6⁉ is very possible, attacking the a2-pawn straightaway. a1) 14. ♘d4 ♖d8 ( 14... ♘xd4 15. cxd4 ♕xa2 16. ♕xa2♗xa2 17. ♗c4 ♗xc4 18. ♖xc4 OOO 19. f4+/= ; 14... ♗xd415. cxd4 ♕xd2+ 16. ♔xd2 ♗xa2 17. ♔c3+/= ) 15. ♘xc6 bxc6 16. ♕c2 OO 17. ♗e2 ♗d4 ( 17... ♕xa2 18. ♕xa2♗xa2 19. ♗xa6+/= ) 18. ♗xd4 ♖xd4 19. O-O ♖df4= a2) 14. ♘g5 ♖d8 15. ♕c2 ♗d5=/+ a3) 14. ♕b2 14... O-O a31) 15. ♗e2 ♕xa2 16. ♕xb7 ♗d5 17. c4 ♖ab8 18. ♕d7♖fd8 19. ♕g4 ( 19. ♕h3⁈ ♗f7 ) 19... ♗f7 20. ♕e4 ( 20.h4 ♖b2−+ ) 20... ♘a5 ( 20... ♗b2 21. O-O! ♗xc1 22.♗xc1=/∞ ; 20... ♖b2 21. O-O ♖xe2 22. ♕xc6+/= ) 21. c6♘b3 22. O-O ♕xe2 23. c7 ♘xc1 24. cxb8=Q ♖xb8 25.♖xc1 ♖d8= a32) 15. a3! 15... ♖ad8 ( 15... ♖xf3 16. gxf3 ♗d5 17.♗e2 ♘e5 18. OO ♘xf3+ 19. ♗xf3 ♗xf3 20. ♕b3+ ♔h821. ♕e6± ) 16. ♗e2 ♘d4 17. ♘xd4 ♗xd4 18. ♗xd4♖xd4 a321) 19. OO ♖a4 20. ♖ce1 ( 20. ♕xb7 ♕xc5transposes. ) 20... ♖xa3 21. ♗f3 ♖b3 22. ♕e2 ♖f6 23.♕e5 ♕xc3= a322) 19. ♕xb7 19... ♕xc5 20. OO ♖a4 21. ♗xa6♖xa3 22. ♗e2 ♖xc3 23. ♖xc3 ♕xc3 24. ♕xe7 ♕f6 , and Black will hold. b) 13... ♕a3! The more forcing option. 14. ♗c4 Perhaps the most critical move, but Black's response is the same as after 14.Be2. ( 14.♗e2 The most obvious move for White, but Black is ready to meet it with 14... ♖d8 15. ♘d4 ♘xd4 16. cxd4 ♗xd4! 17. ♗xd4♖xd4 18. ♕xd4 ♕xc1+ ) 14... ♖d8 15. ♘d4 ♘xd4 16. cxd4♗xd4 , and I think Black holds. 17. ♗xd4 ♖xd4 18. ♕xd4♕xc1+ 19. ♔e2 ♕xh1 20. ♕xh8+ ♔d7 12. dxc5⁉ has also been seen in a number of grandmaster games. 12...fxe4 13. ♘d4 ( 13. ♘g5 ♗f5=/+ ) a) 13... ♘e5 14. ♗e2 ♘d3+ 15. ♗xd3 exd3 16. O-O a1) 16... e5 17. ♕xd3! ♕xa2 ( 17... OO 18. ♕c4+ ; 17...exd4 18. ♕e4+ ♔f8 19. ♕f3+ ♗f5 20. cxd4± ) 18. ♕e4O-O 19. ♘f3 ♕f7 20. ♘g5 ♗f5 21. ♕a4 ♕e7 22. ♖fd1± a2) O-O 17. ♖fd1!+/= b) 13... ♗d7 14. ♗e2 By far the most critical move. ( 14. ♗c4♘e5 ; 14. ♘b3 ♕a4! ; 14. ♖b1 O-O-O⇆ ; 14. ♘xc6 ♗xc6 15.♗c4 ♖d8= ) b1) 14... O-O 15. O-O+/= ♖ad8 16. ♕b2!

b2) 14... O-O-O is what used to be played here, but after 15.♘b3! ( 15. OO ♗xd4! 16. ♗xd4 ♘xd4 17. ♕xd4 ♗c6 18.♕e5 ♕c7= ; 15. ♖d1 ♘xd4 16. cxd4 ♕xd2+ 17. ♖xd2♗c6= ) 15... ♕c7 ( 15... ♕a4 is also possible. 16. OO+/= Giri-Kanarek ) 16. ♕c2 ♘e5 Ding-Zhao Zingchao ( 16... ♗e5 17.♕xe4 ) 17. O-O ♘d3 18. ♖b1 Black's position is quite dangerous. This can't be recommended. b3) 14... ♖d8! I think this is best for Black. b31) 15. ♘b3 ♕a4⇆

b32) 15. ♖d1 O-O⁉⇆ ( 15... ♕xc5 16. ♘e6 ♗xe6 17.♗xc5 ) b33) 15. h4 ♗e6∞ b34) 15. ♘xc6 ♗xc6 16. ♕b2 ♗d4!= b35) 15. O-O 15... ♘xd4 16. ♗xd4 ♗g4! A very important finesse. 17. ♗xg4 ♗xd4 Black holds quite comfortably because he gets a position with opposite-coloured bishops where the somewhat inferior structure is compensated by White's weaknesses on a2 and f2. ( 17... ♖xd4? 18. ♕b2± ) b351) 18. ♕e2 ♗xc5 19. ♕xe4 O-O= b352) 18. ♖fd1 b3521) 18... ♗xc5 19. ♗d7+ ♔f8 20. ♕f4+ ( 20.♕h6+ ♔f7 21. ♕f4+ ♔g7 ) 20... ♔g7 21. ♕e5+♔h6 22. ♔h1 ( 22. ♖d5 ♕b6 ) 22... ♕b6= b3522) e3⁉ 19. ♗d7+ ( 19. fxe3 ♗xc5 ) 19... ♖xd720. fxe3 ♗xe3+ 21. ♕xe3 ♖xd1+ 22. ♖xd1 ♕c7= b353) 18. ♕g5 ♕xc5= b354) 18. ♕b2 ♗xc5 19. ♕xb7 O-O 20. ♕xe4 ♖d2⇆

b355) 18. c6 18... bxc6 19. ♕e2 ♗f6 20. ♕xe4 ♖d621. ♖fd1 ♕e5= 22. ♕c4 ♕b5 23. ♕f4 ♖xd1+ 24.♖xd1 O-O 12... fxe4 13. ♘g5 13. dxc6 exf3 14. gxf3 bxc6 15. ♗c4 ♗h3!∓

13... ♘e5 14. ♘xe4

14. c4 ♕xd2+ 15. ♔xd2 ♗f5 16. ♗xc5 h6∓

14... c4 These lines involve a lot of very beautiful variations and we're following a fantastic game played by Malakhatko and Sutovsky a few years ago. 15. f4

15. d6 ♗f5 16. ♘c5 ♖c8∓

15. ♘g5 ♗f5 16. f4 h6 17. fxe5 hxg5 18. ♗xc4 ♗xe5∓

15. ♘g3 O-O 16. f4 ( 16. ♗e2 ♖d8∓ ) 16... ♘d3+ 17. ♗xd3cxd3 18. ♕xd3 ♖d8 19. O-O ♕xd5=/+ 15. ♗e2 ♗f5

a) 16. ♘g3 ♘d3+ 17. ♗xd3 ♗xd3−+ b) 16. ♘c5 ♖c8 17. ♘e6 ( 17. f4 ♘d3+∓ ) 17... ♗xe6 18.dxe6 ♖d8 19. ♕b2 b5 20. OO ♘d3 21. ♗xd3 ♖xd3∓ c) 16. f3 16... OOO! 17. ♘c5 b6 18. f4 ( 18. g4 ♗d3 19. f4bxc5 20. fxe5 ♖xd5∓ ) 18... ♘g4 19. ♗xg4 ♗xg4 20. h3 ♗f521. g4 bxc5 22. gxf5 gxf5 23. O-O e6 24. d6 ♖hg8 25. ♔h2♗f8∓

15... ♗f5 16. ♘c5! White's best move.

16. ♘f2 O-O-O! 17. fxe5 ♖xd5∓ 18. ♗d4 White seemingly has everything under control and probably thought that was the case until he saw Emil's reply: 18... h5‼ Now the huge threat of ...Bh6 changes the evaluation completely. There's no defence and White's entire position just collapses. White was already almost lost here in Malakhatko-Sutovsky, 2009. 16. fxe5 ♗xe4 17. ♗xc4 is very good for Black: ( 17. d6 exd6 18.exd6 ♕d5∓ ) 17... ♖f8! ( 17... ♕c7⁉ ) 18. d6 or( 18. ♖f1 ♖xf1+19. ♔xf1 ♗xe5 20. ♗b3 ♖c8∓ ; 18. ♕d4 ♗xg2 19. ♖g1 ♗f3∓ )18... ♖d8! ( 18... exd6 19. exd6 OOO ) 19. ♗d4 exd6 20. ♕e3d5→ 16... ♘g4! (c) Rustam Kasimdzhanov. 16... O-O-O? 17. fxe5+− 16... ♘d3+ is the move I was ready to recommend to you and I fully believe White is fine after it, but then I spoke to my old friend and colleague Rustam Kasimdzhanov who suggested I would be foolish not to show you the fantastic 16...Ng4! 17. ♗d4 Once again, this line is the product of Rustam's imagination not mine and I'm very indebted to him for showing it to me first. Now you're seeing it courtesy of him. 17. ♘xb7 ♕b5 18. ♘c5 ♖c8 17... ♗xd4 18. ♕xd4 And now despite the fact that the rook on h8 is hanging with check Black plays 18... ♕a3‼ This position is a draw. 19. ♕xh8+ 19. ♖d1 is possible, but would only be played by a spoilsport, as the subsequent play becomes somewhat boring: 19... ♘f6 20. ♗xc4 ♖c8After the possible 21. ♘d3 ♗xd3 22. ♖xd3 b5 23. ♗b3 ♕c1+ 24.♖d1 ♕xc3+= we end up in an equal endgame which obviously isn't the reason I was showing you this.

19... ♔f7 20. ♕xh7+ ♔f8 White is a whole rook up but he's not better at all because ...Qxc1+ is such a huge threat that White should in fact give an immediate perpetual. 21. ♕h8+ 21. ♘e6+ ♔e8 22. ♕h8+ ♔d7 might not even give White a draw as the king escapes all the checks. ...Qxc1+ is a threat of such immense proportions that Black is already playing for a win. 21... ♔f7=

10.Rb1 a6 11.Rc1 f5 12.e5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗e3 c5 8. ♘f3 ♕a5 9. ♕d2 ♘c6 10. ♖b1 a6 11.♖c1 f5⁉ 12. e5 I think Black should be able to make a more or less forced draw after this move. 12... f4⁉ Sacrificing a pawn to start immediate pressure against the white centre. 12... b5 is a very adventurous move once played by Michal Krasenkow. It's not so easy to refute, but I still think it's not quite sound. 13. ♗xf4 ♗g4 Practice currently says that White is better in this position, but I think Black's play in a very critical game can be improved upon. 14. ♗e2 Other moves aren't threatening at all. 14. ♕b2⁈ is met by 14... ♗xf3! 15. gxf3 ( 15. ♕xb7 ♖b8−+ ) 15...♘xd4 16. ♕xb7 O-O and Black is already fighting for the initiative. 17.♗c4+ ♔h8 18. O-O ♖ab8 19. ♕e4 ♘xf3+ 20. ♕xf3 ♗h6→ 14. d5 ♗xf3 15. gxf3 ♘xe5 Black is perfectly comfortable. 16. ♕e3♕a3! ( 16... ♕c7⁉ 17. ♗e2 OOO⇆ ) 17. ♗e2 ( 17. ♖c2 ♘xf3+‼18. ♕xf3 ♕a4 19. ♗d3 OO 20. c4 ♗h6∓ ; 17. ♖d1 ♘xf3+! 18.♕xf3 ♗xc3+ 19. ♔e2 OO 20. ♕e3 ♕xa2+−+ ) 17... OO 18. O-O♘f7 19. ♖c2 ♕a4=/+ 14. ♘g5 cxd4 ( 14... h6 15. ♘e4 cxd4 16. cxd4 ♕xd2+ 17. ♗xd2♘xd4 18. f4 g5 19. ♖c7 ♖f8⇆ ) 15. cxd4 ♕xd2+ 16. ♗xd2♘xd4 17. ♗c4 h6 18. ♗e3 ( 18. ♘e4 b5→ ) 18... ♖d8 19. ♗xd4♖xd4= 14. ♕e3 ♗xf3 15. ♕xf3 cxd4 16. ♗c4 ♖f8! 17. ♕e4 ♖xf4 18.♕xf4 ♘xe5 19. ♗e2 d3 20. ♗f3 ♘xf3+ 21. ♕xf3 d2+−+ 14. ♗e3 ♖d8 ( 14... ♗xf3 15. gxf3 ♖d8 16. f4+/= ) 15. ♗e2O-O⇆

14... ♖d8!

14... O-O is less precise: a) 15. d5 ♗xf3 16. ♗xf3 ♘xe5= b) 15. OO cxd4 16. ♗c4+ ( 16. cxd4 ♕xd2 17. ♗xd2 ♗xf3 18.♗xf3 ♘xd4= ) 16... ♔h8 17. ♘xd4 ( 17. ♘g5 ♘xe5 18. ♗xe5♗xe5 19. ♘f7+ ♖xf7 20. ♗xf7 dxc3=/+ ) 17... ♘xe5 18. ♗b3♘d3 19. ♕xd3 ♖xf4=

c) 15. dxc5! 15... ♕xc5 ( 15... ♖ad8 16. ♕e3 ♘b4 17. ♗c4+♘d5 18. OO+/= ) 16. OO ♖ad8 17. ♕e3 ♕xe3 18. fxe3 b5 (18... e6 19. ♖b1 b5 20. c4+/= ) 19. c4+/= 15. ♘g5⁉ The main move, and really the only move that creates any problems for Black. 15. ♗h6 ♗xh6 16. ♕xh6 cxd4 17. O-O d3 18. ♗d1 ♗xf3 19.♗xf3 ♕xe5=/+ 15. d5 ♗xf3 16. ♗xf3 ♘xe5 17. ♗xe5 ♗xe5 , and Black is fine. 15. OO cxd4 16. cxd4 ( 16. h3 ♗xf3 17. ♗xf3 d3 18. ♗xc6+bxc6= ) 16... ♕xd2 17. ♗xd2 ♗xf3 18. ♗xf3 ♘xd4 19. ♗d1♗xe5 20. ♖e1 ♗f6 21. ♗b4 ♖d7=/+ 15. h3 can be met by 15... ♗xf3 16. ♗xf3 and besides 16...cxd4 you can and perhaps should start with 16... O-O 17. ♗xc6 bxc6 , maintaining the pressure in the centre for now. a) 18. OO cxd4 19. cxd4 ♕xd2 20. ♗xd2 ♖xd4 21. ♗g5 ( 21.♗e3 ♖a4 22. f4 ♖xa2 23. ♖xc6 a5⇆ ) 21... ♗xe5 22. ♖xc6♗f6 23. ♗h6 ♖a8=

b) 18. g3 Trying to support the bishop and keep Black from breaking through. 18... ♖f5⁉ Black can afford this extra tempo on a waiting move. He now wants to play ...Bxe5. ( 18... cxd4 19. cxd4 ♕xd2+20. ♗xd2 ♖xd4 21. f4 ♖e4+ 22. ♔d1+/= ) 19. O-O cxd4 Only now. 20. cxd4 ♕xd2 21. ♗xd2 ♗xe5= This is an immediate draw. 15. ♗g5 ♖f8⁉ , and Black is ready to start collecting. 15. ♖d1 a) 15... O-O 16. O-O cxd4 17. cxd4 ♕a4!⇆ ( 17... ♕xd2 18.♗xd2+/= )

b) ♗xf3 16. ♗xf3 ♘xe5 17. ♗xe5 ♗xe5 18. O-O cxd4 19.cxd4 ♕xd2 20. ♖xd2 ♖xd4= 15... ♗f5⁉ This is my recommendation. The white centre is about to collapse as there's really no comfortable way to protect it. 15... ♗xe2 was played in the one game that reached this position. 16.♕xe2 a) 16... ♘xd4? 17. ♕c4± b) 16... h6 17. ♘e6 ♘xd4 18. ♕c4± Black is a very long way from equality. c) 16... cxd4 c1) 17. O-O d3 18. ♕e4 ( 18. ♕b2 ♕d5!→ ) 18... ♕d5∓ c2) 17. ♘e6 17... d3

c21) 18. ♕d2 ♔f7 19. ♘xd8+ ♖xd8 20. e6+ ( 20. OO♘xe5=/∞ ) 20... ♔xe6 21. O-O ♔f7 22. ♖fe1 ♕d5=/∞ c22) 18. ♕e4 18... ♔f7 ( 18... ♗xe5 19. ♘xd8± ) 19.♘xd8+ ♖xd8 20. e6+ ♔g8 21. O-O d2 22. ♖b1 ♗xc3 (22... ♕xc3 23. ♗e3+/= ) 23. ♖fd1±

d) ♖f8⁉ A very interesting and strong idea, as otherwise Black is in trouble. I'd even describe this as a "witty" way to solve Black's problems. It looks as though you're allowing a super fork. d1) 17. ♘e6 is met by 17... ♖xf4! 18. ♘xf4 ( 18. ♘xg7+♔f7∓ ) 18... ♗h6 19. g3 cxd4 20. O-O dxc3⇆ , and in this position the pawn on c3 is actually quite strong because it will be supported by the knight on d4. Black is perfectly fine, which I think is what the black player was hoping for. d2) 17. ♗d2! A much better option than the obvious fork on e6, and one played by Salem against Daulyte in the Dubai Open in 2011. I failed to find equality here for Black. d21) 17... cxd4 18. ♘e6 d3 19. ♕e3 Black will be an exchange down and this is far from equal. 19... ♗xe5 ( 19...♔f7 20. ♘xf8 ♔xf8 21. f4 ♕xa2 22. OO e6 23. ♖cd1 )20. OO ♖d7 21. ♘xf8 ♔xf8 22. ♕h3 ♖d6 23. ♕xh7♕xa2 24. ♖cd1 ♕c4 25. ♔h1 , and White was better in Salem-Daulyte, Dubai 13th op 2011 d22) 17... ♕xa2 18. ♘xh7 ♖h8 19. ♘g5 (There's no need for 19. ♘f6+ exf6 20. exf6+ ♔f7 21. fxg7 ♖he8 22. ♗e3♕xe2+ 23. ♔xe2 cxd4 24. cxd4 ♘xd4+ 25. ♔f1 ♘f5 26.♗g5 ♖d7 27. g4 ♘xg7= ) 19... cxd4 20. f4 The bishop on g7 is such a horrible piece that White is much better here. 20...♗h6 ( 20... dxc3 21. ♗xc3 ♕xe2+ 22. ♔xe2+/= ) 21. ♘f3dxc3 ( 21... ♕f7 22. g3 d3 23. ♕e4 ♕f5 24. ♕xf5 gxf525. ♗e3± ) 22. ♖xc3+/= 16. e6 White's one option of note apart from playing Nf3 Bg4 with a repetition. 16. d5 ♘xe5= 16... cxd4 17. cxd4 ♕xd2+ 18. ♗xd2 ♗h6! The last important move. Black is fine here because the white centre is about to collapse. Both d4 and e6 fall in most variations and a draw is by far the most likely result. 19. h4 19. ♘f7 ♗xd2+ 20. ♔xd2 ♖xd4+ 19. ♗e3 ♗xg5 20. ♗xg5 h6 21. ♗e3 ♘xd4 19... ♘xd4 20. ♖c7 20. ♗a5 ♖b8 20... O-O 21. ♖xe7 ♗xe6 22. ♘xe6 ♗xd2+ 23. ♔xd2 ♘xe6+ 24.♔e3 ♖fe8= and a draw isn't far away.

10.Rb1 a6 11.Rc1 f5 12.Bd3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗e3 c5 8. ♘f3 ♕a5 9. ♕d2 ♘c6 10. ♖b1 a6 11.♖c1 f5⁉ 12. ♗d3 The most critical move, and currently perhaps Black's biggest problem in the whole 11...f5 variation. 12... cxd4 13. ♘xd4 13. cxd4 ♕xd2+ 14. ♔xd2 ♖f8! is fine: 15. exf5 ♗xf5 16. ♖c5 (16. ♘g5 ♘xd4 17. ♘xh7 ♖h8 18. ♗xf5 gxf5 19. ♘g5 ♖d8∞ )16... ♖d8 17. ♗xa6 ♘xd4 ( 17... bxa6 18. ♖xc6 ♗e4 19. ♖xa6♗xf3 20. gxf3 ♖xf3 21. ♖a4 ♖h3 22. ♔c2+/= ) 18. ♗xd4 bxa619. ♔c3 ♗xd4+ ( 19... ♔f7⁉ ) 20. ♘xd4 ♗d7 21. f3 ♖f6=Jussupow-Kasparov, Reykjavik 1988 13... ♘e5 My recommendation. 13... O-O is Black's normal response. 14. ♘xc6 ( 14. O-O⁈ ♘xd4 )14... bxc6 a) 15. OO fxe4 ( 15... ♗e6 16. exf5 ♗xf5 17. ♗c4+ ♔h8 18.♗d4 e5 19. ♗e3 ♖fd8 20. ♕e2+/= Rodshtein-Sutovsky, Natanya Richard Riordan GM Rapid 2009 ) 16. ♗xe4 ♗e6 17. ♗xc6 a1) 17... ♖ac8 18. ♗d7+/= a2) 17... ♖fd8⁉ 18. ♕e2 ( 18. ♕e1 ♖ac8 19. ♗b7 ♗c4 20.♗xc8 ♖xc8 21. h3 ♗xf1 22. ♕xf1 ♖xc3 23. ♖xc3 ♗xc324. ♕c4+ ♔g7 25. a4 ♗e5= ) 18... ♖ac8 19. ♗b7 ♗c4 20.♕g4 ♖c7 21. ♖fd1 ♖xd1+ 22. ♖xd1 ♗xa2 23. h3 ♔f7 24.♗xa6 ♖xc3= a3) ♖ad8 a31) 18. ♕e2 ♖c8 ( 18... ♕xa2 19. ♕xa2 ♗xa2 20. c4e6 21. ♗a4 ♖d3 ) 19. ♗b7 ( 19. ♗b6 ♕xa2 ) 19... ♗c420. ♕b2 ( 20. ♕g4 ♖c7 21. ♖fd1 ♕xa2 22. ♕e4 ♗f7= )20... ♖c7 21. ♖fd1 ♕xa2 22. ♕b6 ♗e5 23. ♗e4 ♕b3 24. g3 ♖b8 25. ♕a5 ♕b5= a32) 18. ♕b2 ♖c8= a33) 18. ♕c2 18... ♗c4=/∞ b) 15. ♗c4+ It's important for White to occupy this diagonal. 15...♔h8 16. exf5! ♗xf5 17. OO ♖ad8 18. ♕e2+/= I think this position is very far from equal. It's playable, and I'm not claiming White has a decisive advantage, but it's very, very uncomfortable for Black as his king will forever be weaker than the white king and he also has structural problems. 14. O-O 14. exf5 ♘xd3+ 15. ♕xd3 ♗xf5 16. ♘xf5 ( 16. ♕c4 ♖c8 17.♕b3 ♗d3 ) 16... ♕xf5= 14... ♘xd3 My proposed improvement.

14... O-O isn't entirely satisfactory: 15. exf5 ( 15. ♗c2 fxe4 16.♗xe4 ♘c4 17. ♕c2 ♘xe3 18. fxe3 ♕e5= Matlakov-Negi, Chotowa 2010 ) 15... ♘xd3 16. ♕xd3 ♗xf5 17. ♕c4+ ♖f7 ( 17... ♔h8 18.♘xf5 ♕xf5 19. ♕b4 ) 18. ♕b3 ♕c7 ( 18... ♗e4 19. ♖fd1 ) 19.♘xf5 gxf5 20. ♖fd1+/= 14... f4? was played in a famous game between Grischuk and Kovchan.15. ♗xf4 ♘xd3 16. ♕xd3 e5 It seems as though Black is winning a piece, but in fact after 17. ♘b3 ♕xa2 18. ♖fd1!± it turned out that Black is in a lot of trouble and Alexander won a very nice game. Grischuk-Kovchan, Sochi RUS-tch 2010 15. ♕xd3 e5⁉ This is a very critical position which isn't particularly easy for Black to play. I have to say it's quite risky, but I think it's very playable and I've spent a number of hours making sure this is something Black can do. 16. ♘b3 The main move. 16. exf5 exd4 doesn't work. 17. ♗xd4 O-O∓

16. ♘f3 f4 ( 16... O-O 17. ♕c4+ ♔h8 18. ♘g5± ) 17. ♕c4 The only critical move White can try. 17... ♕b5! ( 17... fxe3 can't be played: 18.♘g5 ♖f8 19. fxe3+− ) 18. ♕c7 ♕d7 19. ♗b6 OO Although this looks scary I'm quite comfortable that Black should hold this endgame. (19... ♕xc7 20. ♗xc7 OO 21. ♗xe5 ♗xe5 22. ♘xe5 ♖e8 23.♘c4 ♖xe4= ) 20. ♖fd1 ( 20. ♕c4+ ♕f7= ; 20. c4⁉ ♕xc7 21.♗xc7 ♖e8 22. ♖fd1 ♗g4 23. c5 ♖ac8 24. ♗d6 ♗f8 25. ♖b1♗xd6 26. cxd6 ♖ed8 27. ♖d5 ♗xf3 28. gxf3 b5 29. ♖bd1 ♖d730. ♖xe5 ♔f7 31. ♖d4 ♖c6 32. a4 bxa4 33. ♖xa4 ♖dxd6 34.♖ea5 ♔g7 ) 20... ♕xc7 21. ♗xc7 ♗g4 ( 21... b5⁉ ) 22. c4 ♖f7!23. ♗d6 ♖d7 24. c5 ♖ad8 25. ♖d5 ♗e6 26. ♖d2 ♗g4 27. ♖b2b5= 16... ♕c7 17. ♕d5 By far the main move. 17. ♗c5 is another critical approach, trying to stop Black from castling.17... ♗e6 18. ♗d6 ( 18. c4 ♖d8 19. ♕e3 f4 20. ♕c3 b6 21.♗a3 ♔f7 22. c5 ♖c8= ) 18... ♕c6⁉ This seems to hold for Black. (18... OOO 19. ♗xc7 ♖xd3 20. ♘c5 ♗c4 21. ♘xd3 ♔xc7+/= ;18... ♗c4⁉ 19. ♕d2 OOO 20. ♖fd1 ♕b6 21. ♖b1 ♗f8 22.♗xe5! ♖xd2 23. ♘xd2 ♕e6 24. ♘xc4± ) 19. c4 ( 19. ♘a5 ♕b520. ♕xb5+ axb5 21. ♘xb7 ♗c4 22. ♖fe1 ♖a7 23. ♘c5 f4 24.a3 ♔f7= ) 19... ♖d8 20. c5 ♕xe4 a) 21. ♕xe4 fxe4 22. ♘a5 ( 22. ♖fe1 ♔f7 23. ♖xe4 ♖he8 )22... b5!= This endgame should be ok for Black. b) 21. ♕c3 21... ♔f7 22. ♖fe1 ♕a4 23. ♗xe5 ♗xe5 24.♕xe5 ( 24. ♖xe5 ♖he8= ) 24... ♖he8 We've arrived at this position by force. 25. c6 White's only serious try, but after 25...♗xb3 26. ♕c7+ Black can play the important 26... ♔f8! ( 26...♔g8 27. ♖xe8+ ♖xe8 28. cxb7 is very strong for White. The difference is that after 28... ♕b4 , threatening mate-in-2 with ...Re1+, White has the additional move 29. ♕c8 , which is quite strong. 29... ♗f7 30. h3 ♔f8 31. ♖d1± ) b1) 27. ♖xe8+ ♖xe8 28. ♕d6+ ( 28. cxb7 ♕b4 29. h3 ♗d530. ♕xh7 ♗xb7 31. ♕h8+ ♔f7 32. ♖c7+ ♖e7 33. ♕h7+= )28... ♔g7 29. cxb7 ♕e4 30. ♖c7+ ♔g8 31. h3 ♗d5 32. f3♕e3+ 33. ♔h2 ♗xb7 34. ♖xb7 ♕e5+=

b2) 27. axb3 27... ♕xc6 28. ♕xh7 ♖xe1+ 29. ♖xe1 ♖e8= I think this is completely equal. 17. c4 OO 18. ♕d5+ ♕f7 19. ♘c5 f4 20. ♗d2 f3! ( 20... ♗g421. h3 ♖ad8 22. ♕xf7+ ♖xf7 23. ♗a5 ♖c8 24. ♘xb7 ♖xb7 25.hxg4 ♖c5 26. ♗d8 ♔f7=/∞ ) 21. g3 ♗h3 22. ♖fd1 b5= 17... ♕f7 White has two move orders to get the same position, either starting with 18.f3 or 18.Nc5. 18. f3 18. ♘c5 O-O ( 18... f4⁉ ) 19. c4 transposes ( 19. ♖b1? ♕xd5 20.exd5 f4 21. ♗c1 ♖d8=/+ ) 18. c4 O-O 19. ♘c5 f4 transposes 18. ♕d6 Once again preventing short castling. 18... ♕e6 19. ♕b4 (19. ♕c7 OO 20. ♖fd1 ♕f7= ) 19... f4 This manoeuvre equalises immediately. 20. ♗b6 f3 Creating a huge threat on the kingside. After21. gxf3 ♕f6 22. ♖fd1 ♗h3 the threats are so strong that White has to give an immediate perpetual with checks on a4 and c4: 23. ♕a4+♔f7 24. ♕c4+ ( 24. ♖d7+ ♔g8 25. ♕c4+ ♗e6 26. ♖d5 h6 ) 24...♔e8= 18. ♗c5 is another option. a) 18... a5⁈ 19. ♕d3! ♕d7 ( 19... ♗e6? 20. ♖fd1 ) 20. ♕e2→ b) 18... ♗e6 is how Black should meet it. 19. ♕d6 ♕d7 ( 19...♖c8 20. exf5 gxf5 21. ♘a5 ♗f8 22. ♘xb7 ♗xd6 23. ♘xd6+♔e7 24. ♘xf7+ ♔xf7 25. ♗d6 e4+/= ) 20. ♕b6 ( 20. ♕xd7+♗xd7 21. ♘a5 b5 22. ♖cd1 ♗e6 23. ♘b7 ♔f7 ) 20... ♕c6Chasing away the white queen. Black of course wants to liquidate into any kind of endgame because his bishops will ensure he's doing ok. 21.♕b4 White avoids the exchange. ( 21. ♕xc6+ bxc6 22. ♘a5 ♔f7 )21... ♗xb3 22. axb3 a5 Forcing the queen even further away. 23.♕a3 ( 23. ♕a4 ♔f7= ) 23... b6 24. ♗d6 ( 24. ♗e7 ♖a7 25.♗h4 ♕xe4 26. ♕d6 ♔f7 27. ♗d8 ♖e8= ) 24... fxe4 Black is ready to play ...Kf7 and finish his development. I think he's very safe here. 25. c4 ( 25. ♖cd1 ♔f7 26. b4 ♖hc8 27. bxa5 bxa5 28.c4 ♕xc4 29. ♗xe5 ♗xe5 30. ♖d7+ ♔g8 31. ♕e7 ♖c7! 32.♕xh7+ ♔f8 33. ♕e7+ ♔g8= ) 25... ♔f7 26. c5 bxc5 27. ♗xc5♖hc8= 18... O-O Necessary. 18... ♗e6 is risky because after 19. ♕d6 ♖d8 20. ♕b4± White is better. 19. ♘c5 Black has an obvious problem. He'd be perfectly fine if he could somehow include the bishop on c8 in the game, but that's currently impossible. This position therefore posed a bit of a problem for me, but I finally settled on 19... ♗f6 ...Rd8 is an obvious threat, and White has two options. 19... ♕xd5 20. exd5 b6 21. ♘a4 b5 22. ♘b6+/= 19... fxe4 20. fxe4 ♕xd5 21. exd5 ♖xf1+ 22. ♔xf1 a) 22... b6 23. ♘a4 b5 24. ♘b6 ♖b8 25. c4+/=

b) 22... b5 23. d6 ( 23. c4 ♗f8 ) 23... ♗f5 24. h3 h5 25. ♗g5♔f7 26. ♖d1 ♗f6 27. ♗xf6 ♔xf6 28. h4± c) ♗f8 23. ♘a4 ( 23. c4 b5 ) 23... b5 24. ♘b6 ♖b8 25. c4bxc4 26. ♖xc4 ♗f5 27. ♔e2+/= 20. ♖fd1 20. ♖b1⁉ is also possible, creating an immediate threat to the b7-pawn. a) 20... ♖d8⁉ 21. ♕xf7+ ♔xf7 22. exf5 ( 22. ♘xb7 ♖d3 23.♗f2 ♖xc3 24. ♘d6+ ♔g8 25. ♖b6 ♗g7 26. ♘xc8 ♖cxc8= )22... gxf5 23. ♖b6+/= b) 20... f4 21. ♗f2 ♕xd5 22. exd5 b5 23. a4 ♗f5 24. ♖a1♖fd8 ( 24... ♖fc8 25. ♖fd1 ♔f7 26. g4+/= ) 25. ♖fd1 ♔f7 transposes c) 20... ♕xd5 21. exd5 b5 The reason for including ...Bf6 and Rb1 is as follows: first of all, White would very much like to have the option of playing c3-c4, when the rook would be more useful on c1 than b1. Secondly, after, let's say, 22. a4 ♖d8 23. ♖fd1 Black has an additional important idea which I think is the best in the position: 23...f4 24. ♗f2 ♗f5 Hitting the rook on b1. Black is ok however White responds. c1) 25. ♘e4 ♗xe4 26. fxe4 ♖dc8 I think this holds by force, although it looks quite tricky. 27. axb5 axb5 28. ♗e1 ♖c4 Black has enough play to equalise. 29. ♖xb5 ♖xe4 30. ♖b6 ♗d8 31.♖b7 ♖a2 32. ♖b8 ♖ee2 33. ♖xd8+ ♔g7 34. ♖d7+ ( 34.♔f1 ♖xg2 35. ♖d7+ ♔h6 36. ♖a7 ♖ac2 37. ♖f7 ♖xh2 38.♔g1 ♖cg2+ 39. ♔f1 ♖c2= ) 34... ♔h6 35. ♗d2 ♖exd2 36.♖xd2 ♖xd2 37. ♔f1 e4 38. c4 e3 39. d6 ♖f2+ 40. ♔e1♖xg2 41. ♖f7 ♖g1+ 42. ♔e2 ♖g2+= c2) 25. ♖a1 25... h5⁉ Trying to stop the important positional idea of g4. ( 25... ♗e7? 26. d6!± ; 25... ♔f7 26. g4 fxg3 27.hxg3+/= ) 26. ♘e4 ♗xe4 27. fxe4 ♖ac8= 20... ♖d8 20... f4 21. ♗f2 ♖d8 22. ♕xd8+ ( 22. ♕b3 ♖xd1+ 23. ♖xd1♕xb3 24. axb3 b5 25. b4 a5 26. bxa5 ♖xa5 27. ♗e1 ♖a2= )22... ♗xd8 23. ♖xd8+ ♔g7 With the centre completely secured I suspect White will be better after the pawns on e5 and f4 fall, as they probably will. 21. ♕xd8+ 21. ♕xf7+ ♔xf7 22. ♖xd8 ♗xd8 23. exf5 gxf5 24. ♖d1 ♗b6 25.♖b1 ♗a7= 21... ♗xd8 22. ♖xd8+ ♔g7 Clearly this looks a bit worrying for Black as he's a little tied up and the rook on a8 isn't yet included in the play. Once again, however, I checked this quite seriously and I'm convinced you can play like this with Black. 23. ♖d5 Perhaps White's most straightforward approach. You can't really maintain the pin along the eighth rank so White goes after the e5-pawn.

23. exf5 gxf5 24. ♘d3 ♕c7 25. ♖d5 ♔g8 26. ♖xe5 ♗d7 27.♗d4 ♖e8 28. ♖ce1 ♖xe5 29. ♖xe5 ♕c4 30. ♘b4 ♕f7 31. a3= 23. ♖cd1 ♕xa2 24. ♗f2 ♕c2 25. ♖f1 f4 26. ♘d3 ♗e6 27.♖xa8 ♕xd3 28. ♖e8 ♕d7 29. ♖a8 ♕b5 30. h4 ♕b3 31. ♖e8a5 32. ♖e7+ ♔f6 33. ♖c7 a4 34. ♗c5 ♗c4 35. ♖a1 b5= 23. c4 b6 24. ♘d7 ♕e7 25. ♗xb6 ♗b7 26. ♘b8 ♖xb8 27. ♖xb8fxe4 28. fxe4 ♗xe4 29. ♗e3 ♕a3 30. ♖e1 ♕xa2 31. ♗f2♕xc4= 23... fxe4 23... f4 24. ♗f2 ♔h6 25. ♖cd1 ♕c7 26. ♘d3 ♗e6 27. ♖c5♕d7 28. ♖xe5 ♖d8 29. ♗h4 ♖f8 30. ♘f2↑ 24. fxe4 b6 25. ♘d3 ♗e6! 26. ♖d6 26. ♘xe5 ♕c7 27. ♖d2 ♔g8 White has won the e5-pawn but the bishop on e6 is important for cutting the most important squares off from the knight on e5. Black isn't going to get mated here and the position is quite unclear. 28. ♗d4 b5 29. ♖f1 ♖d8∞ ( 29... ♖f8⁉ ) 30. ♖df2?b4∓

26... ♕e7 Driving the rook away from the d-file.

26... ♖e8 27. ♘xe5 ♕b7 28. ♗d4 ♔g8 29. ♖e1 ( 29. ♖f1♕xe4 30. ♘d7= ) 29... b5 30. h3→ 27. ♖xb6 ♔g8 Black is now ready to start driving White's pieces away. If White does nothing Black will, for instance, play ...Bc4 and the knight will have to go somewhere. 28. ♘xe5 Black would now be in quite a bit of trouble if not for the fact that there's an immediate draw: 28... ♗h3! Hitting the knight on e5. 29. ♗d4 The only way to comfortably protect the knight. 29... ♕g5 30. ♖c2 ♗xg2 31. ♖xg2 ♕c1+ 32. ♔f2 ♖f8+ 33.♔g3 ♕f4+ 34. ♔h3 ♕h6+= Black has found a perpetual and makes a draw.

1.d4 Nf6 2.c4 g6 3.f3 d5 4.cxd5 Nxd5 5.e4 Nb6 6.Nc3 Bg7 7.e3 0-0 7...Nc6

7...0-0 1. d4 ♘f6 2. c4 g6 3. f3 This move has become exceedingly popular in recent years. It's been played in a World Championship match and numerous elite level tournaments. I've played it with White myself a couple of times, and obviously I've faced it with Black. You're very likely to run into it if you play the Grünfeld with Black these days so I think it's extremely important that you know what you're doing. I'll attempt to guide you through some of the positions you need to know. 3... d5 As this is a Grünfeld series I thought we should stick with the tried and tested method of playing ...d5 at the first opportunity, although Black has a variety of less obvious choices. 3... c5 was played by Gelfand in his World Championship match against Anand and leads to King's Indian/Benoni structures. 3... ♘c6 is another option and a move I've played at least once. It leads to very interesting and unbalanced positions which can be a lot of fun to look at, but I decided not to choose this as the proposed solution to f2-f3. 3... e5 is an oddball variation that was tried by Leko against Kramnik many years ago. 3... e6 is Maxime Vachier-Lagrave's creation, which he twice used against chess players from St. Petersburg - once against Nikita Vitiugov in the decisive last-round game of the Gibraltar Masters and once against me in the Alekhine Memorial. He scored 1.5/2 but I'm not sure the move is as good as the results suggest, so I'm not going to be recommending it either, although it's an interesting move that's also a lot of fun to look at. 4. cxd5 ♘xd5 5. e4 This is basically the point of playing 3.f3 - White forces Black to play ...d5 before the knight appears on c3, so in this position there's no knight trade and the knight has to go back to b6. 5... ♘b6 6. ♘c3 ♗g7 7. ♗e3 O-O The main move at this fork in the road. 8. f4 This is an interesting sideline that I want to cover here. The move White should generally play, 8.Qd2, is covered in 3.f3 Grünfeld: Part 2 . 8. a4⁈ a5 are moves that have been included in some games, but I don't want to waste any time on this as it's quite clearly anti-positional. Once the knight from b8 lands on b4 White will deeply regret what he's done. 8. ♖c1 is another possible waiting move. 8... ♘c6 9. d5 ♘e5 10.♗d4 c6 11. f4 ♘g4 12. ♗xg7 ♔xg7 13. ♗e2 White has developed in a very typical manner - exchanging the dark-squared bishops as soon as possible, driving the knight away and hoping for some advantage in the centre, or to take on c6 with a better pawn structure on the queenside. In this particular position, however, Black has the very important resource13... e5!∞ , indirectly protecting the knight on g4 due to 14. ♗xg4♕h4+ Black is just fine. This has been known about for a number of years so I'm not really covering any new ground here, but it's important to have some general idea of what's going on in these minor sidelines.

