The Secret Life of Bad Bishops - Lund

���-

QUALITYCHESS ---� ---

11

GAMES/CHESS

The Bad Bishop is a notorious villain in chess, but often a perfectly respectable bishop is falsely accused. takes an in-depth look at the bishop - not just dealing with good and bad bishops but also more nuanced cases where

a

"double-edged

bishop" could turn out to be a game-winning hero or a fatal liability. All phases of chess are covered, from opening to middlegame to endgame. The final two chapters test the reader's newly enhanced understanding with carefully chosen exercises and instructive solutions.

Esben Lund, an International Master from Denmark, is a respected author

and coach. His first book for Quality Chess, Rook vs. Two Minor Pieces,

earned excellent reviews.

€19.99

I I QUALITY

ISBN

CHESS

www.q ua I itychess.co.u k

$24.95

978-1-906552-19-0 52495

>

The Secret Life of Bad Bishops By

EsbenLund

Quality Chess .qualitychess.co. uk

www

First edition 20 1 4 by Quality Chess UK Led Copyright © 20 1 4 Esben Lund

The Secret Life of Bad Bishops All rights reserved. No part of chis 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 -906552- 1 9-0 All sales or enquiries should be directed to Quality Chess UK Led, 20 Balvie Road, Milngavie, Glasgow G62 7fA, United Kingdom Phone +44 1 4 1 204 2073 e-mail: [email protected] website: www. qualicychess.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. Skromna 3, 20-704 Lublin, Poland Typeset by Jacob Aagaard Proofreading by John Shaw & Daniel McGowan Edited by David Friedgood Cover design by adamsondesign.com Front Cover Image: Philippe de Champaigne and studio, Triple Portrait of Cardinal de Richelieu © The National Gallery, London. Presented by Sir Augustus Wollaston Franks, 1 869 Back cover Photo by Rune Lund Printed in Poland by Drukarnia Pionier, 3 1 -983 Krakow, ul. Igolomska 1 2

Publisher's Foreword In this Foreword I will introduce the author, and allow Esben to explain what his book is about in the Preface. I have known Esben for about twenty years. Back in 200 1 we wrote a book together, Meeting J.d4, for Everyman Chess. Unlike a lot of other authors, we actually sat in the same room as we wrote, sharing ideas and opinions. The book was no runaway success, but it was decent and we learned a lot from the process. When I co-founded Quality Chess in 2004, Esben told me he had an idea for a book. I did not think twice before offering him a contract, even if the subject, Rook vs. Two Minor Pieces, was a bit technical and unlikely to sell lots of copies. This was exactly how I wanted our new publishing house to be: willing to publish a less commercial book if we felt the content was interesting. We expected the book to find a limited audience, but many friends have told us that they loved it, finding it thought-provoking. One admirer of the book is the Women's World Champion, Hou Yifan, who revealed that she read it as part of her preparation for the 20 1 3 World Championship match against Anna Ushenina, which she won 5Y2-l Y2. Ushenina came well prepared, with top grandmasters to provide her with great openings. But Hou Yifan's focus on the middlegame and endgame proved a sounder strategy. I think many people mistook Rook vs. Two Minor Pieces for an endgame book, when actually it had more material on the opening than the endgame, and overall it was mainly a middlegame book, should you have a deep desire to pin it down. This book might suffer from the same misunderstanding, but at least we have tried to make the title a bit more interesting this time! Ten years have passed and Esben has not only improved as a player, he has also matured (this lovely word that makes growing older into a positive thing) as a writer, as this book shows. Quality Chess is happy to continue our mission of publishing interesting and challenging books, no matter how uncommercial their focus. You might suspect that I am trying to put you off reading this book, but we both know that if you have read this far, you are likely to keep on reading; and you will be rewarded for that decision. Jacob Aagaard Glasgow, July 20 1 4

