Positional Decision Making in Chess - Boris Gelfand - Quality Chess - 2015

Positional Decision Making in Chess by

Boris Gelfand with invaluable help from Jacob Aagaard

Quality Chess www.qualitychess.co. uk

First edition 20 1 5 by Quality Chess UK Ltd Copyright © 20 1 5 Boris Gelfand

POSITIONAL DECISION MAKING IN CHESS All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, electrostatic, magnetic tape, photocopying, recording or otherwise, without prior permission of the publisher. Paperback ISBN 978- 1 -78483-005-2 Hardcover ISBN 978- 1 -78483-006-9 All sales or enquiries should be directed to Quality Chess UK Ltd, Suite 247, Central Chambers, 1 1 Bothwell Street, Glasgow G2 6LY, United Kingdom Phone +44 1 4 1 204 2073 e-mail: [email protected]. uk website: www. qualitychess.co.uk Distributed in North America by National Book Network Distributed in Rest of the World by Quality Chess UK Ltd through Sunrise Handicrafts, ul. Poligonowa 35A, 20-8 1 7 Lublin, Poland Typeset by Jacob Aagaard Proofreading by John Shaw & Andrew Greet Edited by Colin McNab Official FIDE photos by Anastasia Karlovich pages 58, 59, 90, 1 98, 265, 269 Photo page 9 by Nikolaos N tirlis Photos pages 1 1 and 1 1 2 by Valery Levitin Photo page 93 from the Gelfand family album Photo page 9 5 by Shulamit Bushinsky Photo pages 1 35 and 209 by Ariel Van Straten Drawing page 1 34 by Avital Gelfand Cover design by adamsondesign.com Printed in Estonia by Tallinna Raamatutriikikoja LLC

Contents Key to Symbols used Preface Foreword by Jacob Aagaard

1

2

3 4

5

Playing in the Style of Akiba Rubinstein The Squeeze Space Advantage Transformation of Pawn Structures Transformation of Advantages I Cannot Leave My House! Interview with Jacob Aagaard-2012

Name Index Game Index Opening Index

4

5

9 11

59

95 135

209 265

278 282

285

Key to symbols used �

White is slightly better

;

Black is slightly better

± +

White is better

+-+

Black is better White has a decisive advantage Black has a decisive advantage equality

iii

with compensation

�

with counterplay

m

unclear

? ??

a weak move a blunder a good move

!! !? ?!

an excellent move

#

mate

a move worth considering a move of doubtful value

Preface My father was not a chess player himself, but he was sure it was a good idea to teach me chess. Not only because the game was held in high regard in the Soviet Union, but also because he was sure that it develops various skills that could be very useful. Today we have a lot of research that shows that this is indeed the case. He bought Averbakh and Beilin's journey to the Chess Kingdom, which we went through together; a chapter a day. A lot of the other players from my generation, Khalifman, Grischuk, lvanchuk and so on, also read this book early on in their careers. l\scp6ax

UiCR.�Hrr

nyrem CTBHe B max taTHOC KOpon BC fBO

One day when he came home from work, I showed no interest. He was quite puzzled and disappointed by my behaviour. He simply could not understand this shift. Eventually he found out that while he had been at work I had read the book all the way to the end. When I turned six we went on vacation to Crimea, at a lovely seaside resort. At the beach a lot of people used to play chess. I joined in and played a lot of blitz throughout the vacation. Although they were quite experienced players, I still managed to win a few games. One of the players was a friend of my father. He strongly recommended that I should be sent to work with an experienced trainer. This is how my chess career began. My father continued to follow my results and root for me until the end of his life. He kept scrapbooks and built up an extensive library throughout my career.

Position al Decision Making in Chess

6

My father also bought chess books for me. Back then, chess books in the Soviet Union were printed in 50- 1 00,000 copies. Even so, it was not enough to meet the demand. For example, Chess Informant was then printed in 30,000 copies. My father was surprised it was printed in such small numbers. To get my hands on the newest issues I would go to a bookshop that was located just around the corner from where we lived. There I could at times get my hands on books from Eastern Europe, East Germany, Hungary and Yugoslavia. I would work out more of less when the newest issue of Chess Informant would arrive and I knew at what time of the day new books were delivered to the shops. I would then hang around for hours, hoping that this would be the day the new issue would arrive. At that time they had an agreement with a few local chess clubs that they would get a copy for the half a dozen National Masters we had at that time. Otherwise there were only about five copies left for the rest of Minsk! After some years they knew me in the shop and would set aside the latest issue of Chess Informant for me, but only after I had shown my dedication! When I was nine I began a long match with my father, planned to be 30 games. At times he would blunder and I would tell him "Daddy, you are blundering" and would suggest another move he could make. Eventually the match was abandoned after 1 4 games. The following game is of course not a masterpiece seen through the eyes of a grandmaster, but it is still interesting to what extent you can see the style of the grandmaster in the choices of the nine-year-old boy. The two moments I would like you to pay special attention to are moves 1 7 and 24. On move 17 very strong is 1 7.g4!?, preparing 1 8.ll'lfS with a strong attack. But my approach was already then to squeeze my opponent. Probably I would play the same today, only now I would feel less proud about it! On move 24 I exchange the knight on e5 in order to put pressure on the weak d6-pawn. Shortly after this, my father found it hard to withstand the pressure and offered further concessions.

Boris Gelfand -Abram Gelfand Minsk (I O) 1975

l .e4 e5 2.ll'lf3 d6 3.d4 exd4 4.ll'lxd4 �f6 5.tLJc3 !i.e7 6./ic4 0-0 7.0-0 �c6 8 ..ie3 � e5 9 .ib3 h6 10.£4 tLJg6 1 1 .h3 i.d7 1 2.a4 a6 13.�6 c6 14.�g3 i>h7 15.a5 �c7 16.�fl �fab8 •

8

7

6 5

4 3

2 1

a

b

c

d

e

f

g

h

Preface

7

17.� de2 ©g8 18.J.b6 �c8 1 9.�g3 ©h7 20.� d4 �e8 2 1 .©hl J.ds 22.f'5 � e5 23.hds �xd8 8

� ;

7

6 5

4 3

2

1

a

b

c

d

e

f

g

h

24.�f3 �xf3 25J;xf3 �c7 26.gd3 �e8 27.�h4 c5 28.�d5 �d8 29.�g3 gcs 30.� b6 gc6 3I ..id5 gc7 32.gel i.b5 33.c4 i.c6 34 ..ixc6 bxc6 35.e5 �g5 36.�xg5 hxg5 37.exd6 gb7 38.d7 �f6 39.d8=� gxd8 40.gxdS 1-0 Although I was only a small child, the style of play is not so different from today. I showed the game to Vladimir Kramnik and he felt the same way. And when he showed me a game of his from his early childhood, I could also recognize his trademark style in that game.

Hard-to-get books The two most difficult books to get your hands on at that time were My Sixty Memorable Games by Bobby Fischer and 50 Selected Games by Bent Larsen. My father did incredible things in order to acquire these books. First he identified people who had these books. Then he harassed them endlessly to find out what they wanted for them. He managed to exchange his way to 50 Selected Games by providing the owner with a detective story he really wanted. And for My Sixty Memorable Games he paid his way - 25 Roubles. This may not sound of a lot today, but at the time his meagre salary was something like 1 20 Roubles a month. 20% of his monthly income for a chess book!

Rubinstein - by Razuvaev and Murakhveri This book could not have been written without the work of one of the best chess writers, Yuri Razuvaev, and his book on Rubinstein's games, co-authored with Murakhveri. I will write more about this in the next chapter. Jacob and I had a number of conversations about what we wanted to get from our collaboration on this project. I would think like a chess player, eager to show my best games. Jacob would think

8

Positional Decision Making in Chess

like a writer, eager to teach and entertain the reader. As he is leading the pen, his vision will no doubt be the stronger, which is not necessarily a bad thing. There was one thing chat we agreed would make this book more interesting - to reflect on the influence Rubinstein's games have had on my own games. I am not sure the connection will always be as obvious co the reader as it is to me, but I have cried to explain it in the best way possible.

Rubinstein everywhere I am not one of those people who feel a need to pretend chat I am a genius or chat I invented everything on my own. I am proud of my accomplishments and I am proud of my games. The fact chat I had great help and support from a lot of people, such as my father, my trainers Albert Kapengut and Aleksandr Huzman, and most of all from my beautiful wife Maya, is not something I think diminishes my accomplishments; rather I hope chat the fact that these people dear to me have supported me for decades cells something positive about my value as a human being. It is from this viewpoint I write this book. I am past the youthful need to prove co others chat I understand chess, and ready to explain how chess is understood instead. So we have focused on how I chink about chess, in the hope chat it will help aspiring young players to improve, as well as bring joy to chess fans who simply want to enjoy the game more. And the cornerstone of my thought process was laid down by my role model, Akiba Rubinstein. Obviously I always tried to learn from all the great masters, but it was Rubinstein who made the biggest impact on me, closely followed by the games and writing of Efim Geller and Lev Polugaevsky. For this reason we will return to Rubinstein's games - as well as other games that inspired me throughout this book to explain what the reasoning was for my choices when taking the critical decisions. And from there comes the title - Decision Making in Chess.

Judge us on this! Unlike my first book, My Most Memorable Games (Olms 2005) , the intention of this book is not to focus on the accuracy of the moves I made at the board, as shown by subsequent analysis, but on the thought process chat led to me finding chem in the first place. Obviously we have analysed the games and in the process discovered things that made me re-evaluate some games entirely; bur throughout we have focused on the reasons for the decisions and plans I made, and also the limitations of my chinking during the game. I hope the reader will find this approach interesting and that it will help you with your own game. Boris Gelfand Rishon Le Zion, April 20 1 5

Foreword by Jacob Aagaard I am the ghost writer for this book, though the word writer does not fully explain what I have been doing. I have analysed positions, asked questions, recorded che answers, typed in everything and applied my moderate experience with chess writing to improve the structure and order of what had been said. Although this is all work that a writer does, the most important element is missing. The ideas in this book and che reasoning behind them comes from Boris and not me. If you want to know what I think, please buy my books (and I mean that, please buy my books!). I have dreamed about being involved in a project like this for a very long time. As a grandmaster I understand quite a bit of what is happening in top-level games, but obviously my understanding of the game is not at the same level as a World Championship challenger. I have wanted to be able to use my skills as an experienced writer and trainer to ask the right questions and obtain insights from him that you would not get if he was writing the book himself. Actually, it was especially the "obvious" things chat fascinated me about this process. Whenever something was obvious to Boris, I knew that it might not be obvious co many others; and that his explanation would be very instructive.

7he authors in conversation at the Tromso Olympiad in 2014

10

Positional Decision Making i n Chess

This position is the cleanest example.

Boris Gelfand - Daniel Campora Cesme 2004

Boris quickly and confidently made his next few moves, and would not have spent any time explaining chem had I not asked him why he played as he played. The explanation was short, clean and crisp. It was also incredibly illuminating. See more on page 1 1 6. This has been one of the most interesting projects in my career and I want to thank Boris from the bottom of my heart for agreeing to work with me on it. Jacob Aagaard Glasgow, April 20 1 5

Chapter 1 Playing in the Style of Alc6 43 .f4 gxf4 44.gxf4 g6 45.h4 c4 46.a4 d4 47.@e2 d3t 48.@d2 @c7 a

b

c

d

e

g

h

White has made obvious advances, but so has Black. But once again Takacs fails to see the necessiry of changing the course of the game and plays an ugly, passive move. After this his position never truly recovers.

27 ... �dS? Having prepared it, it was necessary to play 27 .. .f4!, when I am unsure about how White should play. One option is 28.exf4 tlJ xd4 29.:9:xe8 @xe8

8 7 6 5 4 3 2 a

h

Despite being a pawn down, Black holds this pawn ending.

Chapter 1

-

This all looks quite hairy for Black and we should not forget Larsen's mantra, Long variation - Wrong variation. It is quite certain that there are improvements for both sides along the way. But what I wanted to illustrate with this long variation was the path the game could have taken, had Black attempted to defend actively. Compare this with how the game developed and you will see that although it looks scary, it was the only sensible way to play the position.

8 7 6 5 4 3 2 1

28.�cS �xc5 29 ..ixcS hc5 30.bxcS

a

b

c

d

e

f

g

h

White has achieved all he can on the queenside. Standard theory on endgame technique tells us that you need to provoke a second weakness. In this case it is logical to do so on the kingside. And this is indeed where Rubinstein turned his attention.

8 7 6 5 4 3 2

34. ©a ile4 3S.g4 g6 36.ilgI � f7

a

b

c

d

e

f

g

h

Rubinstein has achieved what he was looking for all along. White has permanent pressure against b7 and if Black ever manages to play ... b7-b6, the c6-pawn will be weak and there will be an excellent square on c5 available for the white knight. It is possible that Takacs went into this position thinking that he could hold it, as b7 is not that difficult to defend. But then he did not have the advantage of studying Rubinstein's collected games! It was only years later that they were available in book form. 30 ... ©e7 3 1 .:Sb2 ©d7 32.ileb l ©c8 33.©e2

i:!e7

47

Playin g in the Style of Akiba Rubinstein

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

37.h4!? A very concrete move that Rubinstein no doubt calculated deeply. Still, I am not entirely convinced it fits in with "best practice". There was no urgency in the white position, so I think he could have manoeuvred his pieces into better positions before turning to concrete measures. It is possible that Rubinstein was concerned about the consequences of . . . l:%e4-e8-h8, when

Positional Decision Makin g in Chess

48

Black has in turn improved his defence a little, and chat this is why he hurried here.

37 ... gxh4 37 .. .f4 was possible, but White has a big advantage after 38 .E:h l ! and the breakthrough cannot be stopped. 38.gx5 gx5 39J�g7 � d8 40J:�g8 f4? White was much better in any case, but this allows him co win the game in one move.

47.�b3! a5 48.�cl ga8 49.� d3 b5 50.cxb6t ©xb6 5 l .�c5 E!d6 52.a4 gc8

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

53.©g4 ga8 54.©5 ©c7 55J�h7t 1-0

a

b

c

d

e

f

g

h

41 J�h8 4 l . llJ c4! would have won inscancly. 4l. .. fxe3 42.fxe3 ©d7 Ironically 42 . . . E:e7 does not work on account of 43.e4!! with the idea 43 . . .E:xe4 44. llJ c4 and White wins. 43.Etg2 43.lDc4! and again Black can resign. Bue there is nothing objectively wrong with winning slowly, as White did in the game. Black has no councerplay and his pieces are entirely dominated. 43.. J�e8 44.Etxh4 Ete7 45.Eth8 ©c7 46.l:�gg8 gd7 46 . . tll f7 47J'k8t ©d7 48.E:hg8 and White keeps control. .

I see myself as a well-educated player and am always happy when I am able co play games chat show this. The following game is quite interesting from this perspective. In it I managed co play the entire game based not only on one idea, but on the same idea as the above game. The b7/c6-pawns are fixed and vulnerable and I managed co gee a knight co a5.

Boris Gelfand - Alexander Morozevich Astana 2001

Alexander Morozevich is a highly creative grandmaster from Moscow. He has always gone his own way and cried co reinvent the game of chess, which at times is very impressive, but at other times has been a liability for him. He played in the World Championship tournaments in 2005 and 2007 and peaked in the world rankings in 2008 where he was placed 2nd. He has not done as well in recent years, but is still often found either j ust inside or just outside the cop ten. le is well known chat Morozevich played a lot of training games with

Chapter 1

-

Playin g in the Style of Akiba Rubinstein

friends and trainers, in person or online. In 200 1 when this game was played, Morozevich had just emerged in the World elite and this was how he did it. He played rare openings, often provocative and dubious-looking. But he had analysed them deeply and as said, gathered a lot of practical experience in training games. Probably people remember his 1 1 . . .g5 in the Slav, which is now the main line. He played the Chigorin and in the French he popularized some sharp variations with . . . gxf6 in the Burn Variation. His contribution to opening theory in that period was huge and is still felt today, especially through the style of preparation, involving very risky and concrete play, but also creating big problems for the opponent to solve, if he wants to try to refute it. The opponent is forced to think independently and play very energetically from the beginning, as in this game.

1.d4 d5 2.c4 c6 3.tlic3 tlif6 4.tli f3 a6 Again the Chebanenko Variation. At the time this game was played, this variation was quite novel and the ideas of it not yet fully mapped out. At first it was developed by especially Bologan, but later on both Kasparov and Topalov played it for a while, adding a lot of new ideas to the theory. Comparing with this game it makes sense to illustrate Black's idea based on 4 . . . .if5, which is met by: 5.cxd5 ! (5 .Wfb3 Wfb6 is known to cause Black fewer problems; but in the game the a-pawn has moved and the queen would no longer be defended.) 5 . . . cxd5 6.Wfb3

49

Black cannot defend both the b-pawn and cl-pawn with any of the moves that he wants to make. Retreating the bishop is unpleasant, but probably still best. A young Egyptian GM avoided this retreat when playing against me. After 6 . . . Wfb6 7.lll xd5 lll xd5 8.Wfxd5 e6 9.Wfb3 i.b4t 1 0.i.d2 lll c6 1 1 .e3 E:c8 1 2.a3 .ixd2t 1 3 .lll x d2 Black had hardly any compensation for the pawn in Gelfand - Adly, Dresden (ol) 2008 .

5.a4 .if5

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

In the Malakhov game we saw 5 . . . e6. Compared to this, 5 . . . i.f5 is the natural move. In the Slav Defence it is always nice if you can develop the bishop and play . . . e6. The drawback is that it weakens the b7-pawn, which obliges White to play Wfb3 immediately. If Black has time to play . . . e6, he would be able to defend the pawn with . . . Wfc7.

6.�b3 !la7 This might seem incredibly odd, but actually it is quite a common idea in the Chebanenko and one of Morozevich's ideas at the time. He won a good game against Anand in the Dortmund tournament where Anand collapsed, losing four games and winning none.

8 7 6 5 4 3 2 a

b

c

d

e

f

g

h

50

Positional Decision Making in Chess

Anand - Morozevich, Dortmund 200 1 l .d4 d5 2.c4 c6 3 .tll 8 tt:l f6 4.e3 a6 5 .id3 !g4 6.�b3 !x8 7 .gxf3 E:a7

a a

b

c

d

e

f

g

h

8.tt:lc3 e6 9 .Wfc2 !e7 1 0.!d2 0-0 1 1 .0-0-0 tt:l bd7 1 2.cxd5 cxd5 1 3 .e4 'itih8 1 4.e5 tll h5 1 5 .!xh7 b 5 1 6.'itib l E:c7 1 7.Wfd3 !g5

b

c

d

e

f

g

h

The rook is of course badly placed on a7, but Black is banking on this being a temporary inconvenience. If he gets time to play . . . e6 and . . . tt:l bd7, he would be able to develop in a carefree way. Later on White would not be able to prevent the rook from coming back into the game, as we can see from the Anand game. Clearly it is a provocation and it forces White to play very energetically, to go forward and do something. If Black were given time to finish setting up his structure, there would be nothing wrong with his position. He would have no weaknesses and be able to play . . . tt:l e4 or . . . tll h5 with good play. It would be hard to suggest anything sensible for White to do to put pressure on Black.

7.a5 This has a simple point. I want to play Wfb6, which can only be prevented with the concession 7 . . . dxc4. My evaluation of this move was that it would give me an advantage in the long term and even though I knew he had prepared something, I was very optimistic about my chances.

7 ... e6 s.Y;Yb6 Y;Yxb6 9.axb6 gas 1 0.c5 � bd7

Chapter 1

-

Playing in the Style

51

of Akiba Rubinstein

anything wrong with 1 Lif4 either. Maybe it is also a good move. But I wanted to keep things under control, as whenever it gets out of control, Black will have achieved the game he wanted.

1 1 ...J.e7 White also does not need to be afraid of: 1 1 . . .eS 1 2.b4 exd4 1 3 .tll xd4 ig6 1 4.ib2

a

b

c

d

e

f

g

h

This was clearly still part of my opponent's preparation. I was out of book after five moves, while my opponent was still playing really fast.

1 1 .e3 This is the pawn structure I was aiming for. l

1 .if4 is also possible and it is tempting to put the bishop on the other side of the pawn chain, but if I get a knight on a5 and develop the kingside I will have a serious advantage. Therefore it is not so important if my bishop is developed or not; so I quickly get my other pieces out and castle to safety. Getting the knight to a5 is so important that everything else dims in comparison. Another point is of course that Black is not intending to allow me to get everything as I want it, without offering some resistance. We should consider seriously how he intends to deal with this simple plan. Once we do so, it becomes apparent that he is planning on sacrificing a piece at some point for 2-3 pawns and activity. If you know your opponent is planning to sacrifice a piece against you, it makes sense to get your pieces into the game and keep your position compact. These were my reasons and they are all reasonable. But I have to say that I cannot see

a

b

c

d

e

f

g

White enjoys a nice advantage.

h

1 2.J.e2 I could also play 1 2.tll d2, but the idea was to develop before heading for a5 and this is what I did. 1 2 . . . 0-0 I assume that one of his ideas was to exchange the bishop for a knight to prevent it from coming to a5. But it was never really possible. 1 2 . . .ig4 does not work at this point: 1 3.tll d2 ixe2 1 4 .'itixe2 lll xc5 1 5 .dxcS ixc5

52

Positional Decision Making in Chess

White has 1 6.lll a4!, when Black does not have a good way to win the third pawn.

8 7 6 5 4 3 2 a

b

c

d

e

f

g

h

13.�d2! The knight is headed for a5, as planned. White's play in this game is all about timing. If he castled at this point, 1 3.0-0, Black would have enough time to play 1 3 . . . ig4, when after 1 4 .lll d2 ixe2 1 5 .lll xe2 the knight is poorly placed on e2. Black plays 1 5 . . . lll x c5 1 6.dxc5 ixc5. White is potentially still better here, but Black has managed to get three pawns for his piece and has good practical chances. If White wastes time and does nothing, Black will still be able to improve his position.

1 4.lll b3 This looks like it wins a pawn, but Black has an avalanche of tricks. 1 4 . . . ic2 1 5 .lll xa5

a

b

c

b

c

d

e

d

e

f

g

f

g

h

1 5 . . . lll xb6! If White takes the knight, Black gets . . . ib4 in with an advantage. l 6.id2 lll c4! 1 7 .ixc4 dxc4 1 8.0-0 id3 1 9.�fc l The position still looks difficult for Black, but if we dig deeper, we can see that it is actually complicated. White should be careful not to fall for some crafty tricks like: 1 9 . . . �fb8! 20.lll a4?! lll d5 2 1 .lll xc4? ixc4 22.�xc4

Another point concerning this variation is that Black has managed to change the course of the game. I was very happy to have the b7-pawn as a target, so why should I allow my opponent to escape this path so easily? This is one of the main things I learned from Rubinstein.

13 . e5 This is how he planned to play, but in my analysis for this book, I found it not that easy to prove an advantage against: 1 3 . . . aS The first moves I checked turned out to be very tricky. .

.

a

h

22 . . . bS! 23.cxb6 lll xb6 White loses a decisive amount of material. So the best way for White would be to prepare lll b3 by castling short. 1 4.0-0! ic2

Chapter 1

-

Black has to prevent the knight from coming to b3, as Black will no longer have the . . . lll xb6 and . . . ib4 trick, because the rook on a 1 is no longer hanging. Imagine that the first moves were 1 4 . . . e5, then 1 5 .lDb3 .ic2 to provoke this situation, White has 1 6.lll x a5 lll xb6 1 7.cxb6 i.b4 1 8 .lll xc6 and White keeps his advantage.

a

b

c

d

e

f

g

b

c

d

e

f

g

Black is suffering. In the long run he cannot defend the a-pawn. At this point we are of course speculating about what would happen in a game, but we can add a few moves to show a possible course it could take: 1 9 . . . lll f8 20.id2 lll e6 2 1 .1'%da l lll e4 22.lll xe4 dxe4 23 .1'%xa5 1'%xa5 24.1'%xa5 1'%d8 25 .1'%a7

h

1 5 .1'%e l ! I am not sure i f this is too subtle, as this move does lose a tempo when Black takes the bishop on d 1 on the next move. The key idea is that after 1 5 .id 1 id3 1 6.1'%e l lll e4! Black is managing to make things a little murky. White is probably still better, but as said, I want to keep things under control. 15 ... e5 1 6.i.d l i.xd l 1 6 . . . .id3 1 7.lll b3 is entirely in White's favour. 17.1'%xd l 1'%a6 1 8 .lll b3 1'%fa8 1 9.1'%a4

a

53

Playin g in the Style of Akiba Rubinstein

h

So the final conclusion is that after 13 . . . a5 Black will not be able to hold the pawn.

14.0-0

gfes

8 7 6 5 4 3 2 a

1 5.�b3

b

c

d

e

f

g

h

Positional Decision Makin g in Chess

54

One of Black's tricks is that 1 5.b4? loses to 1 5 . . . exd4 1 6.exd4 tll x b6! and Black wins a pawn. 1 5 .b4 would of course be the dream way for White to play, bur the tactics do not work.

1 5 ...iJ8 16.i.d2? It was only when I had the time to go really deep chat I realized that this natural-looking move might be superfluous. I do not think chat it entirely spoils the advantage, bur as can be seen in the notes to the next move, the bishop is actually better placed on c l if Black defends optimally. 1 6.tll a5! This is therefore best. 1 6 . . . gab8 1 7.b4 ge6 17 . . . ti:Jxb6 does not fully work here. White should reply 1 8.dxe5! gxe5 1 9.cxb6 ixb4 20.id2, when Black does not have sufficient compensation. An important point is that after 20 . . . d4? White wins with 2 1 .ti:Je4! or 2 1 .ti:J b l !.

a

b

c

d

e

f

g

h

1 8 .f3! This is the key move. Rather than crying to prove the advantage immediately, White should improve his position as much as possible. At the same time the rook will be passive on b8 and Black will struggle to find squares for all his pieces (the concept of Space Advantage is crucial to understanding Rubinstein's games and we shall have a look at this concept in Chapter 3).

1 8.ti:Jxb7 gxb7 1 9.ixa6 gb8 20.ti:Je2 might look attractive at first, but it is important for White to keep control.

8 �� � �· '.. , 7 � • , , , � ., . �>?.''�"' • �� .1. ,,/J,, J� 6 �r � • �,,• 5 - -%� rw61 � ��� ,,%� 4 �if�r� if��� • ",,�% %w� ,% 3 � � , ,.%w-�� � ": �w-'0 � �7�� � �r,��J 2 -�� � � � � �� -

%_

,

a

b

c

d

e

f

g

h

Here Black can change the course of the game with: 20 . . . ti:J xb6! 2 1 .cxb6 gxb6 22.ic8 id3 23.ixe6 fxe6 24.dxe5 ixe2 2 5 .gel ia6 26.exf6 ixb4 Black has decent compensation. 1 8 . . . h5 1 8 . . . ti:Jxb6 is even worse at this point. Black has to respect White's main threat of l 9.g4! when White wins after both l 9 . . . i.g6 20.dxe5 gxe5 2 1 .f4 and 1 9 . . . i.c2 20.dxe5 gxe5 2 l .ga2! and White wins a piece without any real compensation. 1 9.gd l White is preparing his position slowly. At some point he will strike on b7, a6 or c6, giving Black big problems. Although nothing immediate exists here, it is hard for me to believe chat Black would hold this position in a practical game; which is what counts in the end. Again, it is not easy for Black to change the course of the game.

16 ...i.c2? Morozevich becomes impatient, but the tactics do not work out well for him. The attempt to create counterplay down the e-file does not work. After 1 6 . . . ge? 1 7.ti:Ja5 exd4 1 8.exd4 gae8 White could j ust play 1 9 .i.f3, but more importantly he can

Chapter 1

-

change the nature of his advantage with the tactical strike 1 9.lll x b?! �xe2 20.lll xe2 �xe2 2 l .ic3 . Black has two minor pieces for a rook, but they have no mobility. White will exchange the black rook and win quickly - or slowly. It does not matter; the result is still 1-0. It was only while preparing the material for this book that I realized that the best defensive try for Black is connected with anticipating lll b3-a5 with: 1 6 . . . �ab8! The idea is to play ... h5, ... g6 and ... ih6 to activate the passive f8-bishop. White must still be careful that Black does not manage to sacrifice a piece under good circumstances. l 7.f3!? I believe that this is the most attractive idea. l 7.lll a5 exd4 l 8.exd4 lll x c5 l 9.dxc5 d4 would lead to deep complications. Maybe White is a bit better somewhere, but I do not think this is a sensible way for White to play.