8... ♘c6 9. d5 9. ♘f3 ♗g4 10. d5 ♘a5 11. ♗d4 ♗xf3 12. gxf3 e6 9... ♘a5 10. ♗d4 This is a very well-known position. 10... ♗g4 A treasured memory of mine. I ran into this idea many, many years ago and haven't encountered the position since - in fact I'm very unlikely to encounter it in general - so I'd very much like to share it with you even though I think 10...e5 is a more practical solution. The reason I'm including this is that as well as giving general opening advice I thought that in this series I might try to explain why I'm so fascinated by the Grünfeld and why I keep on playing it year in, year out. This is one of the lines that still brings a smile to my face when I stumble across this file in my Grünfeld database. 10... e5 is the "classical" option, and something Boris Avrukh recommends in his book. I believe it leads to a more or less forced draw. a) 11. fxe5 c6! 12. d6 ( 12. ♘f3 cxd5 13. ♗c5 ♘c6=/∞ ) 12...♘ac4 13. ♘f3 ♗g4 14. ♗xc4 ♘xc4 15. ♕b3 ♗xf3 16. gxf3♘xe5 17. O-O-O ♕xd6 Ward-Howell, Gibraltar 2004 b) 11. ♗xe5 11... ♗xe5 12. fxe5 ♕h4+ 13. g3 ♕e7 b1) 14. ♘f3 ♗g4 b2) 14. b3 ♕xe5 15. ♖c1 f5 ( 15... ♕e7 16. ♗g2 f5 17.♘ge2 fxe4 18. ♕d4 ♗g4 19. ♕xe4 ♕xe4 20. ♗xe4Dokhoian-Krasenkow, Minsk 1986 ) 16. ♘f3 ♕e7 17. e5 ♖d8→ b3) 14. ♕d4 14... ♖d8! ( 14... c5 15. d6! cxd4 16. dxe7♖e8 17. ♘d5 ♘xd5 18. exd5 ♖xe7 19. b4± ) 15. b4 b31) 15... ♘c6 16. ♕c5 ♕g5 17. ♘f3 ♕h5 18. ♗e2♘xe5 19. OO ♗h3 ( 19... ♘xf3+ 20. ♗xf3 ♕e5 21.♘b5 ♗h3 Naslund-Horvath, Budapest 2008 22. ♖fc1! ) 20.♘h4 ♕g5 21. ♖f4 ♘bd7 22. ♕f2 h5∞ Botez-Gara, Mexico City UNAM w Gp C 2012 b32) ♘ac4! b321) 16. ♗xc4 ♕xb4 17. ♘f3 ♕xc4 18. ♕e3 ♖e8!Avrukh ( 18... c6 19. ♖c1 ♗g4 20. ♘d2 ♕b4 21. OO cxd5 22. exd5 ♘xd5 23. ♘xd5 ♖xd5 24. ♘e4± Mamedyarov-Shirov, Baku rapid 2009 ) 19. ♖c1 ( 19. ♖d1♗g4 20. e6 ♗xf3 21. exf7+ ♔xf7 22. ♕xf3 + ♔g7 23.♖f1 ♘d7∞ ; 19. e6 fxe6 20. ♘e5 ♕b4 21. ♘g4 ♖f8 22. ♖b1 ♕e7 23. ♕d4 e5 24. ♕xe5 ♕xe5 25. ♘xe5♗h3=/+ ) 19... ♘d7 20. ♘b1 ( 20. ♕h6 ♖xe5! ) 20...♕a6! ( 20... ♕b4+ 21. ♘bd2 ♘xe5 22. ♘xe5 ♖xe5 23.♖xc7⇆ ) 21. ♕c3 b6! 22. ♕c6 ♕d3∓

b322) 16. ♘f3 16... ♗g4 17. ♗xc4 ♗xf3 18. OO♕xb4 19. e6 ( 19. ♖xf3 ♕xc4 20. ♕e3 ♘d7 21. ♕f4♖f8 22. ♖b1 ♖ae8 23. ♘b5 a6 24. ♖c3 ♕xa2 25.♘a3 ♘xe5 Moskalenko-Konguvel, Barcelona 2004 ) 19...fxe6 20. ♖xf3 ♕xc4 21. ♕f6 exd5 22. ♕e6+ ♔h8 23.♕f6+=

11. ♕d3 The lines arising after this move are the reason I went off on this whole tangent. 11. ♘f3? , with a position that can also be reached if White starts with 10.Nf3 and only plays 11.Bd4 after 10...Bg4, is no good for White, as Black can play 11... ♗xf3 12. gxf3 a) 12... e5 13. fxe5 ♖c8! is an interesting but completely unnecessary alternative: a1) 14. ♘a4 c6 15. ♘xb6 ( 15. d6 ♘d7 16. b4 ♘xe5 17.bxa5 ♕xd6=/∞ ) 15... axb6 16. b4 cxd5 17. bxa5 dxe4 18.♗g2 ( 18. h4 exf3 19. ♗xb6 ♕e7 20. ♔f2 ♕xe5 21. ♖b1 ♕e4∓ ) 18... ♕e7 19. ♕e2 ♕b4+ 20. ♕d2 ♕a4 21. ♕e3♕xa5+ 22. ♔f1 ♖fd8∓ a2) 14. ♕d2 c5 15. ♗f2 ♗xe5 16. f4 ♘bc4 17. ♗xc4 ♘xc418. ♕e2 ♗xc3+ 19. bxc3 b5 20. a4 a6∞ a3) 14. ♕c2 14... c5 15. ♗f2 ♗xe5 16. ♗h3 ♕f6 17. OOO♘ac4 ( 17... ♖c7 18. ♘b5 ♘bc4 ) 18. ♗xc8 ♘xb2 19.♔xb2 ♘c4+ 20. ♔b3 ♘a5+ 21. ♔b2 ♘c4+= b) e6!∓ Once again, Black has the very important threat of Qh4+ and is probably already better here. 11. ♗e2 This is White's second major option. It's important for Black to start with 11... e5 straightaway. There are almost no lines where Black can avoid playing e5 altogether, but here it's important not to take on e2 first. ( 11... ♗xe2 12. ♘gxe2 e5 In this case White has the option of playing 13. dxe6 because the bishop on d4 is now protected. ) 12. fxe5( 12. ♗xe5 ♗xe2 13. ♕xe2 ♗xe5 14. fxe5 ♕h4+ 15. g3 ♕e7 16.♘f3 ♘ac4= ) 12... ♗xe2 Only now, and after, for instance, 13.♘gxe2 ( 13. ♕xe2 ♕h4+ 14. g3 ♕e7 15. ♘f3 ♘ac4= ) 13...♘ac4 it's quite clear that Black's situation has improved in comparison to the normal lines, because the knight on e2 is horribly misplaced it really has to be on f3. 11... e5! 12. fxe5 12. ♗xb6? axb6 13. b4 ♘b3 12. ♗xe5 is better for White, objectively speaking. 12... ♗xe5 13.fxe5 ♘ac4 14. ♕d4 ( 14. ♕g3 After this move Black has the very strong reply 14... ♕g5 , threatening mate-in-one. White can defend against the threat with 15. ♖d1 but now 15... f5!∓ is extremely strong. Black is probably already better in the ensuing complications. ) 14... c515. ♕xc5 ♘xe5=/∞ The open c-file, advantage in development and general piece activity should compensate Black for the single pawn he had to sacrifice. 12... ♘ac4 13. ♕g3 This position is the reason I decided to show this whole line. It seems as though Black is in a lot of trouble. The bishop on g4 is hanging and White is only one move away from completely consolidating and just being a pawn up. 13. h3 is probably the move White should play, objectively speaking. Black will reply 13... ♕h4+ 14. g3 ♕h5 15. ♗g2 ♗d7∞ and although the position is still very unclear it's probably around equality. But this line would be boring... 13... c5 14. ♗xc5 14. ♗f2 ♘xe5

14... ♗xe5 15. ♕xg4 ♘xb2 Black is a whole piece down, but in fact White's queenside is about to collapse and White has problems equalising. 16. ♖c1 16. ♕h3 straightaway is most easily met by 16... ♗xc3+ ( 16... ♖c817. ♗b4 is less clear. ) 17. ♕xc3 ♘6a4 18. ♕d4 ♘xc5 and we've arrived at another picturesque position. White has a choice of two knights that he can seemingly take for free, but after 19. ♕xc5 ( 19. ♕xb2♕h4+ ) 19... ♕h4+ 20. ♔d2 ♕xe4 Black has a lot of activity and is doing well because White's king is so completely unsafe. All these lines are a good example of what Black can hope for if White is a little imprecise in the more ambitious lines of the Grünfeld. 16... ♖c8 16... ♖e8⁉ 17. ♗xf8 17. ♘f3 h5 18. ♕h3 ♖xc5 19. ♘xe5 ♕g5∓

17. ♗b4 ♖e8 ( 17... ♘6a4 18. ♗xf8 ♕xf8 19. ♗e2 ♘xc3 20.♘f3 h5 21. ♕h3 ♗c7 ) 17... ♕xf8 and after, for instance, 18. ♕f3 18. ♘f3 ♗xc3+ 18. ♗e2 ♕h6 19. ♖c2 ♖xc3 18... ♗xc3+ 19. ♖xc3 ♕b4 White will lose an entire rook on the c3-square, leaving him almost lost. 20. ♘e2 ♘2a4 21. a3 ♕a5 22. ♔d2 ♘xc3 23. ♘xc3 ♘a4

7...Nc6 1. d4 ♘f6 2. c4 g6 3. f3 d5 4. cxd5 ♘xd5 5. e4 ♘b6 6. ♘c3♗g7 7. ♗e3 ♘c6 This is a move I played myself against Fabiano Caruana in one of the Grand Prix tournaments. It's an important alternative for Black, and although I'm not recommending it for this particular series I've included some notes on the move. 8. d5 The most obvious reply. 8. ♗b5⁉ is, however, perhaps the most advisable move, as it gives White more flexibility in the upcoming play. 8... OO ( 8... ♗d7 9. ♘ge2♘a5 10. ♗xd7+ ♕xd7 11. b3 ♘c6 12. ♕d2+/= ) 9. ♘ge2 ♘a5 10.b3 e5 ( 10... a6 11. ♗d3 ♘c6 12. ♗c2 e6 13. O-O ♕e7 14. ♕c1e5 15. d5 ♘d4 16. ♖d1+/= KhismatullinYandemirov, Voronezh 2006 )

a) 11. d5 c6 12. dxc6 ♘xc6 ( 12... bxc6 13. ♗d3 ♕h4+ 14.♗f2 ♕e7 15. ♕c2 ♘b7 16. ♘d1 ♗e6 17. OO ♖fd8 18. ♘e3♗f8 19. ♖ad1 ♘c5 20. ♘c1+/= ) 13. ♗xc6 ( 13. ♕xd8 ♖xd814. ♔f2 ♘d4 15. ♖hd1 ♗e6 16. ♗d3 ♖ac8 17. ♘b5 ♘xb518. ♗xb5 ♗f6= ; 13. ♗c5 ♕h4+ 14. g3 ♕h5 15. ♗xf8 ♗xf816. ♗xc6 bxc6 17. ♕d3 ♗h3=/∞ ) 13... bxc6 14. ♕xd8 ♖xd815. ♔f2 ♗f8 16. ♖hd1 ♗a6= b) 11. dxe5 b1) 11... ♕h4+ 12. ♗f2 ( 12. g3 ♕e7 13. f4 ♗h3 14. ♕c2♖fd8 15. ♔f2 a6 16. ♗d3 ♘c6=/∞ Bakker-Gutman, Dresden 2007 ) 12... ♕e7 13. f4 ♖d8 14. ♕c2 a6 15. ♗d3 ♘c6 16.a3± b2) 11... ♗xe5 12. OO ( 12. ♗c5 ♕h4+ 13. g3 ♕h5 14.♗xf8 ♕xf3 15. ♔d2 ♔xf8 ; 12. f4⁉ ♗g7 13. ♗c5∞ ) 12...♕e7 ( 12... a6⁉ 13. ♗d3 ♘c6 14. f4 ♗g7 15. e5+/= ) 13.♕e1 a6 ( 13... ♖d8 14. ♖d1 ♖xd1 15. ♘xd1 a6 16. ♗d3♘c6 17. ♘f2 ♗g7 18. ♕c1 ♗e6 19. f4 ♖d8 20. ♗b1 ♘d421. ♘xd4 ♗xd4 22. ♗xd4 ♖xd4 23. f5→ Boch arov-Timofeev, Khanty-Mansijsk Russia Cup Final 2012 ) 14. ♗d3 ♘c6 15. ♖d1♗d6 16. ♕f2 ♘e5 17. ♗b1 ♗e6 18. h3 f6 19. f4 ♘ed7 20.♘d4 ♗f7 21. ♘ce2 c5 22. ♘f3 c4 23. ♘fd4∞ Bocharov-Timofeev, Khanty-Mansijsk Russia Cup Final 2012 8... ♘e5 9. ♗d4 9. f4 ♘g4 a) 10. ♗b5+ ♗d7 11. ♗d4 ( 11. ♕xg4 ♗xc3+ 12. bxc3 ♗xb513. ♗d4 ♖f8 14. a4 ♘xa4 15. ♕d1 ♘b6 16. ♘f3 ♕d6 17.♗e5 ♕d7 18. ♔f2 c6 19. ♗g7 ♖g8 20. ♗d4 ♘c4∞ Gupta-Zhu Chen, Gibraltar Masters 2012 ) 11... ♗xd4 12. ♕xd4 OO 13.♗e2 ♘f6 ( 13... e5 14. fxe5 ♕h4+ 15. g3 ; 13... c6 14. dxc6 bxc6 ) 14. a4 a5 15. ♗f3 c6 16. ♘ge2 cxd5 17. e5 ♘e4 18.♘xd5 ♘xd5 19. ♕xd5 ♗c6 20. ♕d4 ♖a6 21. ♖c1 f5 22.exf6 ( 22. O-O ♕b6 23. ♕xb6 ♖xb6 24. ♘d4 ♖d8 25. ♖fd1♖xb2 1/2 Khismatullin-Yandemirov, Sochi 2006 ) 22... ♘xf6= b) 10. ♗d4 10... ♘f6 ( 10... ♗xd4 11. ♕xd4 OO tr) 11. ♗b5+♗d7 12. ♗xd7+ ♕xd7 13. ♘f3 OO 14. OO c6 15. dxc6♕xc6 16. ♕e1 ♖fd8 17. ♖c1 ♖ac8 18. e5 ♘fd7 19. ♕h4 ♘f8+/= Gelfand-Timofeev, Eilat 17th ECC 2012 9... O-O 10. f4 ♘g4 11. ♗xg7 11. ♘f3 ♘f6 ( 11... ♗xd4 12. ♕xd4 e6 13. h3 ♘f6 14. OOO exd5 15. exd5 ♕d6 16. g4 ♖d8⇆ ) 12. ♕d2 ( 12. h3 e6 ) 12...e6 13. dxe6 ♗xe6 14. OOO c5 15. ♗e5 ♕xd2+ 16. ♖xd2 ♖fd817. ♗d3 ♘g4 18. ♗xg7 ♔xg7 19. ♖e1 ♗c4 20. h3 ♘f6 21.♖ed1 ♖e8 22. ♗c2+/ = Anand-Caruana, Sao Paulo/Bilbao 5th Masters 2012 11. h3⁈ e5! 12. ♗xb6 ♕h4+ 13. g3 ( 13. ♔d2 axb6∓ ) 13...♕xg3+ 14. ♔d2 axb6 15. hxg4 exf4→

11... ♔xg7 12. ♕d4+

12. ♘f3 c6 ( 12... e6 ) 13. ♕d4+ ♘f6 14. dxc6 ♕xd4 15. ♘xd4bxc6 16. ♗e2 c5 17. ♘b3 c4 18. ♘a5 ♗e6 19. O-O-O ♖ac8 20.♗f3 ♘fd7 Dreev-Smirin, Biel 2002 12. ♗e2 ♘f6 13. ♘f3 c6 14. dxc6 bxc6 15. O-O+/=

12... ♔g8 transposes. 12... ♘f6 13. OOO c6N ( 13... e6⁉ 14. ♘f3 exd5 15. e5=/∞ ) 14.h3 ( 14. dxc6 ♕xd4 15. ♖xd4 bxc6 16. ♘f3 ♗b7 17. ♖d2 ♖ad818. ♗d3+/= ) 14... ♕c7 ( 14... cxd5 15. g4 ♕c7! 16. exd5 ♔g8 17.♗g2 ♖d8∞ ) 15. ♕e5 ♕xe5 16. fxe5 ♘fd7 17. ♘f3 Mamedyarov-Kamsky, Nalchik FIDE GP 2009 12... f6⁉ 13. ♘f3 ( 13. h3 e5 14. ♕d2 ♘h6 15. fxe5 fxe5 16.♘f3 ♘f7 ) 13... e5 a) 14. ♕d2 exf4 15. ♕xf4 ♕e7 16. h3 ♘e5 17. OOO ♗d7 18.♔b1 ♖ae8 19. ♘d4 a6 20. ♗e2 ♕d6∞ Savchenko-Yandemirov, Sochi 2006 ( 20... ♘c8 21. ♗g4! ) b) 14. fxe5? fxe5 15. ♘xe5⁇ c5−+ c) 14. dxe6 14... ♕e7⁉ 15. h3 ( 15. f5 ♖d8 16. ♘d5 ♘xd5 17.exd5 gxf5 18. O-O-O c6∞ GoganovYandemirov, Kazan 34th R. Nezhmetdinov Cup 2012 ) 15... ♖d8 16. ♕g1 ♘h6 17. ♗e2♗xe6 18. ♕f2 ♘f7 19. O-O c6+/= 13. ♗e2 c6 14. ♗xg4 ♗xg4 15. h3 15. f5 a) 15... e5 16. ♕f2 gxf5 ( 16... cxd5 17. h3 d4 18. hxg4 dxc319. bxc3 ♘c4∞ ) 17. h3 ( 17. exf5 ♕g5 18. dxc6 bxc6 19. ♘f3♕g7∞ ) 17... ♗h5 18. exf5 f6 19. dxc6 ( 19. g4 ♗e8! ) 19...bxc6∞ b) 15... cxd5 16. h3 e5 17. ♕f2 transposes. c) gxf5 16. h3 ♗h5 17. exf5 f6 ( 17... e5 18. ♕f2 f6 transposes) 18. ♕h4 ♗f7 19. dxc6 bxc6 20. ♘ge2∞ 15... ♗d7 16. ♘f3 cxd5 17. exd5 e6= Caruana-Svidler, Tashkent FIDE GP 2012

1.d4 Nf6 2.c4 g6 3.f3 d5 4.cxd5 Nxd5 5.e4 Nb6 6.Nc3 Bg7 7.e3 0-0 8.Qd2 Nc6 9.0-0-0 Qd6 10.h4/e5/d5/f4 10.Kb1 10.Nb5

7...0-0 8.Qd2 Nc6 9.0-0-0 Qd6 10.h4 & other moves 1. d4 ♘f6 2. c4 g6 3. f3 d5 4. cxd5 ♘xd5 5. e4 ♘b6 6. ♘c3♗g7 7. ♗e3 O-O 8. ♕d2 ♘c6 This is the move I want to cover here, as it's something I've been doing exclusively in this position. I've never played 8...e5 in my life so I thought I should stick to stuff I'm more comfortable doing myself. 8... e5 is the other major option. 9. d5 c6 10. h4 This has been played in a variety of high-level games and was seen, for instance, in the Anand-Gelfand match. It's a very interesting position and one that's a lot of fun to analyse. 9. O-O-O ♕d6 This is by far the most topical move in this position, and something I've encountered with both colours. I had this position with White against Fabiano Caruana and with Black against Wang Hao in Stavanger in 2013. White has a variety of options. 9... f5 To be honest with you, my original intention was to recommend this move, which I've been playing for a number of years. I even began that video, but then I ran into some very serious chess problems equalising after 10. e5 , so the idea had to be scrapped. Perhaps some day I'll film a video explaining why I think this line is a bit of a problem for Black, but for now we'll concentrate on the alternative. 10. h4⁈ The two main moves in this position, 10.Kb1 and 10.Nb5 , will be covered later, but this is quite a serious possibility. In general, the whole point of the 3.f3 line is to castle long and try and give mate, which distinguishes it from the normal Grünfeld fare where White is perhaps more concerned with positional advancement. 10. d5⁈ ♘e5 drives the black pieces towards the squares they want to occupy. The computer even suggests playing 11. b3 here, which is a good indication that something has gone badly wrong. 11... f5 ( 11... ♕a3+12. ♕b2 ♕xb2+ 13. ♔xb2 f5 ) 10. f4⁈ isn't particularly good either. 10... ♗g4 11. ♘f3 ♖fd8 12. d5Otherwise the centre is about to fall. 12... ♗xc3 is probably the most practical solution, and Black's response is the same regardless of how White takes. 13. ♕xc3 ( 13. bxc3 e6 ) 13... e6 Once again the white centre is under immediate attack and Black is doing fine. 14. ♗c4♘xc4 15. ♕xc4 exd5 16. ♖xd5 ♕e7 10. e5⁈ ♕b4 is completely counterproductive for White, as it drives the queen directly to the squares it wants to occupy. Nc4 will now be a constant threat, as will Rd8, and the bishop has a fantastic square on f5. 10... ♖d8 After this move White has to pay attention to the centre. 11. ♘b5 11. d5 ♘e5 is once again not particularly attractive for White. 12.♕f2 ( 12. ♘b5 ♕d7 13. ♕c2 c6 14. ♘xa7 ♖xa7 15. ♗xb6♖xa2 ) 12... c6 13. dxc6 ♕c7 11... ♕d7 12. h5 a6 13. ♘c3 We're following a game Grischuk and Mamedyarov played in 2010.

13. ♗h6 axb5 14. ♗xg7 ♕d6! 15. hxg6 ( 15. ♕h6 ♖xa2 16. hxg6♕xg6 ) 15... ♕xg6−+ 13... ♗xd4 The much more practical decision. 13... ♘xd4 was the move played by Mamedyarov, which after 14. hxg6 fxg6 15. g4 led to an incredibly crazy position which I think is objectively quite ok for Black, but requires a lot of knowledge and precise "computer-like" play. a) 15... ♕c6 16. ♕f2 ( 16. ♕h2 h5 17. gxh5 ♘a4 18. ♖d3♗e6⇆ ) 16... ♘e6 17. ♖xd8+ ♘xd8 18. ♘e2 ♘f7 19. ♕h4 h620. ♕xe7 ♕d7 21. ♕xd7 ♗xd7 Grischuk-Mamedyarov, Moscow 2010 b) 15... e5 16. ♕h2 14. hxg6 14. ♗xd4 ♕xd4 15. ♕f4 ♕f6 16. ♖xd8+ ♘xd8 17. ♕xc7 ♗d7 is safe for Black both here and after the inclusion of 18. hxg6 fxg6 14... fxg6 15. ♘d5⁉ The critical move in this position. Black can reply in a variety of ways. 15. ♗xd4 ♕xd4 16. ♕f4 ♕f6 ( 16... ♖f8⁉ 17. ♕xc7 ♕b4 18. a3♕c5 19. ♔b1 ♗e6∞ ) 17. ♖xd8+ ♘xd8 18. ♕xc7 ♗d7 , and Black is fine. 15... ♘xd5 15... ♗xe3 is the one "wrong" solution: 16. ♕xe3 ♕e8 At the moment the Nxc7 fork is impossible because of the queen check on d8 after exchanging rooks on d1, but if White finds 17. ♗e2! Black's position is somewhat dangerous, because now Nxc7 is a huge threat, and after 17...♘xd5 18. exd5 ♘b4 19. ♕b3 White gets a lot of initiative on the kingside after driving the knight away from b4. 19... a5 20. a3 ♘a621. ♘h3 ♗xh3 22. ♖xh3+/= 15... ♗g7⁉ is a very decent alternative for Black, although time constraints mean I'm not going to go into any detail about it here. 16. ♗c4 The point of the previous play. 16. ♗xd4 is a somewhat naive move which runs into 16... ♘b6! and Black is just perfectly fine. 17. ♗e3 e5=/+ 16... e6⁉ The most practical solution. 16... ♗xe3 17. ♕xe3 e6 ( 17... ♘b4 18. exd5 ♕f5 19. ♗b3+/= )18. exd5 ♘a5 is fine for Black, but after 19. ♗d3 exd5 and the somewhat strange-looking 20. ♗e4! I think White has enough compensation for the pawns he had to sacrifice: a) 20... ♕f7 21. ♗xd5! ♖xd5 22. ♖xd5 ♘c6!∞ ( 22... ♕xd5?23. ♕e8+ ♔g7 24. ♕e7+ ♕f7 25. ♖xh7++− ) b) 20... d4 21. ♕g5 ♘c6 22. g4 ♕e7 23. ♘h3=/∞ 16... ♗g7 also allows unnecessary complications: 17. ♘h3⁉ ( 17. exd5♘e5 18. ♗b3 a5 is very good for Black. ) 17... e6 18. exd5 ♕e7 19.♘f4⇆

17. ♗xd4 ♘b6 18. ♗xb6 Forced. 18... ♕e7! Perhaps even stronger than just exchanging queens, as after 19. ♕e3 ♖xd1+ 20. ♔xd1 cxb6 Black has a safer king and is currently a pawn up. White can regain the pawn at the cost of tempi by playing 21. ♕xb6 but after 21... ♘e5 Black gets his pieces to good squares and is doing perfectly fine.

10.Kb1 1. d4 ♘f6 2. c4 g6 3. f3 d5 4. cxd5 ♘xd5 5. e4 ♘b6 6. ♘c3♗g7 7. ♗e3 O-O 8. ♕d2 ♘c6 9. O-OO ♕d6 10. ♔b1 One of the two main moves in this position. 10... ♖d8 11. ♘b5 The variations here are extremely tricky and extremely tactical. 11. d5 ♕b4! 11... ♕d7 12. d5 It's already possible for Black to go wrong here. 12... a6! A necessary move, after which White has three options. 12... ♘e5⁈ 13. ♕c2 c6 14. ♘xa7 White was already quite a bit better here in a game between Viswanathan Anand and Shakhriyar Mamedyarov.14... ♖xa7 15. ♗xb6 ♖a8 16. ♘e2 ♖e8 17. ♘c3 AnandMamedyarov, Ajaccio 2011 13. ♘c3 The main move. 13. dxc6 is the least critical move, and leads to a more or less immediate draw. 13... ♕xd2 14. ♖xd2 ♖xd2 15. ♗xd2 axb5 16. cxb7 ( 16.♗xb5 bxc6 17. ♗xc6 ♖b8 18. b3 ♘c4 19. ♔c2 ♖b6 ) 16...♗xb7 17. ♗xb5 f5! Attacking the centre immediately. ( 17... ♖d8 18.♗c1 ♖d4 19. b3 ♖xe4 20. ♘e2 ♖e5 21. a4 c6 22. ♗d3 c5 23.♖d1+/= ) 18. exf5 ♗d5 Th e a2-pawn is hanging and the queenside is very much under attack. In a couple of games that reached this position White immediately bailed out. a) 19. fxg6 ♖xa2 20. gxh7+ ♔h8 21. ♗c3 ( 21. ♗c1 ♖a5 )21... ♗xc3 22. bxc3 ♖xg2 23. h4 c6= b) 19. b3 gxf5⇆

c) 19. ♘e2 19... ♗xa2+ 20. ♔c2 gxf5 and the position is of course completely equal. 21. ♗c3 ( 21. b3 c5 22. ♗e3 c4 23. ♗xb6♗xb3+ 24. ♔d2 ♖b8 25. ♖b1 e6 26. ♔e3 ♗a2 So-Lou Yiping, Ho Chi Minh City Asia Cont-ch 2012 ) 21... ♗c4 22. ♗xc4+ ♘xc423. ♘d4 ♘e3+ 24. ♔d3 ♘xg2 25. ♘xf5 ♘f4+ 26. ♔c4 ♗xc3 27. bxc3 ♔f7 28. ♖e1 ♘g6 Vitiugov-Giri, Reggio Emilia 54th GM 2011

13. ♘a3⁉ A more critical attempt. The reason the knight retreats to this square and not c3 is to prevent Black from using the c4-square as a base for both of his knights, but unfortunately here White runs into tactical interference: 13... ♕e8! A very thematic reply. In this position White has a choice between Qc1 and Qe1. ( 13... ♘e5 14. h3 ) a) 14. ♕c1 ♘a4! 15. dxc6 Two moves are possible for Black. ( 15.♖d2⁈ is simply met by 15... ♘b4 and Black is ready to start attacking the white centre and the queenside. He's already doing very well here. 16. ♕c4 c5 17. dxc6 a5∓ ) 15... ♗xb2 ( 15... ♖xd116. ♕xd1 ♘xb2 is also fine, as can be seen in the 14.Qe1 line. ) 16.♖xd8 ♕xd8 17. ♕c2 b5! White is ok after either 18.Bc1 or 18.Nc4.18. ♗c1 ( 18. ♘c4 ♗g7 19. ♘a5 ♘c3+ 20. ♔c1 ♕d6∓ ) 18...♗xc1

a1) 19. ♕xc1 ♗e6 White is the one who has to worry, and worry quite seriously, about surviving the next 5-10 moves, because Black's attack on the queenside is very strong and White is very poorly prepared for it. 20. ♘e2 ♕d3+ 21. ♘c2 ♖d8 22. ♘f4 ♕c323. ♗d3 ( 23. ♘xe6 ♖d1!−+ ) 23... b4! 24. ♘xe6 ( 24. ♘a1♕e5−+ ) 24... fxe6 25. ♘e1 ♖xd3 26. ♘xd3 ♕xd3+ 27.♕c2 ♘c3+ 28. ♔b2 ♕d4∓ a2) 19. ♔xc1 19... ♕d6 20. ♕d3 ♕xc6+ 21. ♘c2 ♗e6 , and Black is better.

b) 14. ♕e1 b1) 14... ♘a7⁉ 15. h4 ♘b5 is a very thematic idea, in this case leading to equality after 16. ♗xb5 axb5 17. h5 ♘c4 18. ♘xc4bxc4 19. ♘e2 ( 19. hxg6 ♕a4 ; 19. ♗d4 ♕a4 20. a3 c3−+ )19... ♗xb2! ( 19... ♕a4 20. ♘c3 ♗xc3 21. ♕xc3 ♕xa2+ 22.♔c2 is less clear. ) 20. ♔xb2 ♖xa2+! 21. ♔xa2 ♕a4+ 22.♔b1 ♕b3+ 23. ♔c1 ♕a3+= b2) 14... ♘a4! Black's response is the same as after Qc1. 15.dxc6 ♖xd1+ 16. ♕xd1 ♘xb2 17. ♕c2 bxc6 Opening the bfile. 18. ♗c4 ♖b8=/+ 19. ♔c1 For now it seems as though Black's threats have been parried, but after a quiet move like 19... c5Black has a tremendous attack for very little material - he already has two pawns for a piece. This is frankly inadvisable for White. 13... ♕e8 White once again has a choice between Qe1 or Qc1, as Black obviously wants to play ...e6 and hit the white centre straightaway. 13... ♘e5 A very interesting idea of Areshchenko's. 14. ♗d4 a) 14... ♕d6 15. ♕f2! f5 16. ♗e2 ( 16. h3⁉ fxe4 17. fxe4 ♖f818. ♘f3 ♗d7 19. ♗e2+/= ) 16... fxe4 17. fxe4 ♘g4 18. ♗xg4♗xg4 19. ♘f3 ♗xd4 20. ♖xd4 ♖f8 21. h3 ♗xf3 22. gxf3♕c5 23. ♖hd1 ♘c4 24. ♔a1 b) c5⁉ A pawn sacrifice. 15. ♗xc5 ♘ec4 This is a very complex position where Black gets a lot of play for his pawn. Time constraints meant I couldn't cover this position in the video, but you'll find my notes below. 16. ♕c1⁉ ( 16. ♗xc4 ♘xc4 17. ♕c1 ♕c7 18. ♗d4 e519. ♗f2 b5 20. b3 ♘d6 21. ♕e3 f5 22. ♖c1 ♕e7 , and Black was soon better: 23. ♕b6 fxe4 24. ♘xe4 ♗f5 25. ♕e3 a5=/+EljanovAreshchenko, Kiev 81st UKR-ch 2012 )

b1) 16... ♕c7 It's possible to continue in a similar vein to the Eljanov-Areshchenko game: 17. ♗d4 ♗xd4 ( 17... e5 18. ♗f2 )18. ♖xd4 ♕c5 19. ♘ge2 ♗d7 ( 19... e5 20. ♘a4! ) 20. h4 (20. ♔a1⁉ ) 20... e5=/∞ b2) 16... ♘xb2⁉ b21) 17. ♗xb6 ♘xd1 18. ♘xd1 e6! is quite ok for Black. b211) 19. ♕c7 ♕xc7 20. ♗xc7 ♖d7 21. d6 ♗e5 22.♘e3 ( 22. ♘f2 ♗xd6 23. ♗xd6 ♖xd6 ) 22... ♗xd6 23.♗xd6 ( 23. ♗b6 ♗c7 24. ♗xc7 ♖xc7 25. ♘h3 ♗d726. ♗d3 ♖d8 ) 23... ♖xd6 24. ♘e2 ♗d7 25. ♘c1♗a4 with an unclear endgame. b212) 19. ♘e3 ♕d6!= 20. ♗xd8 ♕b4+ 21. ♔c2 ♗d7 22.♗c7 ♕c3+ ( 22... ♖c8⁉ 23. ♔d1 ♗e5 24. ♗d3 ♖xc725. ♕b1 ♕c5=/∞ ) 23. ♔b1 ♕a1+ 24. ♔c2 ♕c3+= b213) 19. ♗xd8 19... ♕xd8 20. ♗c4 ♗d7 21. dxe6fxe6 22. ♗b3 a5 , and despite being a whole piece down Black has excellent compensation. b22) 17. ♕xb2 17... ♘a4 18. ♘xa4 ♗xb2 19. ♘xb2 ♕c7with a very sharp and u nbalanced position which is quite difficult to assess. 13... ♘a5 14. ♗d4 ( 14. b3 ♕d6 ) a) 14... ♘bc4 15. ♗xc4! ( 15. ♕f2? e5 16. ♗c5 b6 17. ♗b4c5 18. ♗xa5 ♘xa5=/+ GustafssonMekhitarian, Reykjavik 2011 )15... ♘xc4 16. ♕c1 ( 16. ♕d3 e5 17. ♗c5 b5 ) 16... ♗xd4 (16... e5 17. dxe6 ♕xe6 18. ♗xg7 ♖xd1 19. ♕xd1 ♔xg7 20.♕d4+ f6 21. ♘d5 ) 17. ♖xd4 b5 18. ♘ge2 ♗b7 ( 18... ♕d6 19. h4 h5 20. g4 hxg4 21. h5 gxf3 22. hxg6 ♕f6 ) 19. b3 ( 19.h4 e5 20. ♖dd1 b4 21. ♘a4 ♕b5 22. h5 ) 19... ♘d6 20. h4h5 21. g4 hxg4 22. h5± b) 14... e5 15. ♗c5 has been played in a number of games. ( 15.♗f2⁉ straightaway may be better rather than prompting Black to play ...b6.) 15... ♘bc4 16. ♕c1 b6 17. ♗f2 ♕d6 18. h4 ♕b4 19.g4 f5 20. g5 fxe4 21. ♘xe4 ♗f5 22. ♗d3 ♖xd5 23. ♗xc4♘xc4 24. ♖xd5 ♗xe4+ 25. fxe4 ♘a3+ 26. ♔a1 ♕xe4 27.♖d2 ♕xh1 28. bxa3 ♖f8 Mamedyarov-Volokitin 14. ♕c1 14. ♕e1⁉ has been played by Levon Aronian, which is a mark of quality straightaway, and requires quite a lot of precision from Black, although I think he's fine. a) 14... ♘a7⁉ , with the idea of playing ... Nb5 later, is thematic in all of these kinds of situations, but in this particular position I think White is better: a1) 15. ♗f4⁈ ♘b5! 16. ♘xb5 axb5 17. ♗xc7 ♘c4 18. ♗xc4bxc4 19. ♘e2 ♗xb2 20. ♔xb2 ♕b5+ 21. ♔c1 ( 21. ♔a1♖xa2+ 22. ♔xa2 ♕b3+= ) 21... ♖xa2 22. ♕c3 ♖d7 leads to a draw: 23. ♘f4 ♖xc7 24. d6 exd6 25. ♘d5 ♕c5 26. ♘f6+♔f8 27. ♘d5=

a2) 15. h4! 15... ♘b5 16. ♘xb5 axb5 17. h5! ( 17. ♗xb6cxb6 18. h5 e6=/+ ) 17... ♘c4 and now, for instance, 18. hxg6fxg6 ( 18... hxg6 19. ♗c1 c6 20. ♕h4± ) 19. ♗d4 I think Black's attack on the queenside simply isn't fast enough to cause White too many problems, so this isn't particularly recommended.19... ♗xd4 20. ♖xd4 e5 21. ♖d1 b) 14... ♘a5 15. h4 leads to very sharp and messy positions which I don't believe Black needs to enter here. c) 14... ♘e5 The play here is very forcing. 15. ♗e2 Preparing for the ...e6 break, as Black has very few other sensible ideas in this position. 15... e6 16. ♗xb6 cxb6 17. f4 White has taken away both the c4 and g4-squares from the knight, and Black needs to react energetically: c1) 17... ♘d7 was played by Fabiano Caruana against Levon Aronian, and after 18. dxe6 ♕xe6 White missed a very strong move. c11) 19. ♘f3 could have been met by 19... ♗xc3! , accepting the double pawn sacrifice ( 19... ♕e8? 20. ♕h4 ♗f6 21.♘g5 ♘f8 22. ♗c4 ♔g7 23. ♕g3 ♗xg5 24. fxg5 ♗e6 25.♘d5 ♗xd5 26. ♗xd5± Aronian-Caruana, Moscow Tal Memorial 2012 ) 20. ♕xc3 ♕xe4+ 21. ♗d3 ♕xf4 I feel Black should be fine here. 22. ♖hf1 ♖f8 23. ♗c2 ♘c5 c12) 19. ♗f3!+/= White is better c2) 17... exd5! This isn't a novelty as it's already been played by Kateryna Lahno against Zhao Xue in a Women's Grand Prix. 18.fxe5 Accepting the sacrifice is necessary if White wants to fight for an advantage. ( 18. ♘xd5 ♘g4 Black was already doing very well in that game. 19. ♗f3 ♗e6 20. ♘e2 ♗xd5 21. exd5 ♘e3 22.♖d3 ♘c4 Zhao Xue-Lahno Jermuk FIDE WGP 2012 ) 18... dxe419. ♖xd8 ♕xd8=/∞ White now needs to play the very precise 20.♗d1 , quickly transferring the bishop to c2, from where it will cover the queenside quite sufficiently. Otherwise the position would be very uncomfortable for White, and I can understand why the Chinese player didn't go for the sacrifice if she hadn't seen Bd1. Now Black actually needs to be quite careful. 20... ♗e6 It's important to start with this move. ( 20... ♗xe5 doesn't seem to equalise after 21. ♘xe4 ♗f5 22. ♘f3 ♕d5 23. ♗c2+/= ) c21) 21. ♘xe4 ♖c8! Black gets his counterplay in as soon as possible and creates the huge threat of ...Qd3+ 22. ♘c3♕d3+ 23. ♔a1 ( 23. ♗c2 ♗xa2+ White is already in trouble here. 24. ♔xa2 ♕xc2∓ ) 23... ♗xe5 24. ♘ge2 b5 25.♘c1 ♕d4 26. ♗b3 b4 27. ♗xe6 bxc3 28. ♗xc8 cxb2+29. ♔b1 bxc1=Q+ 30. ♔xc1 , and here the machine suggests Black might even be slightly better after the surreal 30... h5⁉ , but for humans knowing 30...Qa1+ is a draw should be enough :D c22) 21. ♗c2 21... ♗xe5 Only now does Black take on e5. I think he's fine, as you can see in the analysis below. c221) 22. ♘xe4? ♕d5 c222) 22. ♗xe4 ♗xc3 23. bxc3 ♕d7 24. ♘f3 ♕a4 25.♕e2 ♖e8 26. ♗d3 ♗c4 27. ♕d2 ♕b5+ 28. ♔c2♗xd3+ 29. ♕xd3 ♖e2+ 30. ♘d2 ♕a4+ 31. ♔b1 ( 31.♔c1 ♖xg2 ) 31... ♖xg2∞ c223) 22. ♕xe4 22... ♗xc3 23. bxc3 ♕g5=/∞ ( 23...♕d2=/∞ )

14... ♘a5 Black's plan is once again quite obvious in this position. Although in some situations the knight will come to c4 you mainly want to challenge the white centre with ...e6. White of course wants to stop that. 14... ♘e5 is similar to the previous lines. 15. ♗e2 ( 15. h4 e6 16. h5exd5 17. hxg6 fxg6 is good for Black.; 15. ♗d4 e6 16. f4 ♘g4 )15... e6 ( 15... ♘ec4 16. ♗d4 e5 17. ♗c5+/= ) 16. ♗xb6 cxb6 17.f4 ♗h6 ( 17... ♘d7 18. dxe6 ♕xe6 19. ♗f3+/= ) 18. g4 ( 18. d6⁉b5 19. a3 f6 20. h4 ♘f7 21. h5→ ) 18... exd5 19. g5 ♗g7 20.fxe5 dxe4 21. ♖xd8 ♕xd8 22. ♗d1! , but here White is a bit more coordinated and there's no immediate equality for Black: 22... ♕d4 23.♗c2 ♕xe5 24. ♗xe4 ♗f5 25. ♕c2 ♗xe4 26. ♕xe4 ♕xe4+ 27.♘xe4 ♖e8 28. ♘g3 h6 ( 28... ♖e1+ 29. ♔c2 ♖a1 30. ♘1e2♖xa2 31. ♘c3+/= ) 29. gxh6 ♗xh6 30. ♘f3 ♖e3 31. ♖f1+/= 15. ♗f4 The main move that Black has to concern himself with. This creates an unpleasant threat to the c7-pawn which can't be met by any attractive-looking moves. 15. ♗xb6? cxb6 16. b4 seems to be a way for White to win a piece in two moves, but the position after 16... e6! 17. bxa5 exd5 is probably already better for Black. The king on b1 is completely unsafe and Black gets a lot of play for the piece. 18. ♘xd5 ♗e6−+ 15. h4 e6 a) 16. ♗g5 f6 17. ♗d2 ( 17. ♗f4 ♕f7 ) 17... ♘ac4 18. ♗xc4♘xc4 19. ♗f4 b5!∞ b) 16. ♗f4 16... ♕e7 17. ♗g5 f6 18. ♗e3 exd5 19. ♗xb6cxb6 20. ♘xd5 ♕f7 21. h5 ♗e6= was fine for Black in Rodshtein-Bok, Biel 2012 15... ♖d7 This is the move Black has to play. It looks a bit ugly but it'll work out in the end. 16. ♗h6 The normal move. 16. g3 It's important to note that you have to be prepared to sacrifice material in this line with Black, because in a lot of cases that's the only way to avoid landing in trouble. For instance, here Black needs to play16... c6! 17. ♗h3 cxd5 18. ♗xd7 ♗xd7=/∞ and the compensation for the exchange is at least enough for equality. 16... ♗h8 17. ♘ge2 17. h4 c6 18. h5 cxd5 19. hxg6 fxg6=/+ 17... c6 White now has time to play 18. ♘f4 , preventing Black from opening the centre because taking on d5 will see White land a huge knight on that square. Black instead has a very interesting and important resource: 18... ♖d6! This somewhat strange-looking move actually has a very sensible explanation - Black wants to play ...Bd7 and then ...Rc8, bringing his last two pieces into play. 18... ♕d8 19. ♗e2 is less clear. 19. b3 This move looks a bit ugly, but the queen needs some means of getting away from c1 when a rook lands on c8. This secures the e3-square for the queen as it prevents a knight coming to c4. 19. h4 ♗d7 20. h5 ♖c8 21. hxg6 fxg6

19. ♕e3 ♕d8 20. ♗g5 ♘ac4 21. ♗xc4 ♘xc4 22. ♕e2 ♗xc3 23.bxc3 ♘e5∞ 19... ♗d7 20. ♗e2 ♖c8 21. ♕e3 In this position Black is finally ready to start simplifications: 21... cxd5 22. ♘cxd5 ♘xd5 23. ♘xd5 ♗e6 24. ♖d3 ♗xd5 25.exd5 e6= Black will be in time to completely annihilate the pawn centre and should be fine.