Contents Preface

7

1. Introducing the Double-edged Bishop - Definitions and Limitations

9

1.1 Three opening lines with DEBs

9

1.1.1 The Rubinstein Nimzo-lndian

10

1.1.2 The Berlin Wall

11

1.1.3 The Advance French

12

1.2 Other important factors

13

1.3 Limitations to positions with DEBs

14

1.3.1. Positions with DEB relevance

15

1.4 More about the DEB: Alekhine -Yates

17

1.5 More complex examples

24

1.5.1 Bishop vs. knight

27

1.5.2 Two examples from John Watson

29

1.5.3 A modern grandmaster game: Radjabov - Eljanov

31

1.6 Chapter recap

2. The Exchange Sacrifice

2.1 Case 1: Rook takes DEB - the breakthrough

39

41

42

2.2 Case 2: Exchange sacrifice to strengthen the DEB

44

2.3 Case 3: Exchange sacrifice to weaken the DEB

46

2.4 Chapter recap

59

3. From Opening to Early Middlegame - The Advance French 3.1 Guiding questions

3.2 Theoretical overview 3.3 Chernin's 9 ...�b4!?

3.3.1 The solid 1 l .�xc3

3.3.2 The double-edged 1 l .bxc3!?

61

61 62 67 68 70

3.4 White allows ...g5

71

3.5 Other positions

72

3.6 Answers to questions

78

3. 7 Variation Index

86

3.8 Chapter recap

88

4. Endgame Section

4.1 Rook vs. bishop

89 89

4.1.1 Fortress positions

92

4.1.2 Fortress-like Position (FLP)

97

4.1.3 Fire on Board

98

4.1.4 Complex endgame examples 4.2 Bishop vs. knight

100 111

4.2.1 Zugzwang

111

4.2.2 The other factors:

114

4.2.2.1 The rook's pawn

114

4.2.2.2 The knight dominates

116

4.2.2.3 King and pawn space 4.2.3 Good technique

119 122

4.2.3.1 Pawns and bishop on the same colour

122

4.2.3.2 Bishop check that forces the king to choose a side

123

4.2.3.3 Counterplay is allowed

124

4.2.4 Worst-case scenarios 4.2.4.1 Creative Chess Strategy

128 130

4.2.5 The bishop pair vs. bishop and knight

132

4.2.6 Chapter recap

137

4.2.6.1 Rook vs. bishop

137

4.2.6.2 Bishop vs. knight

138

S. Exercises Introduction Exercises 1-30

6. Solutions

141

141 143

151

Games Index

187

Index of Studies and Positions

189

Name Index

190

Key to symbols used ± +

+-+ =

iii

+!

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 with compensation with counterplay

CD

?? !! !? ?!

#

unclear a weak move a blunder a good move an excellent move a move worth considering a move of doubtful value mate

Bibliography

Engl ish titles

Aagaard: Easy Guide to the Panov-Botvinnik-Attack, Everyman Chess 1 998 Berliner: The System, Gambit 1 999 Dvoretsky & Yusupov: Positional Play, Batsford 2003 Lund: Rook vs. Two Minor Pieces, Quality Chess 2005 Marin: Learn from the Legends, Quality Chess 2004 Romero Holmes: Creative Chess Strategy, Gambit 2003 Shirov: Fire on Board, Volume l, Everyman Chess 1 997 Tisdall: Improve Your Chess Now, Everyman Chess 1 997 Watson: Secrets ofModern Chess Strategy, Gambit 1 999 German titles

Dvoretsky: Die Endspieluniversitdt (Endgame Manual) , Chessgate 2002. Awerbach: Endspiele. Springer gegen Laufer; Turm gegen Leichtfigur. Sporcverlag Berlin 1 989 From Averbakh's endgame work in six volumes. My book includes references such as "Averbakh (473)" and "Averbakh (477)". This allows the reader to distinguish between the positions and also track down the examples in Averbakh's original books. Danish titles

Lund: Skoleskak

1,

forlaget Rosenkilde 2008

Electronic/Internet resources

Chess Base 1 0 Nalimov's endgame database - any position up to 6 pieces including the kings.

Preface You hold in your hands a book that has been long in coming to fruition. It took me years to collect material for the early chapters as well as for the exercise section, and it demanded further time testing it on several chess players of different strengths. When I finally concluded some chapters, I realized from the feedback I received that some of them needed restructuring. In the end I excluded several exercises and also one chapter. My ambition with the book is to introduce the subject of good and bad bishops in a proper way, and that's why the introductory chapter has become slightly longer than usual compared to other chess books. I find existing views of good and bad bishops slightly simplified and misleading. Because the way I am dealing with them is quite a delicate subject, I simply decided not to let the size of the introduction be the main issue, but rather to focus on communicating my view in a proper and understandable way. Both IM Andreas Hagen and FM Sebastian Nilsson told me that, while reading through the introduction, they thought at the beginning: "Now the bishop MUST be bad", while I kept calling it a double-edged bishop (DEB). And the more they read of the introduction, the more they became accustomed to my way of thinki ng. Danish IM Nikolaj Mikkelsen made a nice comment that in many ways grasps and confirms my idea with this project: he said that he is now less afraid that his potentially bad bishop will in fact become bad. The structure of the book is similar to that of Rook vs. Two Minor Pieces - my first book for Quality Chess: I am dealing with a general subject - this time good and bad bishops - and each of the chapters represents a phase in the game. The introductory Chapter 1 is followed by Chapter 2 on exchange sacrifices. This is a quite natural follow-up, since this sacrifice can be a way to exchange the right pieces and isolate the potentially bad bishop further. As we shall see throughout the book, the presence of additional pieces on the board is crucial in determining whether the double-edged bishop will in fact end up as a bad bishop. Chapter 3 deals with the transformation from opening to early middlegame. For that purpose I have chosen to look deeper into a certain variation of the Advance French ( l .e4 e6 2.d4 d5 3.e5), as this opening line leaves both White and Black with double-edged bishops. I realize that not everyone plays this line, but this should not be the main point: the material is specifically chosen beyond j ust the opening theory such that an understanding of the early middlegame is achieved. And by following the good advice from Andreas Hagen to pose questions to the reader throughout the chapter (Dvoretsky likes to do this as well) , this chapter can be seen as a test as to whether the reader has captured the essence of the first part of the book. Please answer each question, and check the solution before you move on, as sometimes the next question follows on from the previous question's solution.