� K �*

•

s

7 � i �'i)� -� i �W" �� �%i"';,?·----% 6 ·� . , % · - •� � 5 :•1; . , .,,%� ?.... . %� �� � ,. � � �% ,,,_}'. . . \,,,,,., �� 4

ttJID � � 3 � ��:r�� - - -· 'w.�� r=----f--- -�

2

a

b

%

c

d

e

f

g

55

Playin g in the Style of Akiba Rubinstein

h

17 . . . ic2!? Black probably has to provoke this concrete action. This is his one chance to give up a piece for three pawns. He should not have been given the chance, but sometimes we make mistakes. We can only reduce the amount of mistakes we make, not eradicate them entirely.

I like White's position after 1 7 . . . g6 1 8 .lll a5 h5 1 9.b4 ih6 20.i.c l ! and I fail to see how Black is going to be able to improve his position. It is likely that White will gradually improve his position on the kingside, while all the time considering sacrifices on a6, b7 or c6. l 8.lll a5 exd4 l 9.exd4 lll xc5 ! This has to be played before White plays b2-b4 and locks down the queenside in a favourable structure forever. 20.dxc5 i.xc5t 20 . . . d4 does not work out well. For example: 2 l .lll e4 lll xe4 22.fxe4 d3 23.if3 ixc5t 24.'kt>h l ixb6 25 .i.f4 and Black is facing unpleasant questions. 2 1 .'kt> h l ixb6 22.�fc l ig6 23 .ifl

s

7

6 5

4

•� � K � •{1, �--J �� ��f% �� �%lJ�£ �! ·� ·. . .. :.•-*-� �, . -�. d . , . . �. �. �� :

,

,

;

�m �m �m 3 � �m�m-�" !'% �� �� ��-----'��z�

2

� - - %���� r�� a

b

c

d

e

f

g

h

Black has three pawns for the piece, but White has a nice structure. He will play b2-b4 at some point and clamp down the black queenside. Still, winning this endgame would take a long time. White would have to eliminate all the black pawns on the queenside and then break through slowly on the kingside. Even so, I like White's position. It might not be easy to win, but it is favourable and he can play on forever. Black's task seems pretty depressing to me.

17.lliaS exd4 It is too late for Black to play passively. After 1 7 . . . �ab8 maybe the simplest is 1 8.�fc l if5

56

Positional Decision Makin g in Chess

l 9.�xa6!? bxa6 20.lll xc6 Ei:bc8 2 1 .lll a5 and the two passed pawns are clearly better than Black's extra piece.

1 8.exd4 tlJxc5 1 9.dxc5 d4

s ,i �

: �L,?. r��,--� � . , ?.� )'0

2 1 . .. tlJe4 White wins in all lines: 2 l . . .Ei:ab8 22.�xf6 gxf6 23. lll x b7! and Black's position collapses. Against 2 l . . . �xc5 White has several options, bur I saw the simple 22.�xf6 gxf6 23.Ei:fc l �xb6 24.lll c4 and White wins a piece. I find the following line quite instructive. Against 2 l . . . �e4 I would play: 22.�xf6 gxf6 23.b4!

�

,%�%.,��f.���� � �- �� �� ��,� �� �· - ?.�ij r,{ � � 5

4

/

,

a

20 ... dxc3 2 1 .i.xc3 It is not clear if Morozevich overlooked something or if he j ust lost patience. Black has not managed to solve the problems with the b7-pawn with his tactical operation. Actually they look more urgent than ever.

� .i �ti

,

� b

c

d

� .l:J� e

f

g

h

White retains the favourable structure. White has no reason to be concerned about doubled pawns in front of the king. We should only care about the things that are truly important. Black's bishop on f8 provides him with no counterplay. White's strategic operations have been entirely successful. We now see that the dark-squared bishop has been bad all of the game and that this has been a big part of Black's problems. In this line White has of course been entirely successful in keeping his best piece, while leaving his opponent with his most impotent piece. Petrosian was known for being great at exchanging the right pieces, bur Rubinstein was not bad at this either.

Chapter 1

-

57

Playin g in the Style of Akiba Rubinstein

is always the chance that something strange could happen. But in this game it did not.

points out a nice trick: 3 1 J'fa5! l::1 b 6 32.l::1 g 5! and everything must go.

23... gabS 24.J.xe4 gxe4

29.ga7 .ie6 30.'Bxc6 i.dS 3 1 .'Bc7 'Bes 32.h3 gf4 33.�d6 ge2 34.6

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

25.f:ldl 25.tll d6 is also good, but the active idea in the game looks nice. 25 ...i.e2 26.gd7 .ig4 After 26 . . . l::1 e7 I had planned 27.l::1 xe7 ixe7 28.l::1 e l ! with the point 28 . . . l::1 x b7 29.l::1 xe2 ixc5 30.l::1 e 8t if8 3 I .ib4 and White wins. .

27.'Bc7 'Bc4 28.'Bxa6 hS 28 . . .ixc5 29.tll xc5 l'hc5 30.b? l::1 b 5 is of course a completely winning position for White.

A human would find some slow plan to convert the advantage, but the computer

a

b

c

d

e

f

g

h

34 .. J�xg2t 35. ©xg2 'Bx6 36. ©gl 1-0 It is hard to guess what went wrong with Morozevich's home preparation for this game; obviously only he can tell. I certainly did not have the feeling of playing a surprising or genius move at any time, though I do think I handled the challenges of preventing his counterplay quite well. Morozevich was one of the first to work deeply with computer programs, so maybe at some point he simply believed the evaluation of the computer, which evaluates the position as acceptable for Black even to this day. Probably his training games also went well and he felt confident enough to try it in a big tournament game. One of the myths of Morozevich has always been that he is very creative and plays with a lot of improvisation. Obviously he is very creative at the board, but we should not forget that this creativity in the opening is based on a lot of home analysis. The public have a tendency to not understand that about the conception of

58

Positional Decision Making in Chess

the most innovative ideas; they do not see how much preparation it requires between tournaments to be creative. Luckily this does not in any way diminish his achievements in this area; being creative at home is very difficult as well, as anyone who has ever cried can testify. My own playing style does not usually bring as much enthusiasm from the chess fans as chat of Morozevich, but in the struggle between dynamics and statics, sometimes someone has to cake the more conservative side. I do not personally chink this makes the games dull, though I am aware chat at times the real battle is going on under the surface, making it harder for some people to comprehend. Hopefully the more verbal explanations in this book will help unlock the thought process behind this kind of strategic game for a lot of people. In general I would like to add chat I am a strong believer in the value of a chess education built on thorough knowledge of the classics. Any attempt co emulate the engines and their 2,000,000 moves a second is doomed co fail. We need to supplement calculation with all other weapons available. And one of these is intuition, which is strongly rooted in pattern recognition. When you have "uploaded" a lot of chess patterns to your brain in your childhood, you will often have a very strong suspicion regarding what the right move is in a position, even though you have no idea why. . .

Moscow 2012

Chapter 2 The Squeeze

Moscow 2012

60

Positional Decision Makin g in Chess

A classically trained player will approach each position based on his understanding of its needs. If he has a lead in development, he will most likely try to attack the opponent's king, for example. If he has a material advantage, he will try to prevent all counterplay and perhaps steer the game towards a winning endgame. The possibilities are many. This is hopefully obvious to everyone, though not everyone thinks this way! The reason why so many trainers recommend that we study the best games of such diverse players as Capablanca, Alekhine, Tai, Petrosian and so on, is that they all excelled in playing different types of positions and it should be our goal to learn to play them all as well as possible. Rather more complex is how we approach having an advantage. Many players try to force their will on a position if they determine that they are better. At times such a concrete way of exploiting the initiative is the right strategy and at other times it is not. Chess is a game with a sizable drawing margin and we need our opponents to make mistakes if we are to win. At times this is achieved by putting concrete problems to them, at other times by asking them to find a way to navigate their way out of the situation they are in. It may not be necessary to harass them too much, and we can just focus on improving our own position slowly. Capablanca talked about the death of chess because of excessive openin g theory, which was already quite a dubious statement a hundred years ago. If only I was able to get the positions he got after the opening with White back then! These days you rarely get an advantage out of the opening with White, let alone a significant one. Top professionals bring deep ideas and analysis to the game as Black and in every

top event some games will end in a draw without the black player having made a single independent move in the game. Grischuk said in an interview that after the Candidates matches in Kazan 20 1 1 , most players changed their approach to the opening. After a number of numbing draws in the Queen's Gambit Declined, some matches were decided in blitz. Since then, most players just want to get a game. It seems a bit of an over­ interpretation, but the trend is definitely going in this direction, though I would hasten to say that things were quite different in my matches in that tournament. At times we are able to get a little something out of the opening. Often the positions are quite complicated and it is hard to determine who is better, but there are also times when our position is simply pleasant. Maybe the opponent has accepted doubled pawns, maybe an isolated pawn. It might not be significant enough for us to call it an advantage, but as long as he has to make a few good moves to prove equality, there is a chance we can set him some practical problems. We apply a little pressure, j ust enough to make his life a bit uncomfortable. In time this might lead to a small inaccuracy - and a few more problems to solve. More problems means more chances to make mistakes. All we have to do is to keep the pressure up - to keep squeezing.

Rubinstein squeezing... Akiba had the most wonderful technique. It is quite natural that he enjoyed slowly improving his position and squeezing the life out of his opponents. The following game made a great impression on me when I was a young boy. To this day I cannot help being impressed by the way Black managed to outplay his opponent

Chapter 2

-

slowly and accurately. I think this game was what Nimzowitsch was thinking about when he said that Rubinstein's trademark was the long plan. In the game he gets a good pawn structure, which makes it easy to improve his position.

David Janowski - Akiba Rubinstein Karlsbad 1907

l.e4 e5 2.�f3 c!LJc6 3.�c3 �f6 4 ..ib5 .ib4

5.0-0 0-0 6.d.3 J.xc3 7.bxc3 d6 8.J.g5 V!!e7

61

The Squeeze 9 ... c!LJ d8

1 0�c4 .ie6! White is not allowed to keep the bishop pair. Rubinstein is indeed slowly neutralizing White's advantages. I I ..ib3 Axb3 1 2.axb3 c!LJe6 13 ..ih4 h6 14J�fe l a6 1 5 ..ig3 Played to put e5 under pressure, but it is a bit slow, giving Black a free hand to transfer a knight to f4. l 5 .d4 made more sense, though Black is fine nonetheless.

15 ... �h5 16.d4 c!LJ ef4

-�-�'-,,.,)m1!� �·� ·�� �

8� 7 .... � 6 i - _, �� 5 � � 4 - n'lr• 3 -tl)� �o- - ---��r� � £� % 2 � . . , ,%

� :I � �8 0 � - 'a' 'Jif ,

a

b

c

d

e

f

g

h

This game was played at a time when opening theory looked nothing like it does today. Most of the games opened l .e4 e5 or l .d4 d5. These days it is quite common for Black players to use dynamics as a way to counter the slight pressure White gets from the start. A hundred years ago people tried to equalize by solving their problems early, which is what Rubinstein manages to do in this game.

9.V!!d2 This is a rather harmless move, but you can understand why White played it. The queen goes to a slightly better square and not much is really happening here anyway. 9.�e l is the main move, with a lot of theory.

a

b

-..,

..,,%

%

'

I

.%

c

d

e

f

g

h

17..txf4?! This is quite a serious mistake that soon leads to a slightly worse pawn structure, creating long-term problems for White. The knight on h5 was not dangerous and the bishop on g3 was defending the king. With the bishop gone, White has to weaken his structure in order to kick the knight away from f4. At this point I prefer l 7.°1We3!? with the idea of �ad l , �d2 and �ed 1 .

17 ... c!LJxf4 1 8.g3 Unhappy with the knight on f4, White commits an ever-so-slight concession.

62

Positional Decision Making in Chess

Rubinstein understood that such changes would matter in the long run and knew how to exploit them.

at that time. It is quite possible that this was the original inspiration for the manoeuvre that Kasparov later popularized in the King's Indian with . . . @h8, . . . lll e7-g8-f6-h5.

18 ... lll g6 19.Yfd3 Yfe6 Preparing to open the f-file. 20.lll d2 f5

a

a

b

c

d

e

f

g

h

21 .d5 This decision looks quite reasonable. After the exchange on f5 White gets a great square on e4 for his knight. On the surface it seems that White has a preferable structure. But if we look deeper, as Rubinstein liked to do, we will see that the knight on e4 is pretty, but ineffective. It will eventually be exchanged and White will be left with an inferior structure, where Black is able to apply some pressure down the f-file, which of course has been weakened with 1 8 .g3 . . . 2 1 . ..Yfd7 22.exf5 gxf5 23.lll e4 gaf8 24.ge3 lll e7! Black starts a regrouping of his knight in order to eliminate the knight on e4, as previously indicated. It takes a few moves, but there is little White can do in the meantime. Remember that we are talking 1 907 here, more than a hundred years ago. Such long manoeuvres were probably entirely unknown

b

c

d

e

f

g

h

25.gdl @hs 26.c4 lll g8 27.f'3 lilf6 28J:Ul. White can play 28.lll c3, but the knight would lose all of its effectiveness. 28 ... lll xe4 29.Yfxe4?! This is quite a normal-looking move, but as we shall see, it fails to anticipate the problems that are approaching. And this is exactly the point. Predicting the coming moves was very difficult; I would even go as far as to say that without the knowledge of chess history that we have today (which among other things is based on knowing this game!), it is not possible to understand the problems White is starting to face. 29.E:xe4 with the idea of We3 seems better. The position is essentially equal. Black has to pay attention to f3-f4 all the time.

29 ... YfdS!! This does not look like anything special. All that happens is that White will have to be a little careful in order to stay out of trouble. But when you know what is coming, you cannot help being mesmerized by Rubinstein's

Chapter 2

-

clarity of thought. The path the queen takes here still feels revolutionary a century on.

63

The Squeeze

I would personally have considered 35 . . . a5!? 36.Wfc3 Wfc5 to keep the control of the b4square, when I prefer Black's position.

36.Yfc3 aS Black cannot afford to lose the fight for the b4-square. 37.Yfxb4 37.c5!? 37... axb4 38.c3 bxc3 39.gxc3 a

b

c

d

e

f

g

h

30.Yfg4 Played against a perceived threat of . . . Wfg5 . 30.�ee l ! Wfb8 3 1 .Wfe3 was best, and the position is still about equal. But who would play something this artificial?

30 ...YfbS! The queen enters the white position from a surprising angle.

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

White has managed to get rid of his doubled pawns and eliminate the intruding black queen. But he is still a tempo away from full equality. It is quite possible that Janowski believed that he was making a draw here, but things are far from that simple.

31.©g2 Yfa7

39 !iaS! Taking control of the a-file means that White still has problems to solve. With his next move White makes a serious mistake that for the first time takes his problems from being j ust optical to being real. ..•

a

b

c

d

e

f

g

h

32J!Uel YfcS 33.Yfe4 Yfb4 34.g l e2 g5f6 35.Yfd3 ©g8

40.©h3? Activating the king, though it should be said that the king is by no means better placed on

Positional Decision Making in Chess

64

g4 than it was on g2. Meanwhile Black has time to make improvements to his structure. 40.c5! was essential. White would be able to use the open c-file to create counterplay against the b7- and d6-pawns and thus keep the game in balance.

8 7 6 5 4 3 2

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

After the move played in the game White will not be able to rid himself of his weaknesses. A temporary inconvenience has shifted and become a permanent problem. The position should still be held with immaculate play, but this is the point of the squeeze: immaculate play is for computers, and chess players are human beings.

40 b6! Preventing c4-c5 . .••

41 .©g4 4 I .b4!? E:a4 42.E:b2 was possibly better, but what is clear is that White is on the defensive and his pieces are being forced into passivity. White will have to avoid making mistakes for a very long time, and some of the problems he will face can be quite considerable.

b

c

d

e

f

g

h

42Jlb2? White decides that his position is safe and that all he needs to do is to wait. Even if a draw was found through deep analysis at some point over the next few moves, it would not change my view that this is the moment where White lost the game. Over the next few moves Black improves his position and his advantage grows and grows. 42.f4! could be considered. After 42 . . . exf4 43.gxf4 White is probably close to being okay, but it is quite understandable that Janowski hesitated about splitting his pawns in this way. Once again this illustrates the challenges White is facing. There are no truly easy decisions.

42 ©h7 43.b4 ©g6 In a few moves the white king will be in trouble on g4. .•.

44Jlbb3 44.©h3 could be tried, but after 44 . . . E:g l ! the king will not manage to get to safety. Again, it is possible that White can hold the draw, but it would be with the edge of his nails. 44 JH"S 45.gc2 ghl ! Rubinstein i s famous fo r his handling of rook endings and this is no exception. ..

41 .. ,gal

a

Chapter

8 7 6 5 4 3 2 1

2

-

65

The Squeeze

53. @d3 gat 54. @d4 gb3! Blocking the white pawns. 55.gf2t @g6 56.gg2t @h6 57.b5 An unpleasant move to make, but also a forced one. 57.:Sa2 b5! and Black wins.

a

b

c

d

e

f

g

h

46.£4 This brilliantly illustrates how things have gone wrong for White. Only five moves ago he rejected making this committal decision, when he was reasonably active. But now he feels compelled to play it, when he is passive and it loses a pawn by force. This is the nature of the squeeze. When you are under pressure you will usually have to make a concession or an unpleasant decision a number of times, and it is so easy to miss the moment when you have to accept that things are going wrong. You then find yourself in a situation where everything has gone entirely wrong. . . Looking a t his options must have made Janowski desperate. After both 46.�h3 :Sh5t and 46.c5 :Sg5t! followed by 47 ... :Sh5t, Black wins a pawn. However, the enigmatic 46.:Sbb2!? is the suggestion from the engines. Black wins the f3-pawn, but White gets to play c4-c5 and can at least fight a little bit. Black should win, but some difficult decisions await.

46... exf4 47.gxf4 h5t 48.�g3 ggl t 49.�fl gg4 Black wins a pawn.

a

b

c

d

e

f

g

h

57 ... ga Played to keep everything under control. This is something I learned from Rubinstein as well. The important thing is not that you win quickly, but that you win safely. This can mean that you spend many more moves winning a position than the computer or even someone else might, but as long as you win, this does not matter. And the way to ensure that you win is to minimize the number of difficult decisions you will have to make in the conversion. And this is exactly what Rubinstein achieves with this move. 57 . . . g5! is the computer's recommendation. White cannot create real counterplay against c7, as h2 would fall and the pawns promote. But to work this out with certainty would require calculation - and calculation always contains a risk of miscalculation. On the other hand, we should also be careful not to play too artificially in order to avoid calculation. Sometimes it only postpones the

66

Positional Decision Making in Chess

moment when a decision has to be made, but does not guarantee that it will not be more difficult.

58.©e4 �ff6 59.ga2 g5 60.ga7 gf4t 60 . . . �f7 was also possible, with the idea of 6 1 .c5 bxc5 62.b6? �f4t and Black wins. 61.©d3 gf7 62.cS!? Seeking counterplay.

Black has to retreat with 66 . . . �g7 to avoid losing to �c7-c6. But after 67.�c6t f2 Wxa2t 4 1 .'tt> g3 Wxb3, where White's advantage is nothing special.

I like 36.e4! with the idea of improving the position of the knight and king: 36 . . . g6 37.lll e3

a

b

c

d

e

f

g

37.h4? Entirely the wrong path. I wanted to free my position, but instead I am weakening it. 37.lll e l was necessary. Black would have to retreat the rook and White keeps most of his advantage. The idea should still be to put the pawn on e4 and improve the position later on, even though the circumstances are a bit poorer now.

37 ... gxh4 38.lll xh4 The computer mentions 38.lll f4, but of course you don't play h2-h4 with the idea of losing a pawn!

h

White will play 38.'tt> g2 and then be ready for E:dl and general consolidation. The conversion of the advantage will probably take a long time, but this has never bothered me. The difficulty of the task is rarely measured in time. I am sure that such a great player as

a

b

c

d

e

f

g

h

38 ... gS? At this point Kasparov misses a very strong move. 38 . . . Wd8! with the idea of . . . E:d2 was

Positional Decision Making in Chess

90

correct. I do not believe I would have allowed such an intrusion, but rather I would have played 39.:gc2. Here Black can take on e3 followed by . . . Wd l t with an inferior, but holdable ending. But strongest is: 39 . . . :gd I t 40.\t>g2 Wg5!

!���!I .,.�-- �� �.� � � �� ���� : �. . . �� :� ��� :��� >,. . !-.la-- - �!� ��r� - - %� �

6 3

2

a

b

c

d

e

f

g

h

White is two pawns up, but cannot easily free himself. Black is of course contemplating sacrificing the bishop on g4 at the right moment, but there are also ideas such as . . . g6 followed by . . . Wh6, when . . . g5 is suddenly threatened. In typical style, the engine thinks the position is "just a draw" - 0.00. The human experience is of course entirely different! If this had happened during the game, I would have thought that all three results were possible. This reminds me of a game I was following live from last year's Candidates tournament. Peter Svidler - Vladimir Kramnik, Khanty-Mansiysk 20 1 4

a

b

c

d

e

f

g

h

4 I .Wc6 Several commentators indicated that this was a mistake, something no one would have been able to work out on their own. Svidler's idea behind this move is nothing short of amazing, only Black has an even more amazing resource to counter it. The engine says that White has a serious advantage if he takes on b4. Obviously Peter understood this, but cried for more. 4 1 . ..e4 Black is playing only moves all the way, which is probably part of the reason why Kramnik did not fully understand his opponent's ideas in advance. He knew he would have to face them soon enough; there was no reason to waste time wondering about which one he would choose in advance. 42.fXe4 8 6 5

4 3

2

a

b

c

d

e

f

g

h

The live commentary was stating that the game would now "just be drawn". Apparently no one had told the players who both looked fully engaged with the game and full of nervous energy! 42 . . . We2 43.m3 :gxd6 44.We8 f6 45 .e5!

Chapter 2

-

91

The Squeeze

Those who followed this game with the computer running not only seemed to fail to understand that this was anything but obvious to the players during the game, they also did not understand how amazingly both players performed in this game!

a

b

c

d

e

f

g

h

This is what Peter had planned. If Black takes on e5 with the queen, White will force a pawn ending by exchanging queens, taking on f8 and winning back the pawn with ic5xb4. Kramnik had not anticipated the defence available to him at this point, but found it after a long think. He did not believe in it, but could not see any refutation, so he played it, hoping for the best. There is a big difference between looking at a position with an engine and having to find your own confidence. 45 . . . f5 46.gxf5 E1f6!! 8

7

6

5 4 3

2

-� ·

� ��� "'""� �� �� �%'"" ��.l"' · · �· -JM � �.... % 8 � • � �...r. .�% 8 � �.� �

�

� . . ·lr� @� : � �� �� 1-=� �� �� -;,.,�� a

b

c

d

e

f

g

h

This looks entirely random, but surprisingly there is no way for White to make progress. 47.©g3 '.We4 48 . .ic5 '.We i t 49.if2 1Mfe4 50.ic5 1Mi'e l t 5 I .if2 lh-Y2

If you want to truly enjoy and appreciate chess, I would recommend that you follow live games without the engine on, as I do.

a

b

c

d

e

f

g

h

39.�g2 Yfe5 39 . . . 1Mfd8 is not as strong now, as White can play 40.E1e l , the main point being that after 40 . . . l:!d2 4 I .l:!e2 White does not have a knight hanging on h4.

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

Positional Decision Making in Chess

92

40.�c2? As happened to me in another game against Kasparov, I blunder on move 40, throwing away the advantage. I was a little nervous at this point; not because I was playing a strong player or because I had a good position, but because I was short of time (seconds!) . It always affects you a bit. Even Grischuk's heart rate goes up! As Alekhine said, "There are no heroes in time trouble." 40.e4 was correct and after: 40 . . . Wi'd6 4 1 .E:e l E:d2 42.E:e2 E:d l t 43.lD e l

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

41. .. gxe3 4 1 . . .E:d2? 42.Wi'c? would give White an advantage, but Kasparov steers the game safely to a draw. 42.�fl �e5 43.�g2 ge2 44.gel gxf2 45.gxe5 gxa 46.� e3 i.d7 47.gxg5t ©f'8 48.ti:)f5 Lf5 49.gxf5 l!z-1/2

a

b

c

d

e

f

g

h

I would have a considerable advantage. The game goes on and White is not well enough placed for us to call it a winning position. A lot of accurate moves would be needed for White to convert his material advantage. An important difference between this position and the position after 36.e4, is that it will not be difficult for Black to find active ideas here, whereas in the earlier position it was very difficult.

40 ...�d6! Finally Black penetrates to the second rank. 41.�el The only way to prevent the intrusion, but it loses a pawn and all of the advantage.

Although he managed to save the game, Kasparov did not come out of it looking great. And this leads us to an important point, because there is no questioning Kasparov's greatness. It was the position which was very difficult to play. In short, it was more difficult for Black to find sufficient counterplay than it was for White to maintain the pressure. One of the reasons why it is so difficult to play a position where you are squeezed is that there will come a time when you need to switch from passive to active defence. If you do it too early, you can find, like Kasparov, that there are no targets for your active counterplay and that you are j ust wasting your time. Or like Janowski, you can miss the moment when you need to do something in order to not be squeezed entirely.

It is also very difficult to cell the difference between positions chat are only slightly worse and chose chat are much worse - both feel uncomfortable. This is why blunders happen so frequently in worse positions. This is not the only time I failed co win a winning position against Kasparov. His resourcefulness in even hopeless-looking positions was quite amazing. (In a totally crazy game - Dos Hermanas 1 996 - I won a piece and was completely winning, but fell for the last trap on move 40.) It does not mean chat the strategy was not successful - only chat chess is really difficult.

A final note The squeeze is the correct strategy in some positions and in others it is not. A classically trained player like myself will be happy co go for this strategy when the position allows it and it gives the best chance of success. The same goes for a player like Topalov, who is not commonly associated with this type of position, but is no less of a classically trained player than myself The difference is chat there are borderline situations where he and I might choose co go in different directions, because of our different interpretation of chess. Bue we are both going in the direction which we feel will give us the greatest chance for success. And at times this is implementing the squeeze. The goal of this chapter has therefore been to present this strategy as one possible way to play, and something we should be happy to do when the position requires it.

Chapter 3 Space Advantage

Te/Aviv 2012- Returning.from the match

96

Positional Decisio n Makin g in Chess

I definitely learned about space advantages from studying Rubinstein's games, as well as other classical games. le is one of my favourite weapons and we shall meet it often in these books, in one form or another. I am not sure if I can add anything new co the concept of space advantage, but I hope I can give the reader a higher level of understanding of the strategy of playing for and with additional space, looking at it from a practical angle.

What is a space advantage? Having more space optically is not the same as having a space advantage. I chink this is one of che things chat tends co confuse some people. fu; I have always understood it, a space advantage occurs when pawns are taking important squares from the opponent's pieces. I chink the easiest way co illustrate this is by showing an absurd example: No space advantage

If we were co add seven pieces co this position, Black would no longer have a fortress - at lease not if White is co move.

White has a space advantage

a

b

c

d

e

f

g

h

1 .lDxg6! fxg6 2 . .ixg6! hxg6 3.h7 and White wins. The inability of Black co defend the weakness on h7 illustrates co a great extent what space advantage is all about. (Obviously this is an absurd example and if Black were co move, he would save the game with 1 . . lDxh6.) .