10.Nb5 1. d4 ♘f6 2. c4 g6 3. f3 d5 4. cxd5 ♘xd5 5. e4 ♘b6 6. ♘c3♗g7 7. ♗e3 O-O 8. ♕d2 ♘c6 9. O-OO ♕d6 10. ♘b5 ♕d7 11.f4⁉ This was played against me in the Stavanger tournament by Wang Hao. It was a bit of a disaster for me as, just as many top chess players are prone to do these days, I pretty much forgot the first move of my opening preparation as I didn't expect to see this line. I had a complete blackout. 11. ♔b1 ♖d8 transposes to 10.Kb1. 11. h4 ♖d8 transposes. 11. ♘a3⁈ e5 12. d5 ♘d4 a) 13. f4 c5! 14. fxe5 ♕e7⇆

b) 13. ♔b1 c6 14. dxc6 bxc6 15. ♕a5 c5! ( 15... ♖b8 16. ♘e2c5 17. ♘xd4 cxd4 18. ♗d2 ♗b7 19. ♗d3 ♕e6 ; 15... ♕b7 16.♘e2 ♖b8 17. ♖d2 ♘d5 18. exd5 ♗f5+ 19. ♔a1 ♗c2 20.♕c3 cxd5 21. ♘xd4 ♖fc8 22. ♗c4 Ivanchuk-Sutovsky, Ningbo 2011 ) 16. ♘e2 ( 16. ♕xc5 ♖b8 ) 16... ♕e7 17. ♘xd4 exd4 18.♗d2 ♗e6→ c) 13. ♘e2 13... c5 ( 13... c6⁉ ) 14. ♔b1 ♕d6 15. ♘c3 a6∞ 11. ♗h6⁉ I played this against Fabiano Caruana myself. It's an interesting try, but by now it's pretty much closed because of the following line: 11...♗xh6 12. ♕xh6 a6 13. ♘c3 ♘xd4 Black has once again won the central pawn. 14. f4 a) 14... c5 gives White additional options after 15. h4⁉ , saving on the Nf3 move. Black will still have some problems to solve. ( 15. ♘f3f6 transposes ) a1) 15... f6 16. h5 a11) 16... ♕e8 a111) 17. hxg6 ♕xg6 18. ♕h2 ♗g4 19. f5 ♕g5+ ( 19...♕g7 20. ♗e2 ♘xe2+ 21. ♘gxe2 ) 20. ♔b1 ♖f7 21.♖d3 ♖g7 22. ♕c7 ♘c4 23. ♕xb7 ♖d8 24. ♕b3♘xb3 25. ♖xd8+ ♔f7 26. ♗xc4+ e6 27. axb3 ♔e7 28.♘h3 Grover-Konguvel, Kolkata 2012 a112) 17. ♖d3! 17... g5 18. fxg5 ♕f7 19. ♘ge2± ( 19.♘f3± ) a12) ♕g4 17. ♘f3 ( 17. hxg6 ♕xg6 18. ♕h2 transposes )

a121) 17... ♖d8 18. ♘xd4 ♖xd4 ( 18... cxd4 19.♗e2+− ) 19. ♗e2 ♕g3 20. ♗f3 ♖xd1+ 21. ♖xd1 ♗e622. hxg6 ♕xg6 23. ♕h4+/= ♕g7 24. ♕f2 ♘c4 25.♕xc5 ♘d6 26. ♔b1+/= a122) 17... g5 18. fxg5 ♘xf3 19. gxf6 ♕g5+ 20. ♕xg5+♘xg5 21. fxe7 ♖e8 22. ♖d6 ♘d7 23. ♘d5 , and White came out ahead: a1221) 23... ♘f7 24. ♖f6+/= ♘fe5 ( 24... b5 25.♖h3 ♘fe5 26. ♖g3+ ♔h8 27. ♖e6± ; 24... ♘g5 25.♖f5 ) 25. ♖e6 b5 26. ♗e2 ♗b7 27. h6± ♘g6 28.♗h5 ♗xd5 29. exd5 ♖xe7 30. ♗xg6 ♖xe6 31.dxe6 hxg6 32. exd7 ♖d8 33. ♖d1 ♔h7 34. a4♔xh6 35. a5± a1222) 23... ♘xe4 24. ♖e6 ♘g5 25. ♖e3 b5 26.h6± ♗b7 27. ♖h5 Hammer-Erdos, Porto Rio Hotel 2012 a2) 15... ♕c7 16. ♘f3± a3) 15... ♕c6 16. ♘f3 ( 16. h5 ♗g4 17. hxg6 ♕xg6 18.♖d3 f5 ) 16... ♕f6 ( 16... ♗g4 17. ♘g5 ) 17. e5 ( 17. h5⁉ )17... ♕g7 18. ♕xg7+ ♔xg7 19. ♘xd4 cxd4 20. ♖xd4 ♗e621. ♗e2 ♖ac8 22. ♔b1 ♖fd8 ( 22... ♘c4 23. ♘d5 ♗xd5 24. ♖xd5 ♘e3 25. ♖d7 ♖c2 26. ♗f3 ♖f2 27. ♗xb7 a5 28.♖c1 ♖xf4 29. ♖xe7 ♖xh4 30. a3 Mamedyarov-Petrosian, Antalya 2004) 23. ♖hd1 ♖xd4 24. ♖xd4 ♖c7= b) f6! It's important to start with this move. b1) 15. h4 e5 16. ♘f3 ( 16. h5 ♕g7 17. hxg6 hxg6 18. ♕h4♗e6 ) 16... ♕g7 ( 16... ♕e7 17. h5 g5 18. ♘xd4 exd4 19.♖xd4 ♗e6 20. ♖d3 ♖ad8 21. fxg5 ♖xd3 22. ♗xd3 fxg523. e5 ♗f5 24. ♘e4 ♗xe4 25. ♗xe4 ♖f2 26. ♖d1 ♕c5+ 27. ♔b1 ♖xb2+ 28. ♔a1 ♖xa2+ 29. ♔xa2 ♕c4+= ) 17.♕xg7+ ♔xg7 18. ♘xd4 exd4 19. ♖xd4 ♗e6 , and Black is fine. b2) 15. ♘f3 b21) 15... c5 16. h4 ♕e8 b211) 17. h5 ♗g4 18. ♘xd4 cxd4 19. hxg6 ( 19. ♖xd4gxh5 20. ♗e2 ♗xe2 21. ♘xe2 ♕g6 22. ♕xh5 ♖ac8+23. ♘c3 ♕xh5 24. ♖xh5 ♖f7= ) 19... ♕xg6 20. ♖xd4♖ac8 21. f5 ♕xh6+ 22. ♖xh6 ♖fd8 Jones-Aronian, London 2012 b212) 17. ♘xd4 17... cxd4 18. ♖xd4 ♗g4 19. f5 , and White retains some pressure. b22) 15... e5 After this move play becomes very forcing and in general Black is just fine. 16. fxe5 This is what I played against Caruana. ( 16. h4 is fine for Black after both Qe7 and 16...♕g7 ) 16... fxe5 17. ♘xe5 ♕d6 18. ♘f3 ( 18. ♘c4 ♕c5 19. ♕e3 ♘xc4 20. ♗xc4+ ♕xc4 21. ♖xd4 ♕c5 22. ♔b1♗e6⇆ ) 18... c5 In this position if Black is allowed to finish development by playing either Bg4 or Be6 his fantastic knight on d4 will ensure that

he's maybe even better, so White has to continue with very forcing play: 19. ♘g5 ♕e7 20. ♘d5♘xd5 21. ♗c4 This is a slight trap. b221) 21... ♕g7 is the most obvious move, but after 22.♕h4! ♔h8 White won't take on d5 but will instead start with23. ♖hf1 , which has already been seen in a game between Sanikidze and Negi. Although this position isn't yet all that bad for Black it's completely unnecessary for him to allow all this.23... ♗e6 ( 23... ♗g4 24. ♗xd5 ♗xd1 25. ♘f7+ ♔g8 26. ♔xd1+− ) 24. ♗xd5 ♗xd5 ( 24... ♗g8!∞ ) 25. exd5 ♖xf1 26. ♖xf1± Sanikidze-Negi, Baden-Baden Classic op 2013 b222) 21... ♔h8! This forces White to take with the pawn.22. exd5 ( 22. ♗xd5 runs into 22... ♕g7 and after 23.♕h4 Black has the very important resource 23... ♖f4 , driving the queen all the way back to e1, when Black is probably already better.) 22... ♕g7 Only now, forcing the queen trade. 23. ♕xg7+ ♔xg7 24. d6 This position was reached in my game against Fabiano. 24... b5 ( 24... ♗f5⁈was Caruana's somewhat inexplicable and imprecise move which even gave me some hopes of winning the game, though I failed to find a very strange-looking resource at one point and the game still ended in a draw. This shouldn't be repeated. 25. b4 b6 26. ♖he1+/= Svidler-Caruana, Thessaloniki GP 2013 ) 25. ♗d5 ♖a7= 26. ♖he1 ♖f5 b2221) 27. ♖e7+ ♖xe7 28. dxe7 ♘e2+ Starting with this move is a very important finesse. ( 28... ♖e5 runs into the very strong 29. ♗e6 when it seems as though White is just winning, as after 29... ♘e2+ White has the d2-square for the king. ) 29. ♔d2 doesn't work as ( 29.♔c2 Moving the king to a light square doesn't help either.29... ♖e5 30. ♗e6 Black can simply take, and after30... ♗xe6 31. e8=Q ♗f5+ he regains the queen on e8 and is completely winning. ) 29... ♖xd5+ 30. ♔xe2♖e5+ wins the pawn on e7 with check. b2222) 27. ♘e6+ As the alternatives fail White is forced to enter a completely drawn rook ending. 27... ♗xe6 28.♗xe6 ♘xe6 29. ♖xe6 ♖d7= When I was preparing for the game against Caruana I had to check that White has equality here, because it looks as though the d6-pawn might be in some danger, but White really isn't risking very much - though neither is he fighting for an advantage. 11... ♕e6 Black has to play this move, hitting the pawns on a2 and e4. 12. ♘c3

12. d5? ♕xe4 13. dxc6 ♗f5 14. ♗d3 ♕xc6+∓ is simply bad for White.

12... ♘c4

12... ♖d8 is less precise: 13. ♘f3 ♘c4 14. ♕e2 a) 14... ♘6a5 15. ♗f2 ♕f6 16. g3 ♕b6 ( 16... ♗g4 17. h3♗xf3 18. ♕xf3 ♕b6 19. b3± ; 16... ♕c6 17. b3 ♘a3 18. ♔b2a6 19. e5 ♘b5 20. ♖c1 ♗g4 21. ♗g2 ♕d7 22. ♘xb5 axb5 23.h3 +/= ) 17. ♘e5 ♘xe5 18. dxe5 ♖xd1+ 19. ♔xd1 ♕c6 20.♔c2 ♗e6 21. ♕b5 ♕xb5 22. ♗xb5+/= b) ♘xe3 15. ♕xe3 b1) 15... ♘a5⁉ b11) 16. ♔b1 ♕b6 b111) 17. ♘d5 ♕d6 18. ♕c3 ( 18. ♕e1 ♘c6 19. ♕h4♗e6 20. ♗c4 ♗xd5 21. ♗xd5 e6 22. ♗xc6 ♕xc6 23.♖he1 ♕b5= ) 18... ♘c6 19. ♗c4 ♗g4 20. ♘e5 ( 20.e5 ♕d7 21. ♘e3 ♗xf3 22. gxf3 ♘xd4 23. ♗d5 c6 24.♕xd4 e6 ) 20... ♗xd1 21. ♘xf7 ♔xf7 22. e5 ♕xd5 23.♖xd1 e6 24. ♗xd5 ♖xd5=/+ b112) 17. ♕e1 17... c6 ( 17... c5 18. dxc5 ♖xd1+ 19.♕xd1 ♕xc5 20. ♕d8+ ♗f8 21. ♘g5 e6 22. ♗b5 ♘c623. ♗xc6 bxc6 24. e5± ) 18. f5 gxf5 19. ♗d3→ b12) 16. d5 16... ♕b6 17. ♕xb6 cxb6 18. ♖d2 ♗xc3 19.bxc3 ♗d7 20. ♗d3 ♖ac8 21. ♔b2 ♖c5 b2) ♘b4 b21) 16. b3 ♕b6 17. a3 ♘c6 18. ♘a4 ♕a5 19. d5 ♘b8 (19... b5⁉ ) 20. ♕c5 ♕xc5+ 21. ♘xc5 c6 22. ♗c4 b6 23.dxc6 ♖xd1+ 24. ♖xd1 ♘xc6 25. ♗d5 ♗g4 = Mamedyarov-Negi, Gibraltar Masters 2012 b22) 16. ♔b1! 16... b5 transposes to Wang Hao-Svidler, Stavanger 2013, covered below. 13. ♕e2 ♘6a5 This is the correct move and leads to a lot of interesting and sharp tactical variations. I could include more computer-generated variations, as they're extremely important in positions like this which can't really be analysed effectively without the machine. However, in my experience understanding purely computergenerated lines is completely impossible unless you produce them on your own with the machine. There's no logic to it, and unless you participate in the creation of those variations it's impossible to memorise them or understand why things happen the way they happen. I suggest you watch the video, study this text and then if some positions catch your fancy and you're not quite sure why the lines say what they say check them by yourself, because that's the only way you can understand what's what in a position as complex and crazy as this one. You'll see why I describe it as crazy in a moment! 13... ♘xe3 is the move I played automatically against Wang Hao. 14.♕xe3 ♘b4 15. ♔b1 a) 15... c5 16. d5 ♕b6 17. a3 b) 15... ♖d8 was my choice 16. ♘f3 I had my first disappointment here.

b1) 16... c5 was my original plan. 17. d5 ♗xc3 A very nice geometrical idea. After 18. bxc3 Black can take on d5 with both the rook and knight and Black is just fine. I was very happy about this until I realised that White can instead play ( 18. ♗c4 A very beautiful move that I wanted to show you purely for its aesthetic value. Black is, to put it mildly, in a lot of trouble here. So I had to scrap this idea. ) b2) 16... b5 was the second choice I settled on, and also an ambitious idea. It pretty much loses, and I did lose the game in a very painful fashion, but there was a beautiful idea behind it. 17.a3 The main move. ( 17. ♗xb5 ♖b8 would of course give Black a lot of play on the queenside ) 17... a5 was my plan. Of course if Black is allowed to play Qb3 he'll have a huge attack, so ( 17...♘a6 An unfortunate necessity, to which Wang Hao responded 18.♗xb5 This is a position that's very painful even to look at. I didn't manage to save it despite trying my best for the next 40 or so moves. 18... ♕b6 19. ♘e5+− Wang Hao-Svidler, Stavanger 2013.) 18. d5 ♗xc3 19. axb4 This capture completely eluded me at first, and once again Black can just resign because of ( 19. ♗c4 is of course no longer possible after 16...b5, which was one of the reasons behind my move. ) 19... axb4 20. dxe6 ♖a1+ For a moment I was happy about this, but unfortunately White can simply play 21. ♔c2 and Black has to resign. ( 21. ♔xa1 ♖xd1+ 22.♔a2 ♗xe6+ is a very beautiful mating variation. ) c) c6⁉ would still have made it possible to get a reasonably playable position. 16. ♖e1⁉ prevents ...Nd5, but Black has gained an important tempo and can play ...a5 or ...b5 as if nothing much has happened. ( 16. a3 ♘d5 ; 16. ♘f3 is well met by 16... ♘d5 and Black has at least managed to solve one of the bigger problems of his position by exchanging off the knight on b4. Otherwise it would soon get hit. 17. ♘xd5 cxd5 18. e5 ♗d7 19. ♗d3 ♖ac8∞ ) 16...♖d8 17. ♘f3 ♕d6 18. a3 ( 18. ♖d1 ♗g4 19. ♗e2 ) 18... ♘a619. e5 ♗f5+ 20. ♗d3 ♕d7 , and Black is alive. 14. ♘f3 c5 The first important fork in the road. 14... b5 15. b3! ( 15. ♗f2 b4 16. ♘d5 ♗b7 ) 15... ♘xe3 16.♕xe3 is good for White. 15. e5 was played by Sergey Karjakin in his game against Anish Giri. 15. ♗f2 is a very important option for White, keeping the bishop alive and hoping to do something in the centre later on. White will also now have the threat of b3 in many positions as Black no longer has the option of ...Nxe3. 15... b5! After this move play becomes very tactical and there are a lot of lines where Black just leaves stuff hanging and goes for the throat. I'll give you an outline of what's going on, and as long as you're not worried about sacrificing one, two, three or it doesn't really matter how many pieces you should be fine here! a) 16. ♘d5 a1) 16... cxd4 17. ♘xd4 ( 17. ♗xd4 ♗b7 ) 17... ♕d7 18.♘b3 a2) ♗b7! is the move to play, completely ignoring the threat of Nc7. 17. b3 ( 17. ♘c7 is met by 17... ♕d6 18. ♘xa8 ♕xf4+ )17... ♗xd5 18. exd5 ♕xd5 19. bxc4 bxc4 In this position the weakness of the king on c1 far outweighs the sacrificed piece. This is at least fine for Black, and probably more. 20. dxc5 ( 20. ♘e5♖ab8 21. ♕f3 ♕e6∓ ) 20... ♕b7 21. ♗d4 ♖ab8∓

b) 16. b3 ♘a3 dxc5 is currently impossible because the c3-knight would be hanging, so White plays 17. ♘d5 but Black once again leaves everything hanging and plays 17... c4! b1) 18. b4 Black has many good choices here, but the one I like most is 18... ♗b7! just completely forgetting about the knight on a5. 19. bxa5 c3!∓ Gaining more space on the kingside before doing anything forced. White is in a lot of danger and Black is already a little better. b2) 18. ♗e1? cxb3 19. axb3 ( 19. ♗xa5 ♗b7∓ and once again White is a piece up but once the c-file is opened play becomes very easy for Black and his attack develops very naturally. White has to be extremely careful not to lose on the spot. ) 19... ♘xb3+ 20.♔b2 ♘c4+ 21. ♔xb3 ♗b7∓ b3) 18. ♘c7 ♕c6 is quite obviously unattractive for White.

b4) 18. ♕b2 This is perhaps the best of three unappealing options for White. 18... ♕xe4 19. ♕xa3 ♕xd5 20. ♕xa5 cxb3 It might be possible for White to hang on for a draw here, but he's not going to enjoy this position either. b41) 21. ♕xb5 b2+ 22. ♕xb2 ♗f5 23. ♖d3 ♖fc8+ 24.♔d1 a5 25. ♗e3 ( 25. ♕b3 ♕c6 ) 25... a4 26. ♕a3♖ab8∓ b42) 21. axb3 21... ♕xb3 22. ♔d2 ♗f5 23. ♕xb5 ♕c2+24. ♔e1 ♕c3+ 25. ♘d2 ♖fc8=/+

c) 16. ♔b1! The sanest and perhaps the safest move for White. 16...♖d8 ( 16... ♕a6⁉ is a decent alternative. ) 17. d5 ( 17. dxc5♗b7 is quite obviously unattractive. ) 17... ♕b6 18. ♕c2 ♖b8We've reached a very strange-looking position where Black currently has no direct threats, but White has none either, and I think generally speaking it's quite clear that Black started first and will be able to develop some sort of attack on the queenside reasonably soon. So I would much rather be Black than White here. 15... cxd4 15... f6 16. d5 ♕b6 17. e6 ♕b4 18. ♘e1 ♘xe3 19. ♕xe3 ♘c420. ♗xc4 ♕xc4 21. ♘c2± 16. ♘xd4 This was Karjakin's move and is the safer option. 16. ♗xd4 leads to a very sharp position where I think Black is fine. 16...b5 a) 17. b4 would simply be met byb 17... ♗b7! 18. bxa5 ♖ac8↑which seems to be very strong, going for the throat straightaway. b) 17. b3 b4 Once again - pieces only go forwards in this line! 18.♘b5 ( 18. ♘e4 ♘a3∞ ) 18... ♗d7! 19. ♘c7 ( 19. bxc4 is a line in which I want to show you what happens. 19... ♖fc8 20. ♘g5♕f5 21. c5 Otherwise the queenside will just completely collapse.21... a6 22. e6 axb5 23. exd7 ♗xd4 24. dxc8=Q+ ♖xc8 Black is a whole rook down, but after 25. ♘e4 ♕xf4+ 26. ♔c2 ♗g7∓ his attack on the dark squares is so strong that I think he might actually be better. ) 19... ♕c6 20. ♘xa8 ♘e3+ 21. ♔b2 Objectively speaking this line is slightly saner for White. The position is unclear.21... ♘xd1+ 22. ♕xd1 ♖d8 23. ♕d2 ( 23. ♗d3 ♗g4 ) 23...♕xa8 24. ♗c5 ♘c6 25. ♕e3 a5∞

c) 17. ♗c5 The best move for White, giving Black a choice. 17...♗h6 ( 17... ♗b7⁉ Black can also start with this move straightaway.18. ♘d4 ♕c8 19. ♗xe7 ♖e8 20. ♗b4 It seems as though White has everything under control but Black now plays 20... ♗h6⇆ and suddenly the central pawns are starting to fall. ) 18. ♔b1 ( 18.♘d5 ♗b7 19. ♘xe7+ ♔h8∓ ) 18... ♗b7 19. ♘d4 ( 19. b3♗xf4 20. bxc4 ♖fc8 21. ♗e3 ♕f5+ 22. ♔a1 ♘xc4↑ ) 19...♕c8 20. ♗xe7 ♖e8 21. ♗b4 ♗xf4 22. e6 fxe6 23. ♘dxb5 This is an extremely unclear position in which I think Black's chances are at least not worse. ( 23. b3 ♘e3∓ ) 23... a6 24. ♘d6 ♘xd6 25.♗xd6 ♗xd6 26. ♖xd6 ♕c5 27. ♕d2 ♖ac8⇆ 28. ♗e2 ♗xg229. ♖d1 ♘b7 30. ♖xa6 ♗h3 16... ♕a6! is an important improvement over 16... ♕b6 as played by Anish Giri. 17. ♘f5! ♘xe3 18. ♘xe3 ♘c6 19.♘ed5 ( 19. ♔b1⁉ ) 19... ♕d8 Anish equalised quite comfortably. 20.g3 ( 20. h3⁉ ; 20. g4⁉ ♗e6 21. h3 ♖c8 22. ♔b1 ♕a5 23. ♕b5♕xb5 24. ♗xb5 ♖fe8 25. ♖hf1 ♔f8= ; 20. ♘f6+ White isn't of course going to take the queen by playing this move, as after 20... exf621. ♖xd8 ♖xd8 the compensation is just too strong. ) 20... ♗e6 21.♔b1⁉ I think this is an important waiting move. ( 21. h4 ♖c8 22. h5♕a5 23. ♔b1 ♖fd8 24. ♗g2 ♖d7 ; 21. ♗g2 ♖c8 22. ♔b1 ♕a5Karjakin-Giri, Wijk aan Zee 2013 ) 21... ♕a5 ( 21... a6 22. h4 ♕a523. h5 ♖fd8 24. ♗g2 ♖d7 ) 22. ♕b5+/= An important resource for White, which wouldn't be an option after Karjakin's 21. Bg2. If Black takes White will recapture with the bishop and retain some pressure. 17. ♗f2 The main move, keeping the bishop alive. 17. ♘c2? ♗f5 18. b3 ( 18. ♗f2 ♖fd8 ) 18... ♖ac8 19. bxc4♘xc4 20. ♗d4 ( 20. ♖d4 b5−+ ) 20... ♖fd8−+ 17. ♘d5 is met by the quite futuristic-looking 17... ♖d8! 18. ♘c7 And now Black really needs to be able to show what he's doing... ( 18. ♘c6♘xc6 19. ♕xc4 ♕xc4+ 20. ♗xc4 ♗g4 21. ♖d2 ♖ac8=/+ ; 18.♘xe7+ is completely bad. 18... ♔h8! 19. ♘xc8 ♖axc8 20. ♔b1♕b6−+ ) 18... ♗g4‼ ...and this is it! Black already had a queen and a rook en prise and he replies by putting one more piece under attack. 19.♕xg4 Black doesn't play ...Nxe3 straightaway here, but first 19...♖ac8! It seems, once again, that Black is fine. It's very difficult to understand why things happen the way they do here, but I still think finding lines like these is very pleasing, even if you do it with the help of the machine - which most probably is what you'll need to do when analysing a position like this. 20. ♘c2 ( 20. ♔b1 ♕b6 ) 20... ♕c6∞ 17... ♖d8⁉ I think this is the strongest move. The position is too complex to cover in any great detail, but I want to show you the main line, which I find extremely beautiful. 17... b5 was supposed to be my original recommendation in this position. a) 18. ♔b1 ♖d8 19. g3 ( 19. g4 ♗b7 20. ♖g1 g5! ) 19... ♗b720. ♖g1 b4 21. ♘cb5 ♖ac8 22. b3 ♗d5↑ b) 18. ♖g1 ♖d8 19. g4 ♗b7 20. ♘cxb5 ( 20. ♘dxb5 ♘b3+ 21.♔c2 ♘bd2→ ) 20... ♖ac8→ c) 18. ♗h4 ♖e8!

d) 18. b3 18... ♘xb3+! Surprisingly, in about 5-6 moves this leads to a very strange, unbalanced endgame where Black has a rook and a pawn against two minor pieces. I think that endgame is holdable for Black, but it's not much fun. 19. axb3 ( 19. ♘xb3 ♗g4! 20. ♕c2♖ac8∓ ) 19... ♕a1+ ( 19... ♕a3+ 20. ♔c2 ) 20. ♔c2 ♕b2+ 21.♔d3 ♖d8! 22. ♕xb2 ( 22. bxc4 bxc4+ 23. ♔xc4 ♕a3!−+ ) 22...♘xb2+ 23. ♔c2 ( 23. ♔e3 b4 24. ♗e2 bxc3 25. ♖c1 f6⇆ )23... ♘xd1 24. ♘xd1 a6 25. ♘c3 ( 25. ♗e2 ♗b7 26. ♗f3♖ac8+ 27. ♔b2 ♗xf3 28. ♘xf3 ♖d3 29. ♖e1 a5⇆ ; 25. ♖g1f6 ) 25... ♗b7 ( 25... f6 26. ♗d3 fxe5 27. ♘c6 ♖d7 28.♘xe5 ♗xe5 29. fxe5 ♗b7 30. ♖d1 ♖c7 31. b4 ♗xg2 32.♗c5+/= ) 26. ♖g1 e6 ( 26... ♖ac8 27. b4 f6 28. ♘e6 ♗e4+ 29.♔b3 ♗d5+ 30. ♘xd5 ♖xd5 31. ♗c5+/= ) 27. ♗e2 ♖ac8 28.♔d3 ( 28. ♔b2 ♗h6 29. g3 ♗f8⇆ ) 28... f6 29. exf6 ♗xf6 30.♗g4 ♖c6 31. ♘e4 e5 32. ♗f3 ♗g7 33. fxe5 ♗xe5 18. g4 This isn't an attacking move as such, but White is preparing to play Bg2. He also takes control of the f5square to prevent ...Bf5 in some lines. 18. ♕f3 ♗h6! 19. ♔b1 ( 19. ♘d5 ♖xd5 20. ♕xd5 ♗g4! ) 19...♕b6 20. b3 ♕b4⇆

18. b4 is an attempt by White to win a piece straightaway, but after18... ♘b3+! 19. axb3 ♕a3+ 20. ♔c2 ♕b2+ 21. ♔d3 the bishop comes to f5 with great effect: 21... ♗f5+ 22. ♔xc4 ♖ac8+−+ 18... g5! Black responds by immediately hitting the white centre. Once again there's a lot that can happen here. 18... b5 allows White to play 19. ♗g2 with tempo. 19... ♗b7 20.♗xb7 ♕xb7 21. ♕e4 19. b4 This main line is the most beautiful and alien-looking of all the lines in this position. It turns out this move wins both knights. 19. b3 ♗xg4 20. ♕xg4 ♘xb3+ 21. ♔c2 ♘xd4+ 22. ♖xd4 ♘a3+23. ♔d2 ( 23. ♔b3 ♕b6+ 24. ♖b4 ♕xf2 25. ♔xa3 ♖ac8 ) 23...♕b6 19. ♘c2 a) 19... ♖xd1+ 20. ♕xd1 gxf4 is also possible, e.g. a1) 21. ♕d8+ ♗f8 22. b3 b5 23. ♕d4 ( 23. bxc4 ♗b7 ) 23...♗b7 24. ♖g1 ♘c6 25. ♕xf4 ♘4xe5 26. ♖g3 ( 26. ♗xb5♘d3+ ) 26... ♘g6=/+ a2) 21. b3 a21) 21... ♗xe5 22. ♘d5 ( 22. ♗d4 ) a22) 21... ♗e6 22. ♘d5 ( 22. bxc4 ♘c6 ; 22. ♘b4 ) 22...♗xe5 23. ♘xe7+ ♔f8 24. ♘f5 ♘c6 25. a4 ♖d8= b) ♗e6 20. f5 ♕c6 21. ♖g1 ♘xb2 22. ♔xb2 ( 22. ♖xd8+♖xd8 23. ♔xb2 ♘c4+ 24. ♔c1 ♘xe5∓ ) 22... ♗c4 23. ♕e4♕xe4 24. ♖xd8+ ♖xd8 25. ♘xe4 ♗xe5+ 26. ♔a3 ♗d5 27.♘g3 ♘c4+ 28. ♗xc4 ♗xc4 29. ♔b4 ♗xa2=/∞ 19... gxf4 20. bxa5 ♕xa5 Everything is left hanging. 21. ♕xc4 ♗xg4 Black is currently two pieces down but he's threatening the rook on d1 and ...Rac8 is also a huge threat.

22. ♖d3 ♖ac8 23. ♕b3 ♗f5 Once again hitting the rook, and giving White a choice. 24. ♖f3 This is the line I wanted to show you, because I think it's beautiful and extremely pleasing to the eye. 24. ♔d2 ♗xd3 25. ♗xd3 a) 25... ♗xe5 is the machine's recommendation in this position, and then 26. ♖g1+ ( 26. ♗xh7+ ♔h8−+ ) 26... ♗g7 Self-pinning the bishop. Black has a rook and four pawns for three minor pieces, and the machine says everything is fine. I spent quite some time just staring at this position without trying to make any moves, attempting to understand anything at all. I completely failed so I can't really recommend this to you. b) ♕xe5 is the much more human solution. 26. ♕b5 More or less forced for White, after which we get an endgame: 26... ♖xd4 27.♕xe5 ♖xd3+ 28. ♔xd3 ♗xe5 29. ♘d5 Unfortunately Black isn't better because of this very important move, hitting the e7-square. After 29... ♔f8 White has 30. ♗xa7 but Black is never worse here and will at the very least make a comfortable draw. 30... e6= 24... ♗e4 25. ♔b2 ♕xe5 25... ♗xf3 26. ♘xf3 is quite obviously wrong as it allows White some semblance of control. 26. ♘xe4 Everything was hanging for White so he has to play this move and start destroying at least some of the enemy combatants. 26... ♖xd4 27. ♗xd4 ♕xd4+ 28. ♘c3 It looks as though White has managed to consolidate, and he's currently a rook and a piece up, but... 28. ♔a3 ♕xe4 29. ♖g1 ( 29. ♗d3 ♕e5−+ ) 29... ♖c5∓

28... ♕e4 Quietly hitting the rook on f3 and setting up pins along both long diagonals. 29. ♕d1 Forced, if White doesn't want to be worse. 29... ♕b4+ 30. ♕b3 Strictly the only move. 30... ♕e4= And this seems to be a forced draw. Once again, I'm not claiming to have given you an entire picture of all the crazy complications which can happen in this line, but I think it creates a pretty decent overview of what you should be aiming for when playing with the black pieces. I hope you've enjoyed looking at these lines - I definitely enjoyed finding them and presenting them to you.

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 7.Bc4 c5 8.Ne2 Nc6 9.Be3 0-0 10.0-0 Bg4 10...Bd7 11.Rc1 11.Rb1

10...Bg4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗c4 I'm now going to cover the Bc4 Grünfeld, a line which has been universally popular over the course of the existence of the Grünfeld Defence. The play here is very interesting and I've always enjoyed looking at this line even if I haven't always enjoyed the results of my analysis. This time I want to do something slightly different. Although there will be quite a lot of actual theory I want to give you an idea of how my views on the line have evolved and how my attitude towards certain positions and how to correctly deal with them has changed. 7... c5 8. ♘e2 ♘c6 9. ♗e3 All of this is quite forced. 9... O-O 9... cxd4 10. cxd4 ♕a5+ 11. ♗d2 ♕d8 This is quite a fancy line that I've tried once or twice against decent opposition. It looks a bit odd, frankly, but the point is that if White doesn't want an immediate repetition - and normally he doesn't - he has to play 12.Bc3. a) 12. d5⁉ ♘e5 13. ♗c3 OO ( 13... ♗d7 14. ♗b3 ♖c8 15.♖c1 ♕b6 16. ♗a4 OO 17. ♗xd7 ♘xd7 18. ♗xg7 ♔xg7 19.O-O ♘f6 Topalov-Mamedyarov, Heraklion 2007 ) 14. ♗b3 ♕b6 15.f4 ♘d7 ( 15... ♘g4⁈ 16. ♗d4 ♕a5+ 17. ♕d2 ♕xd2+ 18.♔xd2 e5 19. h3 !± Kramnik-Svidler, Dortmund 1998 ) 16. ♗xg7♔xg7 17. ♖c1 ♖d8 18. ♖c4 e5 19. ♕a1 f6 20. ♕c3 exf4 21.♘xf4 ♘e5 22. ♖c7+ ♔h8 23. ♖f1 ♗d7 24. ♘e6 ♖dc8 25.♖xf6 ♖xc7 26. ♖f8+ ♖xf8 27. ♕xe5+ Peter Heine Nielsen-McShane, Copenhagen 2003 b) 12. ♗c3 Putting the dark-squared bishop on an unaccustomed diagonal. Play becomes very non-standard and fresh, but I don't think this fully equalises, so I'm not recommending it. 12... O-O 13. O-O (13. ♕d2 ♗d7 14. h4 ♖c8 15. ♗b3 h5 16. f3 e6 17. ♖d1♕e7∞ RadjabovSvidler, Plovdiv 2003 ) 13... ♗g4 ( 13... ♗d7 14.♖b1 b6 15. ♕d2 ♖c8 16. ♖bc1 e6 17. h3 ♖c7 18. ♗b3 ♗c8 19. d5 exd5 20. ♗xg7 ♔xg7 21. exd5 ♘e5 22. d6 Le Quang Liem-Mamedyarov, Khanty-Mansiysk 2013 ) 14. f3 ♗d7 15. ♖b1 (15. ♕d2 a6 16. a4 ♖c8 17. ♖fc1 b5 18. axb5 axb5 19. ♗xb5♘xd4= Van Wely-Mamedyarov, Hoogeveen 2007 ) 15... ♕c7 16.♕d2 a6 17. ♗b3 b5 18. ♔h1 ♕b6 19. f4 ♖ac8 20. f5 b4 21.♗a1 ♘a5⇆ Beliavsky-Svidler, Madrid 1998

10. O-O The main position of the Bc4 line, and one in which Black has a variety of choices.

10. h4 I recommend meeting this advance with 10... ♕a5 11. ♔f1 b612. h5 ♗a6 Grandmaster Ioannis Nikolaidis has played a number of games here with the white pieces, and I believe they sufficiently cover everything that's going on in the position. 13. ♗xa6 ♕xa6 14. hxg6hxg6 15. ♔g1 ( 15. f3 cxd4 16. cxd4 ♖ac8 17. ♔f2 ♘b4 18.♖b1 ♖c4 Murey-Dvoirys, Cappelle la Grande 1999 ) 15... ♖fd8 16.♕d2 ♕a4 17. ♗h6 ♗h8 18. ♗f8! ( 18. ♕e3? cxd4 19. ♕h3 dxc320. ♗d2 ♗e5 21. ♕h7+ ♔f8 22. ♘xc3 ♕c2 23. f4 ♖xd2Nikolaidis-Ivanchuk, Peristeri 2010 ) 18... ♗f6! 19. ♕h6 cxd4 The play that follows is almost forced: 20. f4 dxc3 21. e5 ♘xe5 22. fxe5♗xe5 23. ♗xe7 ♖d1+ 24. ♖xd1 ♕xd1+ 25. ♔f2 ♗d4+ , with a draw in Nikolaidis-Banikas, Vrahati 60th Greek Championship 2010 10. ♖c1 This is a more serious alternative to 10.00. 10... cxd4 11.cxd4 ♕a5+ 12. ♔f1 ( 12. ♗d2 ♕h5 13. d5 ♘e5 14. ♕b3 ♘xc415. ♖xc4 ♕g4 16. ♘g3 b6 is fine. ) 12... ♕a3! This was an important discovery for Black. He wants to play ...Na5, attacking the pride and joy of White's position - the bishop on c4. ( 12... ♗d7 , and then for instance 13. h4 , used to be where the theory of this line would exclusively start. This position is very sharp and very unclear, and I think it appealed to white players a lot because the course of the game is very different to the normal Grünfeld fare. ) a) 13. ♕d2 is the start of a very beautiful line I want to show you.13... ♖d8 14. d5 ♘e5 15. ♗b5 Seemingly creating the very strong threat of Bc5, but Black can just play 15... a6 as after 16. ♗c5 he has 16... ♕f3! I think each of us would like to be able to make a move like this on the board once in our career! 17. ♗xe7 ( 17. gxf3is met by 17... ♗h3+ and Black gets all his material back with interest, so this line isn't very critical. ) 17... ♕xe4 18. ♗xd8♗h3 ( 18... axb5⁉ ) b) 13. h4 reveals the other point behind 12...Qa3. Black now has the very strong 13... ♗g4 as 14. f3? is impossible due to the hanging bishop: 14... ♕xe3 c) 13. ♖c3 White's normal response. 13... ♕d6 14. h4 Black should simply meet this with 14... h5 ( 14... ♖d8 is a move I played against Morozevich in the Astana World Blitz Championship, but after 15. h5 I "failed to control the vehicle" and lost a very pretty miniature. ) 15.f3 ♖d8 16. ♗d5 ( 16. ♔f2 ♘xd4 17. ♘xd4 ♗xd4 18. ♕xd4♕xd4 19. ♗xd4 ♖xd4 20. ♔e3 ♖d8= ) 16... e5! 17. ♗g5♗f6⇆ The play is very complex here, but I believe Black should be fine. 10... ♗g4! Black can, and by now I think should, play this first rather than 10...Bd7. 10... ♘a5 11. ♗d3 b6⁉ I'm proud to say I was the original inventor of this line, mainly due to absolutely murderous jetlag. This was a game I played against Veselin Topalov in the Morelia/Linares tournament, split between Mexico and Spain. It was the Linares leg, so you can imagine that it wasn't even the first jetlag but the second, and it was absolutely unbearable. I couldn't sleep at all and felt completely exhausted, so the idea of preparing for games just didn't appeal to me. I'd turn up at the board and then tell myself, "ok, this is a position where I have to find a move - let's find something." So when Topalov played the Classical Grünfeld against me, as he'd done on a number of previous occasions, I thought, "ok, he probably has something prepared against the lines I played in all those previous encounters, so let's create something new". That's why I played 10...Na5 and then 11...b6, with the idea of continuing12. ♖c1 e5 Since that game this line has exploded to become one of the main variations in the Grünfeld. There have been many crucial developments, it was played in the World Championship match between Topalov and Anand as well as in a number of very high level games outside the World Championship cycle, but I left it altogether after that first game. I thought, "ok, I created this and it now has a life of its own, but it's not much fun for me to join the "clicking party" when everybody's doing it." So it's now living an absolutely separate life without me.

10... b6⁉ I played this against Magnus Carlsen in the 2011 Tal Memorial, and I still believe it's very playable. 11. dxc5 ♕c7 12. ♘d4 The reason I'm not recommending this line, or going into any detail, is that after this move I believe Black holds, but what he holds is one of a number of endgames a pawn down (White even has a choice). Frankly, the idea of holding an endgame a pawn down, even if you do truly believe it's holdable, isn't to everybody's taste. It's ok to do it every now and again, especially in a game where you're not favourite to win to begin with, as was the case in my game against Carlsen (against him a passive position a pawn down also isn't an ideal solution, but there you go!), but in a normal game going for an endgame like this isn't everybody's cup of tea. ( 12. f4Carlsen was soon fighting for equality after this move, although the game was eventually drawn. 12... bxc5 13. ♖b1 ♖d8 14. ♕a4 ♘a515. ♗d5 ♗d7 16. ♕a3 ♖ac8 17. f5 e6 18. ♗f4 ♗e5 19. fxe6fxe6 20. ♗b3 ♘xb3 21. axb3 ♖f8 Carlsen-Svidler, Moscow Tal Memorial 2011 ) 12... ♘e5 13. ♘b5 ♕b8 ( 13... ♕c6 14. ♗d5♕xb5 15. ♖b1 ♕a5 16. ♗xa8 ♗a6 17. cxb6 axb6 18. ♗xb6♕xc3 19. ♗d5 ♗xf1 20. ♕xf1 e6=/∞ Peralta-Alsina Leal, Barcelona 2011 ) 14. ♗e2 ( 14. ♗d5 ♘g4 15. g3 ♘xe3 16. fxe3 a6 17.♗xf7+ ♔h8 18. ♘d4 bxc5 19. ♗d5 ♖xf1+ 20. ♕xf1 cxd4 21.♖b1 ♕a7 22. ♕f7 ♗e6! 23. ♕xe6 dxc3 24. ♗xa8 ♕xe3+=Vachier Lagrave-Gelfand, Tromso World Cup 2013 ) a) 14... a6 15. ♘a3 bxc5 ( 15... b5 16. ♘c2 ♘c6 17. ♘d4♗b7 18. f3 ♖d8 19. ♕c1 ♘xd4 20. cxd4 ♗xd4 21. ♗xd4♖xd4 22. c6 ♗c8 23. ♕c5 ♖d2 24. ♖f2 ♕a7 Kulaots-So, Khanty Mansiysk 2010 ) 16. ♗xc5 ♕c7 17. ♗d4 ♖d8 18. ♕c1♘g4 19. ♗xg4 ♗xg4 20. ♕e3 ♖ab8 21. ♗xg7 ♔xg7 22. ♘c2♖bc8=/+ Bacrot-Vachier Lagrave, Biel 2013 b) 14... bxc5⁉ 15. f4 ( 15. ♖b1 a6 16. ♘d4 ♕c7 17. ♘b3 ♖d818. ♕c2 c4 19. ♘c5 ♘d3 20. ♘xd3 cxd3 21. ♗xd3 ♕xc3 22.♕xc3 ♗xc3 23. ♗c4 ♗d7 Ganguly-Giri, Spanish Team Championship 2012 ) 15... ♘g4 16. ♗xc5 a6 17. ♘a3 ♕c7 18.♗d4 e5 19. fxe5 ♘xe5 20. ♕c1 ♗g4 21. ♗xg4 ♘xg4 22.♕f4 ♕xf4 23. ♖xf4 ♘e5=/∞ Aronian-Grischuk, Kazan Candidates 2011 11. f3 ♗d7 Only now. I was of course aware of this possibility for a number of years, or in fact pretty much for the entirety of my Grünfeld career, but it always seemed to me that this was just a waste of a tempo - why would you force White to make the move f2-f3, which he generally likes to make in this kind of structure? f3 supports the centre and it just seemed somewhat illogical to me to be forcing White into a move which really doesn't seem to have any drawbacks. However, I think it was to my detriment that I never carefully considered the implications of forcing White to play f3. I'll try to explain why that move makes quite a large difference in some lines, but I'll do it by first concentrating on 10...Bd7 and showing you how the understanding of this line developed over the years. 11... ♘a5 leads to very forcing and well-researched lines after both 12.♗xf7+ and ( 12. ♗d3 cxd4 13. cxd4 ♗e6 14. d5 ♗xa1 15. ♕xa1This position has been the subject of very high level theoretical discussions for "donkey's years". I believe it's currently supposed to be a draw, but it never really appealed to me that much because frankly, unless White makes a huge mistake somewhere, there's never any suggestion that Black might be better. White always has enough compensation for some kind of equality or, simply put, a draw. The idea of having to memorise 35-40 move lines to make a forced draw and god forbid you forget something or go mildly wrong in your preparation and simply die without making a move of your own at the board - has never been my idea of fun. So I won't be recommending this line to you, although just for general knowledge you might like to browse through the high level games played in this line over the past ten years. There are a lot of very beautiful variations. )

10...Bd7 11.Rc1 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗c4 c5 8. ♘e2 ♘c6 9. ♗e3 OO 10. O-O ♗d7 This is the move I've generally made in this position, and the move I almost, but not quite, recommend making. What I mean by "almost, but not quite" is that as I said above, Black can, and by now perhaps should, play 10...Bg4 first. 11. ♖c1⁉ This is the move people would just automatically play here when I started playing this position with Black. This is how you were brought up, so to speak - you develop the bishop to e3 in this line and then put the rook on c1, supporting the bishop on c4 in case Black takes on d4. In general, it just looks very natural and harmonious, so why not? 11. h4 b5 12. ♗xb5 ( 12. ♗d3 cxd4 13. cxd4 ♘b4 14. ♗b1 ♗g415. f3 ♗e6 16. a3 ♘c6 ) 12... ♘xd4 13. ♗xd7 ♘xe2+ 14. ♕xe2♕xd7 15. ♕c4 ♖fd8= 11. ♕d2 ♖c8 12. ♖ac1 transposes. 11... ♖c8 12. f4 This move, played by Alexander Beliavsky in a game against Sedlak, is the only one with any kind of separate meaning in this position. 12. ♕d2 White players would also play this on the same general grounds - this is how you complete your development in this line. However, in this particular position Black responds with 12... ♕a5 , already threatening ...cxd4 and liquidating into a very good endgame. a) 13. d5 occurred in one of the very first games I played in this line, in a junior tournament more than 20 years ago. 13... ♘e5 14. ♗b3 I want to draw your attention to an important motif, which is that in this line you almost never want to play ...c5-c4. Instead you generally want to try and put the knight on c4, so here Black plays 14... ♕a6The threat of ...Nc4 creates immediate problems for White. He has to find squares for all his pieces and may be forced to part with one of his bishops. With a knight on c4 it will also be much easier for Black to play for the ...e6 break, because the bishop on b3 is somewhat neutralised. Finally, it's important to note that if you manage to force White to play c4 himself in a position like this then b5 will become a huge factor and Black will get tremendous counterplay. a1) 15. c4 b5 16. ♗h6 bxc4 ( 16... ♗xh6 17. ♕xh6 bxc4 18.♗c2 ♖b8 19. h3 ♖b2 20. ♘g3 ♕a3 21. ♔h2 f6 22. f4♘f7 23. ♕h4 ♕c3 24. ♖f2 ♗a4 25. ♘e2 ♕e3 26. ♗xa4♖xe2 Golovlev-Areshchenko, Alushta Ukrainian Championship 2008) 17. ♗xg7 ♔xg7 18. ♗c2 ♘d3 19. ♖b1 ♖b8 20. ♖fd1♔g8 21. ♘c3 ♖xb1 22. ♖xb1 ♘b4 23. h3 ♕a5 24. ♗d1♘xa2 25. ♕xa2 ♕xc3 Komljenovic-Vorobiov, Cappelle la Grande 2003 a2) 15. ♗g5 15... e6 16. dxe6 ♗xe6 17. ♗d5 ♗xd5 18.exd5 ♖fe8 19. ♘g3 ♘d3 20. ♖b1 c4 21. a4 ♖c5 22. d6♕xd6 23. ♖xb7 ♕d5 Georgiev-Ftacnik, Bled 2002 b) 13. ♕b2 This is possible, and what Michael Prusikin played against me. 13... b6 After this move I believe Black is already quite comfortable and I even managed to win the game I mentioned. 14. f4e6 15. ♖fd1 ( 15. ♘g3 cxd4 16. cxd4 ♖c7 17. e5 ♖fc8 18.♗d2 ♘b4 19. ♘e4 ♖xc4 20. ♖xc4 ♖xc4 21. ♘d6 ♖c7 22.♗xb4 ♕d5 Kaminik-Mikhalevski, Givatayim 2004 ) 15... ♖fd8 16.dxc5 ♗f8 17. ♕b5 ♗xc5 18. ♗xc5 bxc5 19. ♖b1 ♗e8Prusikin-Svidler, Bundesliga 2002

12... ♗g4 The correct move. 12... cxd4 is a forcing alternative for Black. After 13. cxd4 ♗g4 14.d5 he has the very nice geometrical move 14... ♘d4 , which after 15.♗xd4 ♗xe2 16. ♕xe2 ♗xd4+ leads to simplifications... but not equality. This is the kind of position you want to avoid, because although Black has managed to exchange quite a lot of pieces those that remain on the board clearly favour White. The bishop will be very strong on b3 and White will have a serious initiative both in the centre and potentially on the kingside. This just isn't equal and can't be recommended at all. 17.♔h1 ♕a5 18. e5 ♖fd8 19. ♗b3 ♔g7 20. ♖cd1 13. d5 ♘a5 14. ♗d3 e6 Black immediately breaks in the centre. 14... c4 15. ♗b1 e5 ( 15... e6 16. h3 ♗xe2 17. ♕xe2 exd5 18.♖fd1 d4 19. cxd4 ) 16. h3 ♗xe2 17. ♕xe2 b6 18. f5 ♘b7 19. g4g5 20. h4 f6 21. hxg5 fxg5 22. ♔g2 h6 23. ♖h1 ♕f6 24. ♖h5♘d6 25. ♖ch1 Tikkane n-Novotny, Ceska Trebova RC Sport op 2008 15. h3 The main move. 15. c4 b5 As a general rule if you play ...e6 in this system, attacking the d5-pawn, and White replies with c4, it's almost always necessary to play ...b5 straightaway. You can't allow White to keep his beautiful centre for any longer than absolutely necessary. That also holds true here. White might continue: 16. h3 bxc4 17. hxg4 cxd3 18. ♕xd3 exd5 19.♕xd5 ♕xd5 20. exd5 ♖fd8 21. ♖fd1 a) 21... c4 22. ♗xa7 ♖d6 ( 22... c3 23. ♗b6 ) 23. ♘c3 b) ♘b7 You could perhaps argue that White has a slight advantage, but Black's defensive resources should be more than enough to make a draw. 15... ♗xe2 16. ♗xe2 exd5 17. exd5 17. ♕xd5⁉ is best met by 17... ♕xd5 18. exd5 ♖fe8 19. ♔f2 If White could make one more move he would probably be better, but Black is just in time to play the very strong 19... ♖e4 , creating the threat of ...Rce8 and also with the idea of ...Nc4. If White drives the e4-rook away it will go to a4, which is obviously a very good square for it. Black has enough counterplay. 17... ♖e8 18. ♕d2 18. ♖f3 ♕d6 19. ♗f2 a) 19... ♖cd8 20. ♗b5 ♖e4 21. ♕c2 f5 ( 21... ♖xf4 22. ♗g3♖xf3 23. ♗xd6 ♖xc3 24. ♕d2 ♖xc1+ 25. ♕xc1 ♗d4+ 26.♔h1 ♖xd6 27. ♕e1 b6 28. ♕e7 ♖xd5 29. ♕xa7 ♔g7 30.♕xb6 ♗c3+/= ) 22. ♗d3 ♖e7 23. c4 ( 23. ♖d1 c4 24. ♗f1 b625. ♕d2 ♗f6 26. g3 ♖de8 BeliavskySedlak, Murska Sobota SLO-tch 2008 ) 23... b6 ( 23... ♗d4 24. g4 ) 24. g4 fxg4 25. hxg4 b) c4 20. ♕d2 ♖e4 21. g3 b5 22. ♖e3 ♖xe3 23. ♗xe3 ♖e824. ♗f3 ♘b7 25. ♗f2 ( 25. ♕f2 ; 25. ♔g2 ♘c5 26. ♗xc5♕xc5 27. d6 ♖d8 28. d7 ♔f8 ) 25... ♘c5 26. ♖e1 ♖xe1+ 27.♕xe1 ♕f8 28. d6 ♕xd6 29. ♕e8+ ♗f8 30. ♕xb5 ♕d3 18... ♕d6 It's very important Black doesn't allow d5-d6. 18... ♕e7 19. ♖f3 ♕e4 20. f5 19. f5 If Black had been given free rein here he might have considered playing ...f5 himself, boxing in both of White's bishops.