8

Esben Lund - The Secret Life of Bad Bishops

Chapter 4 is devoted to the endgame. Here I have analysed in depth the pure endgames rook vs. bishop and bishop vs. knight. It makes a lot of sense to consider what happens if the additional pieces go off the board and thereby ask the question: "What is the bishop able to do on its own?" This led to some surprising and very useful conclusions. Rook vs. bishop endgames are the sharpest I have ever come across. By entering a pure rook vs. bishop ending, the side with the bishop cannot allow the position to come to a standstill, where only the rook side can improve his position (on the colour inaccessible to the bishop) . If he allows the position to come to a standstill, he should be absolutely positive that it is a fortress. I present and discuss the possible fortresses for comparison. If the fortress-like position (FLP) is not a fortress, then the position is simply lost for the bishop side. It was also beneficial to see in just how many positions i� is better to place the pawns on the same

colour as the bishop - contrary to the simplified rule of thumb that you should place the pawns

on the opposite colour of the bishop. On the same colour the pawns can better be protected by the bishop and this comes in useful if, for instance, your activity is taking place on the kingside, whereas you only wish to defend the queenside from a distance. The concluding Chapters 5 and 6 are devoted to exercises and their solutions, where the reader is able to test if he or she has grasped the essence of all aspects of the double-edged bishop. The exercises are given a level and a recommended time for solving, and the solutions are discussed thoroughly. Thus the middlegame is given special attention in both the introductory Chapters 2 & 3 as well as in the exercise section. The material presented in this book is on a high level, and I believe that players rated 1 900 and above will benefit the most from it. Players eager to improve their play and understanding of the game will of course get a lot from the book as well. As an author and trainer it was important for me to test the material on players who are aspiring for tides (2 IMs and an FM) . I can say with confidence that IMs will benefit a lot from the book on their way to the GM tide, and I also believe that grandmasters can learn a thing or two. I would like to thank all the people involved in this project, especially Andreas Hagen, Nikolaj Mikkelsen and Sebastian Nilsson for giving useful and critical feedback on the whole book from an early stage. This really shaped the book into what it has become. And a special thank-you goes to Jacob Aagaard for making this book possible. For me it was crucial to get an additional good and relevant critique from a grandmaster. In any case, I hope you will find the book beneficial . A note on my name: In 2009 I took a second first name (Silas) that I use in my everyday life, but for reasons of continuity I decided to keep Esben Lund on the front cover. Silas Esben Lund Copenhagen, July 20 1 4

Chapter 1 Introducing the Double-edged Bishop Definitions and Limitations W hen a bishop in a given position can be evaluated as neither good nor bad, it is something in

between - and I shall name this in between "Double-edged bishop" or DEB for short. A DEB has

the potential of both becoming good as well as the risk of becoming bad, depending on the given factors in the position, but for the time being it is neither of the two, and the way the players treat the position will decide the bishop's fate.

Double-edged bishop (DEB): A bishop that cannot yet be declared good or bad, but rather holds the potential of becoming good as well as the risk of becoming bad.

Bad bishop: Analysis of all relevant other factors in the position will show that a bishop has definitely gone from double-edged to bad.

The reason for my self-invented term is that many players and annotators prematurely declare the bishop to be a bad piece. In this declaration process other factors in a given position are often neglected: is a bishop passively placed behind its own pawns always bad? This view is too simplistic, as are other rules of thumb that have found their way into the teaching manuals. This book is dedicated to presenting a more nuanced view on the matter. It is important to be able to spot a DEB in a given position. The strategic fight that arises after the exchange of bishop for knight yields one player the bishop pair and his unique bishop (see below) is often a candidate for a DEB.

Unique bishop: A bishop of a type only possessed by one player (i.e. the opponent might have a bishop, but not one on the same colour squares) . In a short while we shall see that a bishop does not necessarily need to be unique to qualify for DEB - see positions

1 . 1 . 2 and 1 . 1 . 3

below.

We now tum our attention to three examples taken from well-known opening lines.

a

10

Esben Lund - The Secret Life of Bad Bishops

1.1 Three opening lines with DEBs Position 1.1.1 The Rubinstein Nimzo-lndian 1 .d4 �f6 2.c4 e6 3.�c3 i.b4 4.e3 0-0 5.i.d3 d5 6.�a cs 7.0-0 �c6 8.a3 Lc3 9.bxc3 dxc4 IO.i.xc4 'f!c7 I 1.i.b2 e5

a

b

c

d

e

f

g

h

From the perspective of the DEB, the Nimzo-lndian is in general a very interesting opening. Already on move 3 Black indicates that he is willing to part with his dark-squared bishop and thus the bishop pair, and for this trade he usually gets other advantages: either a crippled white pawn structure with doubled c-pawns or time for development if White spends time on a2-a3. The first diagram shows one of the main lines arising after the modest-looking 4.e3. The interesting strategic feature in the diagram position is the unopposed white bishop on b2, and this unique bishop is a DEB. White controls the dark squares extremely well at the moment, also due to his central pawns on c3, d4 and e3. A bishop controls one specific colour complex better than a knight, since the knight changes colour every move: when it is on a light square it controls only dark squares and vice versa.