Most chess games are a baccle of two able minds of similar ability with exaccly the same resources at hand, so it is of course impossible for us co find such an absurd example in grandmaster practice. But this is only because the advantage has been extended g rossly. Lacer in this chapter, we will see how a top grandmaster who has played Board One for a gold-medal-winning team at the Olympiad managed co end up with Black in this position :

a

b

c

d

e

f

g

h

Chapter 3

-

Space Advantage

97

A theoretical explanation My own understanding of space advantage is to a great extent intuitive. I can recognize it when it is there and I know from the example of Rubinstein and others how to handle it. This is quite a natural state for a practical player, who only has to make good decisions, not explain to others how to replicate such decisions. For this reason we looked at chess literature for good descriptions of the concept of space advantage. I am not surprised that we did not find many excellent descriptions of it, but we did find a few, of which the champion by far was Michael Stean's classic Simple Chess, ori ginally published in 1 978, now available in an algebraic edition. His final chapter is on space and starts with this brilliantly clear explanation: "Unlike the ideas expressed so far in this book, space is not an easily definable or recognizable concept. The visual impression you obtain by glancing at a position and estimating who seems to have the lion's share can be misleading. The following is nearer the truth. Any given Pawn structure has a certain capacity for accommodating pieces efficiently. Exceed this capacity and the pieces get in each other's way, and so reduce their mutual activity. This problem of overpopulation is easy to sense when playing a position - it 'feels' cramped. "To take an example, compare Diagrams A and B . A: White h as a space advantage

8 7 6 5 4 3 2 a

b

c

d

e

f

g

h

B: No space advantage

8 7 6 5 4 3 2 a

b

c

d

e

f

g

h

"They do, of course, represent the same position, but with two pairs of minor pieces less in the second case. "In Diagram A Black is terribly congested. There is no way he is ever going to be allowed to play . . . b5, while alternative methods of seeking some breathing space by (after due preparation) . . . e6 or . . . f5 would compromise his pawn structure considerably. White on the other hand can build up at leisure for an eventual e5, safe in the knowledge that so long as he avoids any piece exchange, his adversary will never be able to free his game. "Diagram B is quite a contrast. The size of Black's forces is here well within his position's

Positional Decision Making in Chess

98

'capacity' . As a result there are no spatial problems at all and Black can very quickly seize the initiative by . . . a6 and . . . b5, or even by . . . b5 as a Pawn sacrifice, e.g. l . ..b5 2.cxb5 a6 3.bxa6 :gxa6 with tremendous pressure. "We see from this pair of positions that Black's structure is very good, but his capacity is small. Visually White has a spatial advantage in both cases, but in the second the eye flatters to deceive. In fact, he is grossly overextended. A vast empire requires an army of equal proportions to defend it." (Pages 1 33- 1 34)

Space advantage and counterplay The difference between Stean's two positions is stunning, though we should not be blind to the fact that he removes White's best knight on c3, instead of the less well placed knight on f3, in order to make his point. Still the main difference is quite clear. With fewer pieces on the board, Black is able to prepare a pawn break and open up the position, thus making his artillery valuable. If Black is struggling with limited space, he will often have to make a sad choice between suffering passively or accepting compromises to his pawn structure in order to free himself.

An extreme - and realistic - example

a

b

c

d

e

f

g

h

In my database there are 739 games where this position occurred. I managed to win a game against lvanchuk with the white pieces in this variation, but it is really too early to say anything definite about the position. Both sides have things to be happy about. Maybe after another 800 games we will know more . . . There i s a n argument fo r White having a space advantage, but if we forget about our knowledge of opening theory, we will probably find it more surprising that there is an argument against this as well. With rooks on e8 and c8, a retreat of the queen to b8/a8 and the bishop to ffi, Black will have found good squares for all of his pieces. White will have to run the risk of overextending himself in the pursuit of an advantage. Black's strategy will be to wait and at the right moment blow the position open with . . . b5 or . . . d5 and in some cases . . . e5. I am by no means dogmatic when it comes to space; it does not always come in the shape of an advantage, but it is one of the important strategic concepts in chess we have to take into account. In order to integrate space into our strategic thinking, we need to understand how a space advantage is exploited, so we can see when it is applicable.

Mobility We have seen that when you have more space, it gives you more possibilities to manoeuvre your forces around. For example: you are able to switch your pieces between the two sides of the board faster than your opponent. This can lead to a situation where you enjoy better lines of communication and have better positions for your pieces. You might be able to attack a weakness more times than it can be defended, simply because you have more space

Chapter 3

-

Space Advantage

to line up your pieces, or because the lack of freedom in your opponent's position makes it hard for him to come to the defence of his weaknesses.

Yusupov on how to use a space advantage In Chess Evolution 2 on page 234, Artur starts Chapter 2 1 with the following words: "The side whose pawns are further advanced than those of his opponent often has an advantage in space. This side then has much better lines of communication, which is an effective advantage in positions in which play can be carried out on both flanks. Space is only one element in the position, and it is often less important than, for example, an open file or pawn weaknesses. Nevertheless there are plenty of positions in which the possession of more space is an advantage. The side which has an advantage in space can position its pieces better and regroup them more easily. The side which has less space often has a problem doing that: it has too few good squares available for too many pieces. "Some guidelines for the side which has the advantage in space: 1) Avoid any unnecessary exchanges of pieces, because such exchanges provide more space for the opposing side's remaining pieces. 2) Carefully prepare to open a file, so as to penetrate with your major pieces. 3) Try to prepare a pawn advance, in order to drive the opposing pieces even further back. 4) If the situation permits, you may try to manoeuvre against two weaknesses. Better lines of communication play an important part in this.

99

5) It is important to think prophylactically and to prevent any attempts by your opponent to free himself." I absolutely agree with these five points, but would like to add a sixth point: 6) A space advantage can be increased by a transition into the endgame. If you have a space advantage and have taken control of the position, you could find that exchanging rooks, and especially the queens, will eliminate the opponent's remammg counterplay. A good illustration of this is Gelfand - Campora, found on page 109. hope you have found this theoretical foundation interesting. I would like to close it with the point with which I started it - to me space advantage is first and foremost something I approach intuitively. This does not mean that it does not exist, or that there is not a theoretical understanding of it. It means that for a practical player, a list of rules and well-defined concepts is not as useful as a good feeling for how to handle such positions, through experience, gathered both from using it and from seeing how other players have used it.

Rubinstein and space advantage If you did not notice it at the time, you can go back and look at some of the Rubinstein games we discussed earlier in this book and see his masterly handling of space advantages. But it would be a crime to write about Rubinstein and not examine at least one of his games through the prism of space. I like the next game for many reasons. I like how Rubinstein plays the opening, especially moves 7 and 9. He then plays for a space advantage in the centre with the pawn trans­ formations from moves 11-15, and thereafter masterly uses it to outplay his opponent.

Positional Decision Making in Chess

100

Akiba Rubinstein Oldrich Duras -

Karlsbad 19 1 1

l.c4 e5 2.�c3 �f6 3.g3 .ib4 4 .ig2 0--0 5.�a ges 6.0-0 � c6 •

8 7 6 5 4 3 2

A clever move with a modern scent. It is quite natural to play a3 and b2-b4, but with b2-b3 the bishop comes into the game faster to quickly apply pressure on the centre. At the same time Black is not offered any targets, such as a pawn on b4, for example.

9 d6 I O.i.b2 �xd5 It is hard to see how Black would possibly be able to allow White to keep a strong knight stationed on d5 in perpetuity. Though this does not mean that the exchange on d5 is perfectly timed. Rubinstein now sets out to create a favourable pawn structure. •.•

l l .cxd5 � e7 1 2.e4 c5 1 2 . . . c6 was slightly more accurate, limiting White's options - not that it has any great influence on this game.

7.�d5 I think Rubinstein was first with this idea. He wants to avoid the exchange on c3 under unfavourable circumstances. Obviously Black could have exchanged earlier, but Rubinstein judged, correctly I think, that it matters more here. These days the tlJd5-jump is played in roughly half of the games. 7...JJ8 8.d3 h6 9.b3!?

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

1 3.dxc6 This operation is typical of Rubinstein. After the transformation of the pawn structure, we can see that White has a space advantage, not unlike a lot of positions we see in the King's Indian today (though with the bishop on g7, which gives more chances to create counterplay} . a

b

c

d

e

f

g

h

Once again we see that whenever Rubinstein had a fixed plan, he was not very flexible and

Chapter 3

-

Space Advantage

would continue his plan n o matter what. At this point I would consider whether 1 3.a4!, with the idea liJ d2-c4, might offer White a bigger advantage than the game. But then I would also take into account 1 5 . . . ltJ d4, a move I am sure Rubinstein and Duras rejected as losing a pawn without thinking any further.

13 ... �xc6 14.d4 i.g4 1 5.d5

b

a

d

c

f

e

g

h

At this point I think the black position is starting to get rather sad.

15 ... c!ll e7? I put the question mark here from the perspective of the 2 1 st century. If we imagined that some young player had played this today, we would question his chess education. A modern player would seriously consider giving up a pawn with: 1 5 . . . ltJ d4! 1 6 . .ixd4 exd4 1 7.'?;Vxd4 g6

101

Black has given up a pawn in return for dark-squared counterplay (potentially with opposite-coloured bishops) and will follow up with moves such as . . . .ig7, . . . l:'!c8 and . . . '?;Va5. I am sure that it is enough. To me it looks pretty playable for Black. I have played similar positions in the Sveshnikov many times over the last few years, for example in the World Championship match with Anand. The exchange of both sets of rooks should lead to a draw. Maybe the best thing White can hope for is an ending with rook and bishop versus rook and bishop, though I cannot see a way for White to exchange the queens. But even in this case, all Black needs to do is to keep control of the c-file and he will be OK, though he would still suffer a bit. This is often the case when we play slightly inferior positions: the wrong exchanges will make life difficult for us. There is nothing special in that direction about this position. Of course we cannot criticize Duras for not making this decision. The problem is that the dangers of allowing the opponent to have a space advantage were not fully understood at the time of this game; after all, we are to some extent talking about the pre-Rubinstein era . . . Even 1 5 . . . lD b8!? appears more natural. The knight certainly has lost time returning to its starting position, but at least it has a future on a6 or d7.

1 6.Yfd3 Out of the pin. I find the way Rubinstein played from this point on very instructive. First he improves all his pieces. To start with, both players have a number of pieces in poor positions. Both sides have stupid bishops, the knight on f3 has no function and it is hard to work out where the knight on e7 is headed. But White has a space advantage,

8 7 6 5 4 3 2 a

b

c

d

e

f

g

h

Positional Decision Making in Chess

102

allowing him to regroup his pieces in a smoother way.

16 ...Vfd??! I don't like this move much. The main idea behind this move seems to be to exchange the light-squared bishop. These days our experience in the King's Indian is extensive and we know that we need this bishop to hold the queenside together. But again, we cannot judge Duras on modern terms. At this point we can start to see a consequence of White's space advantage: Black's position is getting rather cramped. Maybe Black should play something like: 1 6 . . . i.d7 l 7.lll d2 b5 This is hardly an ultimate solution, but it seems necessary. 1 8.a4 a6 The computer suggests that Black should take on a4, but this only helps White. Actually, we should forget about computers when we are looking at such closed positions. Here it seems that c4 is a key square. Now imagine that the game continues with: 19 .Ei:fc l Yfb6

a

b

c

d

e

f

g

h

20.a5!? I am not sure I would hurry to play this, but it is quite relevant for our theme to discuss what would happen if we did. 20 . . . Yfb7 2 1 .b4

White is of course better, but compared to my game with Campora later in the chapter, we can see that Black has managed to keep his good pieces. At the moment of writing, June 20 1 4, I am following the Russian Higher League live on the Internet, where something similar happened:

Evgeny Najer - Evgeny Tomashevsky Vladivostok 20 1 4

a

b

c

d

e

f

g

h

After 40 moves m a Closed Ruy Lopez White has the advantage through extra space. But in this case it is not apparent that he can do anything with this extra space. There are no entry-points and no weaknesses within reach. Furthermore White's bishop is tied to the defence of the a5-pawn. All in all, the game should soon end in a draw (as indeed it did) . In our example Rubinstein has freer hands and can start using his extra space to organize an assault on the kingside, presumably connected with f2-f4. And although it is not going to be easy to break Black's defences, it might be possible, due to the smoother manoeuvring White's space advantage gives him.

17.c!lid2 .th3 Black is continuing his plan. There is not really any reason to suggest alternatives.

Chapter 3

-

1 03

Space Advantage

gloss over) , but had tried to slowly improve his position, say with 20 . . . lt:ig6, Rubinstein would presumably have played in much the same way with 2 1 .i.c l i.e7, though here there is a trick.

7 . . 3�·�if��.ef� �� 6 � � 4)%�. � - �%�8

��� ��·rx�-

�� � �ti � 8" � 'ti"• �� '-�� � 3 � '·�� �r%'" .. ·"�r" 2 �� � � � J�;j1.� .. ..� �-�% �:� 5

4 18.a4! Played to secure the c4-square. It is also possible to play 1 8.lt:ic4 with the idea 1 8 . . . b5?! l 9.lt:ia5, which does not improve things for Black. Bue it is quite typical of Rubinstein to spend as much time as possible improving his position. This is a clear sign of his style. 1 8 .lt:ic4 is neither better nor worse than the move Rubinstein chose in the game. It is a bit more energetic, but Rubinstein did not value dynamics that highly. Realistically Black would not change his play. But we should remember Bocvinnik's idea, which is that when you have to play several moves to follow your plan, it is best to start with the moves chat keep your ideas hidden and options open, if only to confuse the opponent. 18 ...J.xg2 19.©xg2 geb8 20.tll c4 b5 I do not like this much. It seems to create extra weaknesses. Black was of course hoping chat b3 would become a target too, but due to the space advantage, it is easy enough for White to organize his forces for the defence of this pawn. At the same time Black's weaknesses on a7 and d6 are more exposed. If Black had not done anything on the queenside (you can argue he already decided to do so on his last move - a point we will quickly

1

a

b

c

d

e

f

g

h

If White plays 22 . .ie3?, Black gets instant counterplay with 22 . . . lt:ih4t! 23.�h l �h3 24.!!g l lt:if3 25.!!g2 �g4 and things are far less clear than they ought to be. Bue we should not be concerned; Rubinstein would never allow any tactics. White can play either 22.h4!? or something like 22.f3 i.g5 23.i.a3. Forcing Black to go back. 23 . . . i.e7

7 Ei�if�ii� 6 .. .. %�� % �.'i. � ...... � L. 8

��� ��-"�-

� '--ti �---ti%·�3 7!§8•V•8� �� ���� 2 ,;,J. . .� ���� '� "··· 5

4

a

b

c

d

e

f

g

h

Probably now I would play 24.:1'!fc l with the idea of doubling on the c-file, playing slowly and improving the position. You can imagine chat in a position with the rook on c3 we then play lt:ie3 (of course not allowing . . . i.g5 with exchanges) with ideas such as lt:if5 , �b5, i.b4 and other moves chat slowly improve the position. For Black the best strategy is probably to

1 04

Positional Decision Making in Chess

simply wait, trying to make sure that he would have some counter-attackin g ideas for when the game finally opens.

21 .axb5 YNxb5 22Jia3!

a

b

c

d

e

f

g

23J:Ua l a6 24.�cl The bishop is rerouted to a better square. The action is on the queenside and the bishop has nothing to do on b2 anymore; it definitely belongs on e3.

h

22 ... �g6 Probably it was already time to put the knight on a passive place like c8. Sure it is not much fun to do so, but it is hard to see what the knight is doing on g6. White's advantage is extensive no matter what; probably the position is already lost for Black, somewhere between ± and +-. When I grew up our evaluation was formed by the system used in Chess Informant, where a slight advantage (t) meant that one side was pushing, and a large advantage (±) meant that nothing conclusive could be seen, but the opponent is in a dire situation. The players from my generation still think like this, while the younger players operate with a different system, which they learned from the computer (+0.20 for example) , which is not very informative in my opinion. Or maybe I am j ust suffering from the afflictions of middle age. (I should probably say that the use of + 1 or +2 as descriptions is now used almost universally among grandmasters. But more or less only when they want to explain how winning the position was that they misplayed!)

26... �e7 We have arrived at the final impressive move. White is obviously doin g well, but the question is how to improve the position further. The knight belongs on d2, where it can defend the b3-pawn. Rubinstein's move keeps an eye on a6 while preparin g to retreat the kni ght.

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

Chapter 3

-

Space Advantage

27.1Yfl ! Played with the idea of exchanging the queen on b5, which is holding the black position together. This is not the only way for White to improve his position, but it is a rather elegant way to do so. White now wins the a-pawn.

105

Black has no counterplay. The white kin g will come to the queenside and White will win.

32.gxa6 gc2 33.g6a2 gxa2 34.gxa2 fi.e7 35.©fl ©f7 36.©e2 ©es 37.©d3 ©d7 3S.©c3 .ids 39.tnc4 !i.c7

Something like 27.lll b2 would be imprecise. Black plays 27 . . . !!ab8 and gains a bit of time to create counterplay.

27 ... tncs 2S.�d2 1Yb4 28 . . . a5 loses a pawn to 29.Wxb5 l:!xb5 30.lll c4. Black has no defensive resources left. 29.1Yc4 1Yxc4

8 7 6 5 4 3 2 1

a

a

b

c

d

e

f

g

h

30.�xc4! The natural move. Rubinstein knows that he will win the a-pawn and ensures that he will keep a passed pawn, which is much more important than unifying the pawns into one super structure. 30.. J:!:abS 3 I .tnd2 �k7 White wins no matter what, of course. The attempt to bring the knight into the game with 3 1 . .. lll a7 is most simply eliminated with 32.ixa7! !!xa7 33.l:!xa6 !!ab7 34.l:!a8 when

b

c

d

e

f

g

h

This is the final instructive moment of the game. White is completely winnin g . He has more space, better pieces and not least, an extra pawn. Still Rubinstein shows his excellent technical feelin g by first improvin g the position on the kingside, even though it was probably already winnin g to simply push the b-pawn. As said before, Rubinstein never rushed unnecessarily; so much so that the mantra promoted by modern books on endgame technique (Shereshevsky, Dvoretsky, Yusupov, Aagaard and Muller} , " Do not hurry" is attributed to Rubinstein.

40.g4 .ids 4IJ'fa6 J.c7 42.h4 /i.ds 43.h5 Fixing the g7-pawn. White is now (over)­ prepared and ready to move the b-pawn up the board. The quality of the preparation is underlined by the fact that nothing really happened towards the end of the game. 43 .. .J.c7 44.b4 gb7 45.gas ©ds 46.©b3 gbs 47.gxbS Lbs

Positional Decision Makin g in Chess

106

8 7 6 5 4 3 2

Boris Gelfand - Magnus Carlsen Moscow 20 1 3

l .d4 � f6 2.c4 e6 3.�f3 d5 4.�c3 � bd7 5 ..if4 dxc4 6.e3 a6 7.a4 .id6 8.hd6 Later I played 8 .Ag5! ? against Dominguez, which we will look at in a lacer volume. .

8 ... cxd6 9.hc4 0-0 10.0-0 d5 1 1 ..id3 b6

a

b

c

d

e

f

g

8 7 6 5 4 3 2

h

48.b5 � e7 49.b6 6 50. gx5 �g8 5 1 ..ifl @cs 52 . .ih4 1-0

No matter who you are, there are only 64 squares on the chessboard And ifl control 40, there will only be 24 left for you - no matter how strong a player you are. Playing for a space advantage is one of my favourite strategies and it is a highly efficient one. Even our newly crowned World Champion had problems dealing with it. Setting the scene before the game: I had won two games in the previous five rounds, against Caruana and Morozevich and was leading the event. In the end I managed to win an additional game (Black against Nakamura) and win the tournament with 6/9, half a point ahead of Magnus, even after missing my chances in this game. The day before the game I saw my old friend Mikhail Shereshevsky, author of the brilliant book Endgame Strategy, whom I had not seen for almost a decade. He suggested chat after my great scare, I should make a draw, as Carlsen is "so strong" . I just laughed. The day you give up the hope of beating young opponents with White, you have truly grown old!

1

a

c

d

e

f

g

h

White has forced one concession on Black: the exchange of his good bishop for White's bad bishop (although it can be debated how poor the bishop really is when it is outside t e pawn chain) . The outcome is a position wnh a very small but also enduring plus for White. The bishop will be passive on b7, while White's bishop on d3 is rather well placed. White plays on without risk. A point is chat defence is harder to play as it includes a fair bit of mind-reading. As we shall see, this is even a difficult task for the highest-rated player of all time.

�

12.�b3 �e7 1 3.l:Ucl .ib7 14.�k2 gfc8 15.gacl �d6 As White's plan will be to advance pawns on both flanks, it makes sense to exchange the rooks. However, at the 20 1 4 Olympiad former World Champion Rustam Kasimdzhanov won

Chapter 3

-

Space Advantage

a great game against Vladimir Kramnik with 1 6.lll e 5!.

16.�a2 gxc2 17,gxc2 :!:k8 1 8,gxcSt i.xc8 19.ti'c3 i.b7

8 7 6 5 4 3 2

107

3 1 .�fl There is no great incentive for White to exchange knights at this point. He needs to have something to squeeze. 3 1 . .. �d6 32.6 � d7 33.g4 f6 34.�fg3 �f'8 35.©fl .icS

a

b

c

d

e

f

g

h

20.b4! White is grabbing space and squeezing Black's minor pieces. 20 ... � e4 2 1 .ti'cl h6 22.h3 ©f'8 23.b5 a5 24.�c3 ti'c7 25.ti'a3t ti'd6 26.ti'cl ti'c7 27.ti'a3t ©e8 28.� e2 g5 29.ti'b2 ©e7 30.�d2 � df6

a

b

c

d

e

f

g

h

Here and previously it made sense for Black to exchange a set of knights, as his position does not have sufficient space for two knights.

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

White has maintained his slight edge. He has extra space and a better bishop still. At some point soon it will be time to open the position on the kingside in order to create weaknesses in the black position. I decided that this made it a good time to exchange queens, though it is by no means obvious that White does not also have winning chances by keeping the queens on the board. It is obvious that the c-file is the best place for the queens. At the moment Black has been allowed to control it, but I do not think he should be allowed to do so uncontested for any longer. By placing my queen on the c-file, I give my opponent a choice of whether to exchange queens or give up the c-file. I play what I consider to be the best move, but at times it is not obvious which path is the right one and in such cases I do the same as everyone else: I play the move that feels right to me on the day. Give me this position on another day (with the same opponent) and I might have waited before making a decision regarding the exchange of queens.

Position al Decision Maki ng in Chess

108

36.�c2 �xc2 I am unsure about this decision. Keeping the queens on would give potential counterplay when the game opens up later on.

.ic2-b3! would force Black to do something immediately.

37.hc2 .td7 38.�c3 .te8 39.f4

a

a

b

c

d

e

f

g

h

Black has by no means been outplayed here, but he is under slight persistent pressure with no clear active options. It is difficult to defend this type of position accurately, especially for humans, and it is therefore less of a surprise that Carlsen errs. The real conundrum is that it is very hard to anticipate when Black should change the pawn structure, or when such a change would be for the worse; when he should be patient and passive, and when he should be active and/ or desperate.

39 ....tf/? This fairly innocuous-looking move is actually a decisive mistake. 39 . . . gxf4 40.exf4 J.f7;t; was required, though White has some pressure still. Here I missed my chance.

40 ..ib3? After the game I found out that I could have won the game with 40.f5!. The threat of

b

c

d

e

f

g

h

After something like 40 . . . e5 4 l ..ib3 lli c4 42.ixc4 dxc4 43.iie l White wins easily. The king goes to c3 and a knight to d2. White wins a pawn and soon after the game. The black knight can only go to h7, f8, d7 and b8. There are no routes to anywhere else. Quite a good illustration of the power of the space advantage. And after 40 . . . exf5 White should not take with the pawn, as I had considered during the game, but instead exchange Black's only good piece with 4 1 .llixf5t! llixf5 42.gxf5 .

a

b

c

d

e

f

g

h

White wins immediately with .ic2-b3 and if . . . iid6 then e3-e4 comes and the d5-pawn falls. It is of course a surprise that after not wantin g to exchan ge kni ghts for a while, this is suddenly the way to win the game. But no one ever claimed that chess was a simple game.

Chapter 3

-

Space Advantage

40 ... gxf4 41 .exf4 c!Lig6 42.£5 I had believed that 42.®e3 would give me a winning position, but now realized that Black has 42 . . . lD h4! generating lots of counterplay. White has no advantage at all.

109

especially queens, will be able to do a lot of damage if they manage to sneak behind a far­ advanced pawn chain, while minor pieces do not gain extra potential from a big void in the same way. Also, they are far less likely to escape from their prison. This of course only matters in positions where there is at least one open file; how are the rooks otherwise going to be able to escape? Bue in most games there is an open file, as for example in this one.

Boris Gelfand - Daniel Hugo Campora Cesme 2004

a

b

c

d

e

f

g

h

42... lDf4 Black has managed to create counterplay and gee his pieces into play. I cried to create some problems for my opponent, but not surprisingly he was able to solve chem rather easily. 43.fxe6 lDxh3t 44.@e3 he6 45.lDxd5t @ds 46.c!Lih5 f5 47.gxf5 c!Lixf5t 4S.@e4 lDg5t 49.@e5 c!Li e7 50.lDhf4 i.gS 5 1 .i.dl hd5 52.c!Lixd5 c!Lif7t 53.@e6 c!Lig5t 54.@e5 lDf7t 55.@e6 c!Lig5t 11z_11z

A clean example In the following game we shall see one of the things chat often comes into question when we are dealing with the concept of space advantage (the guideline I added to Yusupov's five at the beginnin g of the chapter) . When you have managed to squeeze your opponent into only two or three ranks, it is often the case that you want to exchange the rooks and queens, but not minor pieces. Rooks, and

1 .d4 lDf6 2.c4 d6 I have some experience with the . . . ig4 idea chat Campora played in this game. Back in 1 987 I won a nice game with Black against Loginov. This can be found on page 47 in My Most Memorable Games. I also won a game against Aleksandr Huzman two days before, but there I played it with the 1 . . . d6 2. tll f3 ig4 move order. 3.c!Lif3 J.g4 4.lDc3 c6

8 .1.• � �-- � 7 ..lfi"�"...% : "%-.... . • ----% .�•. 6 � N��,,.!Ii � � 5 � � �r!l �� 4 � �.t� • 8 ®'� -- % �� 3 �m -- ��� � �Jii %�W"' �· %' ······ � 2 �tJJ[j�!j �j__{jrfP 1 � �V�.i,-.: a

b

c

d

e

f

g

h

Campora has played this a few times. I think this system is coo passive and chat White gains a definite edge out of the openin g . Exchan gin g

1 10

Positional Decision Makin g in Chess

the good bishop does little to ease the pressure, as we shall see in the game. Such observations might sound very basic, but we should not discard old knowledge j ust because of its age or how easy it is to understand. We should j udge it on whether or not it is useful for us today. In this case it is. If Black plays 4 . . .J.xf3 White can take either way. Against lvanchuk (without . . . tlJ f6) I took with the e-pawn. There is no big reason for this; on a different day, when I had been in a different mood, I might have taken with the g-pawn. White is fighting for an advantage there as well, but at least he has doubled pawns.

5.e4 � bd7 6 ..ie2 e5 7 .0-0 fi.e7 8 ..ie3 0-0

s K- � .J. �-� 7 �� ····· ·Y.-/,�i ·-----· .

6

..... %

.

this is of course a bit dogmatic, it should not be believed religiously. But at least Black should have a serious reason for this exchange, which I do not see here.

9 ... cxd5 Black is making this decision before he is forced to do so. It would be better to keep the flexibility. If Black played something like 9 . . . a6 I would probably play 1 0. tlJ e l ( I O.a4 is anti-positional because of 1 0 . . . a5 and Black has secured a permanent outpost on c5. And after I O.tiJd2 J.xe2 l l .Wfxe2 White does not have the c4-square at his disposal. But it could still make sense, with ideas such as b2-b4 and tlJ d2-b3-a5 .) 1 0 . . .J.xe2 l I .Wfxe2

....

5 llll "JI( • 4 .,,,;z-: .... 3 � �,,m � �ltJ� 2 . .. �. . /,�·0("" ef � 1 � °ii nli�

�.!8'8'• .r� ��

8� ""

'"'

a

�

b

%w.·0

lt� r�

..

. ..

"

c

d

e

f

g

h

a

b

c

d

e

f

g

h

Up to this point I had not spent any time to speak of. There were no difficult decisions to take; it was simply choices between equal options.

With the idea tlJ d3 and potentially even f2-f4. I have put the knight on e l /d3 against the KID many times. If the c4-square is blocked, then d3 is usually the best choice. I simply find the knight better placed on d3 than on d2 in these positions.

9.d5! I cannot say that this was a very difficult decision either. With this move White secures a good pawn structure. The bishop on g4 will sooner or later be exchanged and we are in a King's Indian structure, where Black in principle should do whatever he can to avoid exchanging the light-squared bishop. Okay,

10.cxd5ha We can see one of the problems with exchanging the light-squared bishop in this line: 1 0 . . . a6 l I .tiJd2 1 1 .tlJ e l transposes to the previous note. 1 1 . . . J.xe2 1 2.Wfxe2 b5

Chapter 3 - Space Advantage

Black would also not manage co exchange bishops with l 1 . . .lDe8, when White can cry something like 1 2.i.g4!? and Black is still somewhat uncoordinated.