19... ♕e5 19... ♖e5 20. ♖cd1 20. ♖f3 I believe Black can force equality after both 20...Rcd8 and 20...c4. I'll give you some sample lines - these aren't completely forced, but it's important that Black plays very forcefully. White has a potentially unpleasant initiative, so you need to be very concrete. 20... ♖cd8 20... c4 21. ♖d1 ♕xc3 22. ♕xc3 ♗xc3 23. d6 ♖cd8 24. d7♖e7 25. ♗xa7 The bishop on c3 is hanging and Bb6 is a big idea, but Black can calmly reply 25... ♘c6 , intending to meet Bb6 with ...Bd4+. He has easy equality here. 21. fxg6 hxg6 22. ♖d1 22. c4 would give Black the very important resource 22... ♘c6 , when the knight triumphantly returns to the action. 22... ♕xc3 23. ♕xc3 ♗xc3 24. ♗xc5 ♖xe2 25. ♖xc3 ♖e5= Black is just in time to attack the central pawn on d5. White will find defending it awkward so the most likely result is a draw.

10...Bd7 11.Rb1 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗c4 c5 8. ♘e2 ♘c6 9. ♗e3 OO 10. O-O ♗d7 11.♖b1 This is the main move, and one you're very likely to face if you play the ...Bd7 systems. 11... ♕c7 I think Black has to play this, after which White has at the very least seven possible moves. 11... a6 was what I used to play here for a number of years. The move has a lot of merits, but also two major drawbacks. 12. d5 The more important one for me is that this move is a huge problem. ( 12. dxc5⁉leads to very concrete play where I'm not sure Black gets full equality, although personally I don't mind giving away material. ) 12... ♘e5 13.♗b3 This position is very unpleasant. As I said before, it's normally not to Black's benefit to play ...c5-c4, so he tries to avoid it here, but after13... ♕c7 14. f4 ♘g4 15. ♗c1 it transpires that in order not to be driven off the board completely after h3 and d5 Black still has to play15... c4 , and you end up with a position like the one after, for instance,16. ♗c2 e6 17. h3 ♘f6 18. dxe6 ♗xe6 19. ♘d4 I think you have to try and avoid positions like this at all costs in this system. White will have a tremendous attack on the kingside in most cases, and it will be very, very fast. There's not a lot of counterplay for Black so White can develop his initiative on the kingside completely unimpeded. 19... ♖ad8 20.♗e3 ♖fe8 21. ♕f3 ♗c8 Eljanov-Belov, bereg.ru Internet 2003 22.f5!→ 12. ♘f4 This used to if not worry me, then at least bug me in the old days. 12. ♕d2 a) 12... a6 13. dxc5⁉ a1) 13... ♘a5 14. ♗d5 e6 ( 14... ♗b5 15. ♗d4 e6 16.♗xg7 ♔xg7 17. c4 ) 15. ♗f4 ♕xc5 16. ♗xb7 ♘xb7 17.♕xd7

a2) ♘e5 14. ♗d5+/= b) ♖ad8⁉ b1) 13. ♗f4 e5 14. ♗g5 ( 14. ♗g3 cxd4 15. cxd4 ♘xd4 )14... exd4 15. ♗xd8 ♖xd8 16. cxd4 ♘xd4 17. ♘xd4 ♗xd418. ♕g5 ♗c6 b2) 13. ♗d3 This is the main and correct move for White here, but it transposes to 12.Bd3, which we'll cover later. b3) 13. ♗h6 This move would make a lot of sense for White positionally, but it runs afoul of some tactics. 13... ♗xh6 14.♕xh6 cxd4 15. cxd4 ♗g4 A very strong move, creating the huge threat of ...Nxd4. 16. ♘f4 ( 16. d5 ♘e5 ) 16... ♘xd4 17.♖fc1 ( 17. ♘xg6 runs into the tremendously strong resource 17...♕d6 , when White loses material. ) 17... ♕e5 18. ♖xb7 ♕xe4 12. dxc5 is simply met by 12... ♘a5 ( 12... ♖fd8⁉ 13. ♘d4 ♗e8 14.♘b5 ♕c8 15. ♕c1 ♘e5 16. ♗d5 ♗c6 17. ♖d1 a6=/∞ Fores tier-Karavade, Arinsal 2009 ) a) 13. ♗b5 ♗xb5 ( 13... ♖fd8 14. ♕a4± ) 14. ♖xb5 ♘c4 15.♗d4 ( 15. ♕d3 ♘xe3 16. ♕xe3 ♖fc8=/∞ ) 15... e5 16. ♕d3a6 17. ♖b4 ♘a3 18. ♖b3 exd4 19. ♖xa3 dxc3 20. ♘xc3♕xc5 21. ♘d5 f5 Georgiev-Predojevic, Zlatibor SCG-tch 2006 b) 13. ♗d3 13... ♖fd8 It's important to note that compared to the 11...a6 line Black is very far ahead here because he made the very useful move ...Qc7 instead of ...a6, and ...a6 frankly has negative value. In this position with a pawn on a6 on each and every move Black would have to calculate the idea of c6 followed by Bb6, when White generally wins material. I believe this position should be quite safe for Black because he has the very simple plan of ...e6, ...Bf8 and ...Bxc5, which is surprisingly difficult to counter. So this line also isn't very critical. b1) 14. ♘d4 e6 ( 14... e5⁈ 15. ♘b3 ♗b5 16. ♗xb5 ♖xd117. ♖fxd1 Kouatly-Ftacnik, Lyon 1988 ) b2) 14. ♕c2 14... e6 ( 14... ♗e6 15. ♘f4 ) 15. ♘d4 ♗f8 16.♗g5 ( 16. ♘b5 ♗xb5 17. ♖xb5 a6 18. ♖bb1 ♗xc5 19.♗g5 ♗e7 ) 16... ♖dc8 12. ♕c1 This was played in a very high-level game between Korobov and Areshchenko some years back. a) 12... cxd4! 13. cxd4 ♖ac8 14. ♗d3+/= b) 12... a6 13. ♗f4 ( 13. dxc5 ♘e5 14. ♗d5 ♖ab8 Preparing to play ...e6 and drive the white pieces back. 15. f4 ♘g4 16. ♗d4♗b5 Black had terrific play for the pawn in this position. None of this is particularly forced, but I just wanted to show you what good chess players thought about this position. 17. ♖xb5 axb5 18. e5♘h6 19. ♘g3 ♖fd8 20. ♗b3 b4 Korobov-Areshchenko, Aarhus 2005 ) 13... e5 ( 13... ♕c8 ) 14. dxe5 ♘xe5 15. ♗d5

c) 12... ♖ac8 This is also possible. 13. ♖d1 ( 13. ♗b3 cxd4 14.cxd4 ♘a5 15. ♕a3 ♘xb3 16. ♖xb3 b6 ; 13. ♕a3 ) 13... cxd414. cxd4 a6 Black plans to expand on the queenside. Generally speaking, this position shouldn't be overly problematic for Black. White's play is too vague to create too much uncertainty for Black. (14... ♘a5 15. ♗d3 ♕xc1 16. ♖dxc1 ) 15. ♘f4 b5 16. ♗b3♕b8 ( 16... ♕a5 17. ♕b2 ♖fd8 18. h3 ) 17. ♕d2 ♖fd8 18.♘e2 12. f4 This idea isn't particularly threatening and is by far the easiest move to face. Black simply plays 12... cxd4 13. cxd4 ♘xd4 14.♗xf7+ White's normal move. ( 14. ♗xd4 doesn't win a piece, because after 14... ♕xc4 15. ♗xg7 ♔xg7 16. ♕xd7 Black has 16... ♕xe2 )14... ♖xf7 15. ♗xd4 e5 ( 15... ♗g4⁉ 16. ♗xg7 ♔xg7 17. ♘d4♕d7 18. ♘f5+ ♗xf5 19. exf5 ♕xd1 20. ♖fxd1 b6 21. fxg6 ) a) 16. fxe5 ♖xf1+ 17. ♕xf1 ♗e6 ( 17... ♗xe5 18. ♗xe5 ♕xe519. ♖xb7 ) 18. ♘f4 ♗xa2 19. ♖c1 ♕f7 b) 16. ♗xe5 16... ♗xe5 17. fxe5 ♖xf1+ 18. ♕xf1 ♗c6 Black attacks both of the e-pawns and should be fine. 19. ♕f6 ♖e8 20.e6 ♕e7 21. ♕xe7 ♖xe7 22. ♘d4 ♗xe4 23. ♖e1 ♗d5 24. a4♔f8 ( 24... ♔g7 25. g4 ♔f6 26. g5+ ) 25. ♔f2 ♖c7= 12... ♖ac8⁉ This is my recommendation, and I think it gives Black interesting new possibilities that may mean 12.Nf4 just wasted a tempo. The point of this move is that by playing Nf4 White has "announced" he wants to play Nd5. 12... ♖ad8⁉ 13. ♘d5 ♕c8 14. ♗b5 ♖fe8 15. ♗g5 cxd4 16. cxd4h6 17. ♗xc6 ♗xc6 18. ♘xe7+ ♖xe7 19. ♗xe7 ♖xd4 20. ♕b3♗xe4=/∞ Citak-Kurnosov, Istanbul 2005 12... ♕c8 This was what you were supposed to play, and what I played against Beliavsky in Gibraltar. 13. ♗e2 ♖b8 I made a number of moves along the eighth rank because frankly there aren't too many other moves available for Black. ( 13... ♖d8 14. d5 ♘a5 15. c4 b6 16. ♕c1 e517. dxe6 fxe6 18. h4 ♘c6 19. h5 ♘e5⇆ Riazantsev -Areshchenko, Sochi 2006 ) 14. ♕d2 ♖d8 15. ♖fc1 ♗e8 16. d5 White finally decided to settle the structure. 16... b6 Once again, it's important to note that the knight should vacate the c6square only when completely forced to do so. You don't want to let White play c4, and at the moment c4 would allow ...Nd4, which of course isn't White's idea. The game went on to become very sharp and in the end I even won it, but in general I'm not entirely convinced by this Aikido approach to chess - i.e. trying to use your opponent's strength against him and not doing anything in the centre or attempting any active play until White shows his hand. That's a possible approach, of course, but I'm not convinced it's good here. 17. ♕b2 ♘a518. ♕a3 e5 19. dxe6 fxe6 20. ♗g4 ♗f7 21. ♕a4 ♖d6 22. h4⁈♕e8 23. ♕c2 ♘c4=/+ Beliavsky-Svidler, Gibraltar 2009 13. ♘d5 13. ♗b5 cxd4 14. cxd4 a6 15. ♗e2 b5 16. ♘d5 ♕d8 17. a4bxa4= 13... ♕d8 14. ♖xb7 ♘a5 Hitting both the rook and the bishop, and now ...e6 is a threat for Black. The second part of this mini-series will be dedicated to 12.Bf4 and 12.Bd3. Those are very serious moves and have given me a lot of trouble over the years, so they require a lot of attention. Don't miss

14... cxd4 15. ♗b5 dxe3 16. ♖xd7 exf2+ 17. ♖xf2 ♕xd7 18.♘f6+ ♗xf6 19. ♕xd7 ♗xc3=

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.bxc3 Bg7 7.Bc4 c5 8.Ne2 Nc6 9.Be3 0-0 10.0-0 Bd7 11.Rb1 Qc7 12.Bf4 Qc8 13.Bb3 13.Rc1 a6 14.Qd2 b5 15.Bb3 15.Bd3 13.d5 12.Bd3 Rad8 12...Rfd8 10...Bg4

12.Bf4 Qc8 13.Bb3 & 13.dxc5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗c4 c5 8. ♘e2 ♘c6 9. ♗e3 OO 10. O-O ♗d7 11.♖b1 ♕c7 As I said in Part 1, the two main moves Black has to face here are 12.Bf4 and 12.Bd3, and I'll try to cover them in quite some detail. I'll also attempt to explain why starting with 10...Bg4 11.f3 Bd7 is very beneficial for Black. The differences are much more pronounced and clearer in the 12.Bd3 line, but we'll start by looking at 12.Bf4. 12. ♗f4 ♕c8⁉ White has two major moves in this position, 13.Rc1 and 13.d5. 12... e5 This isn't a bad idea in general, but in this particular case after 13.Bg3 White will be better. a) 13. dxe5 ♘xe5 ( 13... ♗xe5 14. ♗xe5 ♘xe5 15. ♗d5 ♗c616. c4+/= ) 14. ♗xe5 ( 14. ♗d5+/= ) 14... ♗xe5 15. f4 ♗g7 16.♗d5 ♗c6 17. c4 ♖ab8 18. ♖f3 b5 Bick-Mikhailuk, Philadelphia 31st World Open 2003 b) 13. ♗g3 You'll notice that Black plays ...e5 a lot in this line of the Grünfeld, but almost exclusively after ...Na5 and Bd3 are included. It's very important not to allow White control of the d5-square, while here the bishop will land on d5 in the near future, with a very nice position for White. b1) 13... b6 b11) 14. d5⁉ ♘a5 15. d6⁉ ♕b7 16. ♗d5 ♗c6 17. f4∞ b12) 14. ♗d5 b121) 14... ♖ad8⁉ 15. dxe5 ( 15. dxc5 bxc5 ) 15...♘xe5 16. f4→ ( 16. c4 ♗e6 17. ♕c2 ) 16... ♘g4 17.♕d2 c4 ( 17... ♘f6 ) b122) cxd4 15. cxd4 ♖ac8 16. ♕b3 ♕d8 17. dxe5 ( 17.♖bc1 ♕g5! 18. dxe5 ♘d4 19. ♗xf7+ ♔h8 20. ♘xd4♖xc1 21. ♗c4∞ ) 17... ♘xe5

b1221) 18. ♕a3 ♘c4 19. ♕xa7 ( 19. ♗xc4 ♖xc4 20.♕xa7 ♖xe4 21. ♖fd1 ♖xe2 22. ♖xd7 ♖xa2 ) 19...♘d2 20. ♗d6 ♖e8 b1222) 18. f4 18... ♘c6 19. ♖fd1 ♕e7+/= b13) 14. dxc5! 14... bxc5 15. ♕d3 ( 15. ♕d2 ♗e6⁉ 16.♗xe6 fxe6+/= ; 15. f4 ♗g4 16. f5 ♖ad8 , with the idea of ...Na5. ) 15... ♖ab8 16. ♖xb8! ( 16. ♕e3 ♖xb1 17. ♖xb1♖b8 18. ♖d1 ♘a5 19. ♗d5 ♗b5= ) 16... ♖xb8 17. f4± b2) 13... cxd4 14. cxd4 ♘xd4 15. ♕xd4 exd4 ( 15... ♖ac816. ♖fc1 ) 16. ♗xc7 ♖fc8 17. ♖xb7 ♗c6 18. ♗d5 b3) 13... ♖ad8 14. ♗d5 ( 14. ♗h4! ♖c8 15. d5 ♘a5 16. d6♕b8 17. ♗d5± ) 14... cxd4 ( 14... ♗e6 15. ♗xe6 fxe6 16.♕b3 cxd4 17. ♘xd4 ♘xd4 18. cxd4 ♖xd4 19. f3 b6 20.♕xe6+ ♕f7 21. ♕xf7++/= ; 14... ♗g4 15. f3 ♗e6 16. ♗xe6fxe6 17. ♕b3 cxd4 18. ♘xd4± ) 15. cxd4 ♕a5 16. dxe5♘xe5 17. ♕b3 ( 17. ♗h4⁉ ♖b8 ) 17... b5 ( 17... b6 18.♖fc1 ♖c8 19. ♖xc8 ♖xc8 20. ♗xe5 ♗xe5 21. f4 ♕c5+ 22.♔f1 ♗f6 23. ♗xf7+ ♔g7 24. ♗e6 Thorfinnss on-Sutovsky, Caleta 2005 ) 18. ♖fc1 ♘c4 19. ♕b4 ♕xb4 20. ♖xb4 a5 21.♖b3 ♖c8 22. ♗b7 ♖c5 Palliser-Greenfeld, Isle of Man Monarch Open 2006 13. ♗b3 This is an extremely rare move to which Black has a reasonably pleasant choice of how to respond. 13. dxc5⁉ a) 13... ♘a5 14. ♗b5 This is White's plan behind 13.dxc5, and the first occurrence of a line admittedly a minor one - in which it's obvious why it was better to include 10...Bg4 11. f3. In that case ...Qxc5 would now be with check, and Black would win material on the spot. a1) 14... ♗xb5 15. ♖xb5 a11) 15... ♖d8 16. ♕a4 ( 16. ♕b1⁉ ) 16... ♘c6 17. ♖fb1♖d7 18. ♕b3 ♘d8 19. ♗e3 ♕c6 20. f3 e5 21. c4+/= a12) b6 a2) 14... ♖d8 a21) 15. ♗xd7 ♕xd7 ( 15... ♖xd7 16. ♕a4 ♘c6 17.♖fd1 ) 16. ♕xd7 ♖xd7 17. ♖b5 ( 17. ♖fd1 ♖ad8 18.♖xd7 ♖xd7 19. g4 ) 17... ♘c4 18. ♖fb1 ♖ad8⇆ a22) 15. ♕a4⁉+/=

a3) 14... ♗e6 15. ♕a4 ( 15. ♘d4 ♗g4! 16. f3 e5⇆ ) 15... a616. ♕xa5 axb5 17. ♕xb5 ♗xa2 18. ♖b2 ♕e6 19. ♕xb7♗c4 20. ♖e1+/= a4) 14... ♗g4 15. ♗e3 a41) 15... a6 16. ♗d3 ♖d8 17. ♕c2 ♗xe2 18. ♗xe2 e619. ♕b2

a42) 15... ♖d8 16. ♕a4 ( 16. ♕c2 ) 16... ♕c7 ( 16... a617. ♕xa5 axb5 18. ♕xb5 ) 17. ♘f4 a6 18. ♗d3 e6 19. f3( 19. h3 ) a43) ♕c7 16. ♕a4 ( 16. f3 ♖fd8 17. ♘d4 ♗d7 18. ♕a4 ) b) 13... ♘e5 14. ♗xe5 ( 14. ♗d5 ♖b8 15. ♕b3 ♕xc5 16.♗e3 ♕a5 17. f4 ♘g4 18. ♗d4 e6 ) 14... ♗xe5 15. ♕d5 b1) 15... ♕c7 16. ♖xb7 ( 16. f4 ♗e6 17. ♕xe5 ♕xe5 18.fxe5 ♗xc4 19. ♖fe1 ♗a6 ) 16... ♗xh2+ 17. ♔h1 ♗e6 18.♖xc7 ♗xd5 19. ♗xd5 ♗xc7 20. ♗xa8 ♖xa8 21. ♖b1 b2) ♗e6 16. ♕xe5 ♗xc4 17. ♖b2 ( 17. ♖fe1 ♗xa2 18.♖b2 ♗c4 19. ♘d4 ) c) a6 14. ♘d4 ( 14. ♗e3 ♘a5 15. ♗d3 ♖d8 ) 14... e5 13. ♕d3 ♖d8 14. ♖fc1 ♘a5 This has been played quite recently, but it already seems as though White has somewhat misplaced his pieces. He faces the unpleasant choice of whether to exchange his pride and joy - the bishop on c4 - for a knight or a bishop. 15. ♗b5 a6 16. ♗xd7♕xd7 17. ♕g3 ♖ac8= KollyAvrukh, Biel 2002 13... ♘a5 Black could consider alternatives like 13...Rd8 and 13...b6 here - there are a lot of waiting moves - but I wanted to mention this idea because it's just so weird, and in general I like weird ideas! 13... cxd4 14. cxd4 e5 ( 14... ♘xd4⁈ 15. ♘xd4 e5 16. ♗d5+/= )15. dxe5 ♘xe5+/= It's important to note that here, as in many similar positions, this forced sequence doesn't quite equalise for Black. While opening up the long diagonal for the bishop can't be a bad idea in general it's very important to make a subtle distinction between positions in which Black is well-coordinated and therefore fine in this structure, and positions in which he'll come under some immediate fire due to the poor position of his pieces. In this particular case after both 16. ♕d2 and the somewhat faster ( 16. ♖c1 , with the idea of 16... ♗c6 17. ♘d4♖d8 18. ♘xc6 , Black isn't equal at all because he's not in time to get his pieces to the proper squares. ) 14. ♗c2 ♘c4 Trying to trade off White's bishop straightaway by playing ...Na3. 15. ♕c1 White attempts to prevent that.

15... b6

15. ♗d3 ♘a3 16. ♖b2 ( 16. ♖b3? ♗a4 ) 16... c4 17. ♗c2 b5⇆ 15... e5 16. ♗d3! 15... cxd4 16. cxd4 e5 17. ♗d3!+/= 15... b5 16. ♗b3 ( 16. ♗d3 ) 16... cxd4 17. cxd4 ♕c6 18. d5♕f6∞

16. ♗d3 ♕a6!⇆ Black creates a self-pin, and this is the move I wanted to show you. I'm not sure all this is the best way to react to 13.Bb3, but I wanted to mention this quirky little idea for the sake of pure curiosity. For a split second Black is now somewhat passive, but he wants to put the queen on a4, while White can't really attack the knight on c4 with any of his pieces. Once the queen lands on a4 Black will have serious counterplay both in the centre and on the queenside.

13.Rc1 a6 14.Qd2 b5 15.Bb3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗c4 c5 8. ♘e2 ♘c6 9. ♗e3 OO 10. O-O ♗d7 11.♖b1 ♕c7 12. ♗f4 ♕c8 13. ♖c1 One of White's two major options. The other, 13.d5, will be covered later. 13... a6 This is Black's normal response, and at least as of today it seems to be holding. 13... ♖d8⁉ a) 14. h3 b5 15. ♗xb5 ♘xd4= b) 14. ♕b3 e6 15. ♕a3 b5 ( 15... a6 16. dxc5 ♘e5 17. ♗xe5♗xe5 18. ♘d4 ♕c7 19. h3 ♖ac8∓ Hamrakulova-Kiran, Tashkent Asian-chj G18 2007 ) 16. ♗xb5 ♘xd4 17. ♘xd4 cxd4 18. cxd4♕b7 19. ♗d3 ♗c6 ( 19... ♗xd4 20. ♖c7 ♕b6 21. ♖b1♗xf2+ 22. ♔f1 ♕d4 23. ♖b4 ) 20. d5 ( 20. ♖b1 ♕d7 21. d5exd5 22. e5 d4 ) 20... exd5 21. e5 d4 22. ♗a6 ♕d7= c) 14. ♕d2 b5 15. ♗xb5 ♘xd4 16. ♗xd7 ♘xe2+ 17. ♕xe2♕xd7= Kozlov-Kurnosov, Saratov 7th N.Aratovsky Mem 2007 d) 14. d5 14... ♘e5 15. ♗xe5 ♗xe5 16. f4 ♗g7 d1) 17. ♘g3 b5 18. ♗d3 c4 19. ♗b1 a5 20. ♔h1 ♕c5 21.♕f3 ♗e8 ( 21... b4 22. e5 b3 23. ♘e4 ♕b6 24. d6 exd6 25.♘xd6 b2 26. ♖cd1 ♖ab8 27. f5 ♗c6 28. ♕f4 ♕c7 ) 22.♖cd1 ♕a3 23. e5 b4 24. ♘e4 ♖ab8 25. f5 ♗xe5 26. fxg6hxg6 27. ♕h3 ♖b6 28. ♘g5 ♗g7 Krush-Kudrin, Stillwater US Championship 2007 d2) 17. ♕b3 17... ♖b8 d21) 18. d6 e6 19. ♗d3 ( 19. e5 b5 ) 19... b5 20. ♕a3 (20. ♗c2 ♗c6 21. e5 f6 22. ♕a3 fxe5 23. fxe5 ♗xe5 24.♕xc5 ♗d5 ) 20... ♖b6 21. e5 f6 d22) 18. a4 18... b6 19. e5 ♗e8 20. ♘g3 a6 21. ♖cd1 b5Rodshtein-Huzman, Ashdod 2004 13... e5⁉ 13... ♘a5 14. ♗d3 b5 15. ♕d2 ♕b7 16. ♗h6 ♕b6 17. ♗xg7♔xg7 18. h4 h6 19. h5 g5 20. f4 f6 21. e5 Lia ng Chong-Ni Hua, Beijing Chinese Championship 2008 13... ♕d8⁉ This very fresh idea was played by Areshchenko against Bacrot in the analagous position with a pawn on f3, and is discussed in the notes to 10...Bg4 11.f3 Bg4. 14. ♕d2 14. d5 ♘a5 15. ♗d3 e5 16. ♗e3 c4 ( 16... f5 17. c4 f4 18. ♗d2b6 19. ♗xa5 bxa5 20. f3 ♕c7 21. ♗c2 ♕d6 22. ♗a4 ♗c8 23.♖b1 ) 17. ♗b1 b5 18. f4 ♘b7∞

14. a3 b5 15. ♗d5 e6 16. ♗a2 e5 17. ♗xe5 ♘xe5 18. dxe5 c419. f4 ♗g4 20. ♖f2 ♖d8 21. ♕e1 ♗xe2 22. ♕xe2 ♕c5 23. a4♕a3 24. ♕c2 ♗f8 Vaisser-Vachier Lagrave, Aix-les-Bains French Championship 2007 14... b5 White now has a choice. He can play either 15.Bd3 or 15.Bd5/15.Bb3. Those last two moves basically lead to the same position, so it's not so much a choice between them as a question of whether to spend an additional move to "shorten your time trouble". 15. ♗b3 This move, or 15.Bd5, is much sharper and more interesting. 15. ♗d5 e6 ( 15... ♖d8 16. ♗h6 ♗h8 17. ♗g5 ♖a7 18. ♕e3♖e8 19. h4 ♘a5 20. ♘f4 e6 21. ♗b3 f6 22. ♗h6 ♖c7 23. e5♘xb3 24. axb3 ♗c6 25. h5 g5 26. ♘d3 Gordon-Howell, 4NCL 2008 )16. ♗b3 c4 I think this is Black's best options. 17. ♗c2 e5 18. ♗h6♕c7 Transposing to the lines after 15.Bb3. 15... e5 16. ♗h6 c4 17. ♗xg7 17. ♗c2 f6 ( 17... ♗xh6 18. ♕xh6 exd4 19. cxd4 ) 18. ♗xg7♔xg7 19. f4 ♕c7 20. h3 ♖ad8 21. fxe5 fxe5 22. ♕g5 ♕d6 23.d5 ♘e7 24. ♔h1 ♘g8 25. ♘g1 ♘f6 26. ♖f3 ( 26. ♘f3 ♖de8 27.♘h4 ♘h5 ) 26... a5 27. ♖cf1 b4 E.Atalik-Lahno, Kallithea 13th ECC final w 2008 17... ♔xg7 18. ♗c2 ♕c7 This is a very sharp and unclear position, but in general I think Black is doing reasonably ok. In a game I played in the Gibraltar Masters in 2009 David Berczes of Hungary immediately went for the throat with 19. f4 exd4 20. cxd4 20. ♘xd4 ♖ad8 21. f5 f6 20... ♗g4 I have to say this looked incredibly scary to me during the game. 21. d5 21. f5 And it is very scary, because White can play something like this.21... ♗xe2 22. f6+ White tries to give mate immediately. There seemingly isn't a mate, but it has to be checked very carefully before you embark on this. It isn't just my hunch that this is playable, though - I did check this position pretty carefully. 22... ♔h8 23. ♕h6 ♖g8 24. ♖f4g5 25. e5 ♗d3 21... ♗xe2 22. ♕xe2 ♘d4 23. ♕f2 ♘xc2 24. ♖xc2 ♖ac8 24... f6⁉ 25. ♕d4+ 25. e5 ♕c5 26. d6 25... f6 26. ♖e2 26. g3⁉ 26... ♕c5 Berczes-Svidler, Gibraltar 7th Masters 2009

13.Rc1 a6 14.Qd2 b5 15.Bd3 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗c4 c5 8. ♘e2 ♘c6 9. ♗e3 OO 10. O-O ♗d7 11.♖b1 ♕c7 12. ♗f4 ♕c8 13. ♖c1 a6 14. ♕d2 b5 15. ♗d3 I don't think this is a critical line, but it was played by Loek van Wely against me in a Wijk aan Zee tournament some years back. It's really a very traumatic memory for me, because this game was played in Round 12 of 13. After 11 rounds I was +3 and half a point behind the leaders. Generally I do very, very poorly in Wijk and score something around 50%, so this was my one shot at doing ok. In the last two rounds I had Black against Loek and White against a very young Sergey Karjakin. I thought that was quite promising because Loek never shies away from principled battles - I don't think he's ever played for less than a full point against me - and therefore we generally get some double-edged and interesting position where pretty much anything can happen. Then in the last round I was playing the "young buck", who was clearly potentially a very good player but was still just a boy, so I thought I'd maybe do ok. Obviously I scored zero points out of those two games, so that was my one shot at Wijk redemption done with. 15... ♕b7 This was my move, and it's the correct one. 15... e5 a) 16. ♗h6⁉ a1) 16... ♕c7 17. ♗xg7 ♔xg7 18. d5 ( 18. dxc5 ♘a5 19. c4♖fd8 20. ♕e3 bxc4 21. ♗xc4 ♘xc4 22. ♖xc4 ♗b5 23.♖c3 ♕a5 24. ♖fc1 ♖ac8 25. ♘g3 ♕xa2 26. ♕g5± ) 18...♘e7 19. c4 ( 19. f4 f6 ) 19... b4 20. f4 f6+/= a2) ♗xh6 17. ♕xh6 cxd4 ( 17... exd4 18. cxd4 c4 19. ♗b1 ;17... ♕d8 18. f4 c4 19. ♗b1 exd4 20. f5 ♕f6 21. e5 ♕xe522. cxd4 ♕g7 23. ♕f4 ♖fe8 24. f6 ♕f8 25. ♘c3 ♖ad8 26.♖cd1 ♗c8 27. h4 ♘b4 28. h5 ♘d3 29. ♗xd3 cxd3 30.♖xd3 ♗f5 31. ♖g3 ♔h8 32. ♕h4 Dahale-Konguvel, New Delhi 6th Parsvnath op 2008 ) 18. cxd4 exd4 19. ♘xd4 ( 19. ♘f4 ;19. f4 ♕e8 20. f5 f6 21. ♘f4 g5 22. ♘d5 ♕f7 23. h4 ) 19... ♕d8 20. ♘f3 ♕f6 21. ♖c5 ♖ac8 b) 16. ♗e3 b1) 16... c4 17. ♗b1 f5 ( 17... exd4 18. cxd4 ♕d8 19. d5♘e5 20. f4 ♘g4 21. ♗d4 ) 18. d5 ( 18. f4 exd4 19. cxd4fxe4 20. ♗xe4 ♗f5 ) 18... ♘e7 19. d6 ( 19. f4 exf4 20.♘xf4 fxe4 21. ♗xe4 ♘f5 ) 19... ♘c6 20. f4 ♕e8 21. ♘g3 b2) 16... exd4 17. cxd4 cxd4 18. ♘xd4 ( 18. ♗xd4 ♗xd4 19.♘xd4 ♕e8 20. ♘b3 ♕e7 21. ♘c5 ♖fd8 ) 18... ♕b7 19.♘b3 b21) 19... ♖ad8 20. ♘c5 ♕a8 21. f4 b22) 19... ♘e5 20. ♘c5 ♕c6 21. ♗e2 ( 21. f4 ♘xd3 22.♘xd3 ♕d6 23. e5 ♕d5 24. ♖c5 ♕b7 ) b221) 21... ♖ad8 22. ♖fd1 ( 22. ♕a5 ) 22... ♗g4 23.♕xd8 ♗xe2 24. ♖d6 b222) 21... ♗g4 22. ♗xg4 ♘xg4 23. ♘d7 ♕xe4 24.♘xf8 ♗xf8±

b223) 21... ♗e6 22. ♘xe6 ♕xe6 23. ♕d5 ♖ac8 24.♖fd1 ♘c4 ( 24... ♕xd5 25. exd5 ♖xc1 26. ♖xc1♖d8 27. f4 ♘c4 28. ♗xc4 bxc4 29. ♖xc4 ♖xd5 30.♖c8+ ♗f8 31. f5 ) 25. ♗c5 ♕xd5 26. ♖xd5 ♖fe8 27.♗xc4 bxc4 28. ♖xc4 ♖e5= b23) ♗e6 20. ♘c5 ♕e7 21. f4 ♖fd8 22. e5 ♗d5∞ 23.♖fe1 f6 24. e6 f5 25. a4 ♘d4 16. ♗h6 16. dxc5 ♘e5 17. ♖fd1 ♖fd8 18. ♗c2 ♕a7 19. ♗xe5 ♗xe5 20.♕e3 ♕c7 21. f4 ♗g7 22. e5 ♖ac8 23. ♖d5 ♗g4 16... ♕b6! This is the best move in this position. What you shouldn't do is what I did - open the c-file - because the rook will come to c5 in many lines. 16... ♗xh6 I immediately went very badly wrong and got mated by move 25, which is quite difficult to imagine in this position, but anything is possible if you really want it! 17. ♕xh6 cxd4 18. cxd4 Van WelySvidler, Wijk aan Zee 2007 18... ♖ac8 a) 19. h4 ♕a7 ( 19... ♕b6 20. ♖c5 ♗g4 21. ♘f4 ) 20. h5♘xd4 21. ♘f4 ♖xc1 22. ♖xc1 ♖c8 23. ♖d1 b) 19. ♕g5 19... ♕b6 b1) 20. d5 f6 21. ♕d2 ♘e5 22. ♗c2 ♔g7 23. ♘d4 ♖c4 24.♖fd1 ♖fc8 25. h3 g5 ( 25... ♗xh3 26. ♕e3 ) b2) 20. ♖fd1 20... ♖fd8 21. h3 ( 21. ♘f4 ♕xd4 22. ♗xb5♕xe4 23. ♗xa6 ♖b8 24. ♘d5 ♗e6 25. ♘xe7+ ♘xe7 26.♖xd8+ ♖xd8 27. ♕xe7 ♖d2= ) 21... ♔g7 ( 21... ♘xd4 22.♘xd4 ♕xd4 23. ♗xb5 ; 21... f6 22. ♕h6 ) 22. ♘f4 f6 23.♘h5+ ♔g8 24. ♕h6 gxh5 25. e5 ♘xd4 26. ♕xh7+ ♔f8 27.♕h8+ ♔f7 28. ♕xh5+ ♔e6 29. ♕g4+ ♔xe5 30. ♖e1+ ♔d631. ♕f4+ e5 32. ♕xf6+ ♗e6 33. ♕xe5+ ♔e7 17. ♗xg7 ♔xg7 There isn't too much you need to know here - rather you should understand a few things. 18. ♕b2N After this move White is planning to play d5 and c4. 18. ♕e3 cxd4 19. cxd4 e5 ( 19... ♖ac8 20. ♖c5 e5 21. ♖fc1♘xd4 22. ♘xd4 exd4 23. ♕xd4+ ♔g8 Mercadal Benejam-Banet, ICCF 2007) 18. ♗c2⁉ ♖fd8 19. ♕f4 ( 19. ♕g5 cxd4 20. cxd4 f6 ) 19... cxd420. cxd4 f6 21. ♖fd1 ♖ac8 22. ♗b3 a5 ( 22... ♘xd4 23. ♖xd4e5 24. ♖xc8 ; 22... e5 23. dxe5 ♘xe5 24. ♘c3 ♗c6 25. ♘d5♗xd5 26. ♗xd5 ) 23. e5 ( 23. ♗d5 ♘b4 24. ♘c3 e6 ) 23... a4 24.♗c2 ♘b4 25. ♗b1 ♘d5 26. ♕f3 ♖xc1 27. ♖xc1 ♗e6⇆ 18. d5 ♘e5 19. ♔h1 e6

a) 20. c4 bxc4 ( 20... exd5 21. cxd5 c4 22. ♗b1 b4 23. f4♘g4 ) 21. ♗xc4 ♕b4 ( 21... ♘xc4 22. ♖xc4 ) b) 20. f4 ♘xd3 21. ♕xd3 exd5⇆

c) 20. ♘f4 20... ♘xd3 21. ♘xd3 exd5 22. exd5 c4 ( 22... ♕d623. c4 ) 23. ♘e5 ♖ad8

18. ♕f4 cxd4 19. cxd4 f6! This can also be played immediately on move 18. Black is now once again ready either to play ...e5 himself, or to meet 20.d5 with 20...Ne5. a) 20. ♖fd1 ♘xd4 a1) 21. ♘xd4 e5 ( 21... ♕xd4 22. ♗f1 ) 22. ♕g3 exd4 23.♖c7 ♖a7 a2) 21. ♕c7 21... ♘xe2+ 22. ♗xe2 ♕xc7 23. ♖xc7 ♗e8 24.♗f3 ♗f7 b) 20. d5 20... ♘e5 Black has a very comfortable position. 21. ♗c2♖ac8 18... ♖ab8 A very clever little move, planning to meet d5 both immediately, and after 19.Kh1 e5 20.d5, by playing ...c4. 18... ♗g4 19. f3 ( 19. d5 ♘e5 ) 18... ♖fd8 19. ♔h1 ( 19. ♗c4 ♖ab8 20. ♗d5 e6 21. ♗xc6 ♗xc622. d5 exd5 23. exd5 ♗a8 24. c4+ ♕f6 ) 19... ♖ab8 20. f4 ♗g421. d5 19. ♔h1 19. f4 ♗g4 20. ♔h1 ♗xe2 21. dxc5 ( 21. ♗xe2 cxd4 ) 21...♕xc5 22. ♗xe2 ♖bd8 19. e5 ♗g4 20. f3 ( 20. ♗e4 ♗xe2 21. ♕xe2 cxd4 22. ♗xc6♕xc6 23. cxd4 ♕d5 ) 20... cxd4 21. cxd4 ♘xe5 19. ♖b1 e5 20. d5 ♘e7 21. c4 ♕d6 22. ♕c3 ( 22. cxb5 axb5 23.f4 f6 ) 22... b4 23. ♕d2 f6 24. f4 a5 19... e5 20. d5 20. dxe5 ♘xe5 21. c4 ♕f6 20... c4 21. ♗b1 21. ♗xc4 bxc4 makes the point of 18...Rab8 clear - the queen isn't hanging on b6. 21... ♘e7 21... ♘a5⁉ 22. f4 f6 23. fxe5 fxe5 24. ♕b4 ♖xf1+ 25. ♖xf1♖e8 26. a4 ♘b7 22. f4 f6 23. ♕b4 ♖be8 24. fxe5 fxe5 25. ♖xf8 ♔xf8 26. ♖f1+♔g7=

13.d5 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗c4 c5 8. ♘e2 ♘c6 9. ♗e3 OO 10. O-O ♗d7 11.♖b1 ♕c7 12. ♗f4 ♕c8 13. d5⁉ This is the first occurrence of a structure which has to be discussed at some length. It's one White will get if he wants in the Bc4 Grünfeld, so you have to be prepared for it as Black. 13... ♘a5