But the bishop's supenomy on the dark squares at the same time shows a vulnerability on the opposite colour complex - the light squares. Therefore White often makes moves like 1tW d l -e2 and h2-h3 to strengthen his light­ squared control, and he usually tries to avoid the exchange of light-squared bishops. In fact, he should carefully consider every exchange of the remaining pieces on the board (especially the minor pieces) , since the DEB runs the risk of being isolated on its colour complex - and in the worst possible case ends up being a bad bishop. The DEB often shows its potential in cooperation with the other pieces and pawns. It is exactly these strategic pros and cons that make the bishop on b2 a DEB. Black's advantages are a smooth and speedy development and an early initiative that he has been given due to White's time consuming a2a3 - an initiative that he hopes will develop. When we discuss the DEB on b2 in the above position, it reminds us of the simplistic statement that 'the bishop is bad if it is placed behind its own pawns' . But as we shall see shortly, it is not j ust about freeing the bishop on b2 because it is placed behind the pawns c3-d4-e3. Even if it lands on a3, it may not be attacking anything but thin air on the diagonal a3-f8. How to make full use of the DEB for attacking purposes is discussed in the upcoming examples. At this stage I would like to state a simple but important truth about the DEB: The DEB is never entirely isolated on its colour, since the king is always there to support the other colour complex.

Chapter I

-

I ntroducing the Double-edged Bishop

Position 1.1.2: The Berlin Wall l .e4 e5 2.�0 �c6 3.i.b5 �f6 4. 0--0 �xe4 5.d4 �d6 6.i.xc6 dxc6 7. dxe5 �f5 8.WfxdSt iixd8

8 7 6 5 4 3 2 a

b

c

d

·

e

f

g

h

This Spanish line 3 . . . li:Jf6 earned its name 'The Berlin Defence' around 1 890, when it was played in a series of big tournaments in Berlin arranged by the German Chess Federation. When Kramnik successfully used this opening against Kasparov in their 2000 World Championship match, defending this position time and again with the black pieces, the opening was nicknamed 'the Berlin Wall'. After the exchange of queens, Black has lost the right to castle. This means that he will have to spend some time getting full coordination between his pieces. The king is in the centre, but the absence of queens makes a white assault less likely to succeed. There is a fine line between the black king being a liability or a strength in the Berlin Wall; in many situations it proves to be closer to the scene of action than its white counterpart once further pieces go off the board. Black's pawns are doubled on the c-file and White has been given a healthy 4 against 3 pawn majority on the kingside. What makes the Black player go for this opening system, one might ask.

11

The answer is simple: The unique light-squared DEB on c8. White would love to advance his kingside pawns to gain control (especially) over the f5-square, but this goal can hardly be achieved in the near future and the reason for this is mainly due to the power of the c8-bishop. But why is the bishop on c8 a DEB, and not simply a good bishop? In most lines of this opening, White strives to exchange it off, and Black usually avoids it. The answer is again that it's too early to make such a definite evaluation. There are still many other remaining pieces and pawns on the board. In theory, this bishop could still end up as a bad bishop. Imagine that all other pieces beside this bishop and a white knight go off the board, and that White places his pawns in such a way that they cannot be attacked by the bishop. Hitting 'thin air' is very useful when cooperating with other pieces, but the lack of a concrete target in the opponent's position is at the same time a risk. Take a look at this endgame position:

a

b

c

d

e

f

g

h

White can continue with �g3-h4 followed by h2-h3 and g2-g4, when his pawn majority starts moving forward. Meanwhile, Black can do little to create any real counterplay.

Esben Lund - The Secret Life of Bad Bishops

12

In the Berlin Wall, Black in general hopes to reveal his sleeping potential in the position slowly, at the same time controlling the white initiative (especially the break e5-e6 can be dangerous) , and he builds this hope upon the fact that it is very hard for White to spot a clear target in his position - c7 or f7 are so far easily defended. Furthermore, there are no visible outposts for the two white knights. So this difficult strategic position demands accuracy from the White player too.

Position 1 . 1 .3: The Advance French l .e4 e6 2.d4 d5 3.e5 c5 4.c3 �c6 5.�f.3 �ge7 6..id3 cxd4 7.cxd4 �f5 8 .bfS exf5 •

gives Black more space in the centre and on the kingside, and White should look out for . . . g7-g 5, grabbing more space over there. As in other French positions, the d4-pawn is

a possible target in the white position. In my opinion, Black has a fully playable position here. A complex struggle lies ahead and I will use this early middlegame position to delve deeper into the secrets of the DEB (Chapter 3). The bishop pair is generally (and for a good reason) valued highly, so often its possessor will retain it at some cost. In this French Advance variation the price is a worsening of the pawn structure; the same goes for the Berlin Wall. In the Nimzo-lndian above, White's pawn structure is without doubled pawns, but the first player has spent time on the ambitious a2-a3 to obtain the pair of bishops. These examples should give the reader an impression of the nature of the DEB, but the bishop does not necessarily need to be unique to be a candidate for a DEB. The white bishop on cl in both the Berlin Wall and the French Advance variation is not unique, but it has a potential of being a DEB in the future.

a

b

c

d

e

f

g

h

After the exchange on f5 the black pawn structure has been blown into pieces: d5 is isolated, and the f-pawns are doubled. This is the obvious disadvantage for Black after the minor piece exchange. On the other hand, the d5-pawn is not so easy to attack, or to put it cOirectly it is fairly easy for Black to defend, since his DEB on c8 has got itself a home on e6. It does look very passive at e6, but it solidly gives support to all the pawns d5, f5 and f7. Hence, the exchange has scattered the pawns, but given Black an easy development. Also, the new pawn on f5