8 7 6 5 4 3 2

8

7 b

a

c

d

6 e

1 3.E:fc l 1 3.b4 with the idea lD b3-a5 also exposes the c6-square, but Black gets . . . lD b6-c4. 1 3 . . . lD b6 1 4.a4 b4 1 5 . lD d l Threatenin g a4-a5. 1 5 ... a5 1 6.E:c6 Normally Black is already threatening mate when White gets this far; or at least this move is played as an exchange sacrifice. Here Black's position is simply horrible.

s

7

6

5

4

3

2 I

� ��LJ�-J-� � · -'i)�it�i ,,,,,% � �-,,,,,Y,.,,,,%� �� ir• � � �� ��� �"8'� � � %' " / /, �-0 .i. � � m � D

8w/"'"; • tWj�¥ff �

d · ' l.,tir"�f a

111

b

c

d

e

f

g

h

1 1 .J.xa YlYbS Black is trying to exchange his bishop with the . . . i.d8-b6 manoeuvre. If he was successful in doing so, he would probably be more or less okay (as can be seen in the game Johansen Torre, Kuala Lumpur 1 990; though White won there too) . The problem is that White will not allow it.

5

4

3

2

1

a

b

c

d

e

f

g

h

1 2.a4! I am sure chat there is more than one interpretation of this position, but the way I play chess, this is quite simple. I do not wane to allow the exchange of the bishop (at least not this easily) . 1 2 Jks 1 3.aS I spent 1 6 minutes on this move. I needed co calculate, to decide how to rearrange the pieces, and to correctly evaluate the consequences of what will ensue. As we shall see, my long-term considerations were accurate, but I did not gee there in the best possible way. .•

If my arm was twisted and I was forced co give an evaluation of this position with Chess Informant symbols, I believe I would say that White has a clear advantage already. I value the extra space highly and I see no compensation for White's extra space, nor for the absence of the li ght-squared bishop. An important point is that these are long-term features chat basically cannot be changed.

1 3 .tds •••

Positional Decision Makin g in Chess

1 12

I 3 . . . b5?! by the way does not create activity for concrete reasons, which is one of the things I had co check. White plays I 4.lLia2! and the knight arrives on b4 and most likely c6 with decisive effect.

8���� � · 7 ·�' - ---%� �w���----%�

�,·%� i)� : �· -r..7/41 '•8•" !··· 8 �%'"//, ��-'l: �� 3 �wtlfil ..t' " ";-{{j�,,JW{j �� �, , %� 1•i= "" 4

2

�

a

b

c

d

e

f

g

h

14.ie2 We shall see the drawbacks of this move soon enough, but let us first look at the alternatives. A shortcoming of the long plan is that at times you miss alternative ideas.

For example, a very concrete idea was suggested by Lukacs: 1 4.g4!? I would have been happy to find this move, but unfortunately I was nowhere near seeing it. However, this does not mean that it is best. It is based on this tactical variation: 14 . . . h6 1 5 .h4 lLi h7 1 6.g5 ! hxg5 1 7.�g4

8 .1� �.I� �· 7 - - �� �� ���.%.'II -j�

"''"•'i• '"" �� " , %�p ��p r!§ ��t· t• ,%.��%%·���%•�---"��� �o - _:� �---�� �� 2 {' ' %� 1� �, = 6 5

4

--

"

3

a

b

c

d

e

f

g

h

Black is caught in a nasty pin. But rather than playing l 7 .. Jic7? when White has 1 8.lLib5!, Black can sacrifice the exchange with: 1 7 . . . :B:xc3! 1 8.bxc3 Vff c7 With decent compensation.

Chapter 3

-

1 13

Space Advantage

Looking at all the sources, it turns out that White has a stronger way to play. 1 4.�e2! Indicated by Khalifman in his book Open in g far White according to Kramn ik. I am not sure if it is fair to say that his recommendation is influenced by my game, but I will allow myself to be a bit presumptuous, as the strategy seems to be the same. 14 . . . a6 14 . . . tlJc5 l 5.l:%a3 does not improve things for Black. 1 5 .l:%fc l

19 ..ie l! Black will have to retreat with the queen, leaving White with a convincing advantage, as after something like 1 9 . . . tDcd7 20.l:%cb l �c4 2 1 .�xc4 l:%xc4 22 . .ie2 l:%cc8 23.f3 followed by tD c3-a2-b4 Black is going to suffer badly. 1 6.b4 �b7 1 7.�e l tlJ f8 White i s also clearly better after l 7 . . . .ic7 1 8 . tlJ b l .ib8 1 9.l:%xc8t �xc8 20.tlJd2 .ia7 2 1 .l:%c l �d8 22.tlJ fl . The rook is about to arrive on c6. 1 8 . .ie2 l:%c7 1 9. f3 tlJ g6 20.g3 tlJ d7 2 1 .l:%a2 l:%ac8 22.l:%ac2 h6

8 7 6 5 4 3 2 a

b

c

d

e

f

g

h

1 5 . . . b5 This is the main move given by Khalifman, after which White is simply more comfortable. The only chance to gain counterplay seems to be 1 5 . . . b6, but after 1 6.b4 it appears that White is maintaining the pressure: 1 6 . . . bxa5 1 7.bxa5 �b3 ( 1 7 . . . �b4 1 8 ..id2 lDc5 1 9.tlJa4 with a clear advantage for White) 1 8 . .id2 lDc5

a

b

c

d

e

f

g

h

23.�d2 tlJ gf8 24 . .ifl tlJ h7 25.h4 tlJhf6 26 . .ih3 l:%c4 27.tD b l White has a big advantage according to Khalif. The similarities with our main game are quite compelling.

14 b5?! This looks like quite an achievement for Black, but I believed that it was okay to allow this move. As we shall see, it was probably already a serious mistake. ...

Black's best idea was 1 4 . . . a6! to prepare . . . b6. I looked at 1 5 .�b3, but it is not convincing. Black can try a pawn sacrifice with 1 5 . . . b6!? or defend with: 1 5 . . . tlJc5 1 6.�b4 b6! 1 7.axb6 ixb6 1 8.ixa6 l:%xa6 1 9.l:%xa6 tlJxa6 20.Wfxb6 �xb6 2 1 ..ixb6 l:%b8 22.ie3 l:%xb2 23.l:%al llib8 24.f3

Positional Decision Making in Chess

1 14

�

. . . � �� �� ��� �� ���-.--j�r�· · · '� � : �� ����� �� �,,,,�v� �� � � m �·�8-. � ��t' ; , � 2 �.. . . :.� ... r . v.� � ��-�. . ',

6

3

.

.

a

b

c

d

e

f

g

h

White is slightly better with the bishop being better than the knights and the d6-pawn being weak. The white knight will go to d 1 and e3, from where it can go to c4 or f5, putting further pressure on Black. For this reason I looked at the prophylactic waiting move 1 5 .@h l !?, which includes the idea of playing f2-f3 without suffering from any counterplay on the a7-gl diagonal .

a

b

c

d

e

f

g

h

1 5 . . . b6 This should be the most challenging. 1 5 . . . tlic5 1 6.f3 b6 1 7.b4 is not in Black's favour. Nor does it look like there are any other real ways to improve the black position. 1 6.axb6 Here Black has two real options. 1 6 . . .'Wb7 16 . . . a5 ?! 1 7 .f3 .ixb6 l 8 .i.d2 gives White a clear edge.

But also quite decent is 1 6 . . . Lb6 l 7.i.xa6 �e8 1 8 ..ixb6 'Wxb6 1 9.'We2 �eb8 20.ib5 �xa l 2 1 .�xa l tlic5 22.f3 tli b3 with the idea of . . . lli d4. 1 7 .'Wa4 tli c5 18 . .ixc5 �xc5 19 . .ib5 'Wxb6 20 . .ic6 �b8 2 1 .'Wxa6

�� �· .. . v.� � ��-0 �1.,. �M. ��� . .3� . ��. �� ��...."?lft�)r�, �m ��-� �?3)���?3)�.0 2 ��M �:f�8� " ""�. • ""(H �� �

8

s

,

4

.

.

3

... .

a

b

c

. ..

. .

:� d

e

f

g

h

Black has a fair amount of compensation. In both cases ( 1 6 ... 'Wb7 and 16 ... .ixb6) White is to be preferred, but at the same time Black has reasonable chances. I cannot be certain which of the two options I would have chosen (or if I would even have thought of 1 5 .@h I !?); being a pawn up with opposite-coloured bishops where Black might be able to block with a rook on b4 and a bishop on b6, or the ending with level pawns, where White will be able to squeeze for a long time. An argument one way would be that the position where Black gives up a pawn would give him more activity and thus be easier to play. But on the other hand it is also tempting to be a pawn up!

1 5.f'3 Simply defending the e4-pawn. Basically it is the same story as before. Black is planning to exchange his dark-squared bishop and White has a space advantage. 1 5 . .ixb5? �xc3 would be entirely wrong.

Chapter 3

-

Space Advantage

1 5 ... a6 With the e-pawn defended, the b5-pawn in turn has to be defended. Note that 1 5 . . . b4 1 6.tLlb l does not improve things for Black, but only creates more weaknesses in his position.

If Black plays 1 8 . . . tLlh7 with the idea of . . . .ig5, the simplest is probably to play 1 9.h4, which does not harm White's position in any way. 8

7

6

8

5

7

4

6

3 2

5

4

3 2 1

115

1

a

b

c

d

e

f

g

h

16J:Ul Of course I would like to transfer the knight to b4, but at the moment it is not so easy as the a5-pawn is hanging. Of course I could try something with Wfe l and some elaborate manoeuvre, but there are also other problems to solve and it makes sense to solve the easiest one first. Therefore this preparation for doubling rooks on the c-file, as well as the transfer of the bishop to h3, where it will put pressure on Black's position. 16 ...�b7 One of the problems with having limited space is that Black finds it difficult to reorganize his pieces in a meaningful way. Campora chooses to force the exchange of the dark­ squared bishops, but it is time-consuming and White can make significant progress in the meantime. 17.i.fl h6 1 8.g3 i.c7

19.@g2 So, I decided not to put the bishop on h3 after all. This move does not make too much sense. I cannot remember what my intention was, to be honest. White is essentially doing nothing, waiting for Black to play . . . J.b8 so he can reroute his knight. It is worth mentioning that I consider Black's position too difficult to hold already. The best defenders are excellent at posing problems for their opponents, but here it j ust seems too easy for White to continuously improve his position until Black breaks under the pressure. Basically, Black played a passive opening and violated some of the basic ideas of chess: he gave up the fight for space in the centre, he gave up his bishop for the knight. It you repeatedly give the opponent permanent advantages in this way without getting anything in return, you end up in a bad place. The next few moves are automatic.

19 ... .ibs 20.�a2 .ia7 2 1 .ha7 �ha7 22.�b4 � b8

Positional Decision Making in Chess

1 16

8 7 6 5 4 3 2 1

8 111 1.� • ,,%_ 7 6 -� � % ·•· � 5 � 4 '� ��(8'� �� 3 �lZJ� 8 tl] 2 �W,$ 1 DV���

�- - - ��'� '�m-- -·Y.�1-� ��-- --%�1-� r� ·zr-·- - ;� ""' �� �� ��-0 �� :- �/,/,� �� � ��- , ,, , ,

a

b

c

d

e

f

g

h

White's advantage is not up for debate, but he still needs to find a way to break through; especially as the position is rather closed at the moment. Obviously we want to squeeze Black on both flanks, but we should not do this without having a clear plan about how we can improve our position. The next stage in this game is to exchange the rooks and queens. Having done this, Black will have no counterplay whatsoever, allowing me to play on both flanks, which is necessary to exploit the advantage. Going through Rubinstein's games you will find this theme countless times. Someone else told me that the same principle exists in military strategy. It is quite logical; if you are weaker, you want to narrow the battlefield as much as possible and have as few contact points as possible. Both in war and chess, we see this concept played out in the fortress strategy. In this position Black might be able to defend his position on one flank, but as soon as the front becomes wide, he collapses. After the exchanges White will be able to open the game without fearing that something dangerous will slip through the cracks. Also, as White cannot gain control of the c-file, it makes sense to prevent Black from getting it.

a

b

%% ' " "

c

d

e

f

g

h

26.:Sxc7 Apparently it is anything but obvious to some other grandmasters that White should exchange all the heavy artillery in this position, but I did not spend a lot of time on this decision at all. The danger of doing this is of course that Black could be able to set up a fortress and the heavy pieces would be needed to break it. Making such a decision at the board depends a lot on what you believe about the position. I was sure that I could break any attempt and therefore went for it without much hesitation. I should add that it is of course too far into the future to consider which type of fortress Black will try to set up. This is a moment to go with your feeling. I believed that White would win this ending and there were a lot of upsides to it. And I always play according to my beliefs. At the end of the day it is not a matter of life and death; nor is it the right moment to try to find ultimate solutions. 26 ... :Sxc7 27.:Sxc7 �xc7 28.�cl �xcl Black has an unpleasant choice between exchanging queens (ending most realistic hopes of activity) and giving up the c-file. Although I was happy to exchange queens, I would also not have minded keep them on the board if mine was to be the more effective queen!

23.gcl �d7 24J�fc2 gac7 25.�d3 @m 29.c!lixcl c!lifd7 30. d 8 34.'tt> c2 lll d7 3 5 .ih3 'tt> c7 36.lll b3 lll f6 37.lll d2 lll g8 38.lll b l lll e7 39.lll a3 lll c8 40.ifl lll a7 4 1 .f4±

��-•..0,z�� �� ��-...1iz�,j"'0. 7 ., z� ,..,,,,� .... ;� ,j"'' �iii% 6 40• f��-'0· r.• 8

,

�

�

,

.

.

� . % W .... % � :��--��� � � wr

a

b

c

d

e

f

g

h

32.b4! There is a trade-off with this move. On one hand it is no longer possible to put the knight on b4, but on the other hand Black is denied his only good square for the knight and he will feel the squeeze of limited space all the more after the loss of it. Finally, the b5-pawn is fixed. In many positions White could consider sacrificing a piece on b5 followed by queening the a-pawn. I am not sayin g that I would take on b5 (unless it was very clear) , but that it is an additional thing for Black to worry about. 32 ... lC!cd7 After the game I wondered if it was maybe better for Black to put the knight on a4, to avoid getting too cramped. Of course White would keep the advantage, putting the king on c2 to paralyse the knight, and then regrouping the knight to a3. The threats of a sacrifice on b5 would force the black king to the queenside and/or the knight to a7. After this White can start improving his position on the kingside and hopefully win the battle there. Working on this book I looked at how this would actually play out in practice.

.

3 .. ;� ��� �....% �1 ��� �.. ,� � .

2

/

...

b

a

c

d

e

f

g

h

White finally changes the pawn structure. Here Black should probably wait patiently, as a further weakening of the light squares would prove devastating: 4 l . . . f6?! 42.ih3 lll c8 43.ie6 lll e7 44.f5 White's space advantage only grows. Both the black knights have been shut out of play. 44 . . . 'tt> d 8 45.lll b l 'tt> c7 46.lll d2 'tt> d 8 47.lll f3 'tt> c7 48.g4 'tt> d 8

� ��--�.

.. z

� '- �-lJ_., % % � ,� r� Z8 s8 . % -�...% �� 8 � �� � ���� 3 2 �������z� . .. �1 ����

6

�

5

4

�

a

--··

.

b

c

d

e

f

g

h

49.h4 'tt> c7 50.g5 hxg5 5 1 .hxg5 Black is in an awkward situation. If he takes on g5, the knight will quickly show up on e6. If not, the f-pawn will be impossible to defend after: 5 1 . . .'tt> d 8 52.gxf6 gxf6 53.lll h2!

Positional Decision Making in Chess

1 18

can consider putting a rook on c4 or giving a check on a7. Still, I cannot shake the idea that if Black had played . . . f5, I am not absolutely sure I could win the game. I would have to show great technique, put the bishop on c8, manoeuvre the knight around and see what happened. Probably it is winning, but I do not feel the same degree of certainty.

33 ... �£6 34.�d3 � e8 35.f4 After placing all the pieces in their optimal position, White starts to squeeze. 33.i.h3! Another small finesse. After 33.llid3?! f5 ! Black gets the chance to exchange a pawn and create a bit of breathing space. His troubles are by no means behind him, but life is a bit easier for him. It does not make sense to allow it. Black could of course have done this on move 30, as already mentioned.

35 ... £6 Black has to do this, as after 3 5 . . . 'kt>f6 36 . .ic8 he would have to cake on f4 to avoid losing immediately. 36 ..icS �c7 37.@f3 @ds 37 . . . lli d? does not change anything. White plays 38.h4 lli f8 39.h5 and the game will develop much like it did. 38 ..if'5 � d7

a

b

c

d

e

f

g

h

This is the kind of thing chat makes me wonder if I should have played 26.'kt>f2!? before exchanging the heavy artillery, with the intention of playing i.h3 quickly thereafter. This is probably chinking a bit too deeply for a human, but to ponder about such matters retrospectively makes sense, as it helps us to improve our understanding of the game and sharpen our intuition. But after 26.'kt>f2 Black

a

b

c

d

e

f

g

h

A lot of people would be anxious chat they might struggle to break through in such a position, but I was very confident during the game. There are simply too many weaknesses. Also, I chink I managed to find a really strong plan at this point, involving the penetration of the king and manoeuvring the knight to f5 .

Chapter 3

-

Space Advantage

39.h4 �m 40.h5 � e8 4 1 .�fl � c7 42.�dl

8 7 6 5 4 3 2 1

1 19

46 ..ig6! Finally we have arrived at the decisive moment. Black can do nothing about the coming sacrifice on g7. 46... �xg6 47.hxg6 � e8 48.�f5 ©d7 49.©g4 ©d8 50.©h5 ©d7

a

b

c

d

e

f

g

h

�en your opponent does not have any counterplay it makes sense to j ust improve and improve until it is possible to do something active. Of course you always have to be on the lookout for the possibility of fortresses, but in general this is the way I try to do it. Here there is no real chance of a fortress. As I said, there are too many weaknesses.

42 ... � e8 43.�e3 �c7 44 ..ih3 An old habit. Repeating j ust for the sake of repeating. You never know when you will need the extra time and also there is some theory about it being a useful demonstration of power; though I do not think Black's feeling of powerlessness can be any greater than it is here.

8 7 6 1

a

b

c

d

e

f

g

h

5 1 .�xg7 The computer says that taking on e5 first is more accurate, but you only need one way to win. 5 1 ...�xg?t 52.©xh6 f5 53.©xg7 fxe4 54.©fl e3 55.g7 e2 56.g8=°1W el =°IW 57.°1We8f

44... �e8 45 ..if5 �c7

1-0

1 20

Positional Decision Making in Chess

White is obviously winning, taking all the vital black pawns. That it is mate in eight is a bit more surprising, but luckily this is irrelevant for us when we are making decisions.

drawing nine games and losing against Leko. But on Board One sometimes all you have to do is survive, in order that the sharks floating around on the lower boards can do the job.

I think this game is a good illustration of how to win against someone who defends passively from the beginning. The main thing to think about is the possibilities of fortresses and of course counterplay.

Wang Yue plays the Slav/Semi-Slav in what we can safely call a pretty defensive manner. The Slav is of course the team openin g for the Chinese, though they do not always seem to check up on the material available to us in the West. To be honest, they do not really work with foreigners, which means that we know very little about how they work. They have some training camps, but no one really knows what goes on in them. But in general they all have excellent technique and play the Slav, which could be a clue. Wang Yue has a lot of patience and plays defensive positions really well, which is maybe why he has good results in the Chebanenko and Schlechter systems.

The masterpiece The following game would definitely be in a Best Games collection, if that is what I had wanted to write. I have played Wang Yue six times; the first time in the Grand Prix tournament in Sochi 2008. He played great in that Grand Prix cycle and was very close to qualifying for the Candidates Tournament. If he had won from a better position in the last round, he would have. He really impressed me in our first game. I was not in good shape and played the opening poorly, but then I got into a bishop ending and thought I was going to hold the game. But he found an incredible plan, putting the pawns on the same colour as the bishop, winning the game. I saw that a few years later he did the same against Khairullin in the Russian League. Probably he made a real contribution to endgame theory with this game. They will have to be analysed deeply for us to see if this is the case - which is outside the scope of this project. Wang Yue is certainly a high-level player and he has really impressed me on several occasions. Sometimes he plays brilliantly, while in other games he is happy with a draw and plays without any ambition. I don't know what this is about. But his technique is excellent. He is the first Chinese player to get to the top and into the elite tournaments. In 20 1 4 he led the Chinese team to Gold medals in the Tromso Olympiad. He did so by

Boris Gelfand Wang Yue -

Medias 20 1 0

I .d4 d5 2.c4 c6 3.�c3 �f6 4.e3 a6 5.�f3 b5 6.c5 g6

8 7 6 5 4 3 2

a

b

c

d

e

f

g

h

7.�e5 6.c5 is very committal. With this move I grab space, but also fix the structure. Therefore I need the knight j ump to prevent Black from

Chapter 3

-

playing . . . .ig4 i n order to exchange the bishop for knight. If he could do this and follow up with . . . e5, he would be on top. So, at this stage White has a space advantage. It could backfire as White has wasted some time in the opening, but it looks as if it works.

7 Ag7 8.f4 a5 9 .ie2 flc7 This is no longer the way people play. We will look at this after the game. .•.

121

Space Advantage

.

with it first, but it happened in a conversation while we moved the pieces around (on the board, not the screen!), both hands on both sets of pieces. It was truly a collaboration. In a very similar position I played l l .g4?! against Bareev, leading to an early confrontation on the kingside. I got a worse position and lost the game. After this we tried to understand how to play the position, which helped me for this game.

10.0-0 0-0 1 1 .a3! 1 1 J.e6 12 .ia lll bd7 13.lll d3! ...

a

b

c

d

e

f

g

h

This was our idea. Short term, the move simply prevents . . . b4, which could be relevant in some lines. Longer term, it is a slight improvement in the pawn structure, preparing to support the long pawn chain. I could have played this earlier, of course, bur there were other moves that were more pressing; for example castling. Obviously ... b4 is not really a threat at the moment. White replies llia4 and a2-a3, using the b-pawn as a target for the creation of weaknesses. Bue it might be down the line and White is now better prepared for this. The best way to describe my idea is probably that I wanted to play .if3, lli d3 and a3, keeping a space advantage. I had looked at this variation together with Alex Huzman and somehow we came up with this set-up. I say "we" on purpose. Certainly one of us came up

•

a

b

c

d

e

f

g

h

This is the key move. I knew my opponent would place his pieces as he did in the game. So we decided to play for a space advantage. And in doing so I followed the typical plan of not allowing the exchange of pieces unnecessarily. The general idea is to keep the space advantage and then see what happens.

1 3 ...h6? I am not sure about this move. I think his idea was to play . . . .if5 and . . . g5 to find a space for the bishop (h7) . Then he might get good counterplay as . . . b4 could come together with . . . lli e4 or . . . llig4 etc. We will look at the main alternative 1 3 . . . �b?!? after the game.

Positional Decision Makin g in Chess

122

If Black plays 1 3 . . . if5 White avoids the exchange with 1 4.llif2, leading to something like: 14 . . . h5 1 5 .id2 l:!ae8

14.g4! As explained above, White's concept would be ruined if Black were allowed to play . . . if5 and . . . g5, so this move is pretty much forced. 14 ... llih7 I do not think Black has much choice here but to block the kingside. If he played 1 4 . . . h5 1 5 .h3 hxg4 1 6.hxg4 g5 1 7.'it>g2 I would build up on the kingside and Black would be in serious trouble.

b

a

c

d

e

f

g

h

White can play either 1 6.g3;!; or 1 6.h3!?, with the point that after 16 ... h4 1 7.ie 1 the h4pawn is exposed. I am not sure which I would play; it depends on concrete stuff that I have not considered. Either way, White is better. In general I do not like Black's position here. If he was able to play . . . if5xd3 he would get some relief. Then .. .f5 would follow and he would be able to block the position. But this is not going to happen. This is the spirit of this opening: it is really passive - Black gives White a space advantage in return for solidity. But okay, I do not want to make this book a "Beat the Chebanenko" manual, so let's leave it at that. I hope I will not scare people away from playing this against me!

White has won the opening battle. Black did not manage to solve the problem of his light­ squared bishop and is slowly getting squeezed. My next two moves follow from what came before, as A is followed by B .

15.h4 f5 16.g5

8

.:1 m:,,,.,,•� ,�i

�'i'lw� �'i'l � 6 �m'i%mrm'iy,Wf 5 �� rtJ-m\ � j,Jlr» � 4 1

'%

�m ,% �� �m,�,,,,

�%'"/Jtlw.,JtlWJjfj ���lZJif .��� ��1 %"";f" " �w�!n•� �����

3

2

'%

, , , ,J

a

�

.,,,,:

b

c

,,.J

d

e

f

g

h

16 ... hxg5 This might be a mistake, though I was entirely convinced he was going to play it during the game (as sometimes happens) . It was only when I started to analyse the game that I realized that the alternative was playable.

a

b

c

d

e

f

g

h

The big question is what White should do after: 1 6 . . . h5

Chapter 3

-

During the game I thought that it would give me an extra target, by way of a piece sacrifice (though this is by no means the only idea) . But from another point of view, maybe things are not that simple. Black can put his pieces on the queenside and be ready to meet the sacrifice on h5 with a counter-sacrifice: l 7.llie2 if7 l 8.llig3 e6 I 9.id2 E!fb8

� � s i. R ..... .� �WJ.,iii "· · ·

�

� •.�.� 1.�� 5 �� tw;·� r� r� "·· � ��-� : ·····����-�rfl.f!j 2 .. . . � �•. . . ;es Black can still put the rook on e7 and go for that defensive idea. Bur it is very rare that people do this in real life. They decide how to defend and then they stick with it. And so they should, unless they find our that they have missed something entirely.

33 �hS 34.1"1 1h3 This is the typical idea we know from Alekhine. The queen will be placed at the back where it is safest yet still effective. .•.

34 1"1e7 35.°1Wh2 1"1eb7 36.1"1h6 1"1e7 .•.

8

7

6 5

4 3

2 1

a

b

c

d

e

f

g

h

Positional Decision Making in Chess

1 26

It may seem that White has improved his position as much as he can, but this perception is wrong! At this point White has the chance to create a weakness on the other flank. Black cannot hold the a5-pawn and will have to move it.

37 .!LJe2! Efob7 38.tlicl ge7 39.tlib3 a4 After 39 . . . '!Wd8 40.'!Wd2 Black would have to play 40 . . . a4 anyway. Maybe it would make sense to lure the white queen away from the kingside, but it can always return to h2, so Wang Yue probably did not want to wait longer than necessary. •

40 . .!LJcl geb7 4 l ..ie2 ge7 42.tlid3

8

7

6 5

4

I ·

if··� � � � ��--·--'· j_ � - -��m�a��'�m�;=.� �£ �� �jJllrffA �� � · -�llA t" " '

3 � . . . . J�/�

2

1

a

�-���-� �� �---;�m �K\lffi )� ��W/A� ���Wf �@� b

c

d

e

f

g

h

From here the knight can go to both b4 and e5. My thinking is that it makes no sense to go to e5 at the moment, as the bishop on h8 can go nowhere. To allow Black to exchange it would be the equivalent of giving up a piece.

42 geb7 White has improved his position a lot and no further improvements seem possible. Perhaps the king would be safer on c l , but it seems pretty safe on fI as well. For this reason it is time to switch from move-by-move improvement of the position and calculate the consequences of various lines of action. •.•

I decided to break through on the kingside with an exchange sacrifice.

43.tlih4 This comes with the threat of lll xg6 followed by ih5 . And after 43 . . . l:!g7 44.lll e5 White is not only threatening lll hxg6, which seems to be j ust game over, there is also a sort of zugzwang that makes it impossible for Black to prevent ia5 , when White manages to improve his position once again. 43 ....ig?

8

7

6 5

4

3

2

a

b

c

d

e

f

g

h

This is entirely forced. It also forces White to do something. The waiting is finally over.

44J::lxh7! Both the bishops are bad, so it can be a little surprising to learn that the bishop on h7 is the better of the two. The reason is very simple: it has a function! It defends g6. By eliminating it, White can remove the g6-pawn and start to find a way in for the pieces. 44 tlixh7 45 .!LJxg6 Here Wang Yue came up with a strong defensive try that I had missed. I actually believed I was just winning. •.•

45 tlid7! •••

•

Chapter 3

-

Space Advantage

45 . . . 'Wxg6? 46.i.h5 wins for White of course. I believed he would have to retreat, after which I was going to play .ih5 and be doing very well.

46.J.h5 Black's idea is of course that after 46.!!xh7? 'Wxg6 White's advantage is not entirely conclusive. It still makes no sense for me to exchange any black pieces. 46 YHd8 The funny thing is that if Black plays 46 . . . lll df8 the knight on h7 is lost. Compared with the position after 45 . . . lll d7!, this time around there will be nothing hanging on g6, so White just wins the knight. 47.lll xIB 'WxIB 48 . .if3 and after 48 . . . ih8 then 49.g6 is the human move. •••

47.�b4 I think this is an instructive moment. We can now see the advantage of having spent 6-7 moves going to and fro, forcing Black to play . . . a4. At the time we might think that there is no need; we are going to decide the game on the kingside. But when you can improve your position, you should do so! It is rather simplistic to say so, but it is not about saying these easy things, it is about doing them.