13... ♘e5⁉ is also possible, but I always preferred 13...Na5. 14. ♗xe5♗xe5 15. f4 ♗g7 16. e5 ♖b8 ( 16... a6 17. a4 ♕c7 18. a5Farago-Karavade, Budapest 2005 ) 17. a4 ♕c7 ( 17... ♔h8 18. ♘g3♕c7 19. ♘e4 ♗f5 20. ♖e1 b6 21. ♖b2 f6 22. d6 exd6 23.♕xd6 ♕xd6 24. ♘xd6 fxe5 25. fxe5 Wallis-Karavade, Kuala Lumpur 2006 ) 18. ♖a1! ♔h8 19. ♘g3 a6 20. a5 ♗b5 21. ♕d3 f6Harikrishna-Berkes, Paks Marx Mem 2007 14. ♗d3 e5 Black needs to start fighting back for the centre. 15. ♗d2 I believe this is the best move. White's threat of c4 is quite serious. 15. ♗g3 a) 15... c4 16. ♗c2 b6 ( 16... f5 17. f4 exf4 18. ♗xf4 fxe4 19.♗xe4 ♕c5+ 20. ♔h1 ♖ae8 21. ♘g3 ♗xc3 22. d6 ♗g7 23.♗d5+ ♔h8 24. ♘e4 ♕d4 25. ♘g5 Krush-Harika, Beijing 1st WMSG Rapid Pairs 2008 25... ♕xd1 26. ♖bxd1 h6∓ ) 17. ♕d2 (17. h3 ♘b7 18. f4 exf4 19. ♗xf4 ♕c5+ 20. ♔h1 ♖ae8 21.♘d4 ♗e5 22. ♗h6 ♗g7 Antic-Mikhalevski, Sydney op 2007 ) 17...♘b7 18. f4 f6 19. ♔h1 ♘d6 20. ♘g1 ♕d8 21. fxe5 fxe5 22.♘f3= Sakaev-Avrukh, Izmir 2004 b) f5 16. f4 fxe4 17. ♗xe4 ♘c4 18. ♕d3 ♘d6 19. fxe5 ♘xe420. ♕xe4 ♗f5 21. ♕c4 ♗xb1 22. d6+ ♔h8 23. ♖xb1 ♕f5 24.♖e1 ♗xe5 25. ♘d4 ♗xd4+ 26. cxd4 ♖ae8 Carlsen-Navara, Wijk aan Zee A 2007 15. ♗e3 This is of course a natural and possible move. 15... f5 ( 15...c4 16. ♗c2 b5 17. f4 f6 18. ♘g3 ♗h6 19. ♕d2 exf4 20. ♗xf4♗xf4 21. ♖xf4 ♕e8 22. ♖e1 ♕e5 23. ♘e2 ♘b7 24. ♘d4 ♖ae825. ♘f3 ♕d6 26. ♖h4 ♖e7 27. ♔h1∞ Leitao-Mekhitarian, Americana Festival de Xadrez 2007 ) 16. exf5 ( 16. f3 An amazing number of very strong players, including household names, have played f2-f3 either in this position or in very similar positions. It's a concept I want to put on the board and explain in some detail, because I think it's very important to this line. 16... f4 17. ♗f2 c4 18. ♗c2 b5 To say that Black's already winning would perhaps be an overstatement, but not a huge one. The knight on e2 and the bishop on c2 not only don't have any moves currently, they're very unlikely to have any moves in any kind of distant future. Therefore Black is completely free to secure the queenside, transfer the knight to d6 via b7, and then just start a pawn march on the kingside. White is singularly ill-equipped to deal with that because he basically has only two ranks to operate on and none of his pieces have any future. This is a dream position for Black, and I'm really surprised at how many times Black gets it, even in high-level games. 19.a4 a6 20. ♖a1 ♖e8 21. ♔h1 Wells-Rodriguez Lopez, Liverpool 4th EU-ch 2008 ) 16... gxf5 a) 17. ♗g5⁉ b6 18. ♖c1 ♕e8 19. ♘g3 e4 ( 19... ♘b7 20.♗e2! ♗a4 21. ♕d2 ♘d6 22. f4 e4 23. ♘h5 ♗h8 24. ♗h6♕g6 25. ♗xf8 ♔xf8 26. ♔h1± Bernasek-Borisek, Szeged Mitropa Cup 2007 ) 20. ♗b1 ♘c4⇆

b) 17. f3 This move is more normal here. 17... b6 This is a typical position which is very difficult to assess. In general, I think you need to get some experience with this structure so you develop a feel for it. My understanding is that Black should be fine in most cases, getting at least a very playable position. ( 17... c4 18. ♗c2 f4 19. ♗f2 b620. g3 ♗f5 21. gxf4 ♕d7 22. ♗e4 ♖ae8 23. ♘g3 ♗h3 24. f5 ♘b7 25. ♔h1 ♘d6 26. ♖g1 ♔h8 27. ♘h5 ♗xf5 28. ♗xf5♘xf5 29. ♘xg7 ♘xg7 WerleMamedov, Plovdiv IX EU-ch 2008 ) b1) 18. ♘g3? f4

b2) 18. g4 Jan Werle decided to "cut the knot" straightaway by playing this move against Maxime Vachier-Lagrave and trying to open up lines on the kingside. b21) 18... ♘b7⁉ 19. gxf5 ♘d6 20. ♘g3 ♘xf5 21. ♘xf5♗xf5 22. c4 ( 22. ♖b2 c4 23. ♗e4 ) 22... ♗xd3 23.♕xd3 ♕h3 24. ♕e4 ♖f6 25. ♖f2 ♖af8 26. ♖bf1 b22) 18... f4 Cutting off the knight on e2 is a very common theme here. 19. ♗f2 h5⁉ An interesting idea, trying to secure the f5-square for the bishop. Vachier-Lagrave went on to win a very nice game. ( 19... c4⁉ 20. ♗e4 ♘b7 Even here we end up in a situation where if the white knight had been on d2 instead of e2 Black could just resign because his position would be strategically hopeless. With the knight on e2, however, I think Black stands quite well because he's ready to transfer his knight over to d6 and the bishop on e4 will be very difficult to sustain. In general, this is very double-edged and playable for Black. )20. gxh5 c4 21. ♗c2 ♗f5 22. h6 ♗xh6 23. ♔h1 ♗xc2 24.♕xc2 ♕f5 25. ♖g1+ ♔h7 Werle-Vachier Lagrave, Bundesliga 08-09 2008 b3) 18. ♗g5 ♕e8 b4) 18. ♕e1 18... ♘b7 19. ♕h4 b41) 19... ♕d8 20. ♗g5 ♗f6 21. ♗xf6 ( 21. f4 e4 22.♗c2 ♘d6 23. ♘g3 ♗e8 ) 21... ♕xf6 22. ♕xf6 ♖xf6 23.c4 ♘d6 24. ♘c3+/= b42) ♘d6⁉ A precise move. Black will aim to play ...Qd8 and then drive some of White's pieces back. White will try to get something going on the kingside along the lines of Ng3-h5, which obviously shouldn't be allowed unimpeded. 20. ♘g3 ♕d8 21.♗g5 ( 21. ♕h3 e4 ) 21... ♗f6 22. f4 e4 23. ♗e2 ( 23.♘h5 ♗xg5 24. fxg5 exd3 25. ♘f6+ ♖xf6 26. gxf6 ♘e4 ) 23... ♗e8 24. ♗h5 h6= 15... c4 15... f5 16. exf5 ♗xf5 ( 16... gxf5 17. c4 b6 18. ♗xa5 bxa5 19.♗c2 ♖b8 20. ♖xb8 ♕xb8 21. ♗a4 ♗c8 22. d6 ♗e6 ) 17. ♗xf5gxf5 ( 17... ♕xf5 18. ♗e3 b6 19. ♘g3 ) 18. ♕a4 b6 19. c4 ♘b720. ♘c3+/= 16. ♗c2 b6 16... f5 17. exf5 ♗xf5 18. ♘g3 ♗xc2 19. ♕xc2 ♕c5 20. ♗g5♕xd5 21. ♖fd1 ♕e6 22. ♘e4+/= 16... b5 This has been played by serious players, and it's not a bad move.17. f4 f6 ( 17... ♕c5+ 18. ♔h1 exf4 19. ♗xf4 ♖ae8 20. ♕d2♘b7 21. ♘d4 ♘d6 22. ♖fe1 f6 Igla-Lahno, Kallithea 13th ECC final w 2008 ) 18. ♘g3 ( 18. ♗e3 ♘b7 19. ♕d2 ♘d6 20. ♔h1 f5⇆ Kobalia-Areshchenko, Isle of Man Monarch op 2006 ) a) 18... ♕c5+ 19. ♔h1 ♗h6 20. fxe5 ♗xd2 21. ♕xd2 fxe5 22.♕g5 ♖f2 23. ♘f5 ♖f8 24. ♖bc1! ♗xf5 25. exf5 ♘b7 26. h4♖xc2 ( 26... ♖xf1+ 27. ♖xf1 ♘d6 28. h5 ; 26... e4 27. ♖xf2♕xf2 28. fxg6 ♖f4 ) 27. ♖xc2 ♘d6±

b) 18... ♘b7 19. f5 ♘d6 20. ♗e3 ♔h8 21. a4 a6 22. h4 gxf523. ♘xf5 ♗xf5 24. exf5 ♖g8 25. g4→ 17. f4 This is the obvious move, but Black's idea is to meet it with 17... ♕c5+ 17... f6 18. ♘g3 ( 18. ♗e3 ♘b7 19. f5 ♘d6 20. ♘g3 ♗e8 21.h4 ♕d7 22. ♕e2 ♖c8 23. ♖bd1 b5 24. a3 ♕e7 25. ♕g4 gxf526. exf5 ♔h8 27. ♔h1 ♖g8 28. ♕f3 ♗f8 29. ♔h2 ♕g7 30.♖d2 ♖c7∞ Berczes-Bartel, Plovdiv IX EU-ch 2008 ) 18... ♕c5+ 19.♔h1 exf4 20. ♗xf4 ♘b7 21. a4 ♖ae8 22. ♕d2 ♖f7 23. h4 ♗g424. h5→ ShirovAreshchenko, Foros Aerosvit GM 2006 18. ♔h1 exf4 19. ♗xf4 If White could now play Nd4 he'd have a huge edge, but Black is just in time to prevent that. 19. h3 g5 19. ♘xf4 ♖ae8 This really isn't ideal for White. In general, the knight belongs not on f4, but d4. 20. ♕f3 ♘b7 21. ♗e3 ♕e7 22. ♗d4♘d6 23. ♗xg7 ♔xg7 24. ♕g3 ♔h8 25. ♖be1 ♕e5= 19... ♗g4 20. h3 ♗xe2 21. ♕xe2 ♗xc3 White has obvious compensation for the pawn, but in general Black has achieved something here and if he manages to withstand the impending attack on the kingside he'll be a pawn up and doing ok. 22. a4 f6 23. ♖b5 ♕e7 24. d6 ♕e6 25. ♖d1 ♘c6⇆

12.Bd3 Rad8 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗c4 c5 8. ♘e2 ♘c6 9. ♗e3 OO 10. O-O ♗d7 11.♖b1 ♕c7 12. ♗d3 This is perhaps the central and most critical line of the whole ...Bd7 variation. It's also the line in which the difference between 10...Bd7 and 10...Bg4 11.f3 Bd7 is most pronounced. I'll try to explain what Black should do here after starting with 10...Bg4, but I'll do it in a slightly roundabout way. I want to explain my thought processes in this position with a pawn on f2, because I had to play and prepare for this position on a number of occasions against strong players. Obviously a rook has to go to d8 here, as otherwise it's not clear what Black's previous play was all about. 12... ♖ad8 This looks like the more natural move, but for me 13.Qc1 was a very serious problem. 12... a6⁉ a) 13. f4⁉ b) 13. ♗f4 e5 14. ♗g3 ( 14. dxe5 ♗e6 15. ♕d2 b5⇆ ) 14... b515. ♔h1 ♖ae8 16. ♗c2 ♕a7 17. d5 ( 17. dxe5 ♘xe5 18.♗xe5 ♖xe5 19. f4 ♖ee8 ) 17... ♘a5 18. a4 f5 19. axb5 axb5 ( 19... f4? 20. b6 ♕b8 21. ♗h4 h6 22. f3+/= ) 20. f3 ♘c4=/+Begunov-Svidler, St Petersburg 1994

c) 13. dxc5 ♖fd8 14. f4 ♗e6 15. a3 ♖d7 ( 15... ♗c4 16. ♘d4♗xd3 17. ♕xd3 e5 18. fxe5 ♘xe5 19. ♕e2 ) 16. ♕c2 ( 16. f5♗c4 17. ♘d4 ♗xd3 18. ♕xd3 ♘e5 ) 16... ♖ad8 17. ♖fd1 ♗h6=/∞ /\ Ne5 d) 13. ♕d2! 13... b5 ( 13... ♖ad8 14. ♗h6 ♗xh6 15. ♕xh6♗g4 16. f3 ♗c8 17. d5 ♘e5 18. ♕e3 1-0 Khenkin-Mogranzini, Reggio Emilia 49th Intl. 2007 ) d1) 14. ♗h6 ♗xh6 ( 14... e5 15. ♗xg7 ♔xg7 16. d5 c4 17.♗c2 ♘a5 18. f4 f6 19. fxe5 fxe5 20. ♕g5 ♕d6 21. h4 h622. ♕g3 ♘b7 Yevseev-Klimov, St Petersburg 2005 ) 15. ♕xh6cxd4 16. cxd4 ♕b6 17. d5 ♘e5 18. ♕d2 f5 19. ♖bd1♖ad8 EljanovAreshchenko, Sochi Russian Team Championship 2006 d2) 14. f4 d21) 14... ♖ad8⁈ 15. d5 ♘a5 16. f5!→ PH.Nielsen-Ganguly, Tripoli 2004 d22) 14... cxd4 15. cxd4 ♕a5 ( 15... e6 ) 16. d5 ♕xd2 17.♗xd2 ♘d4 18. ♔f2+/= d23) 14... e6 15. dxc5 ♖ad8 16. ♕c2 d24) 14... f5⁉ 15. dxc5 ( 15. exf5 c4 16. ♗e4 ♗xf5 17.♘g3 ; 15. d5 fxe4 16. ♗xe4 ♘a5 17. d6 ♕xd6 18. ♗xa8♕xd2 19. ♗xd2 ♖xa8 ) 15... ♖ad8 16. a4 bxa4 17.♕a2+ ♔h8 18. exf5+/= 13. ♕c1⁉ White aims to put the queen on a3, where it will put a lot of pressure on the black queenside. 13. ♗f4 e5⁉ ( 13... ♕c8 14. d5 ♘e5 15. ♗xe5 ♗xe5 16. f4 ♗g717. c4 e6 18. f5 exf5 19. exf5 b5 ShirovSvidler, Wijk aan Zee A 2007 ) 14. ♗g3 ( 14. ♗g5 exd4 15. cxd4 ♘xd4 16. ♗xd8 ♖xd817. ♘xd4 ♗xd4 18. ♗c4 b5 19. ♗d5 ♗c6 ) 14... cxd4 15. cxd4♕d6 16. d5 ( 16. ♖xb7 ♘xd4 17. ♗h4 ♗c6 ) 16... ♘b4 17. ♕d2( 17. ♗c4 ♖c8 18. ♕b3 a5 19. a3 b5 20. axb4 a4 ) 17... ♘xd318. ♕xd3 b5 19. f4 f5 20. fxe5 ♗xe5 21. ♗xe5 ♕xe5 22. ♘c3fxe4 13. d5 ♘e5 14. c4 e6 ( 14... ♘g4 15. ♗f4 ♗e5 16. ♗xe5 ♕xe517. ♘g3 b6 18. ♗e2 h5 19. f4 ♕g7± Van Hecke- Livshits, Vlissingen 2005 ) 13. ♕d2 ♗c8 14. ♖fc1 ♕d6 15. d5 ♘e5 16. ♗f4 ( 16. c4 e6 17.♘c3 ♘xd3 18. ♕xd3 b6 19. ♗g5 ♖de8 20. ♘b5 ♕d7 21. ♕d2exd5 22. exd5 a6 23. ♘c3 ♕g4 24. h3 ♕xc4 25. ♖xb6 ♕d4 26.d6 ♗e6 27. ♗e3 ♕xd2 28. ♗xd2 ♖b8 29. ♖cb1 ♖xb6 30.♖xb6 Beliavsky-Vachier Lagrave, Paks VI G. Marx Mem 2008 ) 16... f517. c4 b6 18. ♗xe5 ♗xe5 19. f4 ♗xf4 20. ♕xf4 ♕xf4 21. ♘xf4fxe4 22. ♘xg6 hxg6 23. ♗xe4 e6 24. ♖e1 ♔g7 Pinter-Antal, Hungarian Championship 07-08 2007 13. ♕c2 b6 14. ♖fd1 ♘a5 15. ♕b2 c4 16. ♗c2 e5 17. h3 ♖fe818. ♖d2 ♗c6 19. ♖bd1 ♘b7 20. ♕b1 ♕e7 21. dxe5 ♖xd2 22.♖xd2 ♕xe5 Sargissian-Vachier Lagrave, Paks VI G. Marx Mem 2008 13. h3 13... ♕d6 Ideally you'd like to solve all your problems tactically with this move, which leads to a long forced sequence:

13... ♗c8 14. ♖d1 e5 ( 14... cxd4 15. cxd4 ♕a5 16. ♖b5 ) 15.d5 ( 15. dxc5 ♗e6 16. c4 f5 17. ♗g5 ♖d7 18. ♘c3 ) 15... ♘a516. c4 b6 17. ♘c3 f5 18. exf5 gxf5 19. ♗g5+/= Tarnopolsky-Mikhalevski, Ashdod 3rd Festival 2006 13... cxd4 14. cxd4 a) 14... ♕d6 15. ♖d1 ( 15. d5 ♘e5 ) 15... ♘xd4 16. ♘xd4♗xd4 17. ♗xd4 ♕xd4 18. ♗b5 ♕xd1+ 19. ♕xd1 ♗xb5+/= b) ♖c8 15. ♕a3+/= 13... e5⁉ 14. d5 ( 14. dxc5 ♗e6 ) 14... ♘a5 a) 15. ♘g3 c4 ( 15... f5 16. exf5 gxf5 17. ♗g5 ) 16. ♗c2 b617. f4 exf4 18. ♗xf4 ♗e5 19. ♘e2 ♘b7 b) 15. f4 f5 16. exf5 ♗xf5 17. ♗xf5 ♖xf5 18. ♘g3 ♖ff8 19.c4 exf4 20. ♗xf4 ♗d4+ 21. ♔h1 ♗e5 c) 15. c4 15... f5 16. exf5 gxf5 17. f3 b6 18. ♗g5 ♖de8 19.♘g3± isn't a very good version of this structure. 13... ♗e6 14. ♗f4 a) 14... ♗e5 15. d5 ( 15. ♗xe5 ♘xe5 16. ♕f4 ♕d6 17. ♗b5a6∞ Graf-Bobras, Dresden 2006 ) 15... ♗xf4 16. ♕xf4 ♘e5 17.c4+/= b) ♕c8 15. d5 ♘e5 16. ♗xe5 ( 16. ♗c2 ♗d7 17. ♗h6 ♗xh618. ♕xh6 b6 19. ♘g3 ♕a6 20. h3 ♕a5 21. f4 ♕xc3 Antic-Mikhalevski, Las Vegas Masters 2006 ) 16... ♗xe5 17. c4 ♗d7 18.f4 ♗g7 19. e5+/= 14. ♖d1 14. d5 ♘e5⇆

14. ♖xb7 ♗c8 15. ♖b1 cxd4 16. cxd4 ♘xd4 17. ♘xd4 ♗xd4 18.♗c4 ♗e6 19. ♗xd4 ♗xc4 20. ♖d1 ♗xa2 21. ♖b5 f6= 14... cxd4 15. cxd4 ♘xd4 15... ♘b4 16. ♗c4+/= 15... ♖c8 16. ♕b2 ( 16. ♕d2 ♘xd4 17. ♘xd4 ♗xd4 18. ♗f1♗xe3 19. ♕xe3 ♕c7= ; 16. ♗f4 e5 17. dxe5 ♘xe5 18. ♕e3♗c6 ) 16... b6 ( 16... ♘a5 17. ♗f4 ) 17. ♗a6 ♖cd8 18. ♖bc1± 16. ♗xd4 ♗xd4 17. ♘xd4 ♕xd4 With the pawn on f2 - as Black didn't play 10...Bg4 11.f3 Bd7 - this isn't check so White can play 18. ♗b5 ♕xd1+ 19. ♕xd1 ♗xb5+/= Black is a queen down, although he clearly has excellent drawing chances as his queenside will be very secure with the bishop on c6. If his pawns were on e6 and g7 rather than e7 and g6 I think this would be a reasonably easy draw, but with the pawns where they are going for this position never really appealed to me. I felt that if White managed to develop something on the kingside it would become very unpleasant.

12.Bd3 Rfd8 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗c4 c5 8. ♘e2 ♘c6 9. ♗e3 OO 10. O-O ♗d7 11.♖b1 ♕c7 12. ♗d3 ♖fd8 The move I've normally played. 13. ♕d2! I think this is the best move for White in this position, and in fact if it wasn't for this I'd be reasonably satisfied with 12...Rfd8. Somewhat surprisingly, considering how much time I spent analysing this position and fine-tuning my approach, I somehow managed to almost completely ignore this move until it was played against me by Hikaru Nakamura. Then I realised that none of my options were really satisfactory. 13. ♗f4 e5 14. ♗g3 ( 14. ♗g5 exd4 15. ♗xd8 ♖xd8 16. cxd4♘xd4 ) 14... cxd4 15. cxd4 ♕d6 ( 15... ♖ac8 16. d5 ♘a5 17. f3♘c4 18. ♗f2 ♗h6 19. ♗xc4 ♕xc4 20. ♖xb7 ♗a4 21. ♕e1 ♗b522. ♘g3 ♕a4 23. ♕d1 ♕a6 24. ♖xa7 ♗xf1 25. ♘xf1 ♕b5Vitiugov-Timofeev, Sochi Russian Team Championship 2008 ) 16. d5♘b4 17. ♗c4 ( 17. ♕d2 ♘xd3 18. ♕xd3 b5 ) 17... ♖ac8 18.♖c1 b5 19. ♗b3 a5 13. ♕c1 Combatting this move is why I chose 12...Rfd8, because here Black is ready to play 13...Rac8. a) 13... a6⁉ 14. ♕a3 cxd4 15. cxd4 b5 ( 15... ♖dc8 16. d5♕a5 17. ♗c5 ♕xa3 18. ♗xa3 ♘e5 19. ♖xb7 ; 15... ♕a5 16.♕xa5 ♘xa5 17. ♖b6+/= ) 16. d5 ( 16. ♖fc1 ♖dc8 17. ♖c5♕a5∞ ) 16... ♘e5 ( 16... ♕a5 17. ♕xa5 ♘xa5 18. ♗b6+− ) 17.♖fc1 ♕b8 18. f4 ( 18. ♗c2 ♘c4 19. ♕xe7 ♘xe3 20. fxe3 )18... ♘xd3 19. ♕xd3+/= b) 13... cxd4 14. cxd4 ♕d6 15. ♖d1 e5⁉ ( 15... ♘xd4 16.♘xd4 ♗xd4 17. ♗xd4 ♕xd4 18. ♗b5 ♕xd1+ 19. ♕xd1 ♗xb5) b1) 16. dxe5 ♘xe5 17. ♗c4 ♕e7 ( 17... ♕xd1+ 18. ♕xd1♘xc4 19. ♗g5 ) 18. ♗d5 ♗c6 19. ♗c5 ♕h4 20. ♕f4 ♕xf421. ♘xf4 b6 22. ♗e3 ♖ac8 b2) 16. ♕c5⁉ ♕xc5 17. dxc5 ♘a5 18. ♗g5+/= ( 18. ♗d2 b619. ♗xa5 bxa5 20. ♗c4 ♗f8 ) b3) 16. d5 ♘b4 17. ♕a3 ♗f8∞ b4) 16. ♗g5 16... ♖dc8 17. ♕c5 ♕xc5 18. dxc5 ♖c7 19.♘c3+/= c) 13... ♖ac8 c1) 14. ♕a3 cxd4 15. cxd4 ♕a5 It's quite clear that Black's rooks are well-placed for this particular line. The rook on c8 means that Black is already fighting for the c-file, which is what he very much has to do in order to equalise. 16. ♕xa5 ♘xa5 17. ♗d2 ♘c418. ♗xc4 ♖xc4 19. ♖xb7 ♖a4 20. ♗g5 ♔f8 ( 20... f6 21.♗f4 ) 21. ♖fb1 ♔e8 22. ♖1b2 ♗xd4 23. ♘xd4 ♖xd4 24.f3 f6 25. ♗e3 ♖d1+ 26. ♔f2 a6 27. ♗c5 ♖c1 28. ♗a3♖dc8= c2) 14. ♕b2 b5! ( 14... cxd4 15. cxd4 ) 15. ♗xb5 ♖b8= c3) 14. ♖d1 c31) 14... cxd4 15. cxd4 c311) 15... ♕d6

c3111) 16. e5 ♕b8 ( 16... ♕c7 17. ♕b2 ) 17. ♕b2 c3112) 16. ♕b2 c312) 15... e6 16. d5 exd5 17. exd5 c313) ♕a5⁉ 16. ♕b2 ( 16. ♗d2 ♕a4 ; 16. ♕d2 ♕xd217. ♖xd2 ♘a5 ) 16... ♗e6 17. a3 ♖d7 18. ♕b5 ♗a2 (18... a6 19. ♕xa5 ♘xa5 20. d5 ; 18... ♕xb5 19. ♗xb5a6 20. ♗xc6 ) 19. ♖bc1 ♕xb5 20. ♗xb5 a6 21. ♗xc6♖xc6 22. ♖xc6 bxc6 23. ♖c1+/= c32) 14... ♕a5 15. ♖b5 ♕xa2 16. ♖xc5 b6 17. ♖b5♘a5 18. ♖b2+/= c33) 14... ♗e8 15. ♕b2 cxd4 16. cxd4 ♕a5 17. e5+/= c34) 14... a6 15. ♕a3 ( 15. d5 ♘e5 16. c4 b5 ; 15. dxc5♘e5=/∞ ) 15... cxd4 ( 15... b5 16. ♕xc5± ; 15... ♕a5 16.♕xc5 ♕xc5 17. dxc5+/= ) 16. cxd4 ♕a5 17. ♕xa5 ♘xa518. d5 ♘c4! 19. ♖xb7 ( 19. ♗g5 ♘b2 ; 19. ♗xc4 ♖xc4 20. d6 ♗a4 21. dxe7 ♖e8 22. ♖d8 ♖c7 23. ♖xe8+♗xe8 24. ♘f4 ♖xe7 25. ♘d5 ♖xe4 26. ♖xb7 f5= ) 19...♘b2 20. ♖d2 ♘xd3 21. ♖xd3 ♗b5 22. ♖d2 e6 c341) 23. g3 exd5 24. exd5 h5 ( 24... ♗xe2 25. ♖xe2♖xd5 26. ♖c2 +/= ) 25. ♘f4 ♖c1+ 26. ♔g2 ♗c3 27.♖e2 ♖d1 28. ♖c2 ♗e5 c342) 23. h4 exd5 24. exd5 h5 25. ♖b6 ( 25. ♗g5♗xe2 26. ♖xe2 ♖xd5 ) 25... ♗f8 c343) 23. dxe6 c3431) 23... ♖xd2 24. exf7+ ♔h8 25. ♗xd2 ♗xe226. e5 ( 26. ♖e7 ♗b5 ) c3432) 23... fxe6 24. ♖c7 ♖b8 25. ♘f4 ♗e5 26.♖c5 ♗xf4 27. ♗xf4 ♖xd2 28. ♗xd2 ♖d8 29. ♖c2♗d3 30. ♖c1 ♗xe4= 13. ♕b3 ♗e8 14. ♕xb7 ♕xb7 15. ♖xb7 cxd4 16. cxd4 ♘xd4= 13. h3⁉ Topalov played this against me in a tournament in China some years ago. Black has to be very, very concrete, because White is now ready to play f4 next move. a) 13... ♗e8 14. d5 ♘e5 15. c4 e6 ( 15... a6 16. ♘c1 ♖ab817. f4 ♘d7 18. ♕c2 ♘b6 19. ♘e2 ♘a4 20. e5 b5 21. ♘c3Porper-Mikhalevski, Edmonton Canadian Open 2009 ) 16. ♘f4 ♖ab817. ♗e2 ♖d6 18. ♕c2 ♖a6 19. ♖b5 ♗f8 20. a4+/= Topalov-Svidler, Nanjing 2008

b) 13... e6! The best move in the position, and one that's already been played in a high-level grandmaster game. 14. ♗g5 ( 14. e5 ♘e7 15.♗g5 ♗c6 16. ♕d2 ♖d7 ) 14... cxd4! Black should play this immediately. This is a very thematic exchange sacrifice which you'll see in a lot of positions here. ( 14... ♖e8 15. ♕c1 cxd4 16. cxd4♕d6 17. ♖xb7 ♖ab8 18. ♖xb8 ♖xb8 19. ♕c5 ♕xc5 20. dxc5 h6 21. ♗e3 Le Quang Liem-Sasikiran, Moscow Open A 2010 ) 15.♗xd8 ♖xd8 16. cxd4 ♘xd4 17. ♘xd4 ♗xd4 Black has given up the exchange simply to control the dark squares, and the bishop on d4 and to a certain extent the other bishop on c6 will be monsters. Black will have very nice compensation for the exchange. b1) 18. ♕e2 a6 19. ♔h1 ♗c6 20. f4 ♗a7=/∞ b2) 18. ♕c1 ♕e5 19. ♖xb7 ( 19. ♔h1 ♗c6 20. f4 ♕a5 )19... ♗b6 20. ♗a6 ♗b5 21. ♗xb5 ♕xb5 22. ♔h1 ♕a6= b3) 18. ♕c2 18... ♕e5 ( 18... ♕xc2 19. ♗xc2 ♗c6 20.♖fd1 e5 21. g3 a5 22. ♔g2 ♔g7 23. f3 b5 24. ♗b3 a4 25.♗d5 ♗xd5 26. exd5 ♖xd5 27. f4 ♖c5 28. fxe5 ♗xe5 29.♖bc1 ♖xc1 30. ♖xc1 b4 31. ♖c4 b3 32. ♖xa4 bxa2= ) 19.♖xb7 ♗b6 20. ♖b1 ♕d6 21. e5 ( 21. ♗f1 ♕c6!= ) 21...♕d5 22. ♖1xb6 axb6 23. ♕c7 ♗e8 24. ♗b5 ♕d1+ 25.♔h2 ♕d2 26. ♖b8 ♕f4+ 27. ♔g1 ♖d1+ 28. ♗f1 ♕a4 29.♖d8 ♖xd8 30. ♕xd8 ♔g7= 13. f4 This move allows Black a very interesting tactical solution: a) 13... cxd4 14. cxd4 ♗g4 15. d5 ♘d4 16. h3 ♗xe2 17.♗xe2 ♘xe2+ 18. ♕xe2+/= b) 13... e6 b1) 14. f5 exf5 15. exf5 cxd4 16. cxd4 ♕d6 ( 16... ♗xf5 17.♗xf5 gxf5 18. ♖xf5 ; 16... ♖e8 17. ♕d2 ♘e7 18. ♗f4♕c6 19. fxg6 hxg6 20. ♖fc1 ♕d5 GrandeliusHermansson, Reykjavik 23rd Open 2008 ) 17. fxg6 hxg6 18. ♖xb7 ♗e6 (18... ♘xd4 19. ♗c4 ♗e6 20. ♗xe6 ♕xe6 21. ♗xd4 ♗xd4+22. ♘xd4 ♕e3+ 23. ♔h1 ♖xd4 24. ♕b1+/= ) 19. ♕d2 ♘xd420. ♘f4 ♗c8 21. ♖bb1 ♘f5 b2) 14. ♕e1 14... ♗e8 15. ♕f2 cxd4 16. cxd4 ♘xd4 ( 16...♕e7 17. f5 exf5 18. exf5 ♖ac8 ) 17. ♘xd4 ♕c3 18. ♘xe6 (18. ♖b3 ♕xd4 19. ♗a6 ♕xe4 20. ♗xb7 ♗c6 21. ♗xa8♖xa8 ) 18... fxe6 19. ♖b3 ♕c7 20. ♖c1 ( 20. f5 exf5 21.exf5 ♗f7 ; 20. e5 ) 20... ♗c6 21. e5 g5 22. ♗b1 ( 22. ♗e4gxf4 23. ♗xf4 ♗xe4 24. ♖xc7 ♖d1+ 25. ♕f1 ♖xf1+ 26.♔xf1 ♖f8 27. g3 ♗xe5 ) 22... ♖f8 23. ♕c2 ♗h8 24.♖f1→ c) 13... ♗g4 14. d5 e6! c1) 15. c4⁉ c11) 15... ♘d4 16. ♕d2 ♗xe2 17. ♗xe2 b5 c111) 18. ♗d3 exd5 ( 18... ♖ab8 19. ♔h1 exd5 20.cxd5 c4 21. ♖bc1 ♕e7 22. e5 ♘f3 23. ♖xf3 ♖xd5 )19. cxd5 c4 20. ♖bc1 ♕e7 21. ♗b1 c112) 18. ♔h1

c12) ♘b4 16. h3 exd5 17. cxd5 ♘xd3 18. ♕xd3 ( 18.hxg4 c4 ) 18... c4 19. ♕c2 ♗xe2 20. ♕xe2 f5 ( 20... c321. ♕d3 c2 22. ♖bc1 ♖ac8 23. ♖f2 ♕c3 24. ♕d2 ♕a325. ♖e2 ) 21. ♖b5 ( 21. e5 ♖xd5 ) 21... fxe4 22. ♖c5♕d6 23. ♕xc4 b6 24. ♖b5 ♖ac8 25. ♕a4 ♕d7= c2) 15. dxc6 15... c4 Analysis of this position shows that Black's compensation will be fine even with the pawn on b7, where it will survive for a while. 16. cxb7 ( 16. ♕a4 cxd3 17. ♖xb7 dxe218. ♖xc7 ♖d1 19. ♖xd1 exd1=Q+ 20. ♕xd1 ♗xd1 21. ♖b7♖c8 22. c7 ♗e2 23. e5 f6 24. exf6 ♗xf6∓ ) 16... ♖ab8 c21) 17. ♕e1

c211) 17... cxd3 18. ♘d4 ♗xd4 ( 18... ♖xb7 19. ♖xb7♕xb7 20. ♕h4 ♖xd4 21. cxd4 ♕xe4 22. ♕d8+ ♗f823. ♗d2 ) 19. ♗xd4 ♖xb7 20. ♕d2 ♗e2 21. ♖fc1♖db8 22. ♖xb7 ♖xb7 23. ♗f6 ♕b6+ c212) ♖xd3 18. ♘c1 ♖xc3 19. h3 ♗d1 20. e5 ( 20.♗d2 ♗c2 ) 20... ♖a3 21. ♕xd1 ♖xe3 22. ♖f3 c3!=/+ c22) 17. ♗d4 17... cxd3 18. ♕xd3 ♖xb7 ( 18... e5 19.fxe5 ♗xe5 20. ♗xe5 ♕xe5 21. ♘d4± ) 19. ♖xb7 ♕xb720. e5 ♗f5=/∞ 13. d5 ♘e5 14. c4 ( 14. f4 ) 14... e6 15. ♘f4 b6 16. ♗e2 h6 17.♕c1 ♔h7 18. ♘d3 exd5 19. cxd5 f5⇆ Lahner-Meszaros, Frydek Mistek CZE 2005

13... ♖ac8 This is perhaps what I should have played, but Black still runs into problems.

13... a6 I eventually played this after spending quite a lot of time here. This is an "all-purpose" move aimed specifically at meeting 14.d5 Ne5 15.c4 with the ...b5 break, but Hikaru responded correctly: 14. f4! e6 If Black allows White to play d5 he'll just get steamrolled. 15. dxc5 ♘a5 I eventually won a very nice game which I'm quite proud of because it showed a lot of interesting Grünfeld subthemes in the later stages, and in general a victory with Black against a very strong player is something to cherish. Objectively, however, this position just isn't very good for Black. After both Hikaru's move and the much simpler 16...Bd4 Black is just worse, so this can't be recommended. a) 16. ♘d4 e5 17. fxe5 ♕xe5 18. ♕f2 ( 18. ♕b2⁈ ♖ac8 19.♕b4 ♕xc5 20. ♕xc5 ♖xc5 21. ♘e2 ♖c6 22. ♗b6 ♖xb6 23.♖xb6 ♗c6!=/∞ Nakamura-Svidler, Moscow Tal Memorial 2011 ) 18...♗e8 ( 18... ♕e7 19. h4 ♗e8 20. ♖bd1 ♖ac8+/= ) 19. ♗c2 (19. h4⁉ ) 19... ♖ac8 20. ♘e2± /\ Bd4 and Nf4-d5 b) 16. ♗d4! , preparing a direct attack on the kingside 16... ♗f8 (16... ♗b5 17. ♗xb5 axb5 18. ♖xb5 ♘c4 19. ♕d3 ♘a3 20.♖b3 ♗f8+/= ) 17. ♗f6 ♗xc5+ 18. ♘d4 ♖dc8 19. ♔h1→ 13... ♗e8 14. ♖fc1 e6 15. ♗g5⁉ ( 15. ♗h6 ♗h8 16. ♕g5 ♕d617. ♖xb7 cxd4 18. cxd4 ♘xd4 19. ♘xd4 ♕xd4 20. ♕e7 ♗g7 21.♗xg7 ♔xg7 22. ♗f1 ♕xe4 23. ♖xa7 Sundararajan-Sasikiran, Kolkata 3rd Open 2008 ) 15... ♖d7 16. f4 cxd4 17. cxd4 ♗xd4+ 18.♘xd4 ♖xd4 19. ♗f6=/∞ 13... cxd4 14. cxd4 ♕a5 was my initial idea here in my game against Nakamura, but then I realised that after 15. ♕xa5 ♘xa5 , and then the immediate 16. d5 , or Rfd1 and then d5, with Bd2 also an option, this just isn't equal for Black. 14. d5

14. ♗f4? e5 15. ♗g3 cxd4 16. cxd4 ♘xd4 17. ♘xd4 ♕c3 14. f4 e6 15. f5 ( 15. ♕b2 ♘a5 16. dxc5 ♗f8 17. ♗d4 ♗xc5 18.♘g3 ♗xd4+ 19. cxd4 ♕c3 ) 15... exf5 16. exf5 cxd4 17. cxd4♕d6 ( 17... ♗xf5 18. ♗xf5 gxf5 19. ♖xf5 ♕d7 20. ♖g5 ) a) 18. ♖xb7 ♘xd4 19. fxg6 hxg6 20. ♖xa7 ♗e8 b) 18. fxg6 hxg6 19. ♗c4 ♗e6 20. ♗xe6 ( 20. d5 ♗xd5 21.♗xd5 ♕xd5 22. ♕xd5 ♖xd5 23. ♖xb7 f5 ) 20... ♕xe6= 21.♖xb7? ♘a5 22. ♖bxf7 ♘c4 23. ♗h6 ♘xd2 24. ♖xg7+ ♔h8 25. ♘f4 ♖f8!−+ c) 18. ♗f4 18... ♗xd4+ 19. ♘xd4 ♕xd4+ 20. ♗e3 ♕d6 21.♖xb7 ♘e5 22. fxg6 hxg6 23. ♗e4 ♕xd2 24. ♗xd2 ♗c6 25.♗xc6 ♖xd2 26. ♗a4 ♖c4 27. ♗b3 ♖e4 14. ♗h6 ♗xh6 ( 14... cxd4 15. ♗xg7 ♔xg7 16. cxd4 ) 15. ♕xh6cxd4 16. cxd4 ♕d6 17. ♗b5 ( 17. ♖fd1 ♘xd4 18. ♘xd4 ♕xd419. ♗b5 ) 17... ♘xd4 ( 17... a6 ) 18. ♘xd4 ♕xd4 19. ♖bd1♗xb5 20. ♖xd4 ♖xd4 21. ♖e1 ♖c2 22. ♕f4 ♖d6 14... ♘e5 15. c4+/= Although this isn't tragic in itself I think both black rooks are misplaced. Black's major sources of counterplay are the e6/f5 and also the a6/b5 breaks, for which the rook on d8 and especially the rook on c8 are doing absolutely nothing. The rook on c8 would actually be better off on a8, because if the a-file gets opened it would become very active there.

10...Bg4 1. d4 ♘f6 2. c4 g6 3. ♘c3 d5 4. cxd5 ♘xd5 5. e4 ♘xc3 6.bxc3 ♗g7 7. ♗c4 c5 8. ♘e2 ♘c6 9. ♗e3 OO 10. O-O ♗g4 Let's finish by summarising some of the differences if Black correctly starts with this move rather than 10...Bd7. 11. f3 ♗d7 12. ♖b1 ♕c7 13. ♗d3 13. ♗f4 ♕c8 14. ♖c1 ( 14. dxc5? is a mistake because 14... ♘a515. ♗b5 ♕xc5+ comes with check. ) 14... ♕d8⁉ This is a very fresh idea Areshchenko played against Bacrot. It seems to be a very strange idea, as Black's last three moves have succeeded in transferring the queen from d8 to d8 at the apparent loss of three tempi. However, the white bishop isn't really ideally placed on f4. If White plays 14.Be3 we transpose back to something we discussed in Part 1. Black will simply play 14...Rc8 and be very happy with his development. If White plays 15. ♕d2 , as Bacrot did, then 15... cxd4 16. cxd4 ♕a5 is quite a good justification of Black's previous play. 17. ♕xa5 ♘xa5 18. ♗d3 ♖fc8 19. ♖b1 a620. ♖fc1 ♖xc1+ 21. ♖xc1 ♖c8 22. ♗c7 ♘c6 23. ♗b6 e6 Bacrot-Areshchenko, Bundesliga 2012. 13... ♖ad8 14. ♕c1

14. ♕d2 White is almost forced to play another move like this one. Play might continue: 14... ♗c8 15. ♖fc1 In this position it's quite clear that Black has made the correct choice with his rooks, because the bishop is comfortable on c8, the rooks are where they should be and Black can just play, for instance, 15... b6 White needs to show he has a plan here because Black is harmoniously developed and will continue to improve his position. White is somewhat stuck for ideas, and this alone should lead anyone who wants to play this line to start with 10...Bg4. ( 15... ♕d6 is also possible and has been played in a number of games in a position where the white pawn is on f2. With the pawn on f3 White is perhaps slightly more comfortable, but even here this move is interesting for Black. It almost forces White to play 16. d5 , and after 16... ♘e5 Black will start undermining the white centre, with both ...e6 and ...f5 possible in the immediate future. 17. c4 b6 18. ♖b3 f5 19. ♗f4fxe4 20. ♗xe4 e6 21. ♗g3 exd5 22. ♗xd5+ ♗e6 Cvitan-Bosiocic, Sibenik 2011; 15... e6 16. ♗f4 ♕e7 17. ♗g5 ♗f6 18. ♗xf6 ♕xf619. ♗b5 cxd4 20. ♗xc6 bxc6 21. cxd4 ♗a6 Sanchez Ibern-Bacallao Alonso, Barbera del Valles XXXIX Open 2011) 14... cxd4 15. cxd4 ♕d6 Here, after the inclusion of 10...Bg4 11.f3, this move solves all Black's problems. 16. d5 White is forced to play this advance. 16. ♖d1 is no longer an option, because after 16... ♘xd4 17. ♘xd4♗xd4 18. ♗xd4 Black plays 18... ♕xd4+ with check. 16... ♘b4 This move, or 16... ♘e5 will give Black excellent counterplay. It's very clear that in this line you should therefore start with 10...Bg4 11.f3 Bd7.

1.c4 g6 2.Nf3 2.e4 e5 3.Nf3 Bg7 4.d4 exd4 5.Nxd4 Nc6

1.c4 g6 2.Nf3 In this series I also want to give you an idea of what to do when people aren't ready to play ball and try not to play the Grünfeld against you. In this case we're dealing with people who play 1.c4. 1. c4 There's obviously nothing wrong with all the usual replies to this move. I've been a 1...e5 player all my life and nothing should discourage you from playing the classical lines after 1.c4, but I want to give you an idea of a different approach to this position. 1... g6 This is a bit of a "left-field" reply. It aims, first of all, to confuse White somewhat and make him play positions he isn't as accustomed to as he is to the normal positions English Opening players are looking for. Moreover, in many cases this leads to unclear and interesting positions that are fun to look at. The main lines arise after 2.e4 e5 and now either 3.Nf3 or 3.d4, but we first need to cover all the other bases. Obviously there are many moves White can make in this position. 2. ♘f3 This is White's main alternative to 2.e4. 2. d4 is the easiest move for our purposes. We of course simply reply with 2... ♘f6 and we're back where we started in the Grünfeld positions. 2. ♘c3 c5 is very nice for Black. White has "triangulated" himself out of the comfort zone because he can no longer get the proper English positions. 3. ♘f3 ( 3. g3 , switching to playing the 3.g3 English Opening, is of course an option, but we've already achieved a minor victory here as this line isn't a source of worry for Black. ) 3... ♗g7 4.d4 ( 4. e3 ♘f6 5. d4 cxd4 6. exd4 d5 Once again Black gets a fantastic version of the positions he's normally aiming for. ) 4... cxd4 5.♘xd4 ♘c6 White has no time to play e4 and establish the Maroczy Bind. 2... ♗g7 3. g3 This move has some separate value. 3. d4 ♘f6 once again transposes back to the Grünfeld. 3. ♘c3 c5 4. d4 cxd4 5. ♘xd4 ♘c6 is clearly good for Black because White has to choose between the rather sad-looking 6.e3 or 6. ♘c2 , which allows 6... ♗xc3+ 7. bxc3 ♕a5 , and Black is already fighting for an advantage. 3... c5 4. ♗g2 ♘c6 With the knight still on g8 White can't get the comfortable Maroczy positions he's probably aiming for. 5. ♘c3 The only ambitious approach White can try in this position. 5. O-O can be met by 5... e6 6. ♘c3 ♘ge7 and Black is extremely comfortable in this kind of position where he'll be able to play ...d5 in one tempo. In general this is perfectly safe for Black. 5... e6 6. d4 cxd4 7. ♘b5 d5 8. cxd5 exd5 This is now the main move and is quite comfortable for Black.