If White in the diagram position above plays the natural-looking 9.�c3, Black could respond with Chernin's move 9 .. ..ib4!?, and after the further 10 ..i d2 hc3 1 1 . .ixc3

Chapter I

-

I ntroducing the Double-edged Bishop

White's dark-squared bishop on c3 is a DEB. This is due to his own pawns on d4 and e5 placed on the colour of the bishop, as well as the lack of obvious attacking points in the black camp. On the basis of chis variation and the possible exchange of Black's dark-squared bishop for a white knight, it makes sense to treat White's dark-squared bishop as a DEB before chis exchange eventually if ever - takes place. -

The same is the case in the Berlin Wall if Black carries out the exchange i.b4xlll c3. Only then does White get a unique dark-squared bishop, but the knowledge of chis possible exchange of minor pieces allows us to talk about the c l ­ bishop as a DEB earlier on, even i n positions where Black still· holds the bishop pair. The pawn on e5, perhaps even with the f-pawn advanced to f4, indicates possible restrictions of the bishop. On the plus side the bishop can be a monster if White is able to play a timely break with e5-e6, thus opening the diagonal a l -h8 for the bishop. This is exactly why we call the bishop a DEB, since other factors and further play will provide the answer. Theoretically speaking, I could name both these bishops on cl a DEB of the second degree, but admittedly it sounds a bit clumsy. Instead I will give many practical examples of these "hidden DEBs" and soon I believe chat the reader should learn how to spot chem early on in a game and be alert. I shall therefo re simply call the bishop a DEB in both cases. A DEB is not necessarily a unique bishop

In both the French Advance and Berlin Wall above, note chat Black also isolated his own DEB on c8 with the exchange of a pair of minor pieces, so obviously chis decision is double-edged.

13

1 .2 Other important factors The real clue in such positions is to find the other factors chat decide how the potential can be used by both players to obtain their goals, and in a wide range of positions chis goal is often connected to the potential of the DEB. There is some truth in the saying chat the position is no better than the worst placed piece. So it is natural for the player fighting the DEB to exploit its risk of becoming bad, whereas the possessor strives to get the best out of it. For now, I will only point out what I consider the twb main factors to be, namely: 1 ) The remaining complementary pieces on the board 2) Pawn space (Throughout the book I will use the more common term 'space' for far-advanced pawns. The reason for the term 'Pawn space' is to separate it from the term ' King space' - an important concept in the endgame chapter.) Unfortunately, more often than not, the DEB is declared a 'bad bishop' on false grounds and much sooner than it deserves: the human desire for clarity often leads to premature evaluations of a position. The bishop has a one-sided nature as it only operates on one colour complex, and chis, I believe, gives the impression chat a positional verdict of its value is easy to come up with, especially for weaker club players. If chess players in a chat forum want to make a comment on a complex position, often the bishop falls prey to an immediate judgement not built on deep analysis, but on the mere fact chat it is placed behind its own pawns. In comparison, the knight is a far trickier minor piece to grasp due to its often surprising movements. By such premature verdicts other defining factors are more often than not shut out.

14

Esben Lund - Th e Secret Life of Bad Bishops

Take the King's Indian: The bishop on g7 is passive in many lines for a long time, but many years of accumulated 'chess culture' has taught the Black player not to panic: chis bishop does a great job defending the king, and lacer it has chances to widen its scope and join the battle - for instance via the h6-square. Besides chat, there are many remaining complementary pieces in King's Indian positions - and many assaults on the white king have been conducted without the help of the g7-bishop. On the downside, the horrific endgame with an untouchable knight on e4 against the g7bishop is well known to most players: here the black DEB on g7 runs the risk of ending up as a bad bishop. In other words, we have actually forgotten chat the King's Indian bishop is a DEB! My hope is that this book provides a tool for handling chis complex question in a more delicate manner than prior chess literature has been able to, in the end giving the reader a more differentiated view upon the subject: before you judge the bishop good or bad, call it double-edged!

1.3 Limitations to positions with DEBs Is it possible to speak of double-edged bishops in all kinds of positions? le does not always make sense, and in chis section I shall describe relevant and irrelevant positions in connection with the term DEB. We start out with an example from the Nimzo­ lndian opening that goes horribly wrong for White; the Black player is none other than the inventor of the opening:

Hermanis Matisons - Aron Nimzowitsch Carlsbad 1 929

1 .d4 ll:\f6 2.c4 e6 3.lll c3 .ib4 4.lll a hc3t 5.bxc3 d6 6.�c2 �e7 7 ..ia3 c5 8.g3 b6 9.i.g2 .ib7 1 0.0-0 0-0

a

l 1 .lll h4?!

b

c

d

e

f

g

h

White's play in the opening has not been convincing: the bishop sortie to a3 looks strange. Now the knight sortie to the rim allows an unfortunate exchange of lighc­ squared bishops.

l 1 . ..hg2 1 2.@xg2 1 2.fLixg2 looks better, to get the knight back in the game.

12...�b?t 13.@gl �a6 The weak spot in White's position is c4.

1 4.�b3 lll c6 1 5 J:Udl lll a5 1 6.�bS �xb5 17.cxbS White avoided the loss of the c4-pawn, but the weakness of the square remains.