1 27

This is the next critical moment. An amusing thing is that at move 40 my computer (at the time the game was played) gave me + 5 .00 (now it is "only" +3. 50) , but at the same time it could easily have been a fortress, making 0.00 the human evaluation (though I had no fear that this actually was a fortress) . It is therefore not relevant to consider this position from the aspect of computer evaluations. White is better, but he will have to do something to break through. This might reduce the computer's appreciation of the position, but winning the game is about delivering the deadly blow, not about mathematics. Again, this might be a simplistic thing to say, but I think that part of whatever success I have had in chess is because I have a decent feeling for when this sort of advice is useful and I follow it when it makes sense.

48.�h8!! This is probably the hi ghli ght of the game. Still, at depth 26 the computer does not understand this move. Partly I like this move for aesthetic reasons, but of course computers have no understanding of such things! 48 � dfS On 48 . . . @xh8 I had planned 49.if7! lll dIB 50.!!xh7t lll x h7 5 1 .g6. •••

8

47 J:k7 .•

7 6 5 4 3 2 a

b

c

d

e

f

g

h

Here Black has to play 5 1 . . . ih6 j ust to stay in the game. After 52.°Wxh6 !!xf7 53.gxf7 !!xf7 a

b

c

d

e

f

g

h

1 28

Positional Decision Malcing in Chess

54.lll xc6 'W'f6 5 5 .lll e5 our computer evaluation is down to +2. 1 1 , but it is enough! White will play 'W'h5 and .ih4 followed by 'W'e8t, c6 and so on.

49.tll f7! 49 . .if7t gives Black the choice of playing the same line as the previous note. Given the choice I would rather win the exchange back. 49 .. J�xf7 50 ..ixf7t gxf7 5 U:h:h7 \Wes 52.gh3

White is a pawn up, so It is time to exchange queens and broaden the fight on the queenside. The pawn on g5 will stay where it is forever, dominating the black pieces. This is winning slowly, which is a good thing! In such a position there is no reason to hurry. Actually, never hurry! It makes no sense to hurry, as it does not matter if you win in 10 moves or 30 moves. To win without counter-chances is far more important.

52 ... tll g6 53.'!We2 gc7 54.'!Wh5 ©f7 55.'!Wh7 '!Wg8 If Black tries to avoid exchanging queens, White wins by force: 55 . . . We7 56.lll d3 Threatening .ia5 . 56 . . . �a7 57.�h6 lll f8 58.lll e5t 'iii e 8

b

a

c

d

e

f

g

h

59.Wg8 .ixh6 60.gxh6 Wh7 6 1 .Wg5 followed by .ih4. The computer supports this variation, but it is quite human as I worked it out before I got approval!

8 7 6 5 4 3 2 1

�����___;��=-.e6!, when after 4 1 . f4 @d6 42.tt'le5 :B:e2 43.tt'lc4t i>c6 8

7

6 5

4 3

2

a

32 ... e5! 33.fxe5 fxe5 34.�8 E:xa4 35.�xe5t @f6 36.�c6 .ic5 37.E:xh7 E:a2 38.@8 a5 39.h4 a4 8

b

c

d

e

f

g

h

I see no reason why he should be any worse at all. White's play is not forced, but no really eye-catching option exists.

41 .E:a7 Maybe it was stronger to play 4 1 .tt'le5!? .ig7 42.tt'lg4t 'kt>f7 43.h5 gxh5 44.:B:xh5 a3 45 .:B:a5 .if8 46.f4±, but Black also has problems to solve in the game.

7

6 5

4

4I.. ..id6?! Here 4 1 . ..:B:xf2 42.tt'le5 :B:b2! was an interesting defensive try.

3

2

a

b

c

d

e

f

g

h

We are entering the more interesting phase of this game (stating this happily ignores the debate on how to get such a position against Anatoly Karpov) where the pawn structures have disintegrated entirely and the theme has dissipated. Bur chess has a tendency to always be instructive, so I feel it would be sad to ignore the nice little points we find in this endgame.

Chap ter

4

-

� ��������� �-� �

Either Black's passed pawn becomes strong or he gets the chance to exchange the rooks on b4, escaping into a drawn minor-piece ending.

6

�� �m-- - ���r� . .� �- �-���rii �� �� �� � ���� �� �� �� s

42.f4 f;ih2 43.ga6! ©f7 44.� e5t .lxe5 45.©xe5

4

8

7

6

..

b

a

5

c

d

e

. ..

f

g

h

If White plays f4-f5, Black can take on e4. And if White takes on g5 , Black will cake with check and quickly make it to the a-file and secure a draw from there.

4 3

2 1

141

Transformation of Pawn Structures

b

a

c

d

f

e

g

h

It turns out that this was the critical moment of the game, though neither of the players would have known this with any certainty while it was being played.

45 ... ©g7? Karpov could have held the game with 45 . . . gh3!! 46.ga?t @g8.

And after 47.@f6 13xe3 48.@xg6 13g3t 49.@f5 Black would even draw without the a-pawn, but 49 . . . a3 does not ruin his chances.

46J�a7t ©h6 47J�xa4 8

7

6 5

4 3

2 1

a

b

c

d

e

f

g

h

After 47.e4 13xh4 48.13xa4 Black draws easily. White has no way to make progress. If White manages to get his rook co f6, Black j ust has to keep the rook on g4 and be ready with . . . g5! .

a

b

c

d

e

f

g

h

47.. J�xh4 47 . . . 13h3!? is no longer sufficient to hold. White wins after: 48.13a3 @g? 49.13a7t @h6 50.e4 13xh4 5 I .gf7! Defending the pawn and preparing to advance the e-pawn. 5 1 . . .gh I 52.@f6 and White wins. 48. ©f6! gh5 49.e4 gh4 50.e5 gh5

1 42

Positional Decision Making in Chess

King walk to h3 (or a3)

8

7

As this is a chapter on pawn structures, that is of course our main focus, but it would be nonsensical not to draw specific attention to Rubinstein's king march - probably the most famous in chess history before Short Timman, Tilburg 1 992. It is something all the following champions knew well, as can be seen from their games, such as these two examples by the World Champions from 1 985 to 2006.

6 5

4 3

2

a

b

c

d

e

f

g

h

S l .e6 gst 52.©e7 ©g7 53.i>d6 gf'8 S4.ga7t ©f6 55,gd7 1-0 It may very well be that I am the only person outside Hungary who finds this game memorable (and now hopefully a handful of readers!) . I fear that to many it might come across as a bit dry. After all, White got a slight edge from the opening, put Black under some pressure and forced him to solve some problems. Having missed the "easier" ways to solve his problems, eventually Black had to find a study-like move in the rook ending (45 . . . 1':i:h3!!) in order to save the game. Presumably exhausted from four hours of intensive defence and with no one hinting that this was the moment when all energy had to be consumed on saving the game, the World Champion failed the task. I agree that this does not look as exhilarating as Kasparov's unruly demolitions of Karpov in the second half of the 1 980s; but it is also a valid way of playing and on this occasion it was enough to beat the strongest player in the world at the time.

Evgeny Sveshnikov - Garry Kasparov Minsk 1 979

I was following this tournament live in the playing hall. Every day I would go to watch the games and think along with the great players, at times exchanging variations with other spectators. As a student of Rubinstein I was not surprised by Kasparov's play in this game.

34 ...i.cS 35.i.xcS ©xcS 36.©d3 ©b4 37.©c2 ©a3 38.©b l aS 39.©al a4 40.bxa4 ©xa4 4 1 .©bl ©a3 42.©al b4 43.©bl b3 0-1 The second example comes from Kasparov's great rival, who had clearly studied his games, as well as Rubinstein's, of course.

Chapter 4

-

Transformation of Pawn Structures

Saleh Salem Vladimir Kramnik -

Doha 20 1 4

8

7

�xa 43.gxb5 g4 44J�d5 h5 45.�g3 h4 46.� e2 g3 47.fxg3 fxg3 48.b5 gf6 49.g5 gx5 50.ex5 g2t 5 1 [email protected] � d4 52.�gl �xb5 53.a4 �c3 54.�6t ©hi 55.a5 h3 0-1 I could give many more examples of this king march of course, but let us return to pawn structures before this diversion leaves us lost in the woods. Come on, off to Grandma's!

6 5

4 3

The pawn structure relating to the pieces

2

1

1 43

a

b

c

d

e

f

g

h

In a roughly equal position Kramnik activates his king and attacks the weak pawn on h2. White was not able to find a way to meet this in the game, showing the practical value of following in Rubinstein's footsteps.

30 ... @f6 3 U�a7 ©g6 32.� a2 � d8 33.©el ©h5 34.�cl ©h4 35.�b3 ©h3

36.� d4 f4 37J:ta8 ©xh2 38.�5 gf6 39.e4 �c6 40.gxa6 gg6 41 .©fl � e5 42.ga5

These days an average club player will talk with confidence about bad bishops, outposts for knights and so on, in the same way as a layman would talk about evolution or the expansion of the universe. But it was only in the last 1 50 years that these ideas were discovered and explained in an accessible way. A hundred years ago people would commit horrible crimes against their pieces, especially in closed positions. The following game is rather interesting in that respect. White might feel that he succeeded in giving Black doubled pawns, but all he really managed to do was to give Black control of the half-open f-line. Later on, realizing this, he exchanges his good bishop, sitting back with a rather impotent light-squared bishop. But this is not the end. Rubinstein then inexplicably allows the bishop to come back to life (maybe because of time trouble?) , something White does not manage to exploit fully. The game is quite a complicated one, despite the strategic aspects of it and the closed position. Analysing it was great fun and also quite thought-provoking. In the end we have a game that works on many levels: as an exploration of pawn structures and pieces, as an analysis of defensive resources, and finally as a captivating fight!

Positional Decision Makin g in Chess

1 44

Heinrich Wolf- Akiba Rubinstein Te p litz-Schoenau 1 922

l .e4 e5 2.ltla l!lc6 3.lilc3 lll f6 4 ..ib5 .ib4 5.0-0 0-0 6.d3 d6 7.J.g5 .ixc3 This manoeuvre was often played in the past. These days people are more likely to play 7 . . . lLie7!?, not fearing the doubling of the pawns in front of the king.

8.bxc3 Wfe7 9J�el lil d8 10.d4 i.g4 The more modern main line goes 1 0 . . . lLi e6 1 1 ..ic l c5 1 2 . .ifl , when White simply plays a long game with the two bishops. This was revived by Short in the 1 990s and he won some nice games with it.

improvement to the pawn structure seen from Black's point of view. In more games Black has played 1 3 . . . lLi e6, but after 1 4 . .ic l it is not dear that Black has improved his position.

14.l!lxg6 I feel that this is playing into Black's hands somewhat. A modern player would not rush to play this move, as the control over the f4-square should benefit Black. But as so often we find that things are not so simple. White had another option: 1 4.lLif5!? .ixf5 1 5 . .ixf6 'Wxf6 1 6.exf5

1 l .h3 J.h5 1 2.g4 .ig6

b

a

c

d

e

f

g

h

Both sides have their chances.

14... fxg6 1 5 . .ic4t ©h7 a

b

c

d

e

f

g

h

13.lilh4 1 3.d5! is the main line these days. White scores well, though there is no reason to consider Black's position that bad. 13 ... h6! A very interesting decision from Rubinstein, tempting his opponent to take on g6, giving him doubled pawns. But as we shall see, it quickly becomes apparent this is an

a

b

c

d

e

f

g

h

Chapter 4 - Transformation of Pawn Structures White is already facing his first difficult decision. Black's play is easier, but to speak about a real advantage would probably be misleading.

16 ..ih4! 1 6 . .ic l ? would be a bad mistake. If you try to move the pieces around a bit, you will notice that there are dark squares to defend on the kingside and that the bishop has no purpose on c l . I like Black's position; it is not so difficult to work out what he needs to do, while White has a big defensive task ahead of him. 16 ... g5 17 ..a.g3 The bishop is not great, but it is not as bad as some other historical examples. Three immediately come to mind: Winter - Capablanca, Hastings 1 9 1 9

1 45

Sasikiran - Anand, Hyderabad 2002

In all three positions the bad bishop was the decisive factor. Capablanca, Krarnnik and Sasikiran all won, though it could be pointed out that Kasparov found a superhuman way to build a fortress in Winter's position, as described in his My Great Predecessors Volume 1 . This is no criticism of Winter, but merely shows the unlimited possibilities in chess and how we often can find additional resources if we analyse a game deeply.

17 .. �f'l .

a

b

c

d

e

f

g

h

Short - Kramnik, London 20 1 1

a

b

c

d

e

f

g

h

1 8.°fff3 One commentator has quite understandably mentioned 1 8 . .ixf7!? as a serious option. The knights are truly scary the way they jump around on the kingside. But it is not clear that

1 46

Positional Decision Makin g in Chess

it is yet necessary to give up the dream of using the two bishops later on.

18 .. J:laeS 19.'1Ye3 b6 20.J.b5 gds 2 1 .a4 8

7

6 5

4 2

1

b

a

3

a

b

c

d

e

f

g

h

White does not have any clear targets, but with the advance of the a-pawn he aims to double rooks on the a-file and take on b6 to create counterplay. It is not much, but he needs to do something. It is also not as easy as it seems at first for Black to improve his position. Sure, he can bring a knight to f4, but it is not such a great achievement. Everything is protected.

2 1 . .. �hS I am not sure if Rubinstein considered 2 1 . . .a5!?. It is a move with both positives and negatives. It does halt the white advance on the queenside, but it also fixes Black's pawn structure permanently. As we shall see in the game, a lot of changes to the pawn structure can and will happen on both sides of the board. Most importantly, the b6-pawn would become weakened and Black would not be able to play . . . c6, which can be a serious option for him in the game. 22.a5 �g6 23.f3 � f4 24.i.fl @hs

c

d

f

e

g

h

Both players have to some degree executed their main plans for the early middlegame. Black has transferred a knight to f4 and will now have to find a plan for the next ten moves or so. White has a harder choice. He does not have any active plans of his own, so even though his position is objectively fine, it is a bit unpleasant. For example, if he plays 2 5 . h4 Black will reply 25 . . . gxh4 26.J.xh4 g5 27.ig3 h5, and White can start to regret his choice already. And if he prepares an invasion on the a-file, he will be hit hard in the centre: 2 5 .�a4? d5!+

a

b

c

d

e

f

g

h

Black takes over the initiative. The true dangers lie in variations such as 26.exd5 � 6xd5 27.Wxe5 Wd7! 28.�aa l �xh3t 29.J.xh3 �de8 and the queen is trapped.

Chapter

4 - Transformation of Pawn

For this reason the best move is probably the prophylactic 2 5 .Wf f2, trying to keep everything under control. But moves like these are not that easy to find, especially when you have a lot of options. Obviously Black still has a lot of possibilities and the game would continue. It is likely that Rubinstein would have enjoyed such a turn of events as well. We could imagine that Black would transfer the knight from f6 to g6 and maybe later to h4. But this would take time and White would be able to break through on the a-file. Probably this would give him enough counterplay. In the game White went for a more direct approach, which looks poor on the surface, though it actually has some merit to it.

25 ..ixf4?! If you were writing a manual for beginners in the 1 980s you would undoubtedly have criticized this move severely. White is giving up his good bishop, leaving him in a closed position with a bad bishop on fl against a powerful knight. Indeed it is a big concession to give up the dark squares. But luckily chess is more complicated than following basic rules. Yes, they do apply, but they are not the whole story, like gravity fails to explain all of physics. On the plus side for White, we have the closure of the kingside, where Black had serious ambitions, as well as the elimination of Black's best piece and his stronghold of the f4-square. All in all, not an equation which is easy to work out without a few variations. My first impression was that this is a terrible mistake; but now I don't think so. Chess is indeed a difficult game.

25 gxf4 26.Wffl g5 This is a natural continuation. It does not make sense to question it, although other options exist. •••

Str u c tu res

1 47

26 . . . h 5 ? 27.g5 lll d7 28.h4 would be a clear improvement for White. The bishop will be useful on h3.

8 � __ _ __ __ _ _ V, , ,,, ,, � , 7 - --��� 6 t--- - - �!?.'w�A !?.'A%.,'-31):.!?.'A '"'%% - --�w�-

� �� �� �w� ,v.

5

4

3

2

1

d � · . 'a � �ti'•ti'• ----% ' % � � [� � ·-� � � , ·� ""' �� � �%_ , ,,j,,37'6' '�� ,,.,,

�

a

b

� c

d

%

L,,,

, ,,,

��� = f

e

g

h

27.d5?? But this is a horrible mistake. White should have played 27.h4 with the idea 27 . . . h l lll e8 (28 . . . f4 29.ixd7 Wfxd7 30.gxf4 cxb4 3 1 .lll e 4! bxa3 32.Wfxa3 and White keeps an initiative) 29.bxc5 bxc5 30.J.e3 and again the knight does not make it to d6. 27 . . . lll a6 28.b5 lll b8 29.Wf c2 also favours White. 28.ie3 White has won the strategic battle for the c5-square. 28 . . . cxb4 29.axb4 lll d6 This anti-positional idea, suggested by the computer, is not what I would expect Grischuk to play. Anyway, after:

8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

27 . . . b5 (27 . . . lll e8 can now be met by 28.b4 b6 29.J.e3 with some pressure.) 28.b3;t;

27J;ael b6 28.�hl

a

b

c

d

29.b4 tll e8 30.bxc5 bxc5 3 l .�e3 tll d6 32.�xc5 l:%c8 33.J.xd6 ti'xd6

e

f

g

h

It is difficult to find a serious active option for Black here, and Grischuk decided to basically do nothing.

28 ...ti'h5 Another aspect is that both players were already getting low on time. I certainly had more than he did, but what is much more important is that my position is easier to play. Black will have to find a complex plan to create counterplay, while White can improve his position with simple moves. For this reason it also does not make sense to assess the position at this point. White is still improving his position, while Black is not. It makes more sense to decide later on how much better White is.

a

c

d

e

f

g

h

IfWhite takes on f5 (with either piece) , Black has counterplay with . . . ti'g6 and . . . !%xc4. 34.tll b5!? This very concrete line does not end with a clear conclusion. White could also choose to keep things fluid with 34.ti'd3! ?;!;, where one idea is 34 . . . e4?! 3 5 .tll xe4! and Black is under attack. 34 . . . ti'xd5t! 34 . . . ti'g6 3 5 .tll xa7 l:%c7 36.tll c6± is highly unpleasant for Black. 3 5 . cxd5 l:%xc2 36.tll d 6 !%f6 37.�xf5t �xf5 38.tll xf5 !%d2 My general feeling is that Black has good drawing chances, though White can keep putting him under pressure with accurate play.

The following lines show how difficult Black's position is; though it should be said that Grischuk was pretty optimistic, thinking his position was fine. 28 . . . h5 29.tll d l h4 only plays into White's hands after 30.gxh4. 28 . . . tll c7 This could have been tried, involving a pawn sacrifice:

b

39.l:%c l !

Chapter 4

-

Transformation of Pawn Structures

The deep point to Black's play is that 39.E:xe5 ? fails to 39 . . . E:b6!! 40.E:e7 E:bb2, and after 4 1 .tll e3 E:xh2t 42.i>g l E:he2 43.d6 E:xe3 44.E:xe3 i.d4 45.E:fe l E:d2 Black is going to make a draw. 39 . . . E:xd5 40.E:c7 i>h8 4 1 .E:xa7 Followed by g3-g4 to support the knight. The game goes on. During the game I expected him to play: 28 . . . i>h8 The main idea is to avoid all the tactics on f5 , though Black will not be completely successful in this aim. 29.tll d l tll c7 30.b4 cxb4 30 . . . tll e 8 does not really work this time around. White is much better after 3 1 .bxc5 bxc5 32.i.e3 tll d 6 (32 . . . Wb6 33.Wf2 does not improve things) 33 .i.xc5 E:c8 34.i.xd6 Wxd6, and either 3 5 .i.xf5 or 3 5 .tll e3! with a close-to-winning position. 3 1 .i.xb4 3 1 .axb4 would allow 3 1 . . . b5. 3 1 . . .E:c8 32.tll e3 e4 The following piece of analysis is quite fascinating, but I want to underline that it is unlikely that we would see such moves played in a game. Maybe the odds are I % that a top player would find all of this! But as these are the best moves, it would be strange not to include them.

33.E:f4!

1 63

White has a definite advantage and will start to pile up on the f5-pawn. However, he has to act with vigilance. If he hesitates his advantage could easily slip. Black can seek counterplay, but this could easily backfire in sharp tactics. My guess is that I would play 33.'Wf f2 if l had this position in a game. But here Black has an amazing way to gain counterplay: 33 . . . a5 34.i.d2 E:cf8 3 5 .tll g2 (35 .Wf4 tll a6 and the knight comes to c5) 35 . . . f4!! 36.tll xf4 E:xf4 37.gxf4 i.xh3 38.E:gl 1Mfh5 Black has good play on the light squares. 33 . . . tll b5?! A natural-looking move, but it allows a beautiful combination. The lesser evil is 33 ... tll a6 34.E:efl tll xb4 35.axb4 E:cf8 36.c5, which gives White a most pleasant position, although Black is still fighting after 36 . . . i.e5 .

34.tll xf5 !! This amazing tactic might have been difficult to find over the board but, with the support of an engine, everything is clear. 34 . . . E:xf5 34 . . . i.xf5 3 5 .i.xf5 E:xf5 36.Wxe4 tll d4 37.i.c3 and Black is pinned to death. 3 5 .Wxe4 tll d4 36.'Wf d3!! This quiet move decides everything. White is threatening E:xd4. 36 . . . Wf6 37.i.xf5 i.xf5 38 .Wfl 'Wfg6 39.i.c3 i.d3 40.Wf2 tll f5 4 I .E:e6 Wg5

1 64

Positional Decision Making in Chess

4 1 . . .%lfh7 42 . .Axg7t %lfxg7 43.�xf5 and White wins. 42.%lff3 .Ab l 43 . .Axg7t White wins on account of 43 . . . @xg7 44.�g4. All these variations show how Black is struggling to make his counterplay work.

29 ..ig2 'llYg6 Grischuk's main idea was to relieve the pressure on f5, which he considered the main problem. However, White has other options and continues to improve his position. 30.�dl Of course the knight would like to go to f4, but first we have to force Black to play . . . e4, which cannot be achieved without putting the knight on dl first. 30 ... �c7 3 1 .b4 8

6 1 5

�� •.t.E �e ,

"� �

� '0

�% '"//, ��

-----

�

-

;:,,:

, , ,,,;.:�---.�, ,,,;� ii� �% ,,�� �� �---- iD %- 8 �---�� �wdti%� 11�%� 11� '0

�

: rl'� ·'·'·Ii·�� �-�� ,� �

2

a

b

c

d

e

f

g

h

3 1 . .. h5? This is a committal decision, though at the same time a very understandable move. Black is looking for counterplay and advancing the h-pawn is a natural way to achieve this. But in reality, there is no counterplay and the h-pawn simply becomes a target. In the game I managed to win it outright. This is not easy to foresee in time trouble. To call it impossible is

maybe too much, but at least it was too difficult for one of the most gifted players of our time! The problem is of course that when you are short of time, you usually calculate concrete variations, but do not pay too much attention to the strategic aspects of the position. And changes in the pawn structure are all about long-term strategic potential. We have already encountered 3 1 . . .lll e8?! 32.bxc5 bxc5 33.ie3 lll d6 34.ixc5 �c8 35 ..Axd6 %lfxd6 36.lll e3 in almost identical circumstances. Black does not get a better version this time around, but loses the f5-pawn as usual; and with it the game. However, Black probably should have tried: 3 1 . . .e4! ? This has the idea o f defending the c5-pawn with . . . .id4. 32.bxc5 The forcing line 32.if4 lll e8 33.lll e3 lll d6 34.bxc5 bxc5 35 . .Axd6 %l/xd6 36.�xf5 ixf5 37.lll xf5 �xf5 38.ixe4 %lfg6 39.g4 %lfxg4 40.�fl @h8 leaves Black with real drawing chances due to the opposite-coloured bishops. 32 . . . bxc5 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

33.ie3 A reasonable alternative is 33.lll f2 .id4 34.lll h3!?, whereas 33.ic3 lll e8 34.�f4 lll d6 shows exactly what Black is hoping for. 33 . . . id4

Chapter 4

-

After 33 . . . Wd6? 34.g4, Black's centre collapses. At this point it is not obvious how White should continue. He has a lot of pressure, but no clear-cut way forward. What is important is that the �f4-h4 manoeuvre is less effective, as the pawn has not gone to h5 and become weak. The strongest seems to be: 34.Wd2!? Wg7 3 5 .�f4 llJ e8 36.�h4 i.xe3 37.Wxe3 White maintains some pressure.

32 ..ic3!? I played this practical move without going too deeply into the position. I obtain fine squares on e3 and f4 for my pieces, and improve my position easily thereafter. However, it turns out that 32.llJf2! with the idea of llJ h3 was even stronger. Black will later be forced to play . . . e4 all the same and the knight can emerge on f4 as well as g5.

8 7 6 5 4 3 2 1

1 65

Transformation of Pawn Structures

32 . . . Wd6 is not only in violation of the concept of getting the knight to d6; after 33.We2 White is winning a pawn.

33 .ixg7 �xg7 34.bxc5 bxc5 35J�f4! White is reorganizing his pieces in accordance with the new pawn structure. •

35 llJ eS 36."f;Yfl � d6 37 .ifl •••

8 7 6 5 4 3 2 1

•

a

b

c

d

e

f

g

h

White has won the strategic battle. The pawns on c5, f5 and h5 are all targets, while White's pawns are all safe. At the same time White's piece are gravitating towards active squares, from where they will attack the black pawns, while Black is struggling to find good defensive positions for his pieces. Still, winning the strategic battle only matters if you deliver a good technical performance thereafter.

37 Jks? This move is maybe not objectively bad, but then the game was not played by two computers. After the game, Grischuk suggested 37 . . . Wf6 as an improvement, forcing White to make difficult and committal decisions at a point when I was running out of time. The correct way to play is easy to find after the game: 38.Wxc5 �c8 39.Wb4 a5!? Black definitely has some counterplay here, but it does not fully work. ••

a

b

c

d

e

32 ... e4 Black has no alternatives.

f

g

h

32 . . . h4 33.gxh4 e4 34.�e3 is not an improvement for Black. White would put the rooks on g3 and f4 and the knight on e3, squeezing f5 hard.

Positional Decision Making in Chess

1 66

�i �� ��-0� : ��� �, ��-� �% - '8. � .,%.. i B. !n f� • �. %�. . %����•��. . •Y.wr � . . �� � ,�%� � 8

5

4 3

2

..... .

a

•••••

b

c

d

e

f

g

h

Play could continue 40.Wfb3 Wfa l 4 1 .i.e2, when White has a healthy extra pawn. The main line then goes: 4 1 . . .Wfc l 42.Wfe3! Wfxe3 43.tlJxe3 ©f6

� � �i,� r. �� � � � . -�. �� ' 8· '�:.�• • .�. %. !.r� �m. . .v,�-� � . . . %�� ���%1·. . . ·��· · ·"�. . . � � �

6 5

4 3

2

a

b

c

d

e

f

g

h

44.l'l:c l ! l'l:c5 45 .l'l:b l i.c8 46.i.xh5 l'l:b7 47.l'l:xb7 i.xb7 48.i.e2 White now has two extra pawns, though the conversion will require some accuracy. And White had to find all the best moves to get here!

38.i.e2 I suspect that in his time trouble Grischuk missed this idea. 38 .. J�f6 38 . . . ©g8 was again suggested by Grischuk at the press conference. I was planning to play 39.l'l:h4 l'l:h7 40.tiJe3 Wf g5 4 1 .tlJg2!, when the knight is coming to f4. I consider Black's position very difficult.

39.gh4 �e8 40.�f4 Simple prophylaxis, although there is no big reason to avoid 40.i.xh5 Wfe5 4 1 .Wfe3! , when White is completely winning anyway. 40 ... gh6 8

7

6 5

4 3

2

1

a

b

c

d

e

f

g

h

We had passed the time control and decided to spend a good deal of time looking for a clear win, which I succeeded in finding.

41 .gxh5! gxh5 42.i.xh5 �e7 43.g4! The key move that I had to find and evaluate accurately. 43 ... gm It is difficult for Black to find a move, as White is threatening to launch a powerful attack. 43 . . . l'l:h8 44.tlJe3! is also excellent for White. And after 43 . . . fxg4 44.i.xg4 l'l:f8 45.Wf g3, Black is just lost.