8... ♕a5+ was supposed to be very nice for Black and was considered the refutation of White's idea until in a game against Emil Sutovsky in 2012 Hracek played 9.Nd2. a) 9. ♕d2 ♕xb5 10. dxc6 ♘e7! ( 10... ♕b6 11. b3 ♘f6 12.♕xd4 ♘h5 13. ♕a4 ♕xc6 14. ♕xc6+ bxc6 15. ♖b1+/=Savchenko-Bocharov, Olginka RUS-tch 2011 ) 11. ♘xd4 ♕b6 12.OO ( 12. ♘b3 ♘xc6 13. O-O O-O Grachev-Reshetnikov, Taganrog 64th RUS-ch HL 2011 ) 12... OO 13. cxb7 ♗xb7 14. ♗xb7 ♕xb715. ♘b3 ♖fd8 16. ♕f4 ♘d5 17. ♕f3 a5 18. ♘c5 ♕b5 19.♘d3 ♘b4=/∞ Gelfand-Anand, Zuerich Chess Challenge Pairings 5' 2013 b) 9. ♘d2 9... ♕xb5 10. dxc6 ♘e7 11. a4! Black actually runs into some serious problems, and this changed the evaluation of a line which for many years was considered very safe for Black. ( 11. cxb7 This automatic capture would pose Black no problems. ) 11... ♕b4 ( 11...♕b6 12. ♘c4 ♕b4+ 13. ♗d2 ♕xc4 14. cxb7 ♗xb7 15. ♗xb7♖b8 16. ♖c1 ♕a2 17. ♗a6 ♕xb2 18. ♗b5+ ♖xb5 19. axb5♕xb5 20. ♖b1 ♕c6 21. O-O O-O 22. ♗b4+/= Hracek-Sutovsky, Eilat 17th ECC 2012; 11... ♕a6 12. b4!± ) 12. b3! b1) 12... ♘xc6 13. ♗a3 ♕a5 14. OO ♗f8 15. ♘c4 ♕d8 16.♗d6! ( 16. b4 ♘xb4 17. a5 ♘a6 18. ♗d6 ♗xd6 19. ♕xd4O-O 20. ♘xd6 ♕e7 Mikhalevski-Lenderman, Las Vegas 2013)16... ♗xd6 17. ♗xc6+ bxc6 ( 17... ♔e7 18. ♕xd4 ) 18.♕xd4+/= b2) bxc6 13. ♗a3 ♕b7 14. ♘c4 OO 15. OO ♕c7 16. ♖c1 (16. ♗d6 ♕d7 17. ♗e5 ♗a6 ) 16... ♗a6 17. ♘a5!+/= 9. ♘fxd4 This is the only potential problem Black faces in this position. My attempts to equalise immediately failed, so Black first needs to finish his development. 9. ♘bxd4 ♘ge7 is quite harmless. 10. O-O O-O Black has a very simple plan. He wants to play ...Re8 and then ...Nf5, immediately beginning a fight for the d4-square. For instance: 11. ♗e3 ♖e8 12.♖c1 ( 12. ♕d2 ♘f5 13. ♘xf5= was fine for Black in Huzman-Avrukh, ISR-tch 2012 ) 12... ♘f5 13. ♘xc6 bxc6 14. ♗d4 ♘xd4 ( 14...♗a6 Bologan played this in the same European Club Cup and still drew, although I think he suffered a little. 15. ♖e1 ♘xd4 16. ♘xd4 ♗xd417. ♕xd4 ♗xe2 18. ♖xc6 ♗c4 Malakhov-Bologan, Eilat 17th ECC 2012 ) 15. ♘xd4 ♕b6 This is perfectly playable for Black because 16.♘xc6 would be met by 16... ♗xb2 17. ♖c2 ♗f5 and the rook on c2 is a bit stuck. 9. O-O is the latest try by the Master himself, but even here White didn't get very far: 9... ♘ge7 10. ♗f4 O-O 11. ♘c7 ♖b8 12. ♘b5 ♖a813. ♕d2 ♗g4 14. ♘bxd4 ♕b6 15. ♗e3 The game went a) 15... ♘xd4 , and now White would have retained some pressure after 16. ♘xd4⁉ ( 16. ♗xd4 ♗xd4 17. ♘xd4 ♘c6 18. ♘b3♖ad8 was fine for Black: 19. ♖ac1 ♖fe8 20. ♖fe1 ♘b4 21. h3♗e6 22. ♕d4 ♘xa2 23. ♕xb6 KramnikAreshchenko, Tromso World Cup 2013) 16... ♕a6 17. a4+/= b) 15... ♕a6⁉ 16. ♖fd1 ♖fe8 , and Black should be fine: 17. h3♗xf3 18. ♗xf3 ♘e5= 9... ♘f6 9... a6 doesn't equalise fully: 10. ♘xc6 bxc6 11. ♘d4 ♘e7 12. OOc5 13. ♘b3 ( 13. ♘f3 ) 13... c4 14. ♘d4 OO 15. ♗g5 ♕d7 16.♕d2+/=

9... ♘ge7 a) 10. ♗f4 O-O 11. O-O a6 12. ♘xc6 bxc6 13. ♘c3 ( 13. ♘d6♗e6 ) b) 10. O-O 10... O-O 11. ♗e3 isn't particularly attractive because Black will have additional problems. It's much harder for him to play in the same vein with the white knight on b5 rather than f3. 10. O-O O-O 11. ♗e3 Black needs to be precise and act fast in this position. 11... ♘g4 Peter Leko played like this against Boris Gelfand and I think he was correct to do so. 11... a6⁉ is also quite interesting: a) 12. ♘c3 allows White to settle when I think he will be a little better. I don't particularly like this position, although after 12... ♘g413. ♘xd5 ( 13. ♘xc6 bxc6 14. ♗c5 ♖e8∞ ) 13... ♘xe3 14.♘xc6 ♘xd1 15. ♘xd8 ♘xf2 16. ♘xf7 ♖xf7 17. ♖xf2 ♖xf218. ♔xf2 ♗e6 Black should hold. b) 12. ♘xc6 12... bxc6 13. ♘c3 ( 13. ♘d4 c5 14. ♘b3 c4 15.♘d4 ♖e8∞ ) 13... ♖b8 14. b3 ♖e8 is quite ok for Black: 15.♗d4 ♗f8 16. ♖e1 ♗f5 , and Black's active pieces compensate for his somewhat compromised structure. 12. ♗f4 Once again Black needs to be quite fast. 12... a6⁉ I discussed this with Rustam Kasimdzhanov, who is also an expert here, and we think this move is a very good idea for Black. 12... ♘xd4 13. ♘xd4 ♕b6 14. ♘b3 ♗xb2 15. ♖b1 ♗g7 16.♕xd5 Although Leko drew eventually this position isn't much fun for Black and I don't want to recommend it. 16... ♘f6 17. ♕d6 ♗f5 18.♖bc1 ♗e4 Gelfand-Leko, Tashkent GP 2012 13. ♘xc6 bxc6 14. ♘d4 The most natural reply for White is of course to attack the hanging pawns immediately. 14. ♘c3 a5⇆

14... g5⁉ Driving back White's pieces. This is Rustam's idea and I quite like it. 15. ♗c1 Forced. 15. ♘xc6 ♕f6 just loses a piece. 15... ♕f6 16. ♘f3 The critical move. 16. e3 can be met by, for instance, 16... ♖e8∞ and visually it's quite obvious why Black is doing what he's doing. He's driven the bishop away from its beautiful post on f4 and the pawn on e3 is also locking in the bishop and making it hard to reactivate. Black has a very free position and his weakness on the kingside isn't particularly felt. 16... h6 Once again White immediately needs to start proving to Black that ...g5 is an important weakening, so 17. h4 Otherwise Black would just be comfortable.

17... ♖e8 I like the idea of just giving up the g5-pawn. 17... gxh4 is also possible. 18. hxg5 hxg5 19. ♘xg5 19. ♗xg5 ♕xb2 isn't very good for White because both the rook on a1 and the pawn on e2 are hanging. 19... ♖b8 20. ♘f3 a5→ In this position Black is a pawn down with no immediate threats, but on the other hand it's very, very hard for White to develop his pieces in any kind of harmonious manner - I think Black has excellent compensation for the pawn. So this is an interesting proposed improvement on the Gelfand-Leko game if you're feeling a bit adventurous and have nothing against the idea of sacrificing some material for free piece play.

1.c4 g6 2.e4 e5 3.Nf3 Bg7 4.d4 exd4 5.Nxd4 Nc6 1. c4 g6 2. e4 By far the most critical attempt for White and one that requires quite a bit of precision from Black. 2... e5 Black should play this so as not to land in some unpleasant Maroczy Bind positions. White has two main moves here, 3.d4 and 3.Nf3. Let's look at 3.Nf3 first. 3. ♘f3 ♗g7 4. d4 exd4 5. ♘xd4 Black once again has a choice, this time between 5...Nf6 and 5...Nc6. 5... ♘c6 The most obvious move in the position and one I made myself in a game against Bu Xiangzhi in the Khanty-Mansiysk Olympiad in 2010. 6. ♗e3 The main move for White. 6. ♘xc6 bxc6 is quite comfortable for Black because he has a very natural plan. a) 7. ♗e2 lets Black demonstrate his plan: 7... d6 8. O-O ♘e7 9.♘c3 O-O If allowed Black's next moves will be ...c5, ...Nc6 and ...Nd4. If all of that actually happens Black will have a very comfortable position and perhaps even some pressure. 10. ♗g5 ( 10.♗e3 c5 11. ♕d2 ♖b8 12. b3 ♘c6 13. ♖ad1 ♘d4 Lalith-Areshchenko, Gurgon 2009 ) 10... h6 11. ♗h4 ( 11. ♗e3 f5 12.♕d2 ♔h7 13. ♖ad1 fxe4 14. ♘xe4 ♘f5 15. ♗f4 ♘d4 16.♗d3 c5 17. ♘g3 ♕h4 18. ♗e3 ♗b7 Dudukin-Kornev, Vladimir 2004 ) 11... g5 12. ♗g3 ♘g6 13. ♗g4 ♖b8 14. ♗xc8 ♕xc8 15.♕d2 ♕e6∞ Pushkov-Trapl, Dresden EUR-tch Seniors 2010 b) 7. ♘c3 7... d6 8. c5 ( 8. ♗e3 ♘f6 9. ♗e2 O-O 10. O-O♖e8 ) 8... ♘f6 Since White is also undeveloped it's very hard for him to do much about the potential weaknesses in the centre. 9. ♗c4 ( 9.cxd6 cxd6 10. ♗e2 O-O 11. O-O ♖e8 Black is the first to begin active play and I think he's at least very comfortable here. 12. ♗f3♖b8 13. ♖b1 ♗a6 14. ♖e1 ♘d7=/+ Kolev-Rashkovsky, Groningen 1990 ) 9... O-O 10. O-O ♘g4 11. cxd6 cxd6 12. ♗f4 ♘e5 13.♗e2 ♗e6 14. ♖c1 f5= I.Ivanov-Hawkes, Los Angeles 1982 6... ♕f6 Necessary unless you want to find yourself in well-known Maroczy-type positions which aren't so pleasant for Black. 7. ♘xc6 Pretty much forced.

7... dxc6 8. ♘c3 ♕e7 An interesting attempt by Black to get an unbalanced and somewhat assymetric position with active piece play. The first game I saw starting from this position was the game between Loek van Wely and Vassily Ivanchuk in the Foros tournament in the late 2000s. 9. ♗e2 This move played by Bu Xiangzhi against me is, I think, a more judicious approach for White. 9. ♗d3 was Van Wely's move. 9... ♘f6 10. h3 ( 10. OO⁈ ♘g4 11.♗f4 OO 12. h3 ♘e5 13. ♗e2 ♗e6=/+ ; 10. f3 OO 11. OO ♘d7 12. ♕c2 ♘e5 13. ♗e2 ♗e6 14. b3 g5∞ ) 10... ♘d7 11. O-O O-O12. ♕c2 ♘e5 An even more ambitious approach. ( 12... ♘c5⁉ deserves a lot of attention. ) 13. ♗e2 g5 A very attractive idea which you might be familiar with from Benoni structures. Black secures a nice outpost on e5 for his knight. The game was very unclear and I can't really claim that Black is already better, but when faced with this fresh approach Loek started to go quite badly wrong and lost very quickly. 14. ♖ad1 ( 14.f4 gxf4 15. ♗xf4 ♗e6⇆ ) 14... ♗e6 15. b3 ♖fe8 , with very unclear play, Van Wely-Ivanchuk, Foros 3rd Aerosvit 2008 9. ♕d2⁉ f5! 10. exf5 ( 10. f3 ♘f6 11. ♗d3 fxe4 12. ♘xe4 ♘xe413. ♗xe4 ♗e6 14. OOO OO 15. ♕c2 ♗f5 = ) 10... ♗xf5 11.O-O-O ♘f6 12. ♗d3 ♖d8 13. ♕e2 ♖xd3 14. ♖xd3 ♗xd3 15.♕xd3 ♘g4 16. ♗d4 O-O∓ Tratar-Cicak, Campillos IX Malaga op 2006

9... ♘f6 White could now try to keep his edge by playing

10. f3⁉ Strictly speaking I think this is a better move. White doesn't allow Black freedom to jump with his knight. 10. ♕c2 was played against me, and after 10... O-O 11. O-O ♘g4 12.♗xg4 ♗xg4 13. ♖ac1 ♖ad8= I was already very comfortable. I still managed to create some problems for myself later on, but the game was eventually drawn. Bu Xiangzhi-Svidler, Khanty-Mansiysk 39th Olympiad 2010 10... O-O 11. O-O ♗e6 12. ♕c2 ♘d7 12... ♘h5⁉ 13. ♖ad1 White's position is perhaps slightly to be preferred - Black hasn't achieved full equality just yet - but it's still very playable for Black. There's nothing particularly wrong with his position and he can play on the queenside with e.g. ...a5 and ...Nc5, or he can try to break up the centre straightaway by playing ...f5 immediately. 13... f5

1.c4 g6 2.e4 e5 3.Nf3 Bg7 4.d4 exc4 5.Nxd4 Nf6 6.Bd3 6.Nc3 0-0 7.Be3 7.Be2 1.c4 g6 2.e4 e5 3.d4 Nf6 4.dxe5 4.Nf3

3.Nf3 Bg7 4.d4 exc4 5.Nxd4 Nf6 6.Bd3 1. c4 g6 2. e4 e5 3. ♘f3 ♗g7 4. d4 exd4 5. ♘xd4 ♘f6⁉ I think this is the best and most ambitious move here for Black and it's the one I want to recommend. Black's main idea is not to switch to the classical King's Indians but to play ...d5 in one move, saving an all-important tempo on ...d7-d6. I think it actually works and Black manages to achieve that aim quite successfully in all variations, although there are quite a lot of similar-looking but different and important lines which require our attention. White's two main moves now are obviously 6.Bd3 and 6.Nc3. 6. ♗d3 6. e5⁈ deserves only a brief side note, as 6... ♘e4 7. ♕e2 ♘c5 really is inadvisable for White. The pawn on e5 is quite clearly a weakness not a strength. It will require a lot of attention just not to lose it here. 6... O-O 7. O-O c6 A very natural move. 7... d5⁉ is possible straightaway, and regardless with which pawn White captures on d5 Black will follow up with ...c6 and ...cxd5. Black can force the matters even more by starting with 8. cxd5 ( 8. exd5 c6 9. ♗g5cxd5 10. ♘c3 transposes to the main line. ) 8... c6 9. ♘c3 transposes yet again and has the added bonus of avoiding White's additional option of 7...c6 8.e5, although as I explain below that isn't a very dangerous line. 8. ♘c3 8. e5 ♘e8 This is a very normal position for this kind of line. By trying to save on ...d6 Black in many positions of course allows White to play e5, meaning that the knight has to leave f6. The pawn on e5 is then immediately exchanged. For instance, after 9. ♘f3 d6 the pawn will be traded off soon and, in general, the structure where Black has a pawn on c6 against a pawn on c4 with no more pawns in the centre is quite comfortable for Black. 10. ♗g5 ♕d7 11. ♗f4 dxe5 12. ♗xe5 ♗xe513. ♘xe5 ♕d6 14. ♖e1 ♘f6 8... d5 White now has a choice. 9. exd5 The more positional approach. 9. cxd5 cxd5 a) 10. exd5 ♘xd5 11. ♗e4 ♘xc3 12. bxc3 ♕a5 13. ♕b3 ( 13.♖b1 ♕xa2 14. ♗e3 ♕c4 15. ♕f3 ♘c6 16. ♘xc6 bxc6 17.♗xc6 ♗g4 Vokac-Gross, Bratislava 1991 ) 13... ♘d7 14. ♕a3♕xa3 15. ♗xa3 ♖e8 16. ♖fe1 ♗f8 1/2-1/2 Petrosian-Adorjan, Sochi 1977

b) 10. e5 leads to very forcing play. 10... ♘g4 11. f4 b1) 11... f6⁉ is a somewhat cheeky move, considering just how undeveloped Black is, but it's actually still quite playable. 12. f5 (12. ♗e2 fxe5 13. ♘db5 a6 14. ♕xd5+ ♕xd5 15. ♘xd5axb5 16. ♘e7+ ♔h8 17. ♘xc8 ♘xh2 18. ♔xh2 ♖xc8 19. f5gxf5 20. ♖xf5 ♘c6 21. ♖f7 e4= ; 12. exf6 ♕xf6 13. ♘db5 a6 14. h3 ♕b6+ 15. ♔h1 ♘f6 16. ♘a4 ♕d8 17. ♘bc3♘c6=/+ ; 12. h3 fxe5 ) 12... fxe5 13. ♕xg4 e4 14. ♗e3exd3 15. ♕g3 ♘c6∞ b2) ♘c6 A more solid approach. 12. ♘xc6 Forced. 12... bxc6White now has to do something about the big threat of ...Qb6+, after which Black aims to play ...f6, exchange off the e5-pawn and so lengthen the diagonal for the g7-bishop. After the possible 13.♘a4 ( 13. ♕e2 ♖e8 14. h3 ♘h6 15. ♕f2 f6 16. ♘a4 fxe517. fxe5 ♖f8 18. ♕d4 ♘f7 19. ♗f4 ♕h4 20. ♖ae1 g5 21.♗e3 ♘xe5=/+ ) 13... f6 14. h3 ♘h6 15. exf6 ♕xf6 16. ♗e3anoth er typical idea for Black is to play 16... ♗f5 If the light-squared bishops come off the h6-knight will have a very nice outpost on f5. In general I think Black is at least very comfortable here. 17.♗c5 ♖fe8 18. ♖f2 ♗xd3 19. ♕xd3 ♘f5=/+ 9... cxd5 10. ♗g5 h6 It's important to drive the bishop away from g5 before committing to some kind of structure in the centre. 10... dxc4 11. ♗xc4 h6 12. ♗h4 a6 13. ♗b3 ♘bd7 14. ♕c2♘c5 15. ♖ad1 ♕b6 16. ♗c4 ♗g4 17. ♖d2 g5 18. ♗g3+/=Uhlmann-Jansa, Linz 2000 11. ♗e3 White hopes at some point to win a tempo by hitting the h6-pawn with Qd2, but Black can play very forcefully here. 11. ♗f4 doesn't make much sense because the bishop is doing little there. 11... ♘c6 a) 12. ♘xc6 bxc6 13. ♖e1 ( 13. cxd5 ♘xd5 14. ♘xd5 ♕xd5 )13... ♗e6 b) 12. ♘db5 12... ♗g4!∞ 11. ♗h4 is met by 11... ♘c6 12. ♘xc6 Pretty much forced. 12...bxc6 13. cxd5 cxd5 In this position with an isolani on d5 it would be nice for White to have a potential plan of playing Qd2 and then Bh6, but once Black has driven the bishop to h4 that's no longer an option for White. In general Black should be very comfortable in this structure because he has good dynamic play behind the isolated pawn. 14. ♖e1♖b8= 11... dxc4 12. ♗xc4 ♘g4 Exchanging off the bishop. 13. ♘db5 The only ambitious move for White, but after 13... ♕xd1 13... ♗e5 14. ♕xd8 ( 14. h3 ♘xe3 15. fxe3 ♕e7 16. ♘d5 ♕g517. ♘bc7 ♗xh3 18. ♕f3 ♗f5 19. ♘xa8 ♘c6 20. ♘ac7 ♕h4=/∞ )14... ♗xh2+ 15. ♔h1 ♖xd8 16. ♗xa7+/= 14. ♖axd1 ♘xe3 15. fxe3 ♘c6 16. ♘d6 ♘e5 the position after all the trades is quite safe for Black. I checked it for a bit because it looks as though White has some pressure, but in fact after, let's say, 17. ♗d5 ♖b8= Black is just fine here. Play might continue:

18. ♘ce4 ♗e6 19. ♗xe6 fxe6 20. h3 ♖bd8 21. b3 ♖xf1+ 22.♔xf1 b6 23. ♔e2 ♗f8=

3.Nf3 Bg7 4.d4 exc4 5.Nxd4 Nf6 6.Nc3 0-0 7.Be3 1. c4 g6 2. e4 e5 3. ♘f3 ♗g7 4. d4 exd4 5. ♘xd4 ♘f6⁉ 6.♘c3 The most critical move. 6... O-O White once again has a variety of options - the critical ones are 7.Be2 and 7.Be3. 7. ♗e3 7. ♗g5⁈ was once played by Viktor Korchnoi, but it's a very dubious idea. After 7... ♖e8 8. ♕f3 h6 the white bishop actually went back to e3: 9.♗e3 d6 Creating the threat of ...Ng4 and potentially ...Bg4 in some positions. I think David Navara was already better here. Korchnoi-Navara, Prague 2003 7. ♗d3 c6 a) 8. ♗f4 d5 9. exd5 ( 9. cxd5 cxd5 10. ♘db5 dxe4 ) 9...♖e8+ b) 8. O-O 8... d5 transposes to the line we saw after 6.Bd3. 7. g3 is possible but I don't think it's very good because in general in the g3 King's Indian White needs some "peace and quiet" before engaging with Black in the centre. With a pawn already on e4 and the pawn on d4 already exchanged 7... ♘c6 will force White to waste some tempi solving immediate tactical problems. 7... c6⁉ Somewhat more precise than 7...Re8. Black's aim is to play ...d5 with the rook still on f8. 7... ♖e8 is the most natural move in the position, but allows a little trick. 8. ♗d3! White switches back to this line, and Black now has a somewhat unpleasant choice. ( 8. f3 c6 9. c5 b6 would just transpose.) a) 8... d6 9. f3N After this move Black doesn't succeed in getting the kind of positions he generally tries to get in this line. ( 9. ♕d2⁉ ♘g410. ♗g5 ) 9... ♘c6 ( 9... c6 10. OO d5 11. cxd5 ♘xd5 12.♘xd5 cxd5 13. ♖c1 ♘c6 14. ♘xc6 bxc6 15. ♕d2 dxe4 16.♗xe4 ♕xd2 17. ♗xd2 ♗a6= ) 10. ♕d2 b) 8... ♘c6 9. O-O b1) 9... d6 10. ♕d2! ( 10. h3 ♘g4 11. hxg4 ♗xd4 ; 10. f3♘g4 ; 10. ♘xc6 bxc6 ) 10... ♘g4 11. ♗g5 and Black doesn't get the concessions he's looking for. b2) ♘g4 10. ♕xg4 ♘xd4 11. ♕d1 d6 12. ♕d2 ♘e6 13.♖ad1 a5 14. ♗c2 b6 15. b3 ♗b7+/= Cu.Hansen-Nielsen, Middelfart 2003 c) 8... c6 This position has gained some "notoriety" in recent years. White has additional options compared to the normal state of affairs.9. O-O d5 10. cxd5 cxd5 11. ♗b5 This is the move I'm talking about. The fact that the rook is on e8 and the pawn has managed to stay on f2 makes White's position much more stable. Black also needs to solve immediate tactical problems. 11... ♖f8 Pretty much forced.12. exd5 ♘xd5 13. ♘xd5 ♕xd5

c1) 14. a4 a6 15. ♗e2 ♘d7 16. ♖c1 ♘f6 17. ♗c4 ♕d8 18.♘f3 ♕xd1 19. ♖fxd1 ♗d7 20. a5 ♖fc8 21. ♘e5 ♗e6= c2) 14. ♕a4⁉ was played by Evgeny Tomashevsky against Emil Sutovsky, and after 14... ♘d7 he replied with the spectacular 15.♘c6! This position is very interesting, very unclear and a lot of fun to look at. 15... ♕e6 16. ♖fe1 ( 16. ♖ad1⁉ is probably an improvement.) 16... a6 ( 16... bxc6 17. ♗xc6 ♖b8 18. ♗xa7♕f5 19. ♗xb8 ♘xb8 20. ♗e4 ♕g5 21. ♖ab1 ♘d7 22. b4♗d4 23. ♖bd1 ♕f6 24. ♖e2 ♘b6 25. ♕b3 ♗g4 26. ♗f3♗e6 27. ♕c2 ♖c8 Tomashevsky-Sutovsky, Ohrid 2009 ) 17.♖ac1 ( 17. ♘d4 ♕d6= ) 17... bxc6 ( 17... axb5 18. ♕xa8bxc6 19. ♖xc6 ♕e4 ) 18. ♗xc6 ♖b8 19. ♗a7 ♕f5=/+ c3) 14. ♖c1! I believe this is even stronger that 14.Qa4, creating the very unpleasant threat of Bc4. 14... ♕xa2 Black at least tries to suffer for some material, as otherwise his position is just very unattractive. 15. ♗c4 ♕xb2 16. ♕d3 The queen is in trouble and Rb1 is now a big threat. 16... ♕b4 ( 16... ♘a6 17. ♖b1 ♘c518. ♖xb2 ♘xd3 19. ♗xd3 ♗xd4 20. ♗xd4 ♖d8 21. ♖b4a5 22. ♖a4 ♗d7 23. ♖a3± ) 17. ♗xf7+ A huge blow. 17...♔xf7 ( 17... ♖xf7 18. ♖xc8+ is quite clearly unattractive. 18...♖f8 19. ♖b1+− ) 18. ♖c4 It turns out that Black has no particularly good squares for the queen. 18... ♕b2 The best move in the position. ( 18... ♕e7 19. ♕b3!+− Black is already lost because his king can't get back into its corner in time.; 18... ♕b619. ♘f3± The queen will once again land on d5 with check, which will be lethal.; 18... ♕d6 19. ♕b3+− ) 19. ♕e4 Black barely manages to survive by starting to give up his extra material: 19...♔g8 20. ♖xc8 ♘c6 The most likely scenario is that Black will lose the entire queenside and will have to defend a somewhat unpleasant 3 vs 2 pawn endgame on the kingside. 21. ♖xa8♖xa8 22. ♘xc6 bxc6 23. ♕xc6 ♕b8+/= 8. e5 No real problem for Black. 8. ♗d3 d5= is also an attractive version of something we've seen before. 8. ♘b3⁈ ♖e8=/+ 8. c5 This is perhaps the only option that has separate value here, but Black can play 8... ♖e8 forcing White to play 9. f3 and now Black has a very typical breakthrough on the queenside: 9... b6=/+ 10. ♘b3 bxc5Black's easiest option. 11. ♗xc5 d5 Commencing an attack on the e4pawn. This is very comfortable for Black. 12. ♗e2 ♘bd7 13. ♗d4dxe4 14. fxe4 ♘xe4 15. ♘xe4 ♖xe4 16. ♗xg7 ♗a6 17. ♘c1♔xg7 Krasenk ow-Sutovsky, Reykjavik 2003 8... ♘e8 The usual response to e5. 9. ♘f3 d6 I think Black is perfectly fine here. 10. ♗e2 ♕a5 11. exd6 ♘xd6 12. O-O 12. ♕xd6 ♗xc3+ is also fine for Black. 12... ♘f5 This is a very attractive position for Black.

13. ♗d2 ♕c7 14. ♕c1 ♘a6 15. ♗f4 ♕b6=

3.Nf3 Bg7 4.d4 exc4 5.Nxd4 Nf6 6.Nc3 0-0 7.Be2 1. c4 g6 2. e4 e5 3. ♘f3 ♗g7 4. d4 exd4 5. ♘xd4 ♘f6⁉ 6.♘c3 O-O 7. ♗e2 The main option in this position. 7... ♖e8 8. f3 c6 White has a wide choice of moves, the most important of which are 9.Nb3 and 9.Nc2. 9. ♘c2 9. O-O d5= is just very comfortable for Black. 9. c5 is once again met by the strong 9... b6 and after, for instance,10. ♘b3 bxc5 11. ♘xc5 d6 12. ♘d3 d5=/+ Black may already be better and in any case I quite like this position. 9. ♘b3 a) 9... d5⁉ is possible immediately: 10. cxd5 cxd5 a1) 11. ♗g5 h6 12. ♗xf6 ♕xf6 13. ♘xd5 ( 13. ♕xd5 ♘c614. OO ♗e6 15. ♕c5 ♘d4 16. ♘xd4 ♕xd4+ 17. ♕xd4♗xd4+ 18. ♔h1 ♖ac8 19. ♗b5 ♖ed8 20. ♘d5 ♗xd5 21.exd5 ♖xd5 Lalith-Konguvel, Chennai 45th IND-ch 2008 ) 13...♕xb2 14. OO ♘c6 15. ♗b5 ♗e6 16. ♖f2 ♕e5 17. ♖c1♗xd5 18. ♕xd5 Lalith-Kasparov, Balaguer op 2008 a2) 11. exd5 11... ♘a6 12. ♗g5 ♘b4 13. ♕d2 ♕b6 14.♘a4 ♕d6 15. OO ♕xd5 16. ♖fd1 b6 Nyback-Navara, Bled 2002 b) a5 I think it's even stronger to start with this move. b1) 10. c5 a4 11. ♘d2 b5! ( 11... ♘h5 12. g3 b5 ) 12. cxb6d5 , and Black has a very dangerous initiative: 13. ♘xa4 ♘h5 (13... ♘bd7⁉ ) 14. g3 ♘f4!↑ b2) 10. a3 d5! b3) 10. O-O a4 11. ♘c5 a3=/+ is a whole other raft of problems for White, so not very attractive. b4) 10. a4 Forced. 10... d5 Black is very comfortable if White accepts the pawn sacrifice. b41) 11. exd5 cxd5 12. O-O White at least tries to limit the damage. ( 12. cxd5 ♕b6 Black's usual response, and with the pawns on a5 and a4 White has more problems than usual because in many lines the knight on b3 will be hanging - Black clearly has excellent compensation for the sacrificed pawn. ) 12...♗e6 13. c5 ♘c6=/+ This is a typical structure that arises in many lines of the Classical King's Indian, but in this particular case it's clear that Black is miles ahead of the usual situation where he has to fight for the d4-square. White will have a lot of trouble controlling that square here and Black may already be better.

b42) 11. cxd5 11... cxd5 12. ♗g5 ( 12. ♘xd5 ♘xd5 13.♕xd5 ♕xd5 14. exd5 ♘a6 Even this endgame is good for Black because the pawn on d5 won't survive very long. ) 12...h6 13. ♗xf6 ♕xf6 14. ♘xd5 ♕xb2 15. OO ♘c6= Once again fine for Black. 9. ♗g5 has some separate value. a) 9... h6 is less convincing: 10. ♗h4 d5 11. cxd5 ( 11. exd5cxd5 12. O-O ♘c6= Petursson-Sax, Buenos Aires Olympiad 1978 ) a1) 11... cxd5 12. ♘xd5!± ( 12. exd5 g5 13. ♗f2 ♘xd5 14.♘db5 ♗e6 15. OO ♘c6 David-Adorjan, Zalaegerszeg 1991 ) a2) 11... g5 12. ♗g3 cxd5 13. ♘db5 ♘a6 a21) 14. exd5 ♘h5 15. ♗f2 ♘f4 16. OO ♗f5 17. ♗c4 (17. ♕d2 ♖c8 18. ♖fe1 ♘b4∞ M.Gurevich-Jansa, Forchheim 2000 ) 17... ♖c8 18. ♘d4 ♗d7 19. ♘c6+/= Nicholson-McNab, corr 200 a22) 14. ♘d6!± b) 9... ♕a5⁈ 10. ♕d2 ( 10. ♗e3 d5 11. ♘b3 ♕d8 12. cxd5♘xd5 Popov-Khalifman, St Petersburg 1998 ) 10... d5 11. ♘b3♕b6 12. ♗e3 ♕d8 13. cxd5 cxd5 14. exd5± Biriukov-Stepanov Tula 2010 c) 9... ♕b6 The most natural reply, hitting the knight on d4. 10.♘b3 d5 It's important to note that Black has to be prepared to play ...d5 in almost any position here, regardless of how many pieces seemingly control the square. Black should never be worried about playing positions where he's just a pawn down for seemingly nothing concrete yet. For instance, in this position after 11. cxd5 cxd5 c1) 12. exd5 Alexander Khalifman played this against Codenotti at the European Individual Championship in 2011. 12... ♘a6 Once again a very typical move - Black aims to play ...Nb4 or perhaps in some positions ...Nc7, playing against the d5-pawn. The queen on b6 also ensures that White is unable to castle comfortably. 13.♗h4 Aiming to play Bf2 and at least castle. 13... ♘g4⁉ Black could even try this if he's feeling a bit more adventurous. ( 13...♕b4 led to a draw: 14. ♗f2 ♘c7 15. O-O Black will simply win back the pawn by taking with a knight on d5. 1/2 Khalifman-Codenotti, Aix-les-Bains 2011 ) 14. fxg4 White is forced to accept the sacrifice, and after 14... ♗xg4 15. ♔f1 ♖xe2 16. ♘xe2♖e8 Black gets very rich play for the sacrificed material. 17. ♗f2♕b5 18. ♘d4 ♕xb2=/∞ c2) 12. ♗xf6 is simply met by Black taking with the queen. 12...♕xf6 13. ♘xd5 ( 13. ♕xd5 ♘c6 Black has fantastic compensation due to his control of the dark squares. ) 13...♕xb2= Black is also doing very well here. 9... d5 Black's response is the same. 10. cxd5 cxd5 11. ♘xd5 The main move. 11. O-O dxe4 12. ♕xd8 ♖xd8 13. fxe4 ♖e8 14. ♘b5 ♘a6=/+Greenfeld-Leko, Budapest 1996

11. ♗g5 h6⁉ It won't surprise you to hear that this is once again Black's reaction. 12. ♗xf6 ♕xf6 13. ♕xd5 ♗e6 We're following a game played by Gross and Stohl in the very remote past - all the way back in 1991. 14. ♕b5 ♘c6 Black has fantastic play along the dark squares. (14... ♘d7 15. O-O a6 16. ♕d3 ♘e5=/∞ ) 15. ♖d1⁉ is better for White. ( 15. O-O ♘d4!=/+ Here Black was perhaps already better in Gross-Stohl, Rimavska Sobota 1991 ) 15... a6 ( 15... ♕e7 16. OO♕c7 17. f4 a6 18. ♕c5+/= ) 16. ♕b6 ♖ed8 Black gets a lot of play on the dark squares and can never be worse in a position like this. 17.♔f2 ( 17. OO ♘d4 18. ♘xd4 ♕xd4+∓ ) 17... ♘d4 18. ♘e3♕e7=/∞ 11. exd5 ♕b6 White is very uncomfortable with his king in the centre. If12. ♕d4 then the simplest way to proceed is 12... ♘bd7 , and after 13.♕xb6 ♘xb6 the pawn on d5 is about to be lost and Black is slightly better. 14. ♘e3 ♖d8= 11... ♘xd5 12. ♕xd5 ♕c7⁉ An even more ambitious try. 12... ♕xd5⁉ 13. exd5 ♘a6 is also possible, as I think Black gets enough counterplay in the endgame. 14. ♔f2 ♗f5 15. ♘e3 ♗d4 16. ♖d1♗b6 17. ♗b5 ♘b4! 18. ♗xe8 ♖xe8 19. ♔g3 ♗xe3 20. ♗xe3♖xe3 21. ♖ac1∞ 13. ♕c4 ♘c6 14. O-O ♗e6 15. ♕a4 a6→ Black gets fantastic compensation for the pawn because his bishops are hitting the white queenside hard and it's very tough for White to somehow finish his development. This is a very nice position for Black to aim for.

3.d4 Nf6 4.dxe5 1. c4 g6 2. e4 e5 3. d4 Certainly White's best move in this position. Black has to be very precise because White is threatening complete domination in the centre. A lot of finesse is required from Black so as not to let that happen. 3... ♘f6 4. dxe5⁉ 4. ♘c3 exd4 5. ♕xd4 The inclusion of ...Nf6 and Nc3 means that Black can now drive the white queen away from the centre with 5... ♘c6 , and after 6. ♕d1 ♗g7 , Black gets a very comfortable version of the King's Indian. ( 6... ♗b4⁉ ) 7. ♗d3 OO 8. ♘f3 d6 9. OO ♘e5 10. ♗e2♘xf3+ 11. ♗xf3 ♗e6 12. ♗e2 ♘d7 13. ♗e3 Malakhov-Salgado Lopez, Plovdiv 13th EU-ch 2012 4... ♘xe4 There are a lot of very sharp positions here but in general I think Black is doing very well. 5. ♕d5 5. ♗d3 gives Black a choice. a) 5... ♗b4+ 6. ♔f1 ( 6. ♘d2 d5 7. exd6 ♘xd2 8. ♗xd2♕xd6 ) 6... d5 ( 6... ♘c5 7. ♘f3+/= ) 7. exd6 ♘f6 A very strong move. ( 7... ♘xd6? 8. a3 ♗c5 9. ♕e1+ ♕e7 10. ♕c3 ) 8. dxc7♕xc7 Black has at least enough compensation for the pawn due to the somewhat suspect position of the king on f1 and the open centre. 9.♗h6 ♗f8 10. ♕d2 ♗e6 11. ♗xf8 ♔xf8 12. ♕c3 ♔g7 13.♘f3 ♖d8 14. ♘bd2 ♘c6=/∞ b) 5... ♘c5⁉ 6. ♘f3 d6=

5. ♘f3 ♗c5 6. ♕d5 f5 is very strong for Black, as 7. exf6 ♗xf2+ 8.♔e2 ♘xf6 9. ♕e5+ ♔f7=/+ is just bad for White. The bishop on f2 is protected by the fork on g4. 5... f5 5... ♗b4+ is also possible: 6. ♘d2 f5 a) 7. exf6 ♘xf6 8. ♕d3 OO 9. ♗e2 ♖e8 10. a3 ♗f8 11.♘gf3 d6 ( 11... d5 12. cxd5 ♗f5 13. ♕c4 Polugaevsky-Adorjan, Sochi 1976 13... ♕xd5 ) b) 7. ♘f3 c6 8. ♕d3 ♕b6 c) 7. a3 7... c6 ( 7... ♘a6 8. axb4 ♘xb4 9. ♕a5 ♘c2+ 10.♔d1 b6 11. ♕a4 ♘xa1 12. ♘xe4 fxe4 13. ♕xa1 OO 14.♗e3 ) 8. ♕d3 ♕a5 9. ♘f3 ♗c5 10. ♘d4 d6 11. f3 ♗xd4 12.♕xd4 dxe5 13. ♕e3 ♘f6 14. b4 ♕c7 15. ♗b2 O-O 16.♗e2=/∞ 6. ♘c3 6. exf6 ♘xf6 is very comfortable for Black due to his lead in development. 7. ♕d1 ♗b4+ 8. ♘c3 ♕e7+ 9. ♗e3 ♗c5 10. ♕d2d6 11. ♘f3 OO 12. ♗e2 ♗xe3 13. fxe3 ♘bd7 14. O-O ♘c5 6... c6 I like this adventurous approach here. 6... ♗b4⁉ is also ok for Black. 7. ♘e2 c6 8. ♕d3 d6⇆

7. ♕d3 ♘a6 Offering a pawn sacrifice. 8. ♘f3 The safer approach for White.

8. ♘xe4 Accepting the sacrifice gives Black a lot of play: 8... fxe4 9.♕xe4 ♗b4+ 10. ♗d2 ( 10. ♔d1 ♘c5 11. ♕e2 d5 12. exd6+♗e6 is very strong for Black because for the price of two pawns he manages to completely open up the centre. The white king should feel very draughty. ) 10... ♗xd2+ 11. ♔xd2 O-O 12. ♘f3 d6 Black again gets a lot of play against the white king and is doing excellently. 13.♕d4 ♖f4 14. ♕xd6 ♕b6 15. ♖d1 ♗f5 16. ♔e1 ♘c5∓

8... ♘ac5 9. ♕c2 ♘xc3 10. bxc3

10. ♕xc3 is very inadvisable because of 10... ♘e4 with ...Qa5+ to follow. 10... d6 11. ♗g5 ♗e7 12. ♗h6 ♗e6 Here it's just important to remember that Black can also castle queenside, so he doesn't have to worry about the bishop on h6. 13. ♖d1 ♕c7 Black is ready to castle queenside and if anyone's better it's Black.

3.d4 Nf6 4.Nf3 1. c4 g6 2. e4 e5 3. d4 ♘f6 4. ♘f3 I think White needs to respond like this. 4... exd4

4... ♗b4+ is a move I once played against Vassily Ivanchuk, but my opponent soon got an almost winning position because after 5. ♘c3 ( 5.♗d2 may actually be the most unpleasant reply from the black player's perspective as he'll be forced to play some kind of passive position with no winning chances. That would be very much against the spirit of everything I'm trying to show you. ) 5... exd4 6. a3 I replied with the very timid 6... ♗e7 ( 6... ♗xc3+ 7. bxc3 O-O may be better for White, but the game will become very, very sharp. ) 5. e5 Black can get a boring and slightly worse position by playing 5... ♘e4 Play becomes very forced for a while. 5... ♗b4+ 6. ♗d2 ♗xd2+ 7. ♕xd2 a) 7... ♘e4 8. ♕xd4 ♘g5 9. ♘xg5 ♕xg5 10. f4 ♘c6 11. ♕d2( 11. ♕f2 ♕d8 12. ♘c3 OO 13. OO-O d6 14. c5± ) 11... ♕d812. ♘c3± Miles A-Adorjan A/Hastings 1976 b) ♕e7 8. ♗e2 ♘h5 ( 8... ♘g4 9. ♕xd4 h5 10. ♘c3 ♘c6 11.♕d2 ♘gxe5 12. ♘xe5 ♘xe5 13. ♘d5+− ) 9. ♕xd4 ♘c6 10.♕e3 d6 11. exd6 ( 11. ♘c3 dxe5 12. ♘d5 ♘b4 13. ♘xe7Stohl-Navara, CZE-tch 2002 ) 11... cxd6 12. ♘c3+/= 6. ♕xd4 ♗b4+ 7. ♘bd2 7. ♘c3 isn't dangerous at all: 7... ♘xc3 8. bxc3 ♘c6 9. ♕e3 ♗e7The one thing you need to remember here is that castling too early will run into the very dangerous h2-h4h5. 10. ♗e2 ( 10. h4 d6 11. h5♘xe5 12. ♘xe5 dxe5 13. hxg6 fxg6 14. ♕xe5 O-O ) 10... d6 Black should be doing quite comfortably because the moment the e5-pawn is exchanged he'll have a lot more freedom to choose a plan. For instance: (10... O-O⁈ 11. h4 ) 11. exd6 ( 11. O-O O-O ) 11... ♕xd6 is obviously already very nice for Black. 12. ♘d4 ♘xd4 13. cxd4 O-O= 7... ♘g5 7... ♘xd2 8. ♗xd2 ♘c6 9. ♕f4+/= is playable and perhaps not so bad for Black, but not much fun. 9... ♗xd2+ 10. ♕xd2 ♕e7 11. O-O-Ob6 12. ♗d3 ♗b7 13. ♗e4 O-O-O 14. ♖he1 Suba-Kamsky, Brasov ROU-tch 2011 8. ♗e2 The main move. 8. ♘xg5 is possible. 8... ♕xg5 9. f4 ( 9. a3 ♘c6 10. ♕e4 ♗xd2+11. ♗xd2 ♕xe5 12. ♕xe5+ ♘xe5 13. ♗c3 OO 14. O-O-O d6 )9... ♘c6 10. ♕f2 ♕e7 The position becomes very sharp and double-edged, but generally speaking that shouldn't worry the black player at all.11. a3 ♗c5 12. ♕g3 a) 12... ♘d4 13. ♘e4! ( 13. ♕c3 d6 ) 13... ♘c2+ 14. ♔d1♘xa1 15. b4=/∞ b) d6 13. ♘e4 O-O 14. ♘xc5 dxc5 15. ♗d3 ♘d4 16. O-O ♗f5Black has good play. 8. a3 was played by Hikaru Nakamura against me at the Thessaloniki Grand Prix, but it's not a particularly good move because after 8...♘xf3+ 9. gxf3 ♘c6 10. ♕e3 it's very strong to return all the way with the bishop: 10... ♗f8! Much better than 10...Be7. 11. b4⁈ ( 11. ♘e4♗g7 12. ♘f6+ ♔f8 13. f4 ♗xf6 14. exf6 ♕xf6 15. ♗d2 ♕e6 16.♗c3=/∞ ) 11... ♗g7 12. ♗b2 O-O 13. f4 d6 14. O-O-O dxe5 15.fxe5 ♕h4=/+ I was already better here in the game. White could have played a bit better in the previous moves and hope for equality, but no more. Nakamura-Svidler, Thessaloniki GP 2013

8... ♘c6 9. ♕e3 ♘xf3+ 10. ♗xf3 O-O 11. O-O ♖e8 This position is currently considered very safe for Black. 12. ♗d5! Volokitin pointed out this move in his own annotations for Chess Informant to a recent game he played against Vitiugov in the Bundesliga. This is a very critical attempt you need to be aware of because otherwise you might end up in a lot of trouble. I spent some time trying to refine Volokitin's analysis of this position because it's not quite clear how to equalise with Black, but I think I finally worked out a very precise move order which ends in an endgame which although slightly better for White should be quite holdable. 12. ♗xc6 was the move Vitiugov played against Volokitin, and he even went on to lose with White. 12... dxc6 If you could teleport the knight from d2 to f6 White would be winning but the fact that the pawn on e5 is constantly hanging makes that impossible. The game continued: 13. ♘f3( 13. h3 might be played to prevent the natural ...Bg4, but Black instead transfers his bishop to g7. 13... ♗f8 14. ♘f3 ♕e7 15. ♖e1 ♗g7Black is fine because the pawn on e5 is a constant source of worry as, potentially, is the pawn on c4 in any kind of endgame. 16. ♕c3 h6 17.♗f4 g5 18. ♗h2 ♗f5 19. ♘d2 ♖ad8⇆ 20. ♘e4? ♗xe4 21.♖xe4 f5! ) 13... ♗g4 a) 14. ♕b3 a5 15. ♗g5 ♕c8 16. ♗f6 White has seemingly achieved something, but after 16... ♗e7 it turns out that Black is very comfortable attacking the pawn either on e5 or f6 if Black allows the trade. 17. ♗xe7 ( 17. ♖fe1 ♗xf6 18. exf6 ♕f5 ) 17...♖xe7 18. ♖fe1 a4 19. ♕c3 ♗xf3 0. ♕xf3 ♕e6=/+ Black went on to win in Vitiugov-Volokitin, BL 2012 b) 14. ♘g5 ♗f5 15. g4? ♗xg4 16. c5 h6! 17. ♘e4 ♖xe5 18.♕f4 ♖xe4 19. ♕xe4 ♗h3! 20. ♕xb4 ♕d5 21. f3 ♗xf1 22.♔xf1 ♕xf3+ 23. ♔g1 ♖e8−+ c) 14. ♗d2 An attempt to improve on the game. 14... a5 Forcing White's next move. 15. a3 ♗f8 Black is very comfortable and the game could continue: 16. ♗c3 ♕e7 17. h3 ( 17. ♘g5 ♗h6 18.h4 also isn't particularly threatening for Black. 18... ♗e6 19. ♕f4 a4⇆ ) 17... ♗xf3 18. ♕xf3 ♕e6 19. ♕e4 a4= In potential endgames Black might even be better. 12... d6 12... ♘xe5 13. ♘e4 c6 is Black's most obvious reaction, but just loses to 14. ♕f4 , so you can't play this way. 13. ♘e4 ♖xe5 Once again Black has to be very careful. 13... dxe5 14. ♘g5 ♖f8 15. ♘xh7+− just wins on the spot, because15... ♔xh7 runs into 16. ♕h6+ ♔g8 17. ♕xg6+ 14. ♕g3 Opening up the bishop on c1. 14... ♘d4 Absolutely the only move, as otherwise Black is in a lot of trouble. There's a huge threat of ...Ne2+, so White plays 15. ♕d3 , winning a tempo. 15... c6 I think Black has to play this move. 15... ♘e6⁈

a) 16. f4⁉ ♖h5 ( 16... ♖xe4 17. ♕xe4 c6 18. ♗xe6 ♗xe6 19.♗e3 ♗f5 20. ♕d4+/= ) 17. a3 ( 17. ♗d2 ♘c5 ; 17. ♘g3 ♗c5+18. ♗e3 ♖h4∞ ) 17... ♗a5 ( 17... ♘c5 18. ♕e2 ♗a5 19. ♘g5 ♘e6 20. ♗e3→ ) 18. ♘g3 ♗b6+ 19. ♗e3→ b) 16. a3 16... ♗c5 17. ♗d2 It seems as though Black isn't doing that badly, but the more I looked at this position the less I understood how to equalise. ( 17. b4 ♗d4 18. ♖a2 c5 19. ♗e3 ♕e7= ) 17...♗b6 18. ♖ad1 ( 18. ♗c3 c6! ) 18... ♕h4 ( 18... c6 19. ♗xe6♖xe6 20. ♕f3↑ ; 18... ♘d4 19. ♗c3 c5 20. ♖de1 ♗e6 21. f4♖xd5 22. cxd5 ♗xd5 23. ♗xd4 cxd4 24. f5± ; 18... ♗d7 19.♗xb7 ♖b8 20. ♗d5 ♕e7 21. ♗c3 ♘f4 22. ♕f3 ♖f5 23. g4!♘xd5 24. gxf5 ♗xf5 25. ♖fe1 ♘xc3 26. ♘xc3 ♕h4 27. ♘d5♔g7 28. ♕c3+ ♔h6 29. ♘xb6 axb6 30. f3+/= ) 19. ♗c3 ( 19.g3 ♕e7 20. ♖fe1 c6 21. ♗xe6 ♗xe6 22. ♗b4 ♗f5 23. ♕xd6♕xd6 24. ♘xd6 ♖xe1+ 25. ♖xe1 ♗d4 26. ♘xf5 gxf5 27. b3a5= ) 19... ♘f4 ( 19... ♖f5 20. g3 ♕e7 21. b4 ) 20. g3 ♘xd321. gxh4 ♗f5 22. ♘g3+/= 16. ♗xf7+ 16. ♕xd4 cxd5 17. cxd5 is fine for Black after the very important 17...♕e7= 16... ♔xf7 17. ♕xd4 The bishop is completely misplaced on b4 in this position, so you have to be very careful. 17... ♗f5 17... ♕e7 18. ♘g3 ♔g8 19. ♗f4 ♖e6 20. ♖ad1+/= 18. ♘g3 ♕f6 18... ♗c5 19. ♕c3 ♕h4 20. ♗e3 ♖ae8 ( 20... ♗xe3 21. fxe3♔g8 22. ♘xf5 gxf5 23. ♖f3 ♖ae8 24. ♖af1+/= ) 21. ♗xc5 dxc522. ♖ad1 ♕f4 23. b3+/= 19. ♗e3 ♔g8 The most precise move. Black still wants to play ...Bc5 and take on e3, but he's "unpinning himself in advance", preparing for the potential opening of the f-file after ...Bxf3 fxe3. 19... ♖e7 20. ♕d1 ♔g8 is also possible and not that bad for Black. (20... ♖ae8 21. ♘xf5 ♕xf5 22. a3 ♗c5 23. b4 ) 21. a3 ♗c5 22.♕d2 ♗xe3 23. fxe3 ♕e5 24. ♘xf5 gxf5 25. ♖f3 ♖f8 26. ♖af1♖f6 Black should hold, although White will press for a while without risk. 20. ♗f4 ♖e6 21. ♕xf6 ♖xf6 22. a3 ♗c5 23. b4 ♗d4 24. ♖ad1This forces Black to play 24... c5 Something of a concession, but on the other hand a lot of pieces have already come off and I think the endgame that arises after 25. ♘xf5 gxf5 should be quite holdable for Black. For instance, 26. ♖fe1 ♔f7 27. ♖d3 The threat of Rh3 is somewhat unpleasant but can be met by 27... h5 28. ♖h3 ♖h8 White retains some pressure but with accurate defence I think Black should hold this endgame reasonably comfortably. This concludes our overview of 1.c4 g6. Once again I want to point out that there's no real necessity for Black to go into any of this, as 1.c4 e5 is a perfectly natural way to play chess, as is 1.c4 c5. If you have one or perhaps both of those moves in your repertoire already there's absolutely no reason for you to suddenly stop making them, but I think I've provided you with an additional option which might be a good surprise weapon. It's somewhat challenging for the white player because it leads to positions which aren't as wellknown and may come as a surprise to some of your opponents.