17 ... ll:\c4 18 ..icl a6 19.bxa6 gxa6 20.dxcS bxc5

Chapter I

-

I ntroducing the Double-edged Bishop

8 7 6 5

15

be used in a wide range of positions, mainly those of half-open or half-closed character. The definition and the idea of the DEB are universal, but in some positions the concept loses its practical value, and these limitations are the focus of attention now.

4

3 2

a

b

c

d

e.

f

g

h

In this game it is arguable whether the term DEB makes sense at all. I believe it does not, and the reason is the closed nature of the position. The bishop sortie to a3 was a bad idea and Black could counter with the pawn bolster b6-c5-d6: In this case the white bishop was hemmed in by the opponent's pawns and not his own, but with the same effect: the bishop was restricted in its movement. To exaggerate my point one can say that in closed positions all bishops are bad. White's position in the above diagram is bad. Still, this is not due only to his bishop on c l . His scattered pawn structure and bad coordination between the pieces count too. So, to be fair to the bishop on c l , White's position is bad for many reasons, but surely all White's problems accelerated with the exchange of light-squared bishops on move 1 2? In the next section we shall discuss in more detail the (unfair) tendency to make a (bad) bishop the scapegoat for all evils arising from an inferior position.

2 1 .lll g2 lll d5 22.gd3 gras 23.e4 lll e5 White loses a lot of pawns without compensation and therefore resigned. The concept of the DEB is very useful in determining the true value of a bishop. It can

It goes without saying that spotting the DEB is not necessarily the main priority in a razor sharp Najdorf Sicilian with a material imbalance and several pieces hanging. In general, once the positions enter a tactical phase, many of the positional features lose importance to more concrete calculation. Positional rules and guidelines still apply, but they are pushed to the background for a while.

1 .3.1 Positions with DEB relevance D

A

A DEB relevant positions

40%

B DEB irrelevant positions

40%

c g vs . .i endgames

10%

D .i vs. lLl endgames

10%

The pie chart shows a graphic view over the positions relevant for the use of the term DEB. The percentages are meant as an estimate to illustrate my point. Of course no one is able to state the exact number.

16

Esben Lund - The Secret Life of Bad Bishops

A DEB-irrelevant position is either too open or too closed. I do not wish to define the exact notion of 'open' and 'closed' positions. I merely want to make a visual point. To exaggerate this point, one could claim that if the position is too closed, then all bishops are bad, and if the position is too open, all bishops are good! My point is of course that the positions relevant for DEB lie in the span between these two extremes. We could call these positions semi-closed and semi-open, again without giving any exact definitions of what these terms cover. Boch of the endgames E: vs. i. and i. vs. 0i I declare DEB-relevant by definition. Pawn formations do not always matter in these positions with fewer pieces on the board more important is the colour issue: since the remaining pieces have been exchanged, the bishop runs the risk of being isolated on its colour. It is therefore dependent on the king and remaining pawns to control the other colour complex. To avoid these other factors taking over and becoming the most important to discuss, these endgames are declared DEB relevant by definition. These assumptions and definitions are made to show that I am aware of the limitations of the use of the term DEB. In cases arguable whether the term DEB is useful or not, I will discuss it openly. Sometimes the awareness of a possible DEB is important, in case a position is closed but is bound to open up at a later stage of the game. Being able to spot the positional dangers as well as potential advantages often shows the strategic depths of a player: one good example is the game Bacrot - Carlsen in the next chapter on exchange sacrifices. Thus, based on the estimate in the diagram, this book is relevant for 40+ I O+ I 0 60% of all chess positions! =

Now I will turn to the question of how to place one's pawns in connection with the opponent's bishop. Should the pawns be placed on the opposite colour to the bishop to avoid a future attack? Or should they instead be placed on the same colour to try to dominate the bishop? Indeed a difficult question that depends upon many factors.

Sergey Karjakin Vladimir Kramnik -

Moscow 20 1 1

a

b

c

d

e

f

g

h

Position after 39.0ih7.

Let us not discuss how open or closed this position is. Black is in trouble here since White's initiative on the kingside is rolling and he can't make any use of his 4 vs. 3 majority on the queenside. White is about to capture the pawn on g5 and seems to have everything under control. A very delicate matter in the Berlin Wal l is White's pawn formation on the queenside with a4-b3-c4: It does hem in Black's potentially strong bishop, but at the same time the chain is extremely vulnerable once b3 comes under attack. For both players, it is very difficult to foresee the consequences from afar when accepting this pawn formation.

Chapter 1

-

Likewise, the pawn formation h3+g4 also restricts Black's bishop. On move 8 I would call this bishop on c8 a DEB; meanwhile the position has changed and now using the term is arguable. But in the end the notion of the DEB is meant as a working tool to get a better understanding of the position, not as a strict rule that needs to be applied to every position.

39 ... c6? Black could have saved the game with 39. . . fxg4! 40.hxg4 li:'if5t! 4 1 .gxf5 White has to take the knight as it was heading for d4, attacking b3. 4 I . . .i.xf5 42.li:'ixg5 i.c2 picking up at least two pawns when a possible draw is not far off. Kramnik must have been such an opportunity slip.

in

17

Introducing the Double-edged Bishop

1.4 More about the DEB: Alekhine - Yates The following game fragment has been used as a textbook example of a bad bishop in a Stonewall set-up. The seeming clarity of the (technical) stage of the game, where Alekhine slowly but surely builds up his advantage, followed by the nice final mating idea that includes a non-standard king march, has given the game its immortality. But such simplified clarity often benefits the teacher more than the pupil. First, simply play through the game:

Alexander Alekhine - Frederick Yates

time trouble to let

40.lll xgS Now the trick does not work and Black went on to lose .