44. gx5 Lf5 After 44 . . . tlJxf5 45.l'l:gl t 'k!lh8, White has a choice of strong moves. I was contemplating 46.tlJe3 or 46.i.g6. 44 . . . l'l:xf5 looks natural, but White has a forced

Chapter

4

-

Transformation of Pawn Structures

win: 45.�g l t lt>h8 46.°Wh6t and White can either take on h7 or d6 on the next move.

1 67

54 ... �f7 55.h4t ©h5 56.�6 �e5 57.d6 ©g6 58.©f4 ©f6 59.h5 ©e6 60.h6 e3 6 1 .�xe3 ©xd6 62.©6

45J�gl t ©hs 46.Vfh6t .ih7 47.�e3 gf6 47 . . 'Wf6 48.�g6! and Black would have had to resign. .

1-0

Keeping the tension

48 ... gxg6 49.i.xg6 Yfg7 50.Yfxg7t ©xg7 5 1 ..ixh7 ©xh7 52.©g2 ©g6 53.©g3 ©g5

a

b

c

d

e

f

g

h

54.h3! Not the only winning continuation, but the most straightforward. Black has to make a concession.

There are times when it is important to keep the tension. This happens when all the ways you can change the pawn structure are undesirable to you, while the opponent cannot change the pawn structure to his advantage either. This might sound like a rare case, yet in my experience it is anything but. It seems quite common chat both players will try to limit the positive effects of changing the pawn structure available to the opponent. Indeed, outplaying the opponent often means that you combine your ideas in a way that forces him co allow you to change the pawn structure in your favour. It can also mean that you have managed to set the opponent plenty of practical problems and chat he gives into the pressure, only to make a change in the pawn structure that is co your advantage. The latter is what happened in the following complicated game. The possible changes in the pawn structure between moves ten and twenty were numerous, but not many of chem were

1 68

Positional Decision Makin g in Chess

to Black's advantage. In the end he did not manage to find the subtle move that would have kept the tension, which is in some ways quite forgivable, as I did not find it during the game either. But on the other hand, not finding it meant that he lost the game rather quickly thereafter.

Boris Gelfand - Boris Avrukh Ramat Aviv 1 999

My opponent is a well-known author of chess books. His first book, Grandmaster Repertoire 1 J.d4 Volume One became an instant classic upon its publication in 2008 and, I would dare to say, elevated chess books to a formerly unseen level. These days he is more a trainer and writer than a player. He has a very strong classical chess education and strives for dynamic play always, bur based on a solid positional basis. It is an attitude I really like and share, which is one of the reasons I have enjoyed analysing with him on many occasions. Avrukh has beaten many strong players and has been a part of the Israeli national team for many years, being a member of the silver-medal winning team of 2008, as well as the winner of individual gold medals in the 1 998 and 2006 Olympiads. Boris has worked with a lot of the top players: Caruana, Radjabov, Kramnik (before the 2006 match with Topalov) and of course with me on many occasions.

League. But even this does not happen often, and it is rare that the local chess fans have a chance to see me play in Israel. There are some recent signs that things could be improving, but with the past we have had, I am only a reluctant optimist. In my first year in the Israeli League, I won against many strong GMs: Sutovsky, Milov, Avrukh, Har-Zvi, Manor, and drew with Zilberman. It was my best result in the league. I have never played as many as six games again, and never as successfully. Sometimes I made 212, but it is really not the same.

-

This game was played in the Israeli Team Championship, which is more or less the only event in Israel where all the professional players play. Even though there are many grandmasters in Israel, there is not a great chess scene and it is very rare that there are any strong events. When I moved to Israel in 1 998 the only event that made sense to play was a rapid event in Haifa (which ended in 2000) and the Israeli

1 .c4 Avrukh plays the Griinfeld a lot. This was an early morning game, as are the majoriry of games in the Israeli League, so I decided to avoid a theoretical discussion. These days the English Opening is very popular, starting with I .4Jf3 or l .c4. I assume people got fed up with the enormous amounts of theory elsewhere and just wanted to play chess. 1 . tlif6 2.tll c3 dS 3.cxdS tlixdS 4.g3 g6 s.i.g2 tli b6 6.tlio i.g7 7.0-o 0-0 s.d3 tli c6 9.i.e3 eS .•

a

b

c

d

e

f

g

h

This position is similar to a harmless variation of the Sicilian Dragon, where White

Chapter 4 - Transformation of Pawn Structures

1 69

has an extra tempo. Also, in those lines it is standard to put the knight on the preferable square e7, rather than b6.

10.b4 A rare move, but in my opinion the most logical one. When I returned home after the game I was surprised to find only one game in my database with it; but it was a game of the great Leonid Stein, a player I admire a lot. He died in 1 973 at the young age of 38, in tragic circumstances. In his best games we can see how great this loss was. He was truly a fantastic player. When I came to Israel I rented my first apartment from a chess player. After I moved on, it was taken over by a young man who got in trouble with the landlord for playing guitar in all the waking hours of the day. At some point he was called to the landlord's house for a "conversation". When he arrived he saw a chess set and said, "Oh, my uncle was a chess player." "Who is your uncle?" "Leonid Stein." As you can imagine, the attitude of the landlord changed immediately. He stopped throwing epithets at him, although his appreciation for the young man's musical "talent" probably did not increase much.

10 . . . e4? l 1 .lll xe4 .ixa l 1 2 .VNxa l would leave Black desperately weak on the dark squares. He is probably already lost. 1 0 . . . lll xb4? l l ..ic5 does not give Black enough compensation for the exchange, as the semi-open b- and c-files will be useful for the white rooks.

l l .a4 My plan was to advance the pawns, in order to create pressure on the queenside.

Alex Huzman goes to a dentist called Rubinstein. Once I prodded him, asking why he did not go to Tal's son, who is also a dentist and lives in the same town as Alex. "I am a bit afraid of his style. It is better to stay with Rubinstein." There are many famous Jewish chess players; and many of their descendants live in Israel . . .

1 0 ... � d4 1 0 . . . �e8 l l .b5 lll d4 1 2.a4 a6 was played in Stein - Savon, Riga 1 970.

a

b

c

d

e

f

g

h

l l . ...ie6 The most ambitious move, and a good one too. Black had other playable options which

1 70

Positional Decision Making in Chess

we analysed, which I j ust want to show quickly for those who might be interested. 1 1 .. .c6 1 2.lll d 2 .ie6 looks fine. l 1 . . . .ig4 1 2.a5 ixf3 1 3.exf3 lll c8 1 4.a6 bxa6! is quite unclear, but maybe White can fight for an advantage. l 1 . . .lll xf3t!? 1 2 . .ixf3 c6 1 3.b5 lll d5 1 4 . .id2 lll xc3 l 5 . .ixc3 c5 looks easier to play for White, but to call it an advantage might be over the top.

12.tll g5 The critical attempt, attacking a pawn and the bishop. Such moves beg to be played.

1 3 ... a5!? As we shall see below, this was not the most precise way to play. Another idea here was 1 3 . . . .id5! , when White has two main options: 8

7

6 5

3

4

2

i. �

12 ....ib3 1 2 . . . .id5 was also possible, but after 1 3.lll xd5 lll xd5 1 4 . .id2 White would claim a slight edge with the two bishops. This was not Black's idea, of course. 13.flYb l 1 3 .'Wd2 .id5 would equalize; White cannot gain the bishop pair as he did in the previous note.

,.,J'

b

6 5

1

2

a

b

c

d

e

f

g

h

e

f

g

h

. . . . . .J.�• .,'.�. .� • �. %

%

�ffl, A �;�

�,d'

W � �Ll�

. . �,� �a:tm . . a

3

d

c

it. � f�. . . TI � . . . � . %./. . ; 3 �!�-�.lO� . �.,. . . ;_. 8 � . %�

�

2

4

�

a) 1 4.ixd4 This is a critical move that has co be checked, though it does not really work. 1 4 . . . .ixg2! The big test. 1 4 . . . exd4 1 5 .lll xd5 lll xd5 1 6.lll e4 seems to give White a bit of pressure. 1 5 . .ixb6 1 5 .@xg2 exd4 1 6.lll ce4 lll d5 and Black is at least equal. 1 5 . . . .ixfl 1 6 . .ie3

4

5

�

'z

a

1 2.lll d2 lll d5 would equalize immediately. This is Black's main idea; he saves a tempo compared co the variations where he plays . . . c6 first.

%� �·

i ,... %�. . .,v.. · · "� . . . %.. �� . . . ;�. ,.�� .� . �.i'l�· /�� !�. . . %,, -. . . :. . . . ;-m.·.� . m 8 � � ��.,. . . ;a·8"�� � � . . �;� •!m . . ·�

b

c

d

e

f

g

h

1 6 . . . h6! 1 7.lll xf7 !!xf7 1 8 .'Wxfl The point is that Black is in time with: 1 8 . . . e4! If White were able to put the knight on e4, he would have an advantage.

Chapter 4

-

171

Transformation of Pawn Structures

1 9.d4 White can of course cry 1 9. tll xe4 ixa 1 20.Wfxa l , but I cannot see this as playing for an advantage. Things are by no means simple, of course, but if squeezed, Black should be able to give up an exchange in order to neutralize the pressure. 1 9 . . . Wfe7! l 9 . . . ixd4? is not viable, as 20.E:d l E:d7 2 l .Wfh3! would be dangerous for Black; the direct threat is tll b5!.

l 7.tll c5 and the threats on e6, b3 and b7 are too much. b2) 1 4 . . . c6 1 5 .a5 ixe4 1 6.ixe4 tll d7 is of course playable, but I like such positions with the lasting impact of the two bishops. b3) 1 4 . . . a5 gives White an extra option compared to the game: 1 5 .bxa5!? E:xa5 1 6.tll x d5 tll xd5 1 7.ixd4 exd4 1 8.Wfxb7 tll c3 l 9.tll xc3 dxc3 20.E:a2t

7

6 4

3

2 a

b

c

d

e

f

g

h

The position is complex, and Black is at least fine. b) For this reason, I planned to play 1 4.tll ge4.

1

a

b

c

d

e

f

g

h

It would be up to Black to prove that he can hold this position - and it will not be easy to do so!

14.b5!? After this White has a pleasant position, which makes it an easy to move play quite quickly. But when you analyse the game, it makes sense to look at less logical options as well. White had a sharp option in: 1 4.ixb7! E:b8 1 5 .bxa5

a

b

c

d

e

f

g

h

And now the question is whether Black can find something other than 1 4 . . . a5 offering a transposition to the game. b i ) 1 4 . . . f5?! 1 5 .ixd4 exd4 1 6.tll x d5 tll xd5 a

b

c

d

e

f

g

h

1 72

Positional Decision Makin g in Chess

Black can play for compensation with l 5 . . . E:xb7 1 6.axb6 E:xb6 1 7.a5 E:b4!?, or with 1 5 . . . liJd5 1 6 . .ixd5 .ixd5 1 7.°Wd l .ic6. In both cases, Black will have to fight hard in order to make up for the lost pawn. We analysed it for a while and did not find a clear-cut way for Black to play. So, for this reason I think I missed a chance to make Black's life a little more difficulc. le is now sixteen years since this game was played, and I do not remember what I was thinking during it. But it does seem that I was affected by the early hour, and for this reason rejected these complications on principle. I have co say that I do like White's position anyway; I think he has a strategic advantage, and chat for that reason I do not want to press the chaos button. The way the game went proved that I was right from a practical perspective, while the chess could be evaluated differently. But we are back co the main principle of the squeeze: it is much easier co play White's position!

14 ... c6 The natural move, defending b7 and supporting the d5-square. 1 4 . . . liJxa4 is a nice tactic, but the resulcing position after 1 5 .liJxa4 .ixa4 1 6.E:xa4 liJxe2t 1 7.ltih l liJ c3 1 8.°Wc2 liJxa4 1 9.'Wxa4 is horrific for Black.

.i � »,, ,j,% � · � � �� : ,!��l jfj � 5 �8• � r� 4 -lr� �,,, -,,,�---ltS� - - - :w.-0 �?3r0 -� 3 � .i. � 8 � r� � /. , , ; � li -�JZ� �1� �f'=' 8

�

,,,,,

-�-----:�

?3f%

V,� -�

�

2

I

b

a

d

c

f

e

g

h

1 5 ....id5 Again the most logical move. The bishop gees out of harm's way and contests the diagonal. 1 5 . . . cxb5?! This is quite a direct way of playing, but White has great dynamic potential that is revealed in the following lines: 1 6.liJxb5! I 6 . .ixd4 exd4 I 7.'Wxb3 dxc3 1 8 .'Wxb5 liJd5 1 9.E:ac l E:c8 20.liJc5 may look good for White, but Black can equalize with energetic play: 20 . . . 'We7! 2 1 .liJxb7 liJ b4 22.liJxa5 liJ a2 23.E:c2 liJ b4 24.E:cc l E:b8!? 2 5 . liJ c6! liJxc6 26.'Wxc6 'Wxe2 27.a5 'Wxd3 28.a6 With a draw coming. 1 6 . . . liJxe2t 1 7.c;t>h l

1 4 . . . i.d5 1 5 .liJge4 c6 transposes co the game. 14 . . . liJdS?? would unexpectedly strand the bishop in foreign lands after 1 5 ..ixd4 exd6 1 6.liJce4 and White wins.

15.�ge4 le is coo early to change the structure: after 1 5 .bxc6 bxc6 1 6 . .ixd4 exd4 1 7.'Wxb3 dxc3 1 8 .liJe4 liJd5 White has no advantage at all.

a

b

c

d

e

f

g

h

1 7 . . . i.d5 1 7 . . .i.e6 leads to a spectacular forced variation after 1 8 .E:e l , although this is not

Chapter 4

-

Transformation of Pawn Structures

1 73

White's only option. 1 8 . . . lll d4 1 9.lll xd4 exd4 20 . .ig5 f6 2 1 .lll xf6t .ixf6 22.:gxe6 i.xg5 23.:gxb6 :gxf1 24.:gxb7 :gc8 8

:

5

� i ��.1 L-�

�.

� � �r.-.,� -� • ��� ��-��

'8�. �- -� 3 � �-8�- �� 2 �--� - - � � · - - -�� �� -� ��

4

a

b

c

d

e

f

g

h

25 .�b5 Black is still a good way off equalizing. The difference in activity with better white pieces and a safer white king is quite significant. 1 8 ..ic5 :gc8!? l 8 . . .:ge8? l 9.lll bd6 is a disaster. 1 9 . .ixf8 .ixf8 20.�b2! lll d4 2 1 .lll xd4 exd4 22.�b5! After 22.:gfb l ?! .ixe4 23 . .ixe4 .ib4! Black has decent compensation.

a

b

c

d

e

f

g

h

1 6 J!ibS! Intuitively I find this move a bit passive, but all of Black's problems are on the b-file, meaning that this obscure move is probably best. ..

1 6 . . . :gc8 1 7.�b2 with the idea of :gab l would expose the b6-knight somewhat, and mark the last black move as pointless. Actually, my engine has . . . :gbg as a serious candidate in this position, showing that the rook probably belonged there in the first place. 1 6 . . . f5?! This would involve Black taking action before he is ready. l 7.lll g5! The critical move. 1 7 . . . ixg2 1 8.'it>xg2

a

b

c

d

e

f

g

h

The queen is quite active, and Black has to keep his position together while trying to prove compensation for the exchange. In short, I like White.

16Jkl 1 6.�b2!? was also possible, but there is nothing wrong with what I played.

1 74

Positional Decision

This is quite unpleasant for Black. The white queen will come to a2 soon and the following trick does not work: 1 8 . . . We? Instead Black would have to play unnatural moves such as 1 8 . . . tll d 5 1 9.Wa2 ih6, when both 20.h4 and 20.tll f3 look promising for White. It is no big surprise that White is better here, in view of his superior mobilization and the way the black pieces are pinned. 1 9.bxc6 bxc6 20.Wxb6! !%fb8 2 1 .tll d 5! cxd5 22.!%c7 !%xb6 23.!%xe7 Black is struggling in the ending.

Maki n g in Chess Well, to be honest, I am not sure if I made it all the way to the end during the game, or if we made it together in the post-mortem analysis. 1 8 . . . f5 ?! Today we are able to analyse this very accurately with engines. 1 9.bxc6 The critical test. l 9.ixd4? would be a big blunder after l 9 . . . exd4 20.tll x d5 cxd5+, when Black is on the way to c3 via a4. 1 9 . . . bxc6 1 9 . . . ixe4 20.c? Wxc7 2 1 .dxe4 and White has decisive threats. 20.ixd4 exd4

17.�b2 �e7 A prudent move. 1 7 .. .f5?! 1 8.tll x d5 tll xd5 1 9.tll c 5 is pleasant for White, although not a complete disaster for Black.

1 8Jfabl

D

8

. . . . . v.

7

6 a

b

c

d

e

f

g

h

White has achieved the optimal placement of his pieces. Black will have to react to this, which Boris did not do so well.

18 ...J.xe4? During the game I thought that Black was able to solve his problems in a long forced line.

5

4 3

a

b

c

d

e

f

g

h

This arises by force. White now has two serious options that I have analysed deeply.

Chapter 4

-

1 75

Transformation of Pawn Structures

a) The logical move to calculate during the game is the forced line starting with: 24.:B:xc6 fxe4 25 .:B:cc8 Here Black saves himself with: 25 . . . lll d! 26 . .ixe4 lll xe2t 27.�g2 lll c3

a

a

b

c

d

e

f

g

h

I have analysed this and have to accept that White has no advantage here. 28 . .ic6 There are other moves, but I j ust want to give the general idea of why Black draws. 28 . . . 'Wd6! 29.:B:xf8t 'Wxf8 30.:B:xf8t �xf8 3 1 . 'ii f3 'ii e7 32. 'ii f4 'itid6 33 . .ie8 �e7

b

c

d

e

f

g

h

White has to repeat the position, as 34 . .ib5?? lll x b5 3 5 .axb5 a4 36.b6 'ii d6 wins for Black. b) 24.lll d2! lll c3 In my notes this was mentioned as giving Black enough counterplay, but for this book I looked deeper, discovering a clear path to a big advantage. 24 . . . 'Wxe2

c

d

e

f

g

h

25 . .ixd5t! cxd5 26.lll fl and White wins a piece. 25.:B:e l Black is struggling. Th e remaining lines are illustrative, rather than forced. 25 . . . 'Wc7 26.:B:a8

a

a

b

b

c

d

e

f

g

h

26 . . . 'Wb7 26 . . . �g7 27.:B:a6 lll d5 (27 . . ..ib4 28.:B:xc6 'We7 29.e4 also gives White a big advantage) 28 . .ixd5 cxd5 29.lll f3 .ib4 30.:B:fl and Black has problems defending his pieces. 27.:B:xa5 'Wb2 2 8 .lll c4 lll xe2t 29.�fl 'Wc2

Positional Decision Making in Chess

1 76

30.ixc6! Control over the long diagonal (ending in the vital square a8) is more important than anything else. 30 . . . ib4 3 l .!!xe2 Wd l t 32.@g2 Wxe2 33.!!b5 ie l 34.!!b2 Wxd3 3 5 .id5t 'it>g7 36.a5 Black has to give up a piece in order to defend a very bad endgame, probably without real hope. It was only when looking at the game recently that I realized Black had 1 8 . . . ie6!.

19 ... cS This is a big concession. The bishop on g7 can now safely be called "bad" . White's light­ squared bishop is very strong, leaving him with a pleasant position to play. 1 9 . . . We6 20.bxc6 bxc6 2 1 .Wd2 is not devastating by any means, but White would be happy to have the two bishops as a starting point. By the way, 2 1 . . . lD b3? does not work on account of 22.!!xb3 Wxb3 23.!!b l , winning a piece.

20.i.f6 40.gas g5 41 .ghs ©g7 42.gh7t ©gs 43.gd7 1-0

23 ... �e7 24.J.g5! Although positions with opposite-coloured bishops are at times exceptionally drawish, we

What went wrong for Black in this game? Basically he could not find a way to maintain the construction with a pawn on c6 and bishop

Positional Decision Making in Chess

1 78

on d5 and eventually gave in, taking on e4 and played . . . c5, after which his problems took a more permanent form. With time trouble on top of this, he simply collapsed.

Fighting for key squares When there is a difference in strength between the contestants, even when it is as small as in this case, it is quite common that one of the players does things he normally would not do unless forced to. To some extent this is what happened to Avrukh in the previous game; the pressure became too much and he decided to make a concession, as he could not find any alternatives. And to be fair to him, there is no guarantee that such an alternative exists when you are sitting at the board. When players of equal strength face one another, it is common that they fight for key squares, preventing each other's plans and ideas while creating ideas of their own. The outcome of the game is to be found in the depths of long complicated lines and the two players' willingness to enter lines where they are unable to determine the final outcome in advance. The following game is one of the biggest victories in my career, against one of the greatest players I have had the pleasure of competing against.

Levon Aronian Boris Gelfand -

Dresden (ol) 2008

This game was played at the 2008 Olympiad in Dresden, where I and the Israeli team had our greatest result in team chess ever, claiming the silver medals. I had a really good score on Board 1 and missed the gold medal by the narrowest of margins. With a rating performance of 2833, I fell short of Peter Leko's 2834 gold-winning performance! Still, it was an unbelievable result. The silver medal

was much more important, and of course much more difficult to get than the gold medal I achieved with the Soviet Union. The greater importance is not because I am now an Israeli, but because the Soviet team in 1 990 was clearly superior to all the other teams in the event. Important results in 2008 included our victories over Armenia and the Netherlands. Unfortunately we lost to Ukraine, missing out on the gold medal. Levon Aronian is of course one of the greatest players of our time. However, in team events he is usually even stronger. Due not least to his spirit and leadership, the Armenian team managed to win the 2006, 2008 and 20 1 2 Olympiads, and the 20 1 1 World Team Championship, as well as achieving bronze in the 2007 European Team Championship and the 2004 Olympiad. Armenia also won bronze at the 1 992 and 2002 Olympiads, though without Aronian on the team. However, it was from 2006 when Aronian entered the elite and took over Board 1 from Akopian that the team became the most successful in the world. It would not be an exaggeration to say that Aronian is the soul of the Armenian team. I consider him to be the most creative player of our time, both over the board and in opening preparation. So far he has failed in the Candidates tournament, but I hope that Anand's and my results in the Candidates in recent years will inspire him to keep on trying. He has so many qualities that I would not be surprised if he manages to become World Champion one day. I really feel that as long as one wants to work on chess and has the will to progress, there is a possibility of doing so. These days when you read older players complain about younger players that they only know how to push the space bar and have no culture, just think of Aronian. He has a great knowledge of music, literature, arts and culture

Chapter 4

-

Transformation of Pawn Structures

in general. This is another reason why I would be happy if he kept on improving and reached the highest peak. What a World Champion he would be! In a recent interview in New in Chess he said chat whenever he plays me, he tries to impress me. It is friendly competition. We have warm personal relations and work on chess together occasionally. So in every game we try to bring something new. If one player comes up with a good novelty, it is great! This is one of the reasons why we have so many decisive games with each other. It is always an open game where both players want to win.

1 79

Ac this point I avoided 6 . . . E:e8 7.tll d5, as Rubinstein played (see page 1 00 ) . Of course White could have played the knight advance on the previous move, but then Black would not be forced to put his rook on e8 .

7.bxc3 ges 8.d3 I should say chat these days 8 .tll e l is quite popular. Things change and new ideas emerge. It is a lot about keeping the pawn structure flexible. 8 ... e4 9.lLJd4 exd3 10.exd3 lLJxd4 l l .cxd4 d5

1 .c4 e5 2.c!ll c3 lLJf6 3.lLJf.3 lLJc6 4.g3

s �..tSeB �.I 7 &rtm'•"-''Y.--, 6 50 4 3 � �m ,� B'Llt [@%1'�----�m-� rwm_�,,,,,; 2 OA ifiO>< � iO>< if� O A iO>< iffl� iO if{f< 1 �----"-Vm1£B"�

.i

�- ,� " '���il��-�� ,� , ,���� /,� �/, '"//, �� ��-� a

b

c

d

e

f

g

h

A natural choice for Aronian. With this type of pawn structure it is likely chat we will have a long complicated fight, based on deep strategic play. I also have some experience with this system with White, but nowhere near as much as Levon.

4....ib4 In another game I played 4 . . . tll d4 against him. I have also tried 4 . . . d5, getting a good position against him. 5 ..ig2 0-0 6.0-0 hc3

a

b

c

d

e

f

g

h

My preparation was not especially deep. Before the tournament my teammace Boris Avrukh had suggested this system as a possible way co play and I looked at it a little bit. My analysis was really not chat deep, but it seemed a viable option and I went for it. Obviously the main point is chat White has doubled pawns and Black is blockading chem.

1 2J�bl It is not obvious chat this inclusion favours White, but it is quite logical all the same. The following game by the extravagant Michael Basman against former World Champion

1 80

Positional Decision Making in Chess

Mikhail Botvinnik shows some of the strategic ideas of this system: 1 2.J.e3 J.e6 1 3.!!c l h6 1 4.h3 c6 1 5 .%Vd2 %Vd7 1 6.©h2 ©h7 1 7.l:!fe l tlig8 1 8.g4 tli e7 1 9.J.f4 !!ad8 20.c5 g5 2 1 .J.e5 f5 22.gxf5 J.xf5 23.!!e3 tli g6 24.!!ce l l:!e6 25 .J.g3 !!de8

15.h3 Aronian is not seeking an instant clarification in the centre, hoping instead co gee chances later on in the game. One of the main reasons for this move is co cake away the g4-square from the black knight; another is co improve the position with f2-f4 and g3-g4 in order co put pressure on the pieces defending the centre. 1 5 .%Va4 had been played in Norri - Tella, Finland 1 99 5 , when it is not clear whether the bishop is best placed on e6 or f5 .

12 ...h6 Reducing the bishop's scope. The natural idea is now co bring the bishop co e5 as Levon did in the game.

Aronian was not attracted co winning a pawn as Black obtains good counterplay: 1 5 .J.xf6 %Vxf6 1 6.cxd5 cxd5 1 6 . . . J.b?!? 1 7.J.xd5 !!b8! This patient move is the right way co play. White manages co gee some real pressure after: 1 7 . . . J.h3 1 8 .J.xa8 J.xfl 1 9 .J.e4 J.h3 20.%Vh5 J.d7 (20 . . . J.e6 2 1 .%Ve5! and White is better) 2 l .%Vd5 %Ve7

13.i.f4 b6 14.i.e5 c6 An obvious pawn sacrifice, which we will investigate below. 8

7

6

5 4 3

2 1

a

b

c

d

e

f

g

h

White has a real initiative here. With no moves being forced, it is hard co say what should be played, but a possible line is 22.%Vb7 i.h3 23.%Vxe7 !!xe7 24.!!c l g6 2 5 . f4 l:!d7 26.d5 and White has a big advantage. 1 8.%Va4 Black seems co have enough compensation after:

Chapter 4

-

181

Transformation of Pawn Structures

1 8 . . . �h3 1 9.E!:fe l f!:e7

a

a

b

c

d

e

f

g

h

It is not so clear what these doubled pawns can do. Certainly the game goes on and probably White feels a bit more comfortable. It is not as simple co prove equality as I thought during the game. So, generally speaking Black seems to be fine, but with energetic play White can still cause problems with l 5 .�xf6. You really need to dig deeply in order to find these options for White, and this time Aronian did not manage co do so. The way I see it is that in order to find such resources, you have to believe that they are there. I did not believe they were and we can conclude that Aronian also did not. It is hard co blame someone for not seeing something that you did not even believe could be there yourself! (Though of course this is what many chess fans do when they look at top games with the engine running!)

1 5 ... i.e6 The most natural square for the bishop. 16.:Scl The most natural reply. Of course you have to consider 1 6.f4, but I saw an exchange sacrifice that I really liked: 1 6 . . . dxc4! l 7.�xc6 �xh3 1 8.E!:e l

b

c

d

e

f

g

h

Here I had planned to play I 8 . . . E!:xe5 1 9.dxe5 W/d4t 20.mh l l2i g4 2 1 .Wff3 E!:d8 with adequate compensation. After 22.dxc4 l2i f2t 23.mh2 �g4 White has to play 24.Wfg2! to keep the balance, when Black can repeat with 24 . . . �f5 . Instead, after 24.Wfe3 Wfxe3 25.E!:xe3 �f5! Black would win the exchange back with slightly the better chances. Of course it is not likely that you would calculate this far during the game; we are humans after all, but you would go with your feelings, and I instinctively liked this while Aronian clearly disliked it. A closer look with the engines shows that Black has a strong option in l 8 . . . E!:c8!, when after 1 9 .�xe8 l2ixe8 White has to cake back on c4 and after 20.dxc4 f6 the bishop is trapped. The way I play chess I do not have co see everything in advance; I rely a lot on my feeling and on being able to see things in more detail when I am a bit closer. Everyone has their own preference regarding how to balance intuition and calculation and will find the way that suits them best. We all think in both ways, so please remember that we are talking about tendencies and not absolutes when we talk about logical and intuitive players.