1.Nf3 Nf6 2.c4 c5 3.d4 cxd4 4.Nxd4 g6 5.g3 5.Nc3 5.Bg2

1.Nf3 Nf6 2.c4 c5 3.d4 cxd4 4.Nxd4 g6 5.g3 1. ♘f3 Perhaps the most popular Anti-Grünfeld system in existence, and something you definitely need to have some kind of repertoire against. As a Grünfeld player you'll be seeing a lot of it, and in fact I've faced it so much over the course of my career that I even started playing it with White, with reasonable success. A few disclaimers: I think from an objective viewpoint perhaps the best move in this position is 1...c5, but not all of you out there are Sicilian players and recommending you learn a whole new opening against 1.e4 is a bit tricky. So keep in mind that 1...c5 is a very decent option which might leave you a wider choice later on, but if you're not a Sicilian player it's not for you. I also quite like 1...g6, but that's because I'm a Pirc player. If you don't play the Pirc or the Modern Defence then 1.Nf3 g6 runs into 2.e4 and then you're in a bit of trouble. Once again you're stuck playing an opening you don't know - and perhaps an even trickier opening than the Sicilian - so that's also a bit of a problem. 1... ♘f6 I'll stick with this move, which is the one I normally make in this position. 2. c4 c5 People who play 1.Nf3 won't be completely unprepared for this move, but it's a slightly different approach which might interest some of you. 2... g6 This is the main move I've played here and the move normally expected from Grünfeld players. 3. ♘c3 d5 First and foremost, it's much easier for White to prepare because the onus is on him to choose a setup. He can play 4.Qa4+ straightaway, he can play 4.cxd5 Nxd5 5.Qa4+ and there are other things you have to consider. Generally speaking, whilst I don't have anything against this move order in principle I thought I'd show you something else. 3. d4 cxd4 4. ♘xd4 Another disclaimer. There's very little doubt in my mind that from an objective, "scientific" point of view the best way for Black to proceed here is 4...Nc6 5.Nc3 e6. That's a "proper" English Opening where I think Black is currently doing quite well. There's no reason for you not to play that if you already play the English. I, on the other hand, have never played that position with either colour and have almost no idea what's going on. I thought I'd therefore stick to things I have some clue about, or some oddball things which I very much like finding as a black player - small, unpopular lines where you can surprise your opponent. 4... g6⁉ This is just such a move, as long as you're not against having some adventure in your life... 5. g3 For a second I was very happy when preparing this series because I thought White almost has to play this move. 5... ♗g7 And this is familiar ground for me. 5... d5 is tempting, but in fact after 6. ♗g2 a) 6... e5 7. ♘f3 ♗b4+ ( 7... d4 8. O-O ♘c6 9. e3 , and Black has landed in a poor version of the Reti. ) 8. ♘bd2 e4 9. ♘d4O-O 10. ♘c2 ♗e7 11. OO ♗e6 12. b3 ♘c6 13. ♗b2 dxc4 14.bxc4 ♕b6 15. ♖b1 ♕a6 16. ♗xf6 ♗xf6 17. ♘xe4± Giri-Bologan, Biel 45th GM 2012 b) dxc4⁉ 7. ♕a4+ I couldn't find any direct equality for Black, so I'm not recommending this approach to you. 7... ♘bd7 8. O-O ♗g7 9.♖d1 O-O 10. ♘c3⁉→ With serious pressure. ( 10. ♕xc4 ♘b6 11.♕c2 ♘fd5 12. ♕b3 e5 13. ♘b5 ♗e6 14. e4 ♘c7 15. ♖xd8♖axd8 16. ♕c2 ♘xb5 17. ♗g5 ♘d4 18. ♕d1 f6 19. ♗d2♖d7 20. ♘c3 ♘c4 21. ♗c1 ♖c8 Romanov-Nepomniachtchi, Legnica 2013 )

6. ♗g2 O-O 7. ♘c3 Necessary, if White wants to avoid an immediate draw. This is what you should be preparing for in this position if you're playing Black. There's been one major trendsetter in top openings for the past 15 years or so – 7. O-O d5= is a well-known brick wall. 8. cxd5 Pretty much forced. 8...♘xd5 9. ♘b5 ( 9. ♘c3 ♘xc3 10. bxc3 transposes to positions I discussed elsewhere in this series. This is also completely harmless for Black. ) 9... ♗e6 10. ♘d4 ♗c8!= Black is just fine. 11. ♘c3 ♘xc312. bxc3 ♕c7 7... ♕c7 8. b3 d5 This isn't at all new, and I played this position myself against Alexander Grischuk. 9. ♘db5 Kramnik's move. 9. O-O dxc4 10. ♘cb5 ♕d8 11. bxc4 a6 12. ♘c3 ♘g4 13. e3♘e5! A much stronger move. ( 13... ♘c6 was what I expected, and had been well-known for many years. ) 14. ♕e2 ♘bc6 My main concern here was equalising, although eventually I even won the game. Svidler-Grischuk, Astana World Rapid Final 2012 9... ♕a5 10. ♗d2 dxc4 11. bxc4 ♕d8 All very natural. 11... a6 12. ♘d5 ♕d8 a) 13. ♘bc7 ♘xd5 14. ♘xa8 ( 14. ♘xd5 e6 15. ♘c3 ♗d7 )14... ♘c3 15. ♕b3 ♘e4 16. ♗xe4 ( 16. ♖d1 ♘xd2 17. ♕b4♗f5 18. ♖xd2 ♘d7 ) 16... ♗xa1 17. ♘b6 ( 17. ♗h6 ♕a5+ )17... ♘d7 18. ♘xc8 ♘c5 19. ♕b1 ♘xe4 20. ♕xe4 ♕xc8 21.O-O ♗f6 22. ♖b1+/= b) 13. ♘bc3 12. O-O a6 12... ♘c6 13. ♘d5! ♘xd5 14. cxd5 a6 15. ♘c3 ( 15. dxc6 axb516. cxb7 ♗xb7 17. ♗xb7 ♖a7= ) 15... ♘a5 16. ♖c1 ♘c4 17.♗f4 ♗d7 18. d6 ♘xd6 19. ♘d5→ 13. ♘a3 It seems as if Black is doing quite well because he has a more natural pawn structure and no immediate weaknesses. The knight on a3 also looks a bit suspect. On the other hand, Black has something of a weakness on b7 and the bishop on g2 is extremely strong. Black needs to be very precise, and I'm proposing not one but two potential improvements on Radjabov's play. 13... ♘c6⁉ A very natural move. 13... ♗f5 I think this appealed to Radjabov because it stops the rook appearing on b1 straightaway. It seems a natural enough move, but after14. ♘c2 ♘c6 15. ♘e3 the bishop comes under attack and Vladimir got some pull which he eventually converted into a whole point, although Teimour could have defended better. 15... ♕d7 16. ♘xf5 ♕xf5 17.♖b1+/= Kramnik-Radjabov, London Candidates 2013 13... ♘bd7⁉ is also playable, trying to put the knight on c5 rather than c6 and play ...Be6, with quite natural development. Play could continue:14. ♖c1 ( 14. ♘c2⁈ ♘b6! 15. ♘e3 ♘g4 16. ♖c1 ♘xe3 17. ♗xe3♘xc4 18. ♗c5 ♕xd1 19. ♖fxd1 ♘b2 20. ♖d2 ♗xc3 21. ♖xc3♘a4 22. ♖c4 ♘xc5 23. ♖xc5 b6 24. ♖c7 ♖b8 25. ♖xe7 ♗e6= ;14. ♖b1 ♘c5 15. ♗e3⁈ ♕a5 16. ♘d5 ♕xa3 17. ♘xe7+ ♔h8 18.♕d6 ♘fd7∓ ; 14. ♕e1 is quite a strange move favoured by the machine. 14... ♘c5 15. ♗e3 ♕a5 16. ♘c2 The point of the previous play - White is once again preparing Nd5, which has to be parried with16... ♖e8! I think Black is doing fine here as well. 17. ♘d5 ♕xe1 18.♘xf6+ exf6 19. ♖axe1 ♘e4 20. ♘b4 f5 21. ♘d5 ♗e6 22. ♘c7♗xc4=/∞ ) 14... ♘c5 15. ♗e3 ( 15. ♘d5 ♗f5⇆ ) 15... ♕a5 16.♘c2 ♗e6 17. ♕e1 ♖fc8 Defending against the threat of Nd5. 18.♘d5 ♕xe1 19. ♘xf6+ ( 19. ♖fxe1 ♗xd5 20. cxd5 ♘ce4∞ ; 19.♘xe7+ is impossible because of 19... ♔f8 This is a well-known theme from the Dragon. ) 19... ♗xf6 20. ♖fxe1 a5⁉ Not forced, but I think this is the most precise way of playing, preparing for the potential Bxc5 and Bxb7, and in general pushing the pawn towards where it needs to be. I think Black is fine. 21. ♗f3 ♖c7 22. ♘a3 ♖d8 23. ♘b5 ♖cc8=

14. ♖b1 More or less the only try. 14. ♘c2⁈ is met by 14... ♗e6 15. ♘e3 ♘g4⇆ and Black is the first to start concrete play, aiming at White's weaknesses. If anything Black may already be slightly better. 14... ♕d6⁉ This is the move I eventually settled on. 14... ♕a5 15. ♘c2 ♖d8 14... ♗e6 15. ♖xb7 ♘a5 This was my first idea, trying to equalise on the spot, but it leads to quite a strange position. 16. ♖b1 ( 16. ♖b4♖c8= ) 16... ♖c8 17. c5! A very strong move, driving the rook towards the slightly unpleasant c5-square. 17... ♖xc5 ( 17... ♘d7 18. ♘d5± )18. ♘a4 ♖h5 It turns out you have to play this move. Although the machine insists there's nothing wrong with it and Black is still fine I don't really want to recommend this. ( 18... ♖c8 19. ♘b6 ♖c5 20. ♕e1♘c6 21. ♘a4 forces 21... ♖d5 22. ♗c3!+/= ) 19. ♕c2 ♕d6 ( 19...♗f5 20. e4 ♘xe4 21. ♗xa5 ♘xg3 22. ♗xd8 ♗xc2 23. hxg3♗xb1 24. ♗xe7 ♖e8 25. ♖xb1 ♖xe7 26. ♘c4+/= ) 20. ♗b4 ♕e5This is a very picturesque position which is a lot of fun to look at, but I wanted to find something more human. 21. ♖fe1 ( 21. ♘c5 ♖c8 22.♖fc1 ♗d5 ) 21... ♘g4 22. ♗c3 ♕d6 23. ♗xg7 ♔xg7 24. ♕c3+♘f6 25. ♖b6 ♖c8 26. ♕xc8 ♗xc8 27. ♖xd6 exd6 28. ♖d1♗d7 29. ♘b6 ♗e6= 15. ♘c2 There's not really any other move. 15... ♖d8 16. ♗e3 16. ♘d5 ♗e6 17. ♗f4 ♕c5 Although establishing a knight on d5 is one of White's most important ideas in this whole line, in this particular case after 18. ♘ce3 ♖d7 Black is well-placed to meet it. The queen on c5 is very important and can't really be attacked by a piece once the knight lands on e3. Black is ready to play ...Rad8 and then perhaps ...Nh5 in some positions, attacking the bishop on f4. I think this is fine for Black.19. ♕a4 ♘d4 20. ♖fe1 ♖ad8∞ 16... ♕xd1 16... ♘g4 17. ♕xd6 ♖xd6 18. ♘e4 ♘xe3 19. ♘xe3 ♖d8 20.♘c5 ♖a7 21. ♖fd1 ♘d4 22. ♗xb7 ♘xe2+ 23. ♔f1 ♘d4 24.♔g2 e6 25. ♗xc8 ♖xc8 26. ♘b7 ♖c7 27. ♘d6 ♖a8 28. ♖b6+/= 17. ♖fxd1 17. ♖bxd1 ♗f5 18. ♘d4 ♘xd4 19. ♗xd4 ♖ac8= 17... ♗f5 Once again, I analysed this position for some time. It is potentially slightly dangeorus for Black because the pawn on b7 is constantly hanging. 17... ♖xd1+ 18. ♖xd1 ♗e6 19. ♘d5+/= 18. ♖dc1 Black needs to be concrete, but he's fine after 18. ♖xd8+ ♘xd8 19. ♖c1 ♖c8 20. c5 ♔f8 21. ♘d4 ♗d7 22.♘d5 ♘xd5 23. ♗xd5 ♖c7= 18... ♘g4! 18... ♖ab8 19. ♘d5+/= 19. ♖xb7 ♘a5 19... ♗xc3 20. ♗xc6 ♖ac8 21. ♗f3 ♖xc4 22. ♗b6+/= 20. ♘d5 ♘xe3 21. ♘xe7+ ♔f8 22. ♘xe3 ♘xb7 23. ♘7xf5♖ab8! 24. ♘xg7

24. ♗xb7 ♗b2!∓

24. c5 gxf5 25. c6 ♖bc8! 26. c7 ♖d7 27. ♗xb7 ♖dxc7 28. ♖xc7♖xc7 29. ♗xa6 f4 30. gxf4 ♗h6= 24... ♘c5! 25. ♘gf5 gxf5 26. ♘xf5 ♖d2 27. ♗f3 a5! 27... ♖xa2 28. ♖d1 28. ♔g2 a4 , and with total control of the open files Black is at the very least not worse. 29. a3 ♖b3 30. ♖a1 ♖c3

5.Nc3 1. ♘f3 ♘f6 2. c4 c5 3. d4 cxd4 4. ♘xd4 g6⁉ 5. ♘c3! The second most obvious move for White in this position, and the most critical. 5... d5 My intended reply, but White has a very dangerous move: 6. ♗f4! When I stumbled upon this bit of theory I became very doubtful about the whole idea of recommending this to you, but eventually I thought, "people are still playing this with Black, so they must have something". I think this is an interesting sideline which can be played with sufficient preparation. 6. cxd5 is a pedestrian move, and after 6... ♘xd5 7. ♘db5 ♘xc3 8.♕xd8+ ♔xd8 9. ♘xc3 ♗g7 I believe Black is just fine. a) 10. ♗f4 ♗xc3+ White allowed this capture, spoiling the pawn structure on the queenside, in a couple of games where Vadim Zvjaginsev had Black. He won one game and was better in the other, so this isn't a threat to Black. 11. bxc3 f6 ( 11... ♘c6 12. OOO♔e8 13. e4 ♗e6 14. ♔b2 f6 15. ♗b5 ♖c8 16. ♗e3 ♔f7 17.♗a4 ♘a5 Bukavshin-Zvjaginsev, Moscow Aeroflot Rapid qual 2013 )12. OOO ♔e8 13. e4 ♗e6 14. ♔b2 ♔f7 15. ♗b5 ♖c8 16.♖d4 ♘c6 Lintchevski-Zvjaginsev, St Petersburg Chigorin Mem Rapid 2012 b) 10. ♗d2 10... ♗e6 Played by Nikita Vitiugov against Laurence Fressinet in the Alekhine Memorial. It's worth noting that Vitiugov's second for that tournament was Zvjaginsev, so I'm pretty sure he knew what he was doing. 11. g3 ♘c6 12. ♗g2 ♔e8 13. O-O-O⁉ Played against me by Le Quang Liem. I immediately started making mistakes, but the position is quite safe: ( 13. O-O f5 The king will be very safe on f7 and the rooks will get connected. After 14. ♘a4 ♔f7 15.♗xc6 bxc6 16. ♗c3 ♗xc3 17. ♘xc3 ♖ab8 Nikita equalised quite comfortably because he's just in time to start counterplay against the b2-pawn. The weakness on c6 will not be a huge problem. 18. ♘a4c5= Fressinet-Vitiugov,N Alekhine Mem 2013 ) b1) 13... ♖c8⁈ 14. ♔b1 f5? ( 14... f6 , with the idea of 15.♘d5 ♔f7 16. ♘f4 ♗f5+ 17. e4 ♗g4 18. f3 ♗d7 , would still hold. ) 15. ♘d5! ♗d7 16. h4 , and Black was really struggling in Le Quang Liem-Svidler, Tromso World Cup 2013 b2) 13... ♖d8! 14. ♔b1 f5= 6. ♗g5 ♗g7 7. ♗xf6 ♗xf6 8. cxd5 ( 8. ♘xd5 ♗g7 9. ♘b5♘a6=/∞ ) 8... ♕b6 a) 9. e3 ♕xb2 10. ♖c1 ( 10. ♕c1 ♕xc1+ 11. ♖xc1 O-O 12.♗e2 ♖d8 13. O-O ♘d7 ) 10... O-O∞

b) 9. ♘b3 This transposes to a 4.Bg5 Grünfeld, which won't be very attractive to 1.Nf3 players trying to avoid the main Grünfelds by not playing 1.d4. b1) 9... a5⁉ is possible, but White keeps an edge with accurate play: 10. e3 ( 10. ♘d2 ♕xb2 11. ♖b1 ♕xc3 12. ♖c1 ♕a313. ♖xc8+ ♔d7∞ ; 10. ♖b1 ♗f5 11. ♖c1 a4 12. ♘d2♕xb2 13. ♘de4 ♗xe4 14. ♘xe4 ♕b4+ 15. ♘d2 O-O Michel YunisShipov, Dos Hermanas 2004 ) b11) 10... OO 11. ♖b1 a4 ( 11... ♖d8 12. ♘d2 ) 12.♘d2 a3 13. bxa3 ♕a5 14. ♘ce4 ♕xd5∞ b12) 10... a4 b121) 11. ♘d4 ♕xb2 12. ♗b5+ ♗d7⁉ ( 12... ♘d7 13.♕c1 ♕xc1+ 14. ♖xc1 OO 15. ♘xa4 ♖a5 16. ♘c3♗xd4 17. exd4 ♖d8 18. a4 ♘b6 19. O-O ♗d7 20. d6Riazantsev-Turov, Ubeda 2001 ) 13. ♕c1 ♕xc1+ 14. ♖xc1O-O 15. ♔d2 ♗xb5 16. ♘dxb5 ♖a5∞ b122) 11. ♗b5+! 11... ♗d7 12. ♘d4+/= b2) 9... O-O b21) 10. e3 ♖d8 11. ♗e2 ♘c6 ( 11... ♘a6⁉ 12. O-O ♘c713. ♗f3 e6=/∞ ) 12. OO e6 13. ♕d2 ♘e7=/∞ b22) 10. ♖c1 a5 11. ♘d2 a4⁉ ( 11... ♕xb2 12. ♘a4 ♕d413. e3 ♕a7 14. ♕b3+/= ) 12. ♘de4 ♗g7 13. ♕d2 ♗f5=/∞( 13... ♖a5 14. e3 ♖d8=/∞ ) b23) 10. e4 a5 11. a4 ♗d7 12. ♗e2 ♖c8 13. O-O ♘a6=/∞ b24) 10. ♕d2⁉ 10... ♖d8 ( 10... a5⁉ ; 10... ♗f5⁉ ) 11. e4 (11. e3 ♘c6 ) 11... a5 b241) 12. ♖d1 a4 13. ♘d4 ♘c6 ( 13... a3 14. b3♘c6=/∞ ) 14. ♘db5 ♗g4 15. ♖c1 ( 15. ♗e2 ♗xe2 16.♕xe2 a3!= ) 15... ♘d4 ( 15... ♘e5 16. ♗e2 ) 16.♘xd4 ♗xd4=/∞ b242) 12. ♕e3 12... ♘d7! ( 12... ♕xe3+ 13. fxe3 a4 14.♘d4+/= ) 13. ♖d1 ( 13. ♕xb6 ♘xb6=/∞ ) 13... ♕xe3+14. fxe3 ♗h4+ ( 14... a4 15. ♘d4 ) 15. ♔d2 ( 15. g3♗g5 16. ♔d2 ♘b6 ) 15... e6 16. ♘d4 ( 16. dxe6♘e5+ 17. ♔c2 ♗xe6 18. ♗e2 ♖dc8 19. ♔b1 ♗g5 ) 16... ♘b6=/∞ 6... ♗g7! Forced. 6... ♘c6 7. ♘db5 e5 8. ♗g5+− 7. ♘db5 O-O Some kind of disclaimer is once again required. If you're against positions where you have to sacrifice material for no immediate threats but instead aiming for general mayhem, then perhaps this isn't for you. I think it's quite playable, however, and I wouldn't mind playing it once or twice myself if given the chance, so I'm not recommending you something I completely distrust. It's an interesting sideline, especially against someone who's not very well-prepared with White. 8. cxd5

8. e3 is a solid move which allows less craziness. 8... ♕a5 Strictly the only move, but nevertheless a very decent one. Play could continue: 9.a3 ( 9. ♕a4 1/2 Turov-Vocaturo, Eforie Nord op 2010; 9. cxd5 a6 10.♘c7 ♘e4 is very sharp, but I think Black is fine. ) 9... ♘e4! ( 9...dxc4 10. ♗xc4 ♘e4 11. OO a6 12. ♘c7 ♘xc3 13. bxc3 e5 14.♘xa8 exf4 15. ♕b3+/= ) 10. b4 ♕d8 11. ♘xd5 ( 11. ♘xe4 dxe412. ♖c1 ♘c6 White isn't well-developed enough to pose many problems so I think Black's position is quite playable and much better than it looks.13. ♘c7 ♕xd1+ 14. ♖xd1 ♗g4 Black's pieces start coming out. 15.♗e2 ♗xe2 16. ♔xe2 ♖ad8= ) 11... e6 12. f3 exd5 13. fxe4dxe4 14. ♘c7 g5! 15. ♕xd8 ♖xd8 16. ♗xg5 ♖d7 17. ♘xa8♗xa1 18. ♗e2 ♗c3+ 19. ♔f2 ♘c6= 8. ♘c7? ♘h5 just isn't very good for White. 8. ♖c1 ♗d7! a) 9. ♘c7 ♘h5 10. ♘xa8 ♘xf4 11. cxd5 e6 12. d6 ♗c6 13.e4 ♕a5 ( 13... b5 14. ♘c7 b4=/∞ ; 13... ♗e5⁉ ) 14. g3 b5 15.a3 ( 15. gxf4 b4! ) 15... b4 16. axb4 ♕xb4 17. f3 ♗xa8 18.gxf4 ♖d8 19. ♔f2 ♖xd6→ b) 9. e3 9... ♗xb5 10. cxb5 ( 10. ♘xb5? ♕a5+→ ) 10... e6∞ 8... a6 8... ♘h5⁈ 9. ♗g5 ( 9. ♗c7 ♕d7 10. e4 a6 11. ♗xb8 axb5 ) 9...a6 10. ♘d4 h6 11. ♗e3 ♘f6 12. ♘b3+/= Black will need to prove he has enough for the pawn. 9. ♘c7 ♖a7 I know this looks slightly strange, but on the other hand there's now a large onus on White to play very precisely. If he doesn't he might run into trouble. 9... ♘h5? 10. ♘xa8 ♘xf4 11. ♕d2 ♘h5 12. e3 ♘d7 13. ♗e2♘hf6 14. d6 exd6 15. ♕xd6± 10. d6 The main move. 10. ♖c1 is met by 10... b5 and Black can already look to the future with quiet hope. White pretty much has to play 11. d6 exd6 12. ♕xd6♘h5 and the pieces start jumping around. I think Black is at least not worse. 10. ♘e8 ♖xe8 11. ♗xb8 ♖a8 12. ♗e5 ♕b6 Black is already threatening ...Ng4, the pawn on b2 is hanging and I think Black should be able to equalise quite comfortably. 13. ♕b3 ♕xb3 14. axb3 ♖d8 15.♖d1 ♗f5= 10. e3 ♘h5! 10... b5 The following line is pretty much forced for White: 11. dxe7 11. ♖c1 exd6 12. ♕xd6 ♘h5! 13. ♖d1 ♘d7→ 11... ♕xe7 12. ♕d6 12. ♗d6? simply runs into 12... ♕d8 13. ♗xf8 ♕xf8 14. ♘7d5♘xd5 Black is an exchange down - I understand that - but he's so far ahead in development that I'm fairly sure none of you should be concerned about playing a position like this. I think Black is at least not worse. 15.♕xd5 ( 15. ♘xd5 ♗e6 16. ♘c3 ♕c8! 17. ♖c1 ♖d7 18. ♕c2♖c7∓ ) 15... ♗e6 16. ♕d2 ♖d7 17. ♕c1 ♖c7 18. ♕d2 ♕c8 19.♖d1 ♖xc3 20. ♕d8+ ♗f8 21. bxc3 ♕xc3+ 22. ♕d2 ♕a3∓

12... ♕xd6 13. ♗xd6 b4⁉

13... ♖d8 14. ♗e7 ♖d7 15. ♗xf6 ♗xf6 ( 15... ♖axc7 16. ♗xg7♔xg7 17. a3 White has to stop ...b4. 17... ♘c6 18. e3 ♘a5 I looked at this position for quite some time. Black is active, but on the other hand before you sacrifice a whole pawn for some kind of positional advantage you have to make sure you're not losing on the spot. I believe in this case Black is fine and should have enough for equality, although of course not more than equality. ) 16. ♘7d5 ♗xc3+ 17. ♘xc3♖ac7=/∞ 14. ♘3d5 14. ♗xf8 bxc3! 15. ♗xg7 ♔xg7∓

14... ♖d8 15. ♗f4 ♘xd5 16. ♖d1 ♗e6 17. e4 ♗xb2 18. ♘xd5♘c6=

5.Bg2 1. ♘f3 ♘f6 2. c4 c5 3. g3 d5 4. cxd5 ♘xd5 5. ♗g2 5. ♘c3 g6 transposes to the main line. 5. d4⁈ isn't particularly good, because after 5... cxd4 6. ♘xd4 Black can just play 6... e5 7. ♘b5 ♕a5+ 8. ♘1c3 ♘xc3 9. ♘xc3 ♘c610. ♗g2 ♗e6 and it's obvious that Black's pieces are coming out comfortably and he can have no real problems. 11. O-O ♖d8 12. ♗d2♕a6∞ 5... ♘c6 6. d4 6. O-O⁈ e5 Black gets a very favourable version of the English Opening.7. ♘c3 ( 7. d3 ♗e7 8. ♘c3 ♗e6 ) 7... ♗e6! An additional option that's not normally available and leaves Black incredibly comfortable. Normally in the equivalent position he has to play Nd5-c7. If Black isn't forced to play that it can only be to his benefit. 6. ♘c3 g6 transposes to the main line. 6... cxd4 6... ♘f6⁉ is a decent alternative: a) 7. OO cxd4 8. ♕a4 ♕d7⁉ ( 8... ♗d7 9. ♘xd4 ♕b6 10.♘xc6 ♗xc6 11. ♗xc6+ ♕xc6 12. ♕xc6+ bxc6 13. ♘d2+/=Cheparinov-Navara, Sochi 2008 ) 9. e3 d3 10. ♖d1 e5 11. ♘c3♗e7 , and Black is ok, e.g. 12. ♖xd3 ♕xd3 13. ♘xe5 ♕d6 14.♘xc6 O-O 15. ♘xe7+ ♕xe7=/+ b) 7. ♕a4 7... ♗d7 8. dxc5 e5 9. ♘c3 ♗xc5 10. OO or ( 10.♗g5⁉ h6 11. ♗xf6 ♕xf6 12. OO OO transposes to 10.0-0. ) b1) 10... O-O 11. ♗g5 b11) 11... ♗e7 12. ♖fd1 ♘d4 13. ♕c4 ♖c8 14. ♕d3♗f5 15. ♕d2 ( 15. e4! ♗g4 16. ♘xe5 ♗xd1 17. ♖xd1♘e6 18. ♕b1 ♕a5 19. ♗xf6 ♗xf6 20. ♘d7 ♗xc3 21.♘xf8 ♘xf8 22. bxc3 b6 , and Black should hold. ) 15...♘xf3+ 16. ♗xf3 ♕xd2 17. ♖xd2 b6= DorfmanMikhalchishin, Lviv 1984 b12) h6⁉ 12. ♗xf6 ( 12. ♘e4 ♗xf2+! 13. ♖xf2 hxg5∞ )12... ♕xf6 13. ♘e4 ♕e7= b2) 10... h6⁉ 11. ♘d2 O-O 12. ♘de4 ♘xe4 13. ♕xe4 ♗e614. ♖d1 Andersson-Portisch, Reggio Emilia 1985 14... ♕c7 15.♗e3 ( 15. ♘d5 ♖ad8! ) 15... f5⇆

7. ♘xd4 ♘db4 8. ♘xc6 ♕xd1+ 9. ♔xd1 ♘xc6 10. ♘c3

10. ♗xc6+ bxc6 11. ♗e3 e5 12. ♘c3 ( 12. ♘d2 ♗e6 13. ♔c2♗b4 14. ♖hd1 a5 15. a3 ♗e7 16. b3 OO⇆ Ftacnik-Farago, Hastings 1982 ) 12... ♗b4 13. ♔d2 ♗e6 14. ♖hc1 OO 15. a3♗e7 16. ♘a4 ♖fb8 17. ♗c5 ♗g5+ 18. ♗e3 ♗e7= StohlTimman, BL 2000 10... ♗d7 10... ♗f5⁉ 11. ♗e3 g6 12. ♗xc6+ bxc6 13. ♖c1 ♗g7 14. f3 O-ORee-Miles, Wijk aan Zee 1978 11. ♗e3 g6 11... e6⁈ 12. ♖c1 ♗e7 13. ♔c2 O-O 14. ♖hd1 ♖fd8 15. ♔b1+/=Keres-Korchnoi, Moscow 1963 11... OOO 12. ♔e1 ( 12. ♔c1⁈ g6 13. ♖d1 ♗g7 14. ♗xc6♗xc6 15. ♗xa7 ♗xc3 16. bxc3 ♖xd1+ 17. ♔xd1 ♖d8+ 18. ♔e1♖d5=/∞ Milos-Morovic Fernandez, Santiago 1989; 12. ♔c2 e5 13.♖ad1 ♘d4+ 14. ♔b1 ♗c5 15. ♘e4 ♗e7 16. f4+/= Kengis-Hegeler, Gausdal 1991) a) 12... e5 13. ♖c1 ♔b8 14. f4 f5! ( 14... ♗d6 15. ♘b5 ♗b4+16. ♔f2 a6 17. ♘c3+/= ) 15. fxe5 ( 15. ♔f2 e4 16. ♖hd1 g6= )15... ♘xe5 16. ♗f4 ♖e8 17. ♖d1 ♗c6⇆ Kramnik-Kasparov, Frankfurt 2000 b) 12... g6⁉ 13. ♖c1 ♗g7 14. f4 ( 14. ♘b5 ♔b8 15. ♗f4+♗e5= ) 14... ♔b8 15. ♗c5 ♗d4⁉= 11... ♖c8 12. ♖c1 g6 13. f4 ♗g7 14. ♔e1 O-O 15. ♔f2 ♗e6 (15... e5⁈ 16. ♖hd1 ♗e6 17. ♘b5+/= SmirinAlterman, Dresden zt 1998; 15... ♖fd8 16. ♖hd1 ♗e8 17. ♗xc6⁉ ♗xc6 18. ♗xa7♖xd1 19. ♖xd1 ♗xc3 20. bxc3 ♖a8 21. ♗c5+/= ) 16. ♖hd1♖fe8 17. ♘d5 ♗f5 18. ♖c5 e6 19. ♘c3 e5⇆ Andersson-Farago, Rome 1986

11... e5 12. ♖c1 OOO 13. ♔c2 ♗f5+ 14. ♗e4 ♗xe4+ 15.♘xe4 ♗e7 16. ♖hd1 ♖xd1 17. ♖xd1 ♖d8= Andersson -De La Villa, Szirak izt 1987 12. ♖c1 12. ♔c2 ♗f5+ 13. ♔b3⁉ ♖c8 14. ♗xc6+ ♖xc6 15. ♗xa7 ♗e6+16. ♔b4 ♗g7=/∞ Miles-Alburt, Long Beach ch-US 1989 12... ♗g7 We've been following Magnus Carlsen's game against Veselin Topalov from the Norway Tournament in 2013. White perhaps has some slight pull but it's not really anything significant and even Magnus couldn't squeeze water out of this particular stone. I think this is reasonably safe for Black and shouldn't worry anybody. 13. ♔c2 13. f4 OO 14. ♔e1 e5⁈ 15. ♔f2 exf4 16. gxf4 ♗e6 17. b3♖ad8 18. ♘e4 ♗d4 19. ♘c5 ♗xe3+ 20. ♔xe3 ♖fe8 21. ♔f2Andersson-Leko, Ter Apel 1996 21... ♖d4= 13... ♖c8 14. ♖hd1 ♘a5 15. ♗d4 ♗f5+ 16. e4 ♗xd4 17. ♖xd4♗e6 18. b3 f6 19. f4 ♔f7= Carlsen-Topalov, Stavanger 2013

1.Nf3 Nf6 2.c4 c5 3.Nc3 d5 4.cxd5 Nxd5 5.e4 and other 5th moves 5.g3 g6 6.Bg2 Nc6 7.Qa4 7.h4 7.0-0

1.Nf3 Nf6 2.c4 c5 3.Nc3 d5 4.cxd5 Nxd5 5.e4 1. ♘f3 ♘f6 2. c4 c5 3. ♘c3 The main move for White. 3... d5 A more classical approach, with fewer pawns sacrificed by move 7! 3... g6 4. d4 cxd4 5. ♘xd4 d5 transposes to the line a pawn down I showed before. If you think that's fun then the option is still on the table. 4. cxd5 ♘xd5 5. e4 This is the old main theory. 5. ♕a4+⁈ is totally empty: 5... ♘c6 6. ♘e5 ♕c7 ( 6... ♘b4! 7. a3♘c2+ 8. ♕xc2 ♘xe5 9. b4∞ ) 7. ♘xc6 ( 7. ♘xd5 ♕xe5 8. ♘b6 ♖b8 ) 7... ♘xc3 8. dxc3 ♗d7= 5. d4 ♘xc3 6. bxc3 g6 leads to a Grünfeld-type position where White tries to use the fact that the pawn can go to e3, with a secure pawn centre, to play on the queenside. Generally speaking, Black should try to refrain from trading on d4 because then White's pressure along the b and c-files becomes a serious factor, but as long as the pawns are on c5 and c3 Black should be fine. 7. e3 ( 7. ♗f4 ♗g7 8. e3 ♕a5 9. ♕d2 O-O) 7... ♗g7 a) 8. ♗d3 O-O 9. O-O a1) 9... ♘c6 10. ♗a3 b6 11. ♗e4 ♗b7 12. dxc5 ♕c7 13.♘d4 ♘xd4 ( 13... bxc5 14. ♗xc5 1-0 Anand,V (2779)-Krasenkow,M (2651), Calatrava 2007 ) 14. ♗xb7 ♘e2+ 15.♕xe2 ♕xb7 16. cxb6 ( 16. ♖ab1 ♗xc3 17. cxb6 axb6 18.♗c5 ♗f6 19. ♖xb6 1/2 Filippov, V (2605)-Leko,P (2701), Batumi 1999 ) 16... axb6 17. ♗b4 ♖fc8 18. ♖ac1+/= a2) ♕c7 a21) 10. ♖b1 a211) 10... b6 11. ♕e2 ♖d8 12. ♗e4 ( 12. ♖d1 ♗b713. e4 cxd4 14. cxd4 ♘c6 15. d5 ♘d4 16. ♘xd4♗xd4 17. ♗c4 Ribli-Tal, Reykjavik 1988 ) 12... ♗a6! (12... ♘c6 13. d5 f5 14. dxc6 fxe4 15. ♕c4+± ; 12...♗b7 13. ♗xb7 ♕xb7 14. dxc5 ♗xc3 15. e4⁉+/= ) 13.c4 ♘c6 14. d5 f5 15. ♗d3 e5! 16. e4 ♘d4 17. ♘xd4cxd4 18. ♗g5 ♖f8 19. ♖fc1+/= Karpov-Kasparov, Seville 1987 a212) ♘d7 11. e4 e5 12. ♗a3 ♖d8 13. ♕b3 exd4 14.cxd4 cxd4 15. ♖fc1 ♕f4 16. ♗c4 1/2 Karpov,A (2720)-Timman,J (2640), Amsterdam 1985 a22) 10. ♕e2 ♖d8 11. ♖d1 b6 12. ♗b2 ( 12. ♗a3 ♗b713. ♖ac1 ♘c6 14. ♘d2 ♕d7 15. ♘b3 cxd4 16. cxd4 e617. ♗c4 a5 18. ♗b5 a4 19. ♘d2 ♕d5 20. ♘f3 1-0 Razuvaev,Y-Topakian,R, Biel 1994 ) 12... ♘c6 13. ♖ac1♗b7 14. e4 e6 15. ♕e3 ( 15. h4 ♖ac8 16. h5 ♕f4 17.hxg6 hxg6 18. ♗b1 ♘a5 19. ♖e1 ♕g4 1/2-1/2 Kramnik,V-Topalov,V, Wijk aan