London 1 922

1 .d4 lll f6 2.c4 e6 3.lll f3 dS 4.lll c3 �e7 5.�gS 0-0 6.e3 lll bd7 7.l:kl c6 8.'!Wc2 ge8 9.�d3 dxc4 10.hc4 lll d5 1 1 .lll e4

... 1-0

In the diagram position it would be slightly artificial to hang on to the term DEB to describe the bishop on c8, even if the same term was justified on move 8 in the same opening. During the course of the game the character of the position has changed, and in most cases such transformations go unnoticed: It can be difficult to spot the exact point where the change took place. The term DEB is meant as a guideline and a way to spot a possible positional dilemma and react accordingly. It goes without saying the line between 'open' and 'half-open' is vague as well. I only wish to give the reader a refined sensibility of the types of position where it makes sense to talk about a DEB.

a

b

c

d

e

f

g

h

l 1 . .. f5 1 2.he7 '!Wxe7 13.lll ed2 bS 14.�xdS cxdS 1 5.0-0 aS 1 6.lll b3 a4 17.lll cS lll xcS 1 8.'!WxcS '!WxcS 19.gxcS b4 20.gfcl �a6 2 1 .lll eS

18

Esben Lund - Th e Secret Life of Bad Bishops

1 .d4 tll f6 2.c4 e6 3.lll f3 d5 4.tll c3 i.e7 s.i.g5 0-0 6.e3 tll bd7 7 .gel The typical waiting strategy in this opening: White postpones the development of the fl -bishop as he knows it will cost him an additional tempo after . . . dxc4.

7 ... c6 8.Wfc2 ge8 9.i.d3 dxc4 IO.i.xc4 lll d5 I 1 .lll e4!?

a

b

c

d

e

f

g

h

2 1 ..Jfob8 22.£3 b3 23.a3 h6 24.'it>fl 'it>h7 25.h4 !US 26.'it>g3 gfb8 27.gc7 i.b5 28.g lcS .L6 29.g5c6 ge8 30.'it>f4 'it>g8 3 1 .hS i.fl 32.g3 .ia6 33.gfl 'it>h7 34.gcc7 gg8 35.tll d7 'it>h8 36.lll f6 ggfll

8

7

This was a ne w move at the time. Alekhine has aggressive intentions on the kingside and allocates the knight to this side of the board. l l .�xe7 is the continua tion mostly favoured by White. l l .�f4!? lli xf4 1 2.exf4 is a dynamic continuation, where White relies on his better development and central control to outmatch Black's better structure and potentially dangerous bishop pair.

6 5

4 3

2

a

b

c

d

e

f

g

h

37.gxg7! gxf6 38.'it>es 1 -0 Indeed a nice finish: if Black plays either rook to f8, he will be mated. It all looks very convincing, but a closer look reveals that the critical phase of the game is between the first two diagrams, between move 1 1 and 2 1 . After White plays 2 1 . tlie5, the game is practically over. So, let's take another look at the game, focusing on the critical stage of the early middlegame:

a

b

c

d

e

f

g

h

l l...f5?! This might already be a slightly dubious idea. Black had other options on the 1 1 th move: 1 1 .. .�xgS 1 2.lliexgS g6 1 3.h4!? is probably what Alekh ine had in mind. Four years later Yates played:

19

Chapter 1 - I ntroducing the Double-edged Bishop 1 1 .. .1.WaSt 1 2.lt>e2! ? ll'i 7b6 After this move, White's idea is j ustified. 12 .. . f6! l 3 . .ih4 ll'i 7b6 seems like a major improvement for Black. The king on e2 might feel insecure over the next couple of moves, for instance: 14 .El: hd 1 (After 14 ..id3 ll'i b4 1 5 .\Wd2 Black can grab a pawn with 1 5 . . . \Wxa2; while 1 4 . .ib3 ll'i b4 1 5 .\Wb l \Wb5t looks awkward for White after 1 6.'kt>d2) 1 4 . . . ll'i xc4 1 5 .\Wxc4 b6! 1 3 ..ixe? Ei'.xe7 1 4 . .id3 Here Yates once again launched his f-pawn. 14 ... fS ?! 1 5 .ll'ic5 ll'i d7 1 6.a3 ll'ixc5 1 7.\Wxc5 \Wxc5 l 8.Ei:xc5 .id? 1 9. ll'ie5± White had the clearly superior endgame with his king well placed in the centre of the board.

1 2.he7 ffxe7 13.ll'ied2

pawn on e6 and a potential hole on e5. These weaknesses might only be temporary, that is, Black might get rid of them given time. One idea is . . . lt>h8, . . . ll'i 7b6 and a later . . . e6-e 5, but: 2) Black is slightly behind in development as well. This means that the chances for the above plan to succeed have been reduced. Also, the d7-knight has to prevent the knight jump ll'i f3-e5 . I shall have a look at ways for Black to develop the queenside to obtain coordination between his pieces. Thus, it is a simplification to explain Black's miseries with only the so-called 'bad bishop' on c8. For now, we shall more modestly call it a DEB, and only after further analysis can a final verdict be given.