16 . Jk8 After I had played this move I was a bit unhappy with it. .

Positional Decision Making in Chess

1 82

My alternative was 1 6 . . . '?Md?!?, which also makes sense. I am not afraid of the doubled pawns. Without knights I cannot see how White is going to create any real pressure against chem at all. And after 1 7.g4, I can reply 1 7 . . . lll h? with even chances.

If you discarded dynamics, this would be an easy decision to make. Although Black straightens out White's doubled pawns, in return he gains control of the d5-square. The devil is of course in the detail! White has a number of active options chat have to be calculated, the most challenging of them being the one Levon chose in the game. 17 . . . Wd??! would be wrong. White plays: 1 8.g4! dxc4 1 9 . .ixf6 gxf6 20.dxc4 Wd6

a

b

c

d

e

f

g

h

Now 1 8.f4?! f6 1 9.f5 fxe5 20.fxe6 Wxe6 ought to be considered, though we should remember the words of the Armenian Master Eduard Mnacakanian: Black has an extra pawn, if nothing else (see page 204) . So instead White should probably play 1 8 .©h2 or similar, with even chances. As it turns out, the move I made was not bad,

but I did not like entering forced lines. I did not see a concrete problem in them, but being forced co go in one direction is a bad sign.

b

a

c

d

e

f

h

g

2 1 .f5 .id? 22.c5! Black is clearly under pressure. It is not human to allow such a position.

1 8.dxc4 b5!

17.f4!? A very sharp move, threatening f4-f5 immediately. 1 7.�e l would have resembled the 1 6 . . . Wd? lines mentioned above after 1 7 . . . tll h? l 8.g4 %Vd7 l 9.'it>h2, when we have a manoeuvring game with roughly even chances. This is the first critical moment of the game. Ac this point I had to calculate the lines accurately.

17 dxc4! •.•

a

b

c

d

e

f

g

h

Fighting for the d5-square. Not surprisingly, I was not allowed to occupy it.

Chapter 4

-

1 83

Transformation of Pawn Structures

1 9.d5! This is the most concrete option and I would assume that Levon had planned it well in advance. It is also the most principled move, and Aronian has a well-earned reputation for not walking away from a fight. . . 1 9.cxb5 cxb5 seems less challenging. If Black has the time, he will (perhaps after an exchange on c l ) put his bishop on c4. One thing I had planned was to meet 20.'\Wd2 with 20 . . . .ic4! 2 I ..ib7 lll d5!. 8

a

b

c

d

e

f

g

h

At this point I had to calculate some long variations. What I did in the game looked very scary, but when I discovered my 25th move, I knew that I would be out of trouble.

7

6 5

4 3

2

a

b

c

d

e

f

g

h

Black will have great compensation for the exchange on the light squares. Obviously I did not believe that I would be allowed to play this, but you always have to know what you will do against the most threatening ideas, and this was one of the ones I checked. The computer does not think that Black is better here, but maybe it j ust does not understand what is going on? One thing that plays in Black's favour is that the remaining bishops are opposite-coloured, which makes it hard for White to force exchanges.

19 ... cxd5 20.cxb5 20.c5 looks thematic, but I intended 20 . . . lll e4, when after 2 1 . f5 Black has 2 1 . . . f6! 22.fxe6 fxe5 23 . .ixe4 dxe4 where White is probably objectively okay, but still has to j ustify the pawn sacrifice.

20 .. J�xcl 2 1 .Wxcl Wb6t When I returned to the hotel after the game I saw that 2 1 . . .lll e4 was also possible. Play might continue: 22.@h2 '\Wa5 23.'\Wb2 f6 24 . .id4 E:c8

White has to find 25 .E:f3! to get enough counterplay to keep the balance. The position seems rather drawish to me, though it is obvious that it is Black who is setting problems for White here. I think Aronian would have been disappointed to have to play to equalize, as my impression during the game was that he was trying to get an advantage.

�.t. ; � 11 ��� .i.�1 U U 6 - n m1·� �m1·�

22.'it>h2 Yfxb5

. . . . . %.

.

��

� �

�

·� � � �r � , � � -� 8 . u ·u%-�� 32 �U -

�

1

b

a

d

c

f

e

g

h

23.Lf6 Again we can see clearly that Aronian is trying to win the game. If he wanted to draw, he could have forced it with 23.%Va l , when natural play would be 23 . . .if5 24 . .ixf6 gxf6 25. xf6 ie4 26.%Vxh6 ixg2 27.%Vg5t and White can choose who will force a draw.

�

�i �r -� �� � .,. . . . . . %z 6 ·� · .• 5 �1j'·• i ·-�. - ...% � 4 % . 3 � •. • �-� 8 2 8 •. • • J. ·� ..... , . � �

�

.

.

a

b

c

d

e

f

g

h

For example: 27 . . . �f8 28.�xg2 E:eit i9.E:f2 E:xf2t 30.�xf2 %Vb2t 3 1 .�f3 %Vxa2 and it is time to take the perpetual. I am pretty sure that Aronian had missed my 25th move at this point, or he might have considered going for the draw.

23 ... gxf6 24.£5 J.d7

8

U E -� fm·- --

. .

b

a

d

c

f

e

g

h

25.Yfxh6?! I expected this during the game. It is the natural continuation, so to call it a mistake feels wrong. One weird intermediate move White could consider is 25 .a4!? with the intention of preventing . . . ib5 in some lines. However there is always a downside. In this case it i � that the a-pawn quickly comes under attack. After the game I analysed 25 . . . We2 26.%Vxh6 %Vb2 27.E:f4 E:e4. 8

7

6

"' �� �- � · � A �&� . . . .J �.JL�

�%

•

�

5 �-

4

3

i

":-- · ..

·lifr�

.,.8• . 8 . •.1·� "� �- --· · · ... . . v.� 8 -�

% ..

9,;, ,,j

'

.

B � . J;z� ..... % � � � • .

a

b

� c

d

� e

,

f

g

h

I then had 28.E:g4t?! E:xg4 29.hxg4 as a drawing line, but Black has 29 . . . ic6!, potentially followed by ... %Ve5 and . . . a5, when White will have problems to solve. Instead White needs to play 28.a5, when accurate play will keep the balance. But as

Chapter 4

-

Transformation of Pawn Structures

said, this is not what Levon was looking for, and it is only something we have come to understand after analysing the game with computer assistance. I have to be honest and say chat even after analysing the game, the idea seems eerily foreign to me. However, it might have been his last chance to keep the balance.

8 7 6 5 4 3 2 1

28 .i>h2 l:!e4

a

a

b

c

d

e

f

g

h

25 ...Wb2! Maybe Levon had only anticipated 25 . . . ge2?, when after 26.Wxf6 Wb2 27.Wxb2 gxb2 28.g4! the endgame with three split pawns will hold a lot of challenges for Black. 26.©hl 26.l:!f4 l:!e4 transposes to the next note. 26 .. J::l e2 27..bd5? A blunder, played pretty quickly as far as I remember. I do not know what he missed. After the game he was extremely disappointed and we did not really talk about it. I assume chat he missed something pretty simple, as all players do from time to time, even such champions as Levon. During the game I believed that White could force a draw: 27.l:!f4 l:!e i t 27 . . . ixf5 28.l:!xf5 l:!xg2 29.l:!g5t and White has a perpetual.

1 85

b

c

d

e

f

g

h

I believed chat there were several ways for White to make a draw. This was also my conclusion after checking briefly with the computer in 2008. Bue looking deeper now, it appears that things are not so simple. White has no clear way to keep the balance, even though his position might be holdable in the long term. I hope the reader will forgive me for being a bit technical here and showing our analysis in detail. I think it helps illustrate the way White has taken risks and now will have to play many good moves in order to defend. Not an easy task at all. 29.i>h l ! I assumed chat 29.l:!g4t l:!xg4 30.hxg4 would be okay for White, but Black has a very subtle manoeuvre: 30 . . . ib5 3 l .Wf4

Positional Decision Makin g in Chess

186

for White. Understanding Black's 3 1 st move in this variation is by no means easy. Why would forcing the white king forward be an advantage? Only accurate analysis and comparison will show that after 33.g5 ¥Me5 34.¥Mg4 J.e2 it is a vital point that the f5-pawn hangs with check, as otherwise White would secure sufficient counterplay with 35 .¥Mh4, which now loses a pawn. I am by no means convinced that I would have found this during the game! 29 . . . gc4! I did not consider this move in my earlier analysis, but it appears that it causes White quite some difficulties still. On the other hand, after 29 . . . gxf4 30.¥Mxf4 d4 3 1 .¥Md6 i.xf5 32.Wxf6 i.g6 33 . .id5 , Black should give a perpetual.

7

6

-� �� ��-0� �� - - - �� �,-,� � �� ,� �8%� : �� � �� �� �� �-�J� �-� %._j� ��� 8 �- - -�i � �� �,� 8

7

6

3

2

a

b

c

d

e

f

g

h

Black has a number of possible continuations here, which can be analysed deeper and deeper. One of them is: 34 . . . .ie6 3 5 .¥Mxa7t md6 36.Wa6t me5 37.YMfl i.xf5 38 .gf4 i.e4 39.gxe4t dxe4 40.Wf4t me6 4 1 .Wxe4t md6

�� �� ,� �-0i -, , , % �� : ��� �• : �.-,.1.·� �� ��� "-�-%� W!08 �-� '8 �i ������� , ,,, 8

3

5

2

4 2

a

a

b

c

d

e

f

g

h

have spent some time analysing this position in the work for this book, without coming to any final conclusions. The analysis shows that White is struggling, but that he still has plenty of defensive resources. The two main options to look at are a) 30.¥Mh4 and b) 30.gg4t. a) 30.Wh4 ge l t 3 1 .mh2 gc2 leads to a forced continuation: 32.gg4t mf8 33.Wh6t me? 34.¥Me3t

b

c

d

e

f

g

h

We could continue, but it makes sense to end here with a conclusion. White is fighting desperately for survival. One point that it makes sense to confirm is that Black would win if he achieved the realistic double ambition of winning the a-pawn and exchanging queens:

Chapter 4 - Transformation of Pawn Structures After 1 . . .ElcS 2.ib7 fS 3.\tig2 \tif6 4.ia8 Elc3 S .idS Eld3 6.ic4 Eld4 7.ifl \ties 8.\tif2 Eld2t 9.\tie3 Ela2 1 0.ic4 Ela3t l l .id3 Elb3 White is in zugzwang. 1 2.e2 f4 for example, and Black wins. Summing up, 30.1'9h4 is very dangerous for White, although it may still be a decent practical chance. b) So instead he might want to try something else, though 30.Elg4 t Elxg4 3 l .hxg4 ibS is also unpleasant for White.

� �· � ,�, , %� •%� ••• •� �. �•• • � �, ,."� � � ' ,�, ,/. �-� �.�t�@ � � 8

7

%

6

%

%

%

3

%

2

a

b

c

d

� --

e

f

g

h

The main point about this position is that White cannot force a draw immediately: 32.gS ?! Wc l t 33.\tih2 WxgS 34.WxgSt fxgS 3 S .ixdS \tig7

� �� �� '· ' .%. �. �.;,_ ,;,� �wr .:.. :•. � �. �-� �.� �, �r. . . ;m �� �� �� , , 6 5

4

3

%

%

%

" - -%

1 87

give up the bishop for both of the black pawns on the kingside and make a draw with the king in the queenside corner. But I do not believe that Black need allow this. So instead White should try 32.Wf4 with a worse position in which nothing direct exists. To speak of it in terms of lost or drawn makes no sense to a practical player like me. What matters is that White has a real fighting chance, while what matters from Black's perspective is that he is happy to be able to cause White continuing problems.

27 ...i.xf5 2s.Wf4 i.e6

8 7 6 5 4 3 2

b

a

d

c

f

e

g

h

After this I think the position is unholdable, though there is still some play left.

29.i.f3 An interesting line is: 29.ie4 ixh3 30.Wh6 Elxe4 3 1 .1'9xh3 Wxa2 32.Wc8t g7 33.1'9c6

2

a

b

c

d

e

f

g

h

Here both 36.\tig2 ie2! and 36.g4 \tif6 37.\tig3 ie8! with the idea of . . . \ties and . . . f6, make the bishop ending untenable for White. The only hope I can think of is to somehow a

b

c

d

e

f

g

h

Positional Decision Making in Chess

1 88

The first time we looked at this, I said that it would be really shocking if 33 . . . Wf e6? 34.Wfxe6 !!xe6 was not winning. But as it turns out this is indeed the case. After 3 5 .!!al a6 36.g4 . manages to hold by a miracle. White

36.Wg2 Wfe4 and the rook ending features much worse circumstances for White.

29 .. � h2t 30.ci>gl J

�• �.• �%. 7 �la �.1W)·•�- - �� 8

6

��;,�"�

� � ·· • • 3 R � • • -�

z

. . .�. ��� �-� �� h7 3 5 .l:!c?t 'it>h8 The first time I looked through the game I did not manage to find a clear win here, which clearly illustrates what type of challenge we are talking about. 8

7

6 5

4 2

a

b

c

d

e

f

g

h

It turns out that White is winning by force, but he has to find a beautiful tactical idea: 36.llid7!! With the idea of lli f6 and l:!h7#. 36 . . . 'it>g? 37.llide5t 'it>h8 37 . . . 'it>g8

Positional Decision Making in Chess

208

35J�xc8t @h7 36Jk7t @hs 37.tl)es 37.lll d7 works as in the previous comment. 37 e2 The last small trap, hoping that I would check on g6 before taking on e6, after which . . . �h l t would win for Black. •••

For a brief moment I became nervous about 37 . . . exf.2 38.llixe6 fl =lll t, before I realized that the king escapes easily with 39.g3 lt>d7 36.lt>f4 lt>e6 37.lt>g5 lt>e5 38.lt>h6! White has to act quickly, otherwise Black gets counterplay with . . . lt>d4. 38 . . . lt>xf6

39.f4!! 39. lt>xh7 lt>e5 40. lt>g7 lt>d4 4 1 . lt>xf7 lt>c3 42.lt>xg6 lt>xb3 43.f4 a4 would offer Black enough counterplay to save the game. 39 . . . 'itif5 40. lt>xh7 lt>xf4 4 1 . lt>g7 lt>e4 42. lt>xf7 'itid4 43.'itixg6 a4 43 . . . lt>c3 44.h4 and White wins by a tempo. 44.bxa4 lt>xc4 45.f4 lt>b3 46.f5 c4 47.a5! 47.f6 c3 leads to a draw.

a

b

c

d

e

f

g

h

Rubinstein's technique is excellent in what follows, leaving his opponent without any real chances of saving the game.

30.Vfg4 g6 3 1 .Vfg3 Vfd8 32.Yfe5 a4 33.h4 h6 34.©g2 Yfc8 35.©g3 Vfd8 36.©g2 Yfc8 37.Yfxc5 Yfe6 38.Yfe7 Yfc8 39.c5 g5 40.hxg5 hxg5 41 .Vfe3 Vfe6 42.Yfxg5t ©t'8

a

b

c

d

e

f

g

h

43.Yf g7t ©e8 44.VfgSt ©d7 45.Yfg4! ©e8 46.Vfxe6t fxe6 47.©g3 @f7 48.©f4 ©xf6 49.©e4 1-0

Chapter 5 - Transformation of Advantages In the initial stages of the following game, Rubinstein's play is most impressive, yet it is the rook ending that is surely the most interesting part. In a game filled with strategic, tactical and technical challenges, these two giants of early chess history battle their way to move 76. As you would expect, not all decisions on the way were the best possible, but the high quality of the play is undeniable.

Akiba Rubinstein - Alexander Alekhine Karlsbad I 9 1 I

In this game Rubinstein built up a strong centre and space advantage. On the other hand, Black was very solid, so it was not so easy to make something of this advantage.

1 .d4 d5 2.�f3 �f6 3.c4 c6 4.lLJc3 °1Wb6 5.°1Wc2 i.g4 6.i.g5 �bd7 7.e3 lLJ e4 8 ..if4 e6 9 ..id3 °1Wa5 10.0-0 lLJxc3 1 1 .bxc3 i.xf3 1 2.gxf3 dxc4 13.i.xc4 �b6 14.°1Wb3 gds 15.i.e2 i.d6 16.i.g3 hg3 17.hxg3 0-0 1 8.f4

a

b

c

d

e

f

g

h

This is where the opening ends and the middlegame starts. If we look at this with modern eyes, we would not rate White's chances to be that extraordinary. If Black

219

were to play . . . l'!d7 immediately and simply improve his position slowly, he would have a solid and defendable position. Accurate play would still be needed, as White's small pluses are still valid, of course, but the game goes on.

1 8 ... c5? Alekhine has played a lot of natural moves and at this point it looks natural to advance the c-pawn. However, White's centre is very solid so all this move achieves is to weaken the f3-b7 diagonal and the b5-square. 19 ..if3 gd7 20.a4! We can now see the problems caused by l 8 . . . c5. Black's position is suddenly unpleasant. 20 ... � d5 2 I JUc1 g6 22.°1Wb5! White also had the chance to play 22.f5!?, but this was less in keeping with Rubinstein's style.

8 7 6 5 4 3 2 1

a

c

d

e

f

g

h

The next key moment. After this move Black is faced with an unpleasant choice - to exchange queens or to allow White to advance the a-pawn further and put the b-pawn under pressure.

22 ...°1Wxb5?

Positional Decision Makin g in Chess

220

For tactical reasons this is the inferior choice. I have the feeling that Alekhine assessed the resulting ending as solid and defendable, but Rubinstein shows that this is not the case. 22 . . .Vfic7 23.�ab l is unpleasant for Black, but it was the lesser evil.

23.axb5 cxd4 24.cxd4? The most obvious positional move, but here Rubinstein missed a chance to improve his position greatly. I have already mentioned this once; Rubinstein was not very flexible and would rarely deviate from his strategic goals. He could have played: 24.c4! lli b6 (24 . . . lli e7 25.�xa7) 25 .c5! lli c8 26.exd4

a

b

c

d

e

f

g

h

With a technically winning position. I am not sure if it was Nimzowitsch or Lasker who said about Rubinstein that he could at times be too slow, but on the other hand he would never force matters prematurely.

a

b

c

d

e

f

g

h

27.ixd5! It looks odd that White gives his strong bishop away, but as Black has to take with the pawn, his structure is weakened further and fixed in an awkward position. The transformative power of White's advantage is clear. 27... exdS 28.gacl White enjoys a big advantage. Black has weak pawns on a7, d5 and h7 to look after. Rubinstein loved to have this kind of permanent advantage where he could massage his opponent indefinitely. And then he was of course famous for his skills in rook endgames. 28 ... gfdS 29. @fl @e7 30. @e2 gd6 3 l .g6c3 g6d7 32. @d3 gas 33.gc6 gd6

24... b6 I am pretty sure that Alekhine believed that his position was a fortress, with j ust one weakness and a lack of entry points in the position. But Rubinstein manages to outmanoeuvre him.

a

b

c

d

e

f

g

h

Chapter 5

-

Transformation of Advantages

Up to this point Rubinstein has played a great game, but from here on his play is less impressive. Now, more than a hundred years later, it is of course difficult to guess why. Maybe he had not slept well; maybe it was something he ate. The possibilities are endless when once you get going. But the most accurate statement is simply that he did not handle the endgame well on this occasion. This is not a j udgement on his strength, as top grandmasters today also play badly on occasion, which reminds me of Grischuk saying that he did not fully understand why I am so infatuated with Rubinstein's games. In his kind words, I am apparently a much better player than Rubinstein was. Even if you were thinking this was the case, it certainly was not the case when I bought Razuvaev's book! Or in Jacob Aagaard's words: It was only by climbing on top of Akiba's shoulders that I was able to j ump into the world elite and make my own mark on modern chess history - however faint.

34.@e2? I have no idea why Rubinstein allowed Black to exchange a set of rooks. It made more sense to wait and try to improve the position further.

22 1

36.8 Another careless move? Rubinstein appears not only blind to his opponent's chance to create counterplay with the a-pawn, but he also weakens the e-pawn. However, as we shall see, Rubinstein had a deep prophylactic idea. An important point is that after 36.@d3 Black can force a draw with 36 . . . l3c8!. The pawn ending is surprisingly a draw even after: 37.l3xc8 ©xc8 38.f3 ©b7 39.e4 a5 40.bxa6t ©xa6 4 1 .eS

� ������� �� , =� ,�

. . . ?.� �W% �� � � �jpj� �� 4 - � � � �� � �-�%%��� . , . , . %� ��-zr� �� �� �� 6 5

a

b

c

d

e

f

g

h

Usually we would assume that White would have great winning chances if he was able to exchange his e-pawn for the black b-pawn and get his king to c6.

34.13c7t l3d7 3 5 .l37c2 and White keeps a big advantage. He can always choose to exchange rooks when it best suits him.

34 .. J�xc6 35J�xc6 @d7

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

Surely Black would end up in zugzwang? But in this case Black would actually be winning with a series of thematic breaks with . . . h5, ... g5 and . . . f4, allowing the h-pawn to run up the flank to promotion. White can stay back a

b

c

d

e

f

g

h

Positional Decision Makin g in Chess

222

with the king and avoid defeat, but there is no advantage to speak of. Rubinstein realized this and thus prepared to push his e-pawn with 36.f3 . Black would now lose the pawn ending after: 36 . . . :B:c8? 37.:B:xc8 c;t>xc8 38.e4!

b

a

c

d

e

f

g

h

37 . . . lt>c7!! This is the really surprising move. The black king needs to get across to the b-file so it can attack the b-pawn. 38 .:B:c6t lt>b7 39.lt>d3 a4 40.c;t>c2 The white kin g is in time to block the a-pawn, but Black can play either 40 . . . :B:e8 or 40 . . . :B:a5 4 1 .:B:c5 a3 42.lt>b l a2t b

a

c

d

e

f

g

h

Black does not have time to create counterplay on the queenside and White can slowly penetrate on the kingside. However, there is another drawback that the great men apparently did not notice.

7

6 5

4 2

5 4 3 2

b

a

c

d

e

f

g

h

43.lt>al :B:a3 Black gains enough counterplay to equalize.

37.i>d3 ge7 38.g4 ge6?!

a

b

c

d

e

f

g

h

36 .. JfoS? Alekhine misses his chance to create counterplay. 36 . . . a5! would have led to a draw. This move was originally proposed by Spielmann, but the saving idea was not known until our age: 37.:B:xb6

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

Chapter 5

-

Alekhine is apparently convinced that White's last move made the pawn endin g tenable for him. And Rubinstein agrees!

39.gcl?! But just because two legends think alike, it does not mean that they are right. White wins after: 39.l:!xe6! @xe6

@d6 45.f5 h4 46.f6 and the pawns no longer need support from the white kin g. 42.e5 t @e6 The protected passed pawn is a decisive asset. White wins in a further 20 moves in a logical and forced sequence. 43.@e3 @f7 44.@f2 @e7 45.cilg3 c;!/f7 46.@h4 @g6

a

40.gxfSt! 40.e4? dxe4t 4 I .fxe4 fxg4! leads to a draw with both kings being pacified by two connected passed pawns. 40 . . . gxf5 40 . . . @xf5 4 I .@e2 and the white king comes to the kingside and wins. 4 1 .e4 @d6 4 1 . . .dxe4t 42.fxe4 fxe4t 43.@xe4 h5 could be what Alekhine was relying on.

�

6 5

4 3

2

-� �� �� �� � =� �� . . . ?.. . . . . . 7. � �� �� � �� �� -% ��)'% �£ � 1���7. � � . 7.� � � �� �� �� � B � � �� �� ..

.

.

a

b

c

.....

..

d

e

f

g

h

223

Transformation of Advantages

But White can advance his pawns decisively before dealing with the h-pawn: 44.d5t

b

c

d

e

f

g

h

47.e6! @f6 48.@h5 @xe6 49.@g5 ! cile7 50.@xf5 @f7 5 I .@g5 @g7 52.f5 @f7 53.@h6 @f6 54.@xh7 @xf5 5 5 . @g7 @f4 56.@f6 @xf3 57.@e5 @e3 5 8.@xd5 @d3 59.@e5 @c4 60.d5 @xb5 6 1 .d6 @c6 62.@e6 White wins. It is easy to understand why Rubinstein played as he did. In the game he retained the advantage and allowed Black to make mistakes under pressure. Going for the pawn ending without the certainty provided by the en gine is not an easy choice. Calculating these lines would be difficult, even for today's top players .

39 .. J:le7 40J:lhl ci>e6 White retains a serious advantage in the rook ending. It is possible that Black could draw it with accurate play, but in practice the position is hard to defend. 4Ukl ci>d7 42.gel gf7 43.flal ci>d6 44.Ekl ci>d7 45.gc6 gm

Positional Decision Makin g in Chess

224

8 7 6 5 4 3 2

52.ci>h4? 52.gxh5 gxh5 53.\!?h4 was much stronger. Black cannot defend the h-pawn successfully, as after 53 . . .l;h? 54.\!?g5 h4 5 5 .@xf5 h3 56.E:c l the counterplay with the h-pawn is an illusion. White's material advantage decides the game. •••

a

b

c

d

e

f

g

h

Rubinstein has shuffled his rook about a bit and is now ready to go with the king to the kingside to force Black to make concessions. Again we can see that although he has missed some of the subtleties, Rubinstein's vision for the rook ending of a better pawn structure where Black has no counterplay has dominated the game.

46.ci>e2 �U'7 47.ci>fl �us 48.ci>g3 ges 49.gc3 ge7 50.ci>h4 It is not feasible for Black to allow the king to come to g5, but it is not pleasant moving the h-pawn either. 50 h6 5 1 . ci>g3 h5 ..•

8 7 6 5 4 3 2 1

52 gh7? This often happens in chess and you will also be able to see it in my games in this book. Once a player has decided co go down one path, his opponent does not consider a possible change of path. Here Black should have played 52 . . . hxg4 53.fxg4 E:h7t with a transposition to the game. 53.ci>gS? And White again could have won the game with 53.gxf5 gxf5 54.l!?g5! . But as said, both players were mentally committed in a certain direction and probably made the moves rather quickly. 53 fxg4 54.fxg4 hxg4 55.ci>xg4 ghl •••

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

a

b

c

d

e

f

g

h

White of course has the advantage in this rook endgame. He is more active and has only two weak pawns, while Black has three. The problem is chat Black's rook is fully in the game, securing him a lot of counterplay. With accurate play, Black can hold.

Chapter 5

-

56.@g5 White also had another try, winning a pawn with 56.l'!c6 l'!gl t 57.Wf3 l'!fl t 5 8.We2 l'!b l 59.l'!xg6 l'!xb5 60.l'!g7t We6 6 1 .l'!xa7. However, Black is ideally placed and White's pawns are not that great. A deeper study shows, not surprisingly, that Black makes a draw with decent play. 56 J�bl

8 7 6 5 4 3 2 1

•.

b

a

d

c

f

e

g

225

Transformation of Advantages

Black is barely hanging on, but hanging on nevertheless. 62.l'!b3 Again White's dream plans do not work: 62.Wg7 l'!g l t 63.Wf8 l'!fl 64.f7?? We6 and Black wins. 62 . . . a4! 63.l'!xb6 l'!xe3 64.l'!a6 a3 Black is active enough to secure a draw. 65 .'it>g6 65 .Wf8 l'!d3 and Black draws. 65 . . . l'!g3t 66.'i!?f5 l'!f3t 67.We5 'i!?c7 68.'i!?e6 Wb7 69.l'!a4 Wc6

h

57.ga3 57.Wxg6 l'!xb5 58.f5 This also looks tempting, but Black is in time. 58 . . . l'!b l 59.'i!?f7 White could easily push it too far: 59.f6 We6 60.l'!c6t Wd7 If White returns with the rook, a draw by repetition seems likely. A catastrophe awaits after 6 l .f7?? l'!gl t 62.Wh5 l'!fl and it is Black who wins. 59 . . . a5 60.l'!a3 l'!fl 6 1 .f6 l'!e l

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

57 J::lxb5?? At the same time an obvious-looking move and a horrendous blunder. After this the rook ••

a

b

c

d

e

f

g

h

226

Positional Decision Making in Chess

is entirely out of play on b5, caking far too long co come back into the game. White wins a pawn and manages to make something great out of his f-pawn. 57 . . . e6! This would have held the game. White has virtually no advantage. For example: 58.�xa7 �g l t 59.h6 59.h4 �h i t 60.g3 �b l wins back the pawn under desirable circumstances. 59 . @ f6 60.h7 g5! .