Zee 1998 20. ♖cd1 cxd4 21. cxd4 ♘c422. ♗c1 b5 23. g3 ♖e8 24. ♔g2 f5 25. ♘g5 1/2 Kramnik,V (2790)-Topalov,V (2740), Wijk aan Zee 1998 ) a221) 15... ♖d7N 16. h4 h6 ( 16... ♖ad8 17. h5+/= ) 17.♗a3 ♖ad8 a2211) 18. dxc5 bxc5 a22111) 19. ♗xc5? ♘e5! 20. ♘xe5 ( 20. ♗d4?♘xd3 21. ♖xd3 e5−+ ) 20... ♗xe5 21. ♖b1♗xe4 22. ♕xe4 ♕xc5∓ a22112) 19. ♗e2=

a2212) 18. h5 18... g5 19. e5 ♘e7!∞ Filippov-Lautier, Rethymnon 2003 a222) 15... ♖ac8 16. h4 h6 17. ♗e2 ♕e7 18. h5 g5 19.♗a3 ♗f8 20. dxc5 ♕f6 21. e5 ♕f4 Portisch-Kasparov, Madrid 1988 a23) 10. a4 ♘c6 11. ♗a3 b6 12. ♖b1 ( 12. dxc5 bxc5 13.♖b1 ♖b8 14. ♖xb8 ♘xb8 15. ♕b3 ♗e6 16. ♗c4 ♗xc417. ♕xc4 ♘d7 1/2 Salov,V (2675)-Ivanchuk,V (2730), Amsterdam 1996 ) 12... ♖d8 13. ♕e2 ♘a5 14. dxc5 bxc515. ♖fd1 ♗d7 16. ♗b5 a6 17. ♗xd7 ♖xd7 18. ♖xd7♕xd7 Korchnoi-Sax, Wijk aan Zee 1991 a24) 10. ♗a3 10... ♘d7 11. ♕e2 b6 12. e4 ♗b7 13.♖fd1 ♖fd8 14. ♖ab1 ♖ac8 15. ♕e3 ♘f6 16. ♗b2 e617. h3 ♕c6 18. ♘d2 ♕a4 19. a3 ♗a6 Portisch-Tal, 1975 b) 8. ♗e2 OO 9. OO b6 10. a4 ♘c6 ( 10... ♗b7 11. a5 ♕c712. ♕b3 e6 13. ♗b2 ♘d7 14. a6 ♗c6 15. c4 cxd4 16. exd4♖ac8 17. ♖fd1 ♖fd8 1/2 Beliavsky,A (2640)-Sakaev,K (2629), Herceg Novi 2000 ) 11. ♗a3 ♗f5 12. ♘d2 ♕c7 13. e4 ♗d7 14.♘b3 ♖ad8 15. ♕c2 ♘b8 16. dxc5 ♗xa4 17. cxb6 ♕xb6Bareev-Lautier, Ubeda 1997 c) 8. ♗b5+ c1) 8... ♗d7 is also possible: c11) 9. a4⁈ ♕a5! c111) 10. ♗xd7+ ♘xd7 11. ♗d2 OO 12. OO e5 13.♕c2 ( 13. e4 exd4 14. cxd4 ♕a6 Timman-Krasenkow, Amsterdam 2004 ) 13... ♕a6 14. d5 e4! 15. ♕xe4 ♘f6Polugaevsky-Vaganian, Moscow URS-ch 1983 c112) 10. ♕b3 10... cxd4 11. exd4 OO 12. OO ♕c7⁉ (12... ♗e6 13. ♕a3 a6 14. ♗e2 ♘c6 15. ♗e3 ♗d5 1/2 Petrosian-Smejkal, Rio de Janeiro 1979 ) 13. ♖e1 e6 14.♘g5 ♘c6∞ Anand-Leko, Linares 1999 c12) 9. ♖b1 ♗xb5 10. ♖xb5 b6 11. OO ♘d7 12. e4 OO13. ♗g5 h6 14. ♗h4 ♕c7 Piket-Van Wely, Monaco 2000 c13) 9. ♕b3 OO 10. OO ♕c7 11. ♗a3 ♗e6⁉ 12. d5♗g4 13. ♗e2 ♘d7 14. h3 ♗xf3 15. ♗xf3 ♖ab8 16.♗b2 b5 17. ♖fd1 c4 18. ♕c2 ♘c5=/+ Karpov-Topalov, Monaco 1998

c14) 9. ♗d3 9... OO ( 9... ♗c6 10. OO OO 11. ♖b1♘d7 12. e4 e6 13. ♕e2 a6 14. ♗e3 b5 15. ♖bd1 ♕c7⇆Kramnik-Navara, Prague 2008 ) 10. OO ♘c6 11. ♖b1 ♕c712. h3 ♖fd8 13. ♕e2 ♗e8 14. ♗a3 b6 15. ♖fc1 ♖ac8 16. ♗a6 ♖b8 17. ♘d2 ♘a5 18. ♘b3 ♘xb3 19. axb3♗c6 Karpov-Leko, Polanica Zdroj 1998 c2) 8... ♘d7 9. O-O O-O 10. a4 Or the immediate ( 10. ♕e2a6 11. ♗d3 e5 12. ♖b1 ♕a5 13. ♕c2 b5 14. a4 ♖b8 15.axb5 axb5 1/2 Portisch,L (2610)-Miles,A ( 2610), Bugojno 1986 ) c21) 10... b6 11. a5 ♗b7 12. a6⁉ ( 12. axb6 axb6 has been seen in a number of high-level games, but it's completely harmless for Black. ) 12... ♗e4 13. ♗b2 ♖c8 14. ♕e2 e615. c4 ♕c7∞ c22) 10... ♕c7 c221) 11. e4⁉ ♘f6 12. ♖e1 ♖d8 13. ♗a3 ( 13. h3 b614. ♗g5 ♗b7 15. ♗d3 h6 16. ♗d2 ♖ac8 17. a5 e6∞ )13... b6 c2211) 14. ♕b3 e6 15. a5 ♗b7 16. e5 ♘h5⁉ ( 16...♘d5 17. a6 +/= Timman-Svidler, Wijk aan Zee 2004 ) 17.♘g5 ♘f4⇆

c2212) 14. ♗d3 14... ♗b7 15. ♕d2 e6 16. h3♖ac8∞ KorchnoiKamsky, Tilburg 1991 c222) 11. ♗a3 11... b6 12. a5 ♗b7 has been seen in two of my own games, stretching across almost a quarter of a century: 13. ♕e2 ( 13. ♕a4 ♘f6 14. a6 ♗d5 15. ♖fe1♘e4 16. ♖ac1 ♖ad8 17. ♗d3 f5 18. c4 ♗c6 19. ♕c2e6⇆ Frolov-Svidler, Leningrad 1990 ) 13... ♘f6 14. a6♗c6 ( 14... ♗e4⁉ ) 15. ♗xc6 ♕xc6 16. ♘e5 ♕c7 17.♕f3 c2221) 17... ♖fc8 18. ♖ab1 e6 19. ♖fc1 ♘e8 20.♘c4 ( 20. ♘d3 ♕d7∞ ) 20... ♗f8!∞ Bologan-Svidler, Tromso WCup 2013 c2222) e6⁉ 18. ♖ab1 ♖ac8 19. ♖fc1 ♖fd8 is perhaps a more harmonious set-up, with Ne8-d6 to follow. 5. e3 is an interesting move, but only in connection with 7.h4. a) 5... g6⁉ was played by Magnus, but I suspect he was purposely trying to provoke his opponent to play sharper: 6. ♗b5+ ( 6. ♕b3⁉ is a very promising move here. ) 6... ♗d7 7. ♕b3 ♘xc3 8. ♘e5 e69. bxc3 ♗xb5 10. ♕xb5+ ♘d7 11. ♖b1 b6 12. c4 ( 12. d4 ♗d6 ) 12... ♗d6 13. ♘xd7 ( 13. ♗b2 ♗xe5 14. ♗xe5 a6 15.♕b2 ♘xe5 16. ♕xe5 OO ) 13... ♕xd7 14. a4 O-O-O 15. a5♕xb5 16. ♖xb5 ♗c7 was eventually drawn in Andreikin-Carlsen, Moscow Tal 2013 b) 5... ♘c6 is a decent choice. b1) 6. d4 e6 ( 6... cxd4 7. exd4 g6 transposes to another fashionable line: 8. ♕b3 ♘xc3 9. ♗c4 ♘d5 10. ♗xd5 e6 ) 7.♗d3 leads to Nimzo-QGD positions.

b2) 6. ♗b5 6... e6 White has to choose between QGD-type structures after 7.d4 and 7. ♘e5 ♗d7 8. ♗xc6 ♗xc6 9. ♘xc6bxc6 10. b3 ( 10. ♕a4 ♕d7 11. b3 ♘b4 12. OO ♗e7 13.♗a3 Polugaevsky-Andersson, Biel 1990 13... a5∞ ) 10... ♘xc311. dxc3 ♕xd1+ 12. ♔xd1 OOO 13. ♔c2 ♗e7 14. e4 h5 15. ♗e3 g6 16. ♖ad1 ♔b7 17. ♖d3 c4! 18. bxc4 c5 , and Black held in Kramnik-Anand, Monaco 20th Amber Blind 2011 c) 5... ♘xc3 6. bxc3 g6 , and if White wants to avoid transposing into positions discussed in notes to 5.d4 he should try 7.h4!? c1) Some high-level games after 7. ♗b5+ continued c11) 7... ♗d7 8. a4 ( 8. ♕b3 ♘c6 9. ♗a3 b6 10. ♖b1♗g7 11. d4 OO 12. OO ♘a5 13. ♕a4 ♗f5 14. ♖bc1♕c7 15. dxc5 bxc5 16. ♖fd1 ♖fd8∞ Speelman-Sasikiran, Gibraltar Masters 2011 ) 8... ♗g7 9. OO OO 10. ♗a3♕c7! ( 10... b6⁈ 11. d4 a6 12. ♗e2 ♘c6 13. ♕c2 ♕c714. dxc5+/= Bareev-Kasparov, Cannes rapid 2001 ) 11. d4♗xb5 12. axb5 ♘d7 , and Black is ready to play either a6 or a5, with good play. c12) 7... ♘d7 8. a4 ♗g7 9. a5 OO 10. OO ♘f6 11. d4♗f5 12. ♕b3 ♖c8 13. ♗b2 ♘e4 14. ♗e2 ♕c7 15. c4cxd4 16. exd4 ♖fd8∞ Gelfand-Van Wely, Moscow 2004 c2) 7. d4 ♗g7 transposes. c3) 7. h4⁉ 7... ♗g7 8. h5 ♘c6 9. ♗c4⁉ ( 9. ♖b1 ♕c7 10.d4 cxd4 11. cxd4 OO 12. hxg6 hxg6 13. ♕c2 ♗e6∞ Popov-Vorobiov, Tomsk 2006 ) 9... a6 10. d4 b5 ( 10... e5 11. hxg6hxg6 12. ♖xh8+ ♗xh8 13. ♕b3 ♕e7 14. ♗a3 exd4 15.cxd4 ♘xd4 16. ♘xd4 ♗xd4 17. ♖d1± Lerner-Sideif Sade, Frunze 1979 ) 11. ♗b3 , and Black's position feels a bit insecure. 5... ♘b4 6. ♗c4 6. ♗b5+ has for many years been known to be a draw after 6... ♘8c67. d4 cxd4 8. a3 dxc3 There's a lot of theory here but nobody plays this line any more with the white pieces because Black has finally learned how to make a draw. 6... ♘d3+ 7. ♔e2 ♘f4+ 8. ♔f1 ♘d3⁉ This very funny looking move was played by Fabiano Caruana against Dmitry Jakovenko. As things stand right now this seems to be quite possible, and if this move is possible it closes down the entire line. This has to be the way players with Black will keep playing. 8... ♘e6 is possible, and nowadays White usually replies with 9. b4 This is an extremely exciting position which is a lot of fun to analyse. 9... g6 (9... cxb4 10. ♘e2=/∞ ) 10. ♗xe6 ♗xe6 11. bxc5 ♗g7 12. d4∞ is a very interesting position I might cover at some point in the future, but the most trendy reply here is 8...Nd3. 9. ♕e2 9. ♕b3 e6 a) 10. ♕b5+⁈ ♗d7! 11. ♕xb7 ♘c6 a1) 12. ♗xd3 ♖b8 13. ♕a6 ♘b4 14. ♕c4 ( 14. ♕xa7 ♖a815. ♕b7 ♘xd3 ) 14... ♘xd3∓ 15. ♕xd3? ♗b5 a2) 12. ♘b5 12... ♖b8 13. ♕c7 ♘xc1 14. ♕xd8+ ♔xd8 15.♖xc1 ♘a5=/+ b) 10. ♘a4 ♘b4=/+

c) 10. ♗b5+ 10... ♘c6 11. ♘a4 ♘b4 12. a3 ♕a5 13. ♗e2♘a6 14. ♘c3 ♗e7 15. d3 OO 16. ♗e3 ♖d8 17. ♖c1 ♗f6 , with excellent play for Black in Jakovenko-Caruana, Sochi 20th RUS-tch 2013 9... ♘xc1 10. ♖xc1 e6 A sharp, unclear position - Black will obviously be fine if he finishes development, so White needs to be very energetic. Play may continue 11. h4⁉ 11. ♗b5+ a) 11... ♗d7⁉ 12. d4 ( 12. g3 a6 13. ♗xd7+ ♕xd7 14. d4 cxd415. ♘d5 d3! 16. ♕e3 exd5 17. exd5+ ♗e7 18. d6 ♘c6 19.dxe7 ♘xe7= ) 12... cxd4 13. ♘xd4 a6 , and the beautiful 14.♘d5⁉ ( 14. ♗xd7+ ♘xd7= ) 14... ♗d6 15. e5 runs into the no less stunning 15... ♗xe5! 16. ♕xe5 OO‼ 17. ♘c7 ♗xb5+ 18.♘dxb5 ♘c6 19. ♕g3 ♖b8! ( 19... ♖c8 20. ♘e8 ♖xe8 21.♘d6 ) 20. ♘c3 ♖c8=/+ b) ♘d7 12. e5 a6 13. ♗d3 b5 14. ♗e4 ♖b8 15. g3 ♗e7 16.♔g2 O-O 17. ♖hd1 ♕b6∞ 11. d4 cxd4 12. ♘b5 ( 12. ♗b5+ ♗d7 13. ♘xd4 a6 transposes to 11.Bb5 Bd7. ) 12... ♘c6 13. ♗d5 d3!∓

11... a6 12. e5 ♘c6 13. h5 h6

13... b5 14. ♗d3 ♗b7 15. h6 g6 16. ♘e4± 14. ♖h3 b5 15. ♗d3 ♗b7 16. ♗e4 , with very sharp play in Sutovsky-Nepomniachtchi, Poikovsky 2013.

5.g3 g6 6.Bg2 Nc6 7.Qa4 1. ♘f3 ♘f6 2. c4 c5 3. ♘c3 d5 4. cxd5 ♘xd5 5. g3 The main move for players with White nowadays. 5... g6 There are a lot of move orders here and I'll try to cover more or less everything on moves 6, 7 and 8 to give you a complete picture of what's what - what you should be aiming for and what you should be avoiding. 6. ♗g2 The main move. 6. d4 ♗g7 7. ♗g2 ♘xc3 8. bxc3 cxd4 9. cxd4 ♘c6 10. O-O O-Otransposes to 1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.g3 ( Ponomariov-Svidler, 2011), which I covered elsewhere in this series. 6. h4 is met by 6... h6 Generally speaking, I'm suggesting you play ...h6 against h4 in every single case. I'm not sure this "weakening" of the black kingside can be exploited, and in fact there are many cases, as you'll see later on, where the inclusion of h4 h6 gives Black an additional stabilising option in the centre of playing ...Bc8-e6, because Ng5 is no longer a possibility for White. ( 6... ♘c6 7. h5 ♗g7 8. ♗g2 ) a) 7. ♕b3 ♘b4 8. ♘e5 ♗e6 9. ♕a4+ ♘8c6 10. ♗g2 ♗g7Aiming for some sacrifices and compensation. This move and 10...Qa5 are both fine for Black. ( 10... ♕a5 11. ♘xc6 ♕xa4 12. ♘xa4♘c2+ 13. ♔f1 ♘xa1 14. b3 ♗g7 15. ♘a5∞ ) 11. ♘xc6 ♘xc6 12. ♗xc6+ bxc6 13. ♕xc6+ ♔f8 14. ♕xc5 ♗d4=/∞ b) 7. ♗g2 is pretty much obligatory for White, and after 7... ♘c6that transposes to the main line. c) 7. ♕a4+ 7... ♘c6 8. ♘e5 ( 8. ♗g2 ♗g7 9. h5 g5 10. ♘e5♗xe5 11. ♗xd5 ♗d7 ) 8... ♘xc3 9. dxc3 ♕d5 10. ♘xc6 ♗d7You'll see this position in a slightly different version with pawns on h2 and h7 in 6.Qa4+ as well. This is well-known to be completely harmless for Black. 11. e4 ♕xc6 12. ♕xc6 ♗xc6 13. f3

6. ♕a4+ ♘c6 7. ♘e5 is adequately met by 7... ♘xc3 8. dxc3 ♕d59. ♘xc6 ♗d7= 6. ♕b3 is, as usual, met by 6... ♘b4! It's useful to form some sort of patterns here for the various replies to queen jumps on the queenside. Qb3, for instance, is almost always met by ...Nb4. a) 7. ♘e5 is met by 7... ♗e6 8. ♕a4+ ♘8c6 9. a3 ( 9. ♗g2♕a5 ) 9... ♘c2+ 10. ♕xc2 ♘xe5 11. ♗g2 ♗g7 12. ♗xb7 ♖b8 13. ♗g2 O-O=/∞ b) 7. ♗g2 ♗g7 8. O-O O-O 9. d3 ♗e6 10. ♕d1 ♘8c6 , and Black is doing fine. c) 7. a3 ♘4c6 8. d3 ♗g7 9. ♗g2 O-O 10. O-O ♘a5 11. ♕d1♘bc6∞ d) 7. ♘e4 7... ♗g7 It's obviously very important not to allow Qc3, so Black happily sacrifices the c5pawn. ( 7... b6 8. ♗g2 ♗e6 9. ♕c3f6 10. a3 ♘d5 11. ♕c2 ♘d7 12. d4± Polugaevsky-Furman, USSR 1971 ) 8. ♘xc5 ♕a5 Black has fantastic play for the pawn. d1) 9. ♕c4 was seen in an old game between Miles and Korchnoi.9... ♗f5 ( 9... ♘8a6⁉ 10. ♘xa6 ♘xa6=/∞ would have been very strong. Black has fantastic play for a single pawn. ) 10. e4 ♗g411. h3 ♘8a6 12. hxg4 ♕xc5 13. ♕xc5 ♘xc5 14. ♗b5+ ♔f815. ♔e2 ♘xe4= Korchnoi-Miles, Tilburg 1986 d2) 9. a3 9... ♘8c6 The simplest way to proceed. ( 9... ♘4c6 is also quite playable. 10. ♕c4 b5 11. ♕h4 b4 12. ♘d3 ♘a6 13.♗g2 ♗d7 14. OO ♖c8 15. ♘de1 ♘c5 16. ♘c2 ♘b3 17.♖b1 ♕c5=/∞ Polugaevsky-Bronstein, SU-ch 1971 ) 10. ♕c4 O-OBlack has fantastic play for the pawn and shouldn't be at all worried. ( 10... b5 11. ♕e4 f5 12. ♕b1 ♘d5 13. b4 ♕d8 14. ♗b2e5 15. ♗g2+− Kuligowski-Schmidt, Warsaw 1978 ) 11. ♗g2 ( 11.♖b1 ♗f5 12. axb4 ♘xb4 13. e4 ♖ac8−+ ) 11... b5 12. ♕b3♗f5 13. e4 ♗g4 14. O-O ♗xf3 15. ♗xf3 ♘a6 16. ♘xa6♘d4 17. ♕d1 ♕xa6=/∞ Forintos-Bukic, Ljubljana 1981 6... ♘c6 It's important to start with this move, after which White once again has a wide range of options. 6... ♗g7 is unpleasant for Black as it allows White the option of 7.♘g5! ♘xc3 ( 7... e6 8. ♘ge4 The c5-pawn is hanging, so Black has to defend it somehow. 8... b6 Black would be fine if he was in time to play ...Nc6 but unfortunately 9. d4 is an important blow here. I can't really recommend this position to anyone playing Black as it's quite dangerous.)8. dxc3 ♕xd1+ 9. ♔xd1 OO 10. ♗e3 c4 11. ♔e1 h6 12. ♘e4♗d7 13. f4 ( 13. a4 ♗c6 14. ♔f1 ♘d7 15. ♘d2 ♖fc8 16. a5 b517. axb6 axb6 18. ♖xa8 ♗xa8 19. ♗xa8 ♖xa8 20. ♘xc4 ♖a2Sargissian-Sutovsky, Antwerp 2009 ) 13... ♗c6 14. ♔f2 ♘d7 15. a4♖fc8 16. ♖hd1 f5 17. ♘d2 ♘f6 18. h3 b6 19. ♗f3 ♖ab8 20.♗d4+/= Sarg issian-Howell, Antwerp 2009 7. ♕a4 7. ♕b3 ♘db4 Once again this is the response to the queen coming to b3. ( 7... e6 8. OO ♗g7 9. d3 OO 10. ♕b5 ♕b6 11. ♕c4 ♘xc312. bxc3 ♘a5 13. ♕a4 ♗xc3 14. ♗h6 ♗xa1 15. ♖xa1 ♘c6 16.♗xf8 ♔xf8 17. ♕f4=/∞ )

a) 8. a3 can be met by 8... ♘a5 , with the aim of putting the other knight on c6. For instance: 9. ♕d1 ( 9. ♕a4+ ♗d7 10. ♕d1♘bc6 11. d3 ♗g7 12. ♗e3 c4 ) 9... ♘bc6 Black is just one move away from completing his development, when the weakness on b3 may be more telling than the somewhat weird-looking knights on c6 and a5.10. b4 cxb4 11. axb4 ♘xb4 12. ♕a4+ ♘bc6 ( 12... ♘ac6 13.♗a3± ) 13. ♘e5 ♗g7 14. ♘xc6 ♘xc6 15. ♗xc6+ bxc6 16.♕xc6+ ♗d7 17. ♕e4 O-O 18. O-O ♗f5=/+ Panno-Kavalek, Biel 1977 b) 8. O-O 8... ♗g7 9. d3 b1) 9... OO 10. ♗e3 b6 11. ♕d1 ( 11. a3 ♘a5 12. ♕d1♘bc6 ) 11... ♗b7 12. ♖c1 ♖c8 13. ♕a4 ♘a5 14. a3 ♗c615. ♕d1 ♘d5=/+ Djuric-Langeweg, Formia 1994 b2) ♗e6 10. ♕d1 OO 11. ♘g5 ( 11. ♗e3 b6 12. ♕d2 ♖c813. ♗h6 f6 14. ♖fd1 ♗xh6 15. ♕xh6 ♘d4 16. h3 ♕d7 17.♔h2 ♖fd8 18. ♘h4 ♕e8 19. ♖d2 ♕f7 20. ♗e4 ♕g7Uhlmann ,W-Schmidt, Halle 1976 ) 11... ♗d5 ( 11... ♗d7 12.♗e3 ) 12. ♘ge4 ♗xe4 13. ♗xe4 e6= 7. d4 The problem with this is that we now have a knight on c6, and after 7... ♗g7 8. O-O we have a choice to make. a) 8... O-O 9. ♘xd5 ♕xd5 10. ♗e3+/= b) 8... ♘xc3 is the most natural move, but after 9. bxc3 cxd4White won't take with the pawn but the knight: 10. ♘xd4 This pretty much forces Black to go for the position after 10... ♘xd4 11. cxd4♕xd4 12. ♖b1 This is considered to be a draw and probably is a draw, but it's a draw which Black will have to make with a series of precise moves. He won't have any chances for counterplay whatsoever, so this isn't particularly to my liking. 12... O-O 13. ♗e3 ♕c4 14.♗d5 b1) 14... ♕c7 15. ♕b3 ♕e5 16. ♖fd1 ( 16. ♗f4 ♕f6 17.♗xb7 ♗xb7 18. ♕xb7 ♕e6 19. a4 ♖ad8 20. ♕xa7 ♖a8=Skatchkov-Vokarev, Sochi 20th RUS-tch 2013 ) b2) ♕a6 15. ♕b3 ( 15. ♕d3 ♕xd3 16. exd3 ♖d8 17. ♗xb7♗xb7 18. ♖xb7 ♗f6 19. ♖xa7 ♖xa7 20. ♗xa7 ♖a8Smirnov-Giri, Sochi 13th RUS-tch 2012 ) 15... ♕xe2 16. ♗c5♗e6 17. ♗xe6 ♕xe6 18. ♕xe6 fxe6 19. ♖xb7 b21) 19... ♖f5 20. ♖c7 ♖d5 21. ♖e1 ( 21. ♗xe7 ♖d2 22.a4 ♗d4= ) b22) a6 20. ♖fb1 ♖fc8 21. ♗xe7 ♖c2 22. ♖d1 ♖xa2 23.♖dd7 ♖b2 24. ♖bc7 ♗e5 25. ♗d6 ♗xd6 26. ♖xd6♖f8= Erdos-Jovanovic, HUN-tch 2012 c) 8... cxd4⁉ This is an interesting alternative and my proposed solution here. 9. ♘xd4 ♘xd4 c1) 10. ♗xd5 e6 ( 10... O-O 11. ♗e3 e5 ) 11. ♗g2 The piece on d5 has to go back. ( 11. ♗e3 ♘f5 is clearly good for Black. )11... O-O transposes. c2) 10. ♘xd5 It doesn't really matter with which piece White takes. 10... e6! c21) 11. ♗e3 ♘f5 is likewise good for Black. 12. ♘b4 ♕xd113. ♖fxd1 ♘xe3 14. fxe3 ♗xb2 15. ♖ab1 ♗c3 16. ♘a6♗a5 17. ♖dc1 ( 17. ♗xb7 ♗xb7 18. ♖xb7 O-O ) 17...O-O 18. ♘c7 ♖b8 19. ♘a6 ♖a8= c22) 11. ♘c3 11... O-O I think Black should equalise quite comfortably with accurate play.

c221) 12. ♗e3 e5 is very good for Black. He wants to develop the bishop on c8 somewhere towards the e6-square.13. b3 This prevents 13...Be6 immediately, but it's fine for Black to waste one more tempo on preparing it. ( 13. ♖c1would be met by 13... ♗e6 straightaway, and 14. ♗xb7wouldn't be very good because of 14... ♖b8 ) 13... ♖b8 14.♖c1 ♗g4 Provoking h3. 15. h3 ♗e6 I think Black is quite fine here because the knight on d4 is very hard to drive away. In general, I like Black's harmonious setup. c222) 12. e3 ♘c6 Black is fine. The "half a tempo" White has isn't significant and I think the position is just completely equal. 13. ♕b3 ♘a5 14. ♕b4 ♕b6! 15. ♕xb6 axb6 16.♖d1 ♗xc3 17. bxc3 ♘b3= c223) 12. ♗d2! This is the one move that gave me some trouble, but I finally settled on a move order which I think pretty much equalises. c2231) 12... ♗d7 13. e3 c22311) 13... ♘b5 14. ♗xb7 ♖b8 15. ♗g2+/= c22312) 13... ♘c6 14. ♘e4 ♕e7 15. ♖c1 ( 15.♕b3 f5 16. ♘c3 ♖fd8 17. ♖ad1 ♗e8 ) c22313) ♘f5 14. ♕b3 c2232) 12... ♕b6 13. e3 ♘c6 14. ♘a4 ♕b5 15.♕b3 ♕xb3 16. axb3 ♖d8 17. ♗c3 ♖b8 18.♖fd1+/= c2233) 12... ♕e7 c22331) 13. ♕c1 can be met by sacrificing the pawn on b7: 13... ♗d7 14. ♗xb7 ( 14. ♗g5 ♕c5 ) 14...♖ab8 15. ♗g2 ♖fc8 This position is quite obviously very decent for Black. c22332) 13. e3 13... ♘c6 14. ♕b3 ♖d8 15. ♖fd1 ♗d7 ( 15... ♘a5 is premature because of 16. ♕b5 )16. ♖ac1 ♗e8 ( 16... ♘a5 would be a blunder because 17. ♘d5 is almost winning.; 16... ♖ab8 17.♘e4 ♗e8 18. ♗c3 ♖xd1+ 19. ♖xd1 ♖d8 20.♖c1 ♗xc3 21. ♕xc3 e5 22. h4 ) 17. ♘e4 b6 Black is now ready to start swapping down and the position after the possible 18. ♗c3 ♖xd1+ 19. ♖xd1 ♖d8 is perfectly fine for Black. He has no weaknesses and the somewhat inferior bishop on e8 compared to the bishop on g2 shouldn't be that much of a factor in a position with very few pieces left on the board and no real structural weaknesses. 20. ♖c1 ♘a5 21. ♕c2 ♘c622. ♗xg7 ♔xg7 23. ♕c3+ e5 24. f4 f6 25. a3exf4 26. gxf4 ♕e6 27. ♘g5 ♕e7=

7. ♘g5 is perhaps the worst of White's options and can be met by 7...e6 After 8. ♘ge4 the obvious difference is that the pawn on c5 isn't hanging and you can play 8... f5⁉ 9. ♘xd5 exd5 10. ♘c3 ♗e6 I think Black may already be better because of his huge spatial advantage. 7... ♗g7 8. ♕c4 8. O-O O-O transposes to a later line. 8. ♘g5 can be met by 8... e6 9. ♘ge4 ♘b6 10. ♕b5 O-O Once again the pawn on c5 is just abandoned because White wastes too much time trying to collect it. a) 11. ♕xc5? f5 ( 11... ♘d4=/∞ ) 12. ♘d6 ♘d4 13. O-O ♘c2 14.♖b1 ♗d4−+ b) 11. ♘xc5 11... ♘d4=/∞ 12. ♕d3 ♕e7 13. ♘b3 ( 13. b4♖d8=/∞ ) 13... e5 We're following a game played between Lputian and Timman in Wijk aan Zee 2000. Once again Black has fantastic play for only a single pawn. Generally speaking, in this line you shouldn't worry too much about having to sacrifice the c-pawn for 3 or 4 tempi because you get a lot of compensation. 14. OO ( 14. ♘xd4 exd415. ♘d5 ♘xd5 16. ♗xd5 ♗h3=/∞ ) 14... ♗e6 ( 14... ♗f5 ) 15.♕b1 ♖fc8=/+ Lputian-Timman, Wijk aan Zee 2000 8... ♘db4 9. O-O ♕a5 The most precise move, creating the huge threat of ...Be6. 9... ♕b6⁉ 10. ♘e4 10. a3 ♗e6 11. ♕b5 ♕xb5 12. ♘xb5 ♘a6 13. d3 ♖c8 10... O-O Regardless of how White takes the pawn Black will get fantastic play against the white queenside because his bishops will be attacking the pawns on a2 and b2. 11. ♕xc5 11. ♘xc5 b6 12. ♘b3 ♕h5 13. d3 ♗e6 14. ♕e4 ♖ac8 15. ♗d2a5∞ 11... ♕xc5 12. ♘xc5 b6 13. ♘e4 ♗e6 14. ♘c3 14. ♘fg5 ♗xa2 14... ♖ac8 , and Black has fantastic compensation for the pawn. 15. d3 15. ♘g5 ♗f5 16. d3 ♘d4 17. ♘ge4 ( 17. ♔h1 ♖xc3 18. bxc3♘xe2 ) 17... ♗g4 18. f3 ♗e6 19. ♗d2 ♘dc2 20. ♖ac1 ♗d4+ 21.♔h1 ♘e3 15... ♘c2 16. ♖b1 ♘6b4 17. a4 17. ♗g5 ♘xa2 18. ♘e4 ♖fe8 17... ♘a2 18. ♗d2 18. ♘e4 ♘xc1 19. ♖bxc1 ( 19. ♖fxc1 ♗a2 ) 19... ♗xb2 18... ♗xc3 19. bxc3 ♘xc3 20. ♗xc3 ♖xc3=/+

7.h4 1. ♘f3 ♘f6 2. c4 c5 3. ♘c3 d5 4. cxd5 ♘xd5 5. g3 g6 6. ♗g2♘c6 7. h4⁉ Once again Black responds 7... h6! In this position we can finally have a detailed look at what's going on if White tries to exploit the potential weakness on the kingside. Let's have a look at some of White's options. 8. O-O This is the most principled reply. 8. d3 ♗g7 9. ♗d2 Once again we see a position in which it's possible to play 9... ♗e6! because Ng5 isn't an option. Black is planning to play something like ...Rc8 and then perhaps ...b6 and ...Qd7. Once he's done stabilising his position in the centre he'll decide whether he wants to castle short or play with his king in the centre. For White the move h4 is also a worry if he wants to castle kingside, so this also isn't a huge problem for Black. 10. ♖c1 ( 10. OO b6 ) 10... ♖c8 ( 10... b6 11. a3 ♕d7 12.♕a4 ♖c8 13. OO OO 14. ♘xd5 ♗xd5 15. b4 ♘e5 16. ♕xd7♘xd7 17. bxc5 ♘xc5 18. ♗b4= Loginov-Biriukov, St Petersburg 2007) 11. ♘xd5 ♕xd5 12. ♗c3 OO 13. a3 ♘d4 14. OO ♕d7 15. b4♗d5 16. e4 ♗c6 17. bxc5 ♗a4 18. ♕d2 ♘b3 19. ♕b2 ♘xc1 20.♗xg7 ♘xd3 21. ♕a1 ♖fe8 22. ♗xh6 e5 23. ♗e3 f6∞ Larsen-Sosonko, Tilburg 1979 8. ♕b3 e6 9. O-O ♗g7 8. h5 g5 9. O-O ♗g7 10. ♘xd5 ♕xd5 11. d3 ( 11. d4 cxd4 12.♘xg5 ♕c5 13. ♘e4 ♕xh5 Black is just a pawn up for not very much, so this is quite obviously good for him. ) 11... ♕d6 It's possible to go after the h5-pawn straightaway, but I think this move is more judicious. (11... g4 12. ♘g5 ♕f5 ) 12. ♗e3 a) 12... b6 13. d4 b) 12... O-O 13. ♖c1 ♘d4 ( 13... b6 14. d4 ) 14. ♗xd4 cxd415. ♘d2 ♕b4 c) ♖b8 Preparing b6 and, if necessary, ...Bb7. Once Black has made all of his preparations he'll probably castle, although he could also continue playing with the king in the centre. The pawn on h5 will potentially be a weakness for the remainder of the game. 13. ♕a4 (13. ♘d2 b6 ) 13... ♗d7 14. ♕a3 b6 15. ♖fd1 ♖c8 16. d4cxd4 17. ♕xd6 exd6 18. ♘xd4 ♘xd4 19. ♗xd4 ♗xd4 20.♖xd4 ♖c2 21. ♖xd6 ♔e7 22. ♖ad1 ♗e6= 8... ♗g7 9. ♘xd5 9. d4 cxd4 10. ♘xd4 ♘xd4 11. ♗xd5 O-O It's unclear whether h4 or h6 is the more damaging - the c8-bishop now has a very promising future on g4. 12. ♗e3 ♘f5 13. ♗f4 ♕b6 9. d3 ♗e6 9. ♕a4 ♗e6! 9... ♕xd5 10. d3 O-O I think this is best, facing the dangers "head-on". 10... ♕d6 11. ♗e3 ♖b8 12. ♘d2 b6 13. ♗f4 e5 14. ♘c4±Matjushin-Biriukov, St Petersburg 2002 10... ♗d7 11. ♗e3 ♖c8 12. d4 11. ♗e3 Now the difference becomes obvious. With a pawn on h6 White is threatening Qc1 with a double blow, which has to be addressed. 11... ♕h5 In this position the variations do become very sharp, but I think Black is doing well and I'm happy recommending this to you. White has two main options here. Actually, since recording the video I had a chance to try this out in practice - in the 2nd game of my World Cup match vs Radjabov.

12. ♖c1 Complete mayhem begins if White plays this move. This position actually required some thought because play becomes very sharp. 12. ♕c1 b6 13. ♘h2 ♗d7 14. ♗f3 ♕e5 15. ♗xh6 It seems as though White has made some progress by exchanging the h6-pawn for the b2-pawn after 15... ♗xh6 16. ♕xh6 ♕xb2∞ , which on the face of it has to be good for White, but in fact if you imagine Black returning the queen to g7 and then playing something like ...Rad8, and then ...Nd4, you'll realise that structurally the pawn on h6 is not such a huge deal, whereas Black is quite ready to dominate in the centre and will probably be better in every single endgame. This is a very unclear position that I'm quite happy recommending for Black. 12... b6 13. d4 The main move. 13. ♘g5 seems like an attractive knight jump that threatens the knight on c6 and also some kind of trap for the queen on h5, but it runs into the very strong 13... ♗g4 and now once again White has more hanging than Black. ...Bxe2 is a huge threat and also the knight on g5 will eventually hang, so this is fine for Black. 13. b4 ♘xb4 13... ♖d8! 13... ♗b7 14. dxc5 ♗xb2 15. ♖b1± 14. ♘d2 Unpinning the d-pawn and also threatening the knight on c6. 14... cxd4 A very beautiful continuation. 15. ♗f4 Otherwise Black is clearly doing ok. Up to this point we were both playing very fast, but my next move took Teimour by surprise. 15... d3! It seems as though Black is about to lose a piece on c6, but in fact he gets a lot of counterplay. I checked this position quite carefully because it does look a bit strange and dangerous for Black. I needed to be careful I'm not just recommending something that loses by force. In fact I'm fairly sure Black is not worse anywhere at all here and in many cases he'll start playing for a win if White is not very precise. 15... e5 16. ♗xh6 ( 16. ♗xc6 ♖b8 17. ♗xh6 ♗xh6 ; 16. ♗f3♕f5 17. ♗xc6 ♖b8 18. e4 ♕f6∞ ) 16... d3 ( 16... ♕xh6 17.♗xc6 ♖b8 18. ♘e4+/= ; 16... ♗xh6 17. ♘e4! ♕f5 18. ♖xc6♗e6 19. ♘d6→ ) 17. exd3 ♕xd1 18. ♖cxd1 ♗xh6 19. ♗xc6♖b8 20. ♘c4 f6+/= 16. ♖e1! 16. exd3 ♗g4 17. ♕a4 ♘d4! 18. ♗xa8 ♘e2+ 19. ♔h2 ♘xc1=/+ 16. e3 ♕xd1 17. ♖fxd1 ♗g4! 18. ♗xc6 ( 18. ♖f1 ♗e2 19. ♗xc6♖ac8 ) 18... ♗xd1 19. ♖xd1 ( 19. ♗xa8 ♗c2∓ ) 19... ♖ac8 20.♗b7 ♖c2=/+

16. ♕a4 ♘d4 17. ♗xa8 ♘xe2+ ( 17... ♗d7 18. ♕xa7 ♘xe2+ 19.♔h2 ♘xf4∞ ) 18. ♔h2 ( 18. ♔g2 ♗d7 19. ♗c6 ♗xc6+ 20. ♖xc6♖d4 21. ♖c8+ ♗f8 22. ♕e8 ♘xf4+ 23. gxf4 ♕g4+= ) 18... ♗g419. ♗c6 ( 19. ♕xa7 g5 ) 19... g5 20. ♗e3 gxh4 21. ♔g2 ♗d4!∓

16... ♘d4 17. ♗xa8 ♗g4⁉

17... ♘xe2+ This move was also possible, but I liked the text move better. 18. ♔g2 ( 18. ♔h2 ♗g4∓ ; 18. ♖xe2 dxe2 19. ♕e1 ♗g420. ♗e4 ♗xb2 21. ♖c2 ♗d4=/∞ ) 18... ♗g4 19. ♗f3 ♗xf3+ 20.♘xf3 ♘xf4+ 21. gxf4 ♕g4+ 22. ♔h1 ♕h3+ 23. ♘h2 ( 23. ♔g1= )23... d2 24. ♖g1 ♕xh4 25. ♖c2 ♕xf4 26. f3 h5=/∞ 18. f3

18. ♗e4 ♘xe2+ 19. ♖xe2 dxe2 20. ♕e1 ♗xb2 18... dxe2 19. ♕a4 ♖xa8 20. fxg4 ♕xg4 21. ♔g2 b5 This leads to an immediate draw, but my move, 21...Rd8, is also quite good. The game was eventually drawn after an intense fight. 22. ♕b4 22. ♕a3 e5 23. ♘f3 exf4 24. ♘xd4 ♗xd4 25. ♕f3 ♕xf3+ 26.♔xf3 fxg3= 22... e5 23. ♗e3 ♗f8 24. ♕c3 ♘f5 25. ♗f2 ♗b4 26. ♕c6 ♖d827. ♘e4 ♘xh4+ 28. ♔g1 ♘f3+ 29. ♔g2 ♘h4+=

7.0-0 1. ♘f3 ♘f6 2. c4 c5 3. ♘c3 d5 4. cxd5 ♘xd5 5. g3 g6 6. ♗g2♘c6 7. O-O White's main move. 7... ♗g7 8. ♕a4 This worried me for a while. 8. ♕b3 ♘c7 The simplest response. Generally speaking c7 is a very decent square for the knight. 9. d3 O-O a) 10. ♘e4 b6 b) 10. ♕c4 b6 11. ♕h4 Ng5 is a big threat in this position, but it's very adequately met by 11... ♕d7 ( 11... ♗b7 12. ♗h6 ♖b8 13.♘g5 ♕d4 ) 12. ♗h6 can be met by 12... ♕g4 , starting to swap off pieces. I think Black is doing quite well. c) 10. ♗e3 10... b6 The knight on c7 protects the rook on a8 so there are no tactical jumps with the f3-knight. Black is very comfortable. 11. ♖ac1 c1) 11... ♘e6 12. ♘g5 ♘ed4 13. ♗xd4 ( 13. ♕d1 ♗b7 14.♘f3 ♕d7 15. a3 ♖fd8 16. ♗xd4 ♘xd4 17. ♘xd4 ♗xg2 18.♔xg2 ♗xd4 19. ♕d2 a5 Reshevsky-Kirov, Nice 1974 ) 13...♘xd4 14. ♕a3 ♗d7=/∞ c2) ♗e6 12. ♕a4 ♘d4 13. ♗xd4 cxd4 14. ♘e4 a5∞ 8. ♘g5 e6 9. ♘ge4 b6 a) 10. d3 OO 11. ♗g5 f6 12. ♗d2 ♘de7 ( 12... ♗b7 13. a3♕d7 14. ♖c1 ♖ad8 15. ♘xd5 exd5 16. ♘c3 d4= Speelman-Van der Wiel, Plovdiv 1983 ) 13. a3 f5 14. ♘g5 h6 15. ♘h3 ♗b7 16.b4 cxb4 17. axb4 ♕d7 18. ♕a4 ♖fd8 19. ♘f4 g5 20. ♘h5♗h8 21. h3 ♘d4 22. ♕d1 ♗xg2 23. ♔xg2 ♖ac8∞ Smyslov-Miles, Tilburg 1977 b) 10. ♘xd5 exd5 11. ♕a4 ♗d7 12. ♘c3 ♘d4 13. ♕d1 ♗c6 c) 10. ♕a4 c1) 10... ♗b7⁉ 11. ♘xd5 exd5 12. ♘c3 OO 13. ♗xd5 ( 13.♘xd5 ♘d4 14. ♘c3 b5 15. ♕d1 ♗xg2 16. ♔xg2 b4∓Alfredsson-Wikman, cr 2000 ) 13... ♗xc3 14. ♗xc6 ♗xc6 15.♕xc6 ♗g7 16. d3 ♖e8 17. ♖e1 ♕d4=/∞ c2) 10... ♗d7 11. ♘d6+ ( 11. ♘xd5 exd5 12. ♘c3 ♘e7 13.♕f4 ♗c6 14. b4 OO 15. bxc5 d4∓ Timman-Polugaevsky, Breda 1979 ) 11... ♔e7 12. ♘de4 ♖e8 13. d3 ♔f8 14. ♗g5 f6 15.♗d2 ♘ce7 16. ♕c4 ♗c6 17. a3 h6∞ /\ f5, Bauer-Vachier Lagrave, Chartres 2005

8. d4 transposes. 8. ♘xd5 ♕xd5 9. d3 O-O 10. ♗e3 This is all theory that's been known about for many decades. a) 10... ♗xb2 11. ♖b1 ♗g7 12. ♕a4 ♕d7 13. ♘g5!+/= A very important trick, after which I'm not sure Black is entirely equal. ( 13.♗xc5 was played by Kramnik against Grischuk in a blitz tournament, and after 13... b6 14. ♗e3 ♘d4 Black was fine. 15. ♕d1 ♗b716. ♘xd4 ♗xg2 17. ♔xg2 ♗xd4= Kramnik-Grischuk, Moscow 2012) b) 10... ♗d7! The simplest way to proceed with Black. 11. ♘d4♕d6 12. ♘xc6 ♗xc6 13. ♗xc6 ♕xc6 14. ♖c1 ♕e6 This is a very well-known position in which White for many years has tried claiming some kind of moral advantage. By now, however, I think it's universally accepted that White doesn't really have anything. 15.♖xc5 ♕xa2 16. ♖b5 b6= 8... O-O! 8... ♘b6 I used to play this as I was so worried about 8...0-0 9.Qc4. 9.♕b5 ♘d7 Although this isn't tragic for Black I don't quite like it because the knight doesn't really belong on d7 and you have to work quite hard here not to be worse. I later realised this is uncalled for. 9. ♕c4 9. ♕b5 ♕b6 is perfectly fine for Black. 10. ♕xb6 ♘xb6 11. d3 c412. dxc4 ♘xc4 , and Black was already slightly ahead in Ljubojevic-Ivanchuk, Monaco 1993 9... ♘xc3 9... ♘db4 10. ♕xc5 b6 11. ♕g5 10. dxc3 10. bxc3 b6 11. ♘g5 ♕d7= 10... ♕b6! It turns out Black is completely fine. 10... b6 11. ♘g5 ♕c7 ( 11... ♗b7 It seems as though Black has everything under control, but this is a very wellknown trap that I once fell into in a blitz game. White plays 12. ♘e6! fxe6 13. ♗xc6 and suddenly Black is completely lost. ) 12. ♕h4 h6 13. ♘f3 g5 14.♗xg5 hxg5 15. ♘xg5 ♗f5 16. e4 ♗g6 17. f4 f6 18. ♘e6 1/2 Hertneck-Miles, Dortmund 1986 11. ♘g5 This was the move that used to worry me. 11. ♕h4 allows Black to seize the d-file with 11... ♖d8 , and as the game between Vitiugov and Gao Rui in the Russian-China match showed Black is just very comfortable. Generally speaking, you shouldn't be worried if you're Black here because White's setup looks a bit artificial and Black is very safe. ( 11... ♖e8 12. ♖b1 ♘e5 13. ♘xe5 ♗xe5 14.b4 cxb4 15. ♗e3 ♕a6 16. ♕xb4 ♖b8 Uhlmann-Smejkal, Trencianske Teplice 1979 ) 12. ♖e1 ♗d7 13. ♖b1 h5 14. ♘g5♘e5 15. ♗f4 ♗c6 ( 15... ♗f6⁉ ) 16. ♗xe5 ♗xe5 17. ♕c4 ♗e8=( 17... ♖f8⁈ 18. ♗xc6 ♕xc6 19. ♖bd1 ♗f6 20. ♘e4 ♖ac8 21.♖d5 b6 22. ♖ed1+/= Vitiugov-Gao Rui, CHN-tch 2012 ) 11... h6! A nifty little trick I want to share with you. 11... ♗f5⁈ 12. ♘e4 ♖fc8 ( 12... ♘a5 13. ♕xc5 ♗xe4 14. ♕xb6axb6 15. ♗xe4 ♘b3 16. ♖b1+/= ) 13. ♕xc5 ♘d4 14. ♕xb6♘xe2+ 15. ♔h1 axb6 16. ♗e3±

11... ♘e5 12. ♕h4 h6 was the kind of position I'd tried to equalise for many years. 13. ♘f3 ( 13. ♘e4 g5 ) 13... g5 14. ♕e4 Black has to prove that the weakening on the queenside doesn't really matter in this position. Perhaps it doesn't, and I think I had some analysis that showed that Black can survive with precise play, but then I realised that none of this is absolutely necessary. 14... f5 ( 14... ♘xf3+ 15. ♗xf3 ) 15.♕a4 is playable for Black, but 11...h6 is much easier. 12. ♘e4 It seems like the c5-pawn is gone, but now... 12... ♗e6 13. ♕xc5 f5! and it turns out White is absolutely forced to play 14. ♕xb6 axb6 15. ♘d2 and Black just wins the pawn back straightaway. 15... ♖xa2 At the very least Black is no worse. 16. ♖xa2 ♗xa2 17. e4 e6= So I think we've covered all the move orders White can aim for after 3.Nc3 d5 4.cxd5 Nxd5 5.g3 and I think this should allow you to play this position with a certain degree of confidence.

I hope you’ve enjoyed my series of videos and eBook on the Grünfeld Defence. I’ve put a lot of effort into this project, trying to make it as informative and entertaining as possible. Clearly the things I’ve shown will be put to the test and over the course of time some of them will perhaps be refuted or cast into some doubt. I want to ensure you, though, that we checked all of these things very carefully, and at the time of production I stand by every single line I’m giving. Once again, I hope you’ve enjoyed this presentation. This has been Peter Svidler for chess24.