.

b

a

c

d

e

f

g

h

Black manages to gee rid of one of his weaknesses - and this time without losing it! 6 I .fxg5t �xg5 62.�b7 �h5t 63.gB �h3 Black will manage to liquidate to a theoretically drawn ending.

8 7 6 5 4 3 2

b

c

d

e

f

g

60 :axe3 6 1 .f6 gg3t 62.h7 :aa 63.f7 gf4 64.g7 gg4t 65.@f6 gf4t 66.@gS JUI 67.g6 g81 t 6s.f6 got 69.g7 g81t 70.f'8 gdl 71 .es gelt 72.ds go 73.gd?t c6 74.e8 gf4 75.ge7 b5 •••

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

76.gc7 Accurate to the end! If Black scares giving checks again, the white king will go up and pluck off the d5-pawn before finally winning the black rook. 1-0 By no means a flawless game, but you can say the same about a lot of my best games, so please do not lee the precision tools we have at our disposal today diminish your appreciation of the great skills Rubinstein and Alekhine possessed. This reminds me of a recent press conference from Scavanger where Alexander Grischuk once again displayed his entertaining wit. In a more serious moment he made the following comment on the increased use of the word blunder in the language of modern chess.

58Jha7t d6 59.@xg6 gb3 60.f5

a

Rubinstein could not care less about the e-pawn.

h

"Actually, I chink this is why people chink chat players forty years ago played better than nowadays. Now the computer

Chapter 5

-

Transformation of Advantages

immediately shows the mistakes, It's red, and they think OK, they are completepatzers. But in fact in many of those situations where the computer immediately shows that it is a blunder, it would have taken a week to debate it in the press; Igor Zaitsev would find some amazing resource and so on. But now the computer just shows it in two seconds."

Transformations in my own games

227

last two rounds instead of drawing them both. In the game against Shirov in the penultimate round, he was completely winning, as is well known. But also in the last round he was winning against Ljubojevic, though we only know this today when the computers point out the amazing winning combination. In the Chess Informant annotations from the time of the game, the move that throws the win away was awarded an exclamation mark. . .

Things are of course never easy in a practical encounter against a resourceful opponent. In the rest of this chapter I will give four examples from my own games. They all show different aspects of transforming the position in an attempt to fight for - or keep - the advantage. Things are rather subtle in places, but I hope that the deeper meaning will still shine through. In the first game we shall see how I managed to transform the advantage of the first move into a slightly preferable structure. My opponent played decently for quite a while, but the pressure of being a little worse, no matter what he did, eventually led him to make a small mistake, making it possible for me to improve the structure a bit more. Then a few things happened in the game, but eventually the better structure was decisive.

Boris Gelfand Branko Damljanovic -

Istanbul (ol) 2000

Branko Damljanovic is one of the strongest Serbian players of the past. I know him quite well, as we played for the same club, Agrouniverzal, once in the Serbian Cup and many times in the European Club Cup. The peak of his career came in 1 990, where he played brilliantly in the lnterzonal and would have been a Candidate had he won one of the

a

b

c

d

e

f

g

h

The game ended 37.l:'!:e7t? d l .ixb l 1 5 .E:xb l lli f6 1 6.llib5

8

5

4 3

2

a

b

c

d

e

f

g

h

And for quite some time it seemed co me chat White is doing well. Bue then I found 1 9 . . . h5 20.g5 lli e4!! 2 1 . fxe4 Wfxe4 and the queen is forking the white rooks! [Boris laughs infectiously.] Maybe I spent coo much time solving chose positions you sent co me!

7his is a beautiful line. Yes, yes. Then I moved on co other moves. So I calculated all of these lines and in the end I missed l 7.Wff2 and the position is so bad after this that I had co resign. It was j ust a blunder.

How did you react to this blunder emotionally? I was pretty calm. Maybe too calm. I do not know why, but I was pretty calm. OK, it can happen. It is a long match and I decided that it would be better j ust to forget about it and concentrate on the next game; and this is what I cried to do.

Positional Decision Maki ng in Chess

272

To me Game 9 seemed to be the most critical game ofthe match.

I probably agree.

Maybe move 19 [where Boris played 1 9.c5] was the critical moment ofthe game?

Do you think you wanted to be too much in control when you played this move? I shall tell you what happened. First of all it was not possible to calculate everything to the end. Secondly, I was fully aware that fortresses are Vishy's bread and butter. He did it very often, even in our mutual games. I knew there was a serious risk that this would happen. But I also did not see how I should continue if l played 1 9.a3 and he simply waited. So I thought, "No matter what I do, there is risk involved." Maybe I would have to play c4-c5 in worse circumstances. So I invested a lot of time trying to calculate the endgame and I decided that with the damaged pawn structure on the kingside and the pawn on a6, I would create another weakness on the kingside and I would break through. I was totally sure that it was won from an assessment point of view.

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

Appendix

-

I Cannot Leave My House!

273

And maybe with 33.g4 I would have been able to do it. I am not completely sure. I have not had time to analyse it deeply yet. I believe that if l had been able to put my pawn on f5 I would have won. But I played imprecisely and did not manage to achieve this. I simply missed that he could put his pawn on f5 .

But ofcourse 1 5 . . . ixB was also a very strange mistakefo r Anand to make. Of course. But I think Anand even admitted someplace that he mistook the position for that of an old Karpov game. [Portisch - Karpov, Bugojno 1 978 - actually Black could transpose to this game with 1 5 . . . h6! 1 6 . .id2 l:!fe8 1 7.l:!e l and now 1 7 . . . ixB! 1 8.Wfxf3 e5 with even chances.] He intended to play 16 ... e5, but then realized that l 7 . .if5! wins the exchange. I was completely shocked, not realizing immediately what was going on.

This is also one of the moments that make people say that '�his is not the same Anand as we used to know ': These things happen. We will see what happens in the next few tournaments. Before his match against Kramnik in 2008 he played really badly in Bilbao [two losses and eight draws, ending in last place] , much worse than before this match. But then he won against Kramnik and against Topalov, and did well in so many tournaments. So I don't see a pattern of decline so far.

Do you think that you were a bigger challenge to Anand than Kramnik and Topalov? I think I created more problems for him than they did. I am not sure of the exact reason; maybe I was more successful in my opening preparation. Kramnik got caught in this Meran line twice and the match was simply over. Things were less easy against Topalov of course. I should say that the thing I find really interesting about the match is that you can make a grand strategy for the match in a way you cannot do in a tournament. You can push forward your strategy throughout the match, which makes it more interesting for me than a tournament. In matches the players' style and how they interact with each other is much more important than in tournaments as well. Don't get me wrong, tournaments are also interesting, but matches are played only rarely, which is perhaps why I cherish them so much. For example: before the match I analysed Anand's games deeply and realized that he was fantastic with the initiative and that I simply should not give it to him. And it worked.

Do you think the public like matches less because you needfewer decisive results to win? Too many people follow chess for us to generalize. People like different things. This is not the Eurovision Song Contest. We do not have to follow the majority view. A lot of people like chess. Some prefer tournaments, others matches, some rapid chess and so on. Fortunately we have all kinds of events at the moment so everyone can choose to follow the event that suits them best.

274

Positional Decision Making in Chess

Unfortunately the great majority of the public is not interested in chess at all. Let us for example take these complaints about the number of draws. Last year I played in Tai Memorial [Moscow 20 1 1 ] , which was maybe the most interesting tournament in the last few years based on the quality of the games. Then three rounds in a row there were more or less only draws, because everybody defended excellently and big advantages were not converted. People started to complain, but people who watched the games appreciated it. Or let me give you another example, an excellent example. Last year Shirov started with 3 Yz/4 in Lublin. Then he saved an ending two pawns down in almost a hundred moves against Sasikiran and in the final round he sacrificed the queen against Grachev, a game that ended with a perpetual check.

Shirov Grachev -

Lublin 20 1 1

I .d4 e6 2.e4 d5 3.�c3 �f6 4.eS �fd7 5.£4 c5 6.�f'3 �c6 7 ..le3 cxd4 8.�xd4 .icS 9.'ilYd2 10.0-0-0 hd4 1 1 .hd4 'ilYaS 1 2.h4 gbs 1 3J�h3 b5 14.£5 �xd4 1 5.f6 b4 16.'ilYgS �f5 17 ..id3 h6

0-0

18.LfS hxg5 19.hxgS bxc3! The only move that does not lose. 20 ..ih7t 20.E:dh l would have created more problems for Black, but Shirov clearly trusted that his opponent knew what to do and did not want to let him do it over the board! Black draws only with: 20 . . . cxb2t 2 1 .b 1

Appendix - I Cannot Leave My House!

275

Ja -*- � �· , �� � ••• � - -··" --- - -"� � io �� '"; f�t� .... . %� ..... %� . %� �� �� �� �� .:t � � �!@ �� ��� 2 8� 8 -� 8� 8 .:t � 8

7

6 5

4

..

..

3

-�

a

b

c

d

e

f

g

h

2 1 . . .'Wxa2t! 22.'it>xa2 b l ='Wt 23.!!xb l !!xb l and now White has to give a perpetual check now to avoid defeat.

20 ... @hs 2 1 ..ig6t @gs 22.i.h?t @hs 23.i.d3t @gs 24 ..ih?t 1/2-1/2 On a chess site they wrote that "Shirov started with 3 Y2/4 and cruised to victory with three draws." I was really laughing. And the j ournalist; how dare he write this? He said: "Oh, I had not time to check the games, so I decided to j ust write something." Unfortunately the people who watch the games do not have time to comment on them, so we are left with these people making comments that "the games are boring" and things like this. So the people who did not watch the games were unhappy. What can you do? But I can say in this match that we both did everything we could to win the match. But sometimes you cannot get an interesting game. Like Game 5 where I had this novelty in the Sveshnikov and equalized.

My assessment of the match was that when both the players are relying on surprises with Black and have an extra halfyear to prepare it is very difficult for the player with White to get an advantage in the firstfew games. Exactly! Actually this was a sort of revelation to me during the match. Before it I could not imagine that it would be so difficult to get anything in the opening. Even though Anand played a side-line it was very difficult to get an advantage, because he had analysed it deeply with computers. You need to spend a comparable amount of time in a line to be able to create problems.

Let us talk about the tie-breaks. It seems to me that the decisive factor was that Anand changed his openings and you did not? Probably! ? I still have to revisit it to make up my mind. But also, if your opening has served you well? You can play something you have not analysed deeply, but when Anand played l .e4 in the match, he was ready to meet all kinds of openings. So I relied on the openings I knew better. I could have played another opening, for example the Najdorf, and got trapped there. I had to

276

Positional Decision Making in Chess

make a choice and all choices had advantages and drawbacks. So I chose as I did. But even the way it came I had my chances. Of course I had to spend too much time in Game 2 finding an antidote to this interesting idea - 1 1 .tll a 5! - which we had not predicted in our preparation. But in Monaco I managed to win far more complicated rook endings and hold far more difficult endgames than Game 2, also playing with increments of 1 0 seconds.

I was surprised in Game 2 that you did not give up your bishop for the pawn in order to defend rook against rook and knight. Somehow I could not see a way to get it. I was also hesitating; should I look for it or should I make a fortress? I really spent too much time hesitating. I should have made up my mind more decisively. Could I really have forced this endgame?

Yes, a Jew times. But a few times I could also have created a fortress.

I would think that a harmless theoretical position would be a better choice than a fortress, which you need to understand But a theoretical position, you just know. Exactly! But you know, I would make a draw eight or nine times out of ten from this position. But this time it did not happen. And Game 3 I would probably win 999 from 1 OOO, but still this did not happen. In Game 4 I also had the advantage, but I should have played more patiently, advancing on both flanks and maybe it would have been enough to win the game.

What were you actually thinking when you played :gh7?

8 7 6 5 4 3 2 1

a

b

c

d

e

f

g

h

[The game went 59.gh7?? i>d6 60.i>g3 i>e6 1h-1/2. 59.�g3 j ust wins.]

Appendix - I Cannot Leave My House!

277

It is hard to explain. I have studied rook endings all my life. At age nine I knew the theory of the ending with 3 vs. 3 plus a pawn on the queenside. I knew all the ideas. And when I was twelve I studied Levenfish & Smyslov's and Minev's books and others very deeply, so I cannot say that I am not familiar with rook endings. I know a lot about it as well as the Vancura position [black king on b7, rook on c6, checking the white king perpetually] , which he obviously cannot get here. Probably I had some black-out, believing that he would get the Vancura position after 'kt>g3. I don't know. Rook endings give me a lot of pleasure. Recently there was a brilliant article in the Russian magazine 64 by Ilya Odessky with some really deep ideas, which I studied for days. In one of my training sessions before the match I spent a full day j ust trying to understand what was going on to increase my understanding of rook endgames. Even before the match I took care of my rook endgame knowledge, but what happened here has nothing to do with rook endings.

I was thinking it was just nerves. No; you could not imagine how calm I was. And probably this was the problem - I was too calm. I really do not know. For me it is a bit disappointing that the match was lost in a rook ending, which I studied and enjoy so much.

It seemed that after you lost the tie-break you were very emotional and very disappointed? Well, of course I was disappointed. I felt that I had played a great match, so of course it was disappointing to lose it in this way. But I realized that anything could happen. You need to be ready for anything when you play the World Championship.

I have a tricky question: what mattered most to you, to become World Champion or to play the match? This is actually a good question. Probably it interested me more to play the match and this was a problem. When I was ten years old a journalist came to the chess club where I studied. He asked us all what our dreams were. I remember it like it happened j ust now. One guy said he wanted to become a National Master and one guy said he wanted to be a Grandmaster. And they both succeeded. But I answered that I j ust wanted to keep on enjoying playing chess and I still enjoy it. So maybe this is a disadvantage. Some say that to become World Champion you have to have an obsessive desire to become so. But I am not obsessed with titles. The biggest challenge was to try to defeat the World Champion in a match. I really did not care if I would get the title or not. The only issue is that if I got the title there would be another great match two years down the road. I want to say that the last few months have been the most interesting in my life. And now I will have to win London 20 1 3 to get another match.

And you are going to do this? I hope so, I will start preparing pretty soon.

Name Index A

1 , 2, 3, 9, 1 0, 1 05 , 22 1 , 265 42, 1 9 1 49 24, 60, 92, 1 25 , 1 29, 1 30, 2 1 9, 220, 222, 223, 226, 254 Ambroz 69 Anand 33, 34, 39, 49, 50, 68, 75, 77, 1 0 1 , 1 4 5 , 1 53, 1 78, 1 92, 1 94, 1 97, 202, 203, 204, 207, 208 , 2 1 2, 254, 266, 267, 268 , 269, 273, 275, 3 5 , 83, 1 78, 1 79, 1 80, 1 8 1 , Aronian 1 83, 1 84, 1 89, 1 97, 239, 269 Averbakh 5 , 1 54, 228 Avrukh 8 1 , 1 68, 1 78, 1 79, 255 Aagaard Adams Adly Alekhine

B

Bareev Bas man Beaumont Begun Beilin Bo golj ubow Boleslavsky Bolo gan Botvinnik Bronstein Burshtein

28, 29, 1 2 1 , 246 1 79, 1 80 240 38 5 244 2 1 6, 23 1 32, 34, 49, 245 27, 8 5 , 1 03, 1 80, 2 1 0 2 1 6, 23 1 27, 28

c

Campora Canal Capablanca Carlsen Caruana Chajes Chebanenko Cohn

1 0, 99, 1 02, 1 09, 1 1 5, 2 1 6 24, 2 5 , 26, 27 60, 84, 1 29, 1 37, 1 45 , 1 52 28, 3 5 , 77, 82, 1 06, 1 08, 1 92, 229, 254, 2 5 5 , 26 1 27, 1 06, 1 68 1 30 32, 33, 34, 39, 49, 1 20, 1 22, 1 92, 1 94, 1 95 , 2 5 5 1 37, 1 38 , 1 39

D

Damlj anovic D avid Dodon Domin g uez D u n is Duras Dvoretsky

227, 228 37, 6 1 34 1 6, 1 06 37 1 00, 1 0 1 , 1 02 27, 1 05 , 1 48

E

Eljanov Euwe

3 5 , 36, 1 30, 1 3 1 25

F

Fischer Flohr Fressinet Fridman Furman

7, 1 2, 33, 38, 68 84 254, 256 254 33

Name I ndex

279

G

K

24 1 32, 239 Gavrikov 8, 1 2 Geller Georgiev 32 Giri 27 229 Glauser 1 54, 228 Glek Grachev 274 Grischuk 5, 1 2, 60, 67, 92, 1 53, 1 54, 1 5 5 , 1 57, 1 58, 1 59, 1 62, 1 64, 1 65, 1 66, 22 1 , 226, 228, 254 191 Gurevich

3 5 , 246 Kapengut 8 Karj akin 27, 35 Karpov 1 2, 40, 82, 83, 84, 85, 1 39, 1 40, 1 4 1 , 1 42, 205, 244, 245 , 254, 26 1 , 273 Kasparov 33, 34, 49, 62, 82, 83, 84, 8 5 , 87, 88, 89, 92, 94, 1 42, 145, 1 93, 2 1 2, 244 Khairullin 1 20 Khalifman 5, 1 1 2, 1 1 3, 228 Khusenkhojaev 252 Korchnoi 33, 238 Kosteniuk 12 Kozul 239 Kramnik 7, 42, 67, 83, 90, 9 1 , 1 07, 1 1 3, 1 43, 1 45, 1 68, 1 93, 1 94, 244, 254, 273 Kuzubov 34

Gaerths

H

Har-Zvi Hiibner Huzman

I

lvanchuk

1 68 245, 246 8, 66, 1 09, 1 2 1 , 1 69, 1 97, 204, 239, 270 5, 1 4, 1 5, 1 6, 1 9, 20, 23, 83, 1 1 0, 1 32, 246

Johansen

L

Larsen Lasker

J Janowski

Kam sky

Lautier

6 1 , 63, 64, 65 , 92 111

Leko Leven fish Lj uboj evic Loginov Lukacs Lysyj

7, 1 2, 47, 86 1 3, 220 12 1 20, 1 78, 205, 229 2 1 4, 2 1 5 , 277, 282 227 1 09 1 12 34

280

Positional Decision Making in Chess

M

R

28, 29, 32, 34, 36, 39, 49 1 68 Manor 2 1 6, 23 1 Marin 24 1 Meissner 229 Milos 1 68 Milov 277 Minev 1 82, 204 Mnacakanian 246 Morovic Fernandez 4 5 , 48, 49, 50, 54, Morozevich 56, 57, 5 8 , 84, 1 06 32, 1 32, 1 33 , 1 36, 229 Movsesian Muehlebach 69 Muller 1 05 , 1 48 Murakhveri 7, 1 2

68, 69, 70, 7 1 , 1 68 Razuvaev 7, 1 2, 22 1 1 39, 1 40 Ribli 8 1 , 1 54 Roiz Rubinstein 3 , 7, 8 , 1 1 , 1 2, 1 3 , 1 4, 24, 25, 26, 27, 44, 4 5 , 47, 52, 54, 56, 60, 6 1 , 62, 64, 65, 66, 96, 97, 99, 1 00, I O I , 1 02, 1 03 , 1 04, 1 05 , 1 1 6, 1 37, 1 38 , 1 39, 1 42, 1 43 , 1 44, 1 46, 1 47, 1 49, 1 52, 1 53 , 1 58, 1 69, 1 79, 1 92, 200, 208, 2 1 0, 2 1 1 , 2 1 2, 2 1 3, 2 1 4, 2 1 5 , 2 1 6, 2 1 7, 2 1 8, 2 1 9, 220, 22 1 , 222, 223, 224, 226, 236, 270 244, 245 , 246, 2 5 1 Rublevsky

Malakhov

N

Naiditsch Naj er Nakamura Nepomniachtchi Nimzowitsch Norri

0

Odessky Orarovsky

p

Petrosian Philidor Piker Polugaevsky Portisch

1 54, 269 1 02 1 06 27 27, 6 1 , 86, 1 92, 2 1 1 , 220 1 80

Radjabov

s

Salem Sasikiran Savon Schlechter Shereshevsky

Shirov Short

1 3, 1 4, 27, 1 20, 1 2, 34, 227, 254, 1 2 1 , 1 42, 1 44,

Smirin

277 33

Smyslov Sokolov Spassky

68,

Stean Stein

33, 56, 60, 8 5 , 1 58, 204, 2 1 0, 2 1 2 1 36 85 8 , 89, 229 273

Sutovsky Sveshnikov Svetushkin Svidler

81, 101,

1 43 1 4 5 , 274 1 69 2 1 6, 2 1 7 1 05 , 1 06 274, 275 1 4 5 , 246 8 1 , 252 277 254 1 53 , 204 97, 98 1 69 1 68, 252 1 42, 275 34 90, 228

Name Index

T

45, 46, 47 35, 60, 1 69, 274 Tal Tella 1 80 69, 1 42 Timm an 1 2, 1 02, 1 94 Tomashevsky Topalov 1 2, 49, 83, 94, 1 68, 2 1 2, 267, 273 69, 1 1 1 Torre 238, 239, 240, 242 Tseitlin Takacs

v

Van Wely

Volkov

67, 7 1 , 72, 76, 77, 78 1 30, 1 3 1

w

Wang Hao Wang Yue Wells

Winter Wolf

y

Yusupov

z

Zaitsev

Zilberman

28 1

27 1 20, 1 26, 1 32 240 145 1 44, 1 52 83, 99, 1 05, 1 09 227, 244 1 68

Game Index Boris Gelfand - Abram Gelfand, Minsk

Gelfand - Campora, Cesme

2004

Gelfand - Shirov, Polanica Zdroj

( 1 O) 1 975

1 998

Akiba Rubinstein - Carl Schlechter, San Sebastian

1912

Boris Gelfand - Vassily Ivanchuk, Russian Team Championship, Dagomys

Gelfand - lvanchuk, Lvov

2000

2000 2009

2009

Gelfand - lvanchuk, Polanica Zdroj lvanchuk - Dominguez, Linares

Esteban Canal - Akiba Rubinstein, Karlsbad

1 929

2008 2005

Vladimir Malakhov - Evgeny Bareev, Russian Team Championship, Dagomys Boris Gelfand - Vladimir Malakhov, Russian Team Championship, Sochi

Robert James Fischer - Tigran Petrosian, Buenos Ai res ( 1 )

Gelfand - Karj akin , Moscow (blitz)

20 1 2 2005 (ol) 1 974

20 1 3

Elj anov - Malakhov, Eilat

David - Dunis, Porz Barcares Karpov - Unzicker, Nice

Akiba Rubinstein - Sandor Takacs, Budapest

1 926 200 1

Boris Gelfand - Alexander Morozevich, Astana

Gelfand - Adly, Dresden (ol)

2008 200 1

Anand - Morozevich, Dortmund

1 907 2003 Am broz - Muehlebach , Bad Ragaz 1 99 1 Gelfand - Radj abov, Enghien-les-Bains 2003 Carlsen - Anand, Sochi (6) 20 1 4 Carlsen - Anand, Sochi ( 1 1 ) 20 1 4 Boris Gelfand - Garry Kasparov, Linares 1 994 Kasparov - Yusupov, Linares 1 993 Flohr - Capablanca, Moscow 1 935 Svidler - Kramnik, Khanty-Mansiysk 20 1 4 Akiba Rubinstein - Oldrich Duras, Karlsbad 1 9 1 1 Evgeny Najer - Evgeny Tomashevsky, Vladivostok 20 1 4 Boris Gelfand - Magnus Carlsen, Moscow 20 1 3 Boris Gelfand - Daniel Hugo Campora, Cesme 2004 Boris Gelfand - Wang Yue, Medias 20 1 0 Capablanca - Treybal , Karlsbad 1 929 Alekhine - Chajes, Karlsbad 1 923 Eljanov - Volkov, Yerevan 20 1 4 David Janowski - Akiba Rubinstein, Karlsbad Boris Gelfand - Loek van Wely, Plovdiv

1 97 1

6 10 12 13 14 15 15 16 24 28 32 33 35 36 37 40 45 48 49 50 61 67 69 71 75 77 82 83 84 90 1 00 1 02 1 06 1 09 1 20 1 29 1 30 1 30

283

Sergei Movsesian - Wang Yue, Istanbul

Gelfand - Ivanchuk, Elancourt

20 1 3

Gelfand - Movsesian , Polanica Zdroj

20 1 2

2000

1 909 1 980

Erich Cohn - Akiba Rubinstein, St Petersburg Zoltan Ribli - Anatoly Karpov, Amsterdam

Sveshnikov - Kasparov, Minsk Salem - Kram nik, Doha

20 1 4

1 979

Heinrich Wolf - Akiba Rubinstein, Teplitz-Schoenau

1919

Winter - Capablanca, Hastings

20 1 1 Hyderabad 2002

1 922

Shore - Kramnik, London Sasikiran - Anand,

Boris Gelfand - Alexander Grischuk, Beij i ng

Gelfand - Naidi tsch , Wij k aan Zee

20 1 4

Boris Gelfand - Boris Avrukh, Ramat Aviv

Stein - Savon, Riga

1 970

1 999

Levon Aronian - Boris Gelfand, Dresden (ol)

Bocvinnik - Basman , Hastings Norri - Tella, Finland

1 995

1 966

Adams - Gurevich, Dresden (ol)

20 1 3

2008

2008

Boris Gelfand - Viswanathan Anand, Moscow

(2) 20 1 2 (4) 20 1 2 Gelfand - Anand, Moscow ( 6) 20 1 2 Gelfand - Aronian, Tashkent 20 1 4

(7) 20 1 2

Gelfand - Anand, Moscow

Gelfand - Anand, Moscow

Gelfand - Aronian, Paris/Sc Petersburg

20 1 3

1 928 191 1 Akiba Rubinstein - Carl Schlechter, Berlin 1 9 1 8 Akiba Rubinstein - Alexander Alekhine, Karlsbad 1 9 1 1 Boris Gelfand - Branko Damljanovic, Istanbul (ol) 2000 Damlj anovic - Lj ubojevic, Manila 1 990 Leko - Carlsen, Nice (rapid) 2008 Glauser - Polugaevsky, Havana (ol) 1 966 Boris Gelfand - Mark Tseitlin, Tel Aviv 1 999 Korchnoi - Tseitlin, Leningrad 1 973 Wells - Beaumont, Edinburgh 1 989 Meissner - Gaerchs, Germany 1 992 Boris Gelfand - Sergey Rublevsky, Palma de Mallorca 2008 Bologan - Rublevsky, Saint Vi ncent 2005 Gelfand - Hilbner, Munich 1 994 Akiba Rubinstein - Aron Nimzowitsch, Berlin

Aki b a Rubinstein - Grigory Levenfi.sh, Karlsbad

1 32 1 32 1 36 1 37 1 39 1 42 1 43 1 44 145 1 45 1 45 1 53 1 54 1 68 1 69 1 78 1 80 1 80 191 1 92 1 93 1 93 1 94 1 95 1 97 21 1 214 216 219 227 227 229 229 238 238 240 24 1 244 245 246

Kamsky - Short, Linares

( 1 ) I 994

Morovic - Rublevsky, Khanty-Mansiysk Bareev - lvanchuk, Dortmund

1 992

Khusenkhoj aev - Smirin, Tromso (ol)

( I . I) 2009

20 I 4

Boris Gelfand - Laurent Fressinet, Canada de Calatrava (rapid)

(3) 20 I 2 Gelfand - Anand, Moscow (7) 2 0 I 2 Gelfand - Anand, Moscow (8) 20 I 2 Gelfand - Anand, Moscow (9) 20 1 2 Portisch - Karpov, B ugoj no 1 978 Shirov - Grachev, Lublin 20 I I Gelfand - Anand, Moscow ( 1 5) 20 1 2 Gelfand - Anand, Moscow

2007

246 246 246 252 254 266 269 270 272 273 274 276

O p ening Index 6

Philidor Defence

6 1 , 1 44

Four Knights Game French Defence

2 1 4, 274

Sicilian Defence

33, 67

Queen's Gambit Accepted

28, 1 37, 244

Queen's Gambit Declined

45, 82, 1 06

Semi-Tarrasch Defence Slav Defence Chebanenko Slav Chebanenko Semi-Slav Catalan System Griinfeld Defence

13 49, 2 1 9 , 254 32, 34, 49, 50, 1 20, 1 30, 1 32 1 92, 269 1 39 1 4 , 239

King's Indian Defence

1 53, 228

Old Indian Defence

1 09, 2 1 6

Nimzo- lndian Defence English Opening Hedgehog System

21 1 1 5 , 1 00, 1 68, 1 79, 238 24