1 3 bS?! •••

8

13 ... ll'i b4? 1 4.\Wxf5 b5 1 5 . .ib3 simply loses a pawn.

6

An interesting try is 1 3 . . . b6!? Black wants to develop his queenside without creating too many weaknesses. This is a solid and realistic approach at this stage. 1 4.0-0 Less convincing is 1 4 . .ixd5 cxd5 l 5.\Wa4 a5. 1 4 . . . .ib?

7

5

4 3

2

a

b

c

d

e

f

g

h

From this point it might be possible to show a line or an idea for Black that gives a playable position, but from a practical viewpoint the position is much easier to play for White. Strategically, White is playing almost without risk, whereas Black has several potential headaches. The bishop on c8 is only one of them: 1 ) The move with the f-pawn has weakened the black position, given him a backward

a

1 5 .e4

b

c

d

e

f

g

h

Esben Lund - The Secret Life of Bad Bishops

20

Let us instead consider a more straightforward approach for White: 1 5 .�xd5 cxd5 1 6.Wa4 Wd6 This is far more solid for Black than in the game. A sample line to show a natural development of events gives an interesting endgame: 17.l'!d a5 1 8 . 2'%fc l l'!ec8 1 9.l'!xc8t l'!xc8 20.l'!xc8t �xc8 2 1 .lLieS lLixe5 22.dxe5 Wd7 23.Wxd7 �xd7 24.lLif3 cj;>f7 2 5.ltJd4t

a

b

c

d

e

f

g

The most common worst-case scenario is chat of knight vs. DEB. In the endgame section, however, I will deal with the endgame rook vs. bishop as well, since the exchange sacrifice is a central motif in many positions with DEBs - which will be evident later in chis introduction and especially in Chapter 3, which is devoted to chis theme. In rook vs. bishop endgames I shall name the bishop a DEB by definition as explained in the last section, as the main problem for the lone bishop is the same: isolated on its rnlour, the king and rook of the opponent have a free hand to penetrate on the other colour complex. Back to the main continuation after l 5.e4:

h

Does White have chances to win chis endgame? The first player has a strong knight outpost on d4 and space in the centre with the pawn on e5, but compared to the similar worst-case scenario presented in the Berlin Wall endgame at the beginning of the introduction, White does not have a pawn majority and therefore no clear plan for advancement. A way to penetrate into Black's position is not apparent at the moment, as Black can place his a-, b-, g- and h-pawns on dark squares. Of course White can make a risk-free winning attempt, but my best guess is chat it is harder for White to win than for Black to draw, although Black will have some tough hours ahead defending. Worst-case scenario: A case where the remaining pieces have been exchanged and the DEB has been isolated and may therefore be called a bad bishop. But even so, many such worst-case scenarios can be defended to a draw with accurate play.

a

b

c

d

e

f

g

h

1 5 ... fxe4 l 6. ltJxe4± This looks pleasant for White with the isolated pawn on e6 and his better coordination, but at least he will have to work harder to increase his advantage. l 6 . . . h6 1 7.l'!fe l l'!f8 Although better for White, nothing tangible is yet apparent, and Black might come up with some threats of his own.

14.hdS cxd5 15.0-0 Th e pawn move . . . b5 h a s given Black new potential problems: the weakness of the c5square and a potential target for White on b5. White still enjoys a lead in development, and the e6-pawn/e5-square issue also still remains.

Chapter I - In troducing the Double-edged Bishop

a

a

b

c

d

e

f

g

h

Thus, so far the advance with the b-pawn seems only to have added to Black's problems. The more modest pawn move to b6 would have reduced the obvious weaknesses in the position, and while trying to catch up in development, Black could hope to slowly neutralize White's initiative and hold the position together. Now Black has to thread even more carefully than before.

1 5 aS ...

1 5 . . . b4?! dominates.

1 6.\Wa4!

and

White

totally

1 5 . . .i.aG!? This was worth a try - Black is hoping to catch up in development and coordination. In the lines to follow, White always seems to keep a small edge - at least from a practical point of view. But our interest in this position is to come to a deeper understanding of the DEB: when does it cross the fine line of no return, when we can put a verdict upon it as being bad? If no final verdict can be found, the bishop simply remains a DEB. 1 6.\Wc6 llib6 1 7.\WcS Practically forcing the exchange of queens. l 7 . . . b4 l 7 . . . \Wf6 1 8.\Wa3! highlights Black's bad piece coordination on the queenside. 1 8.Ei:fe l llic4 1 9.\Wxe? Ei:xe7 20.b3

b

c

d

e

21

f

g

h

20 . . . llia3! 2 1 .llieS i.b5 22.llidf3 22.Ei:cS a5 23.Ei:ec l i.e8 24.Ei:c8 Ei:ea7 ts similar 22 . . . aS 23.Ei:cS i.e8 By keeping an extra pair of knights on the board, Black could argue that White has a superfluous piece since both his knights aim at the e5-square. But the real benefit for Black is that with an extra knight he is able to attack the white queenside pawns on a2 and b3. 24.Ei:ec l llib5 25.Ei:c8 Ei:ea7 8

a

b

c

d

e

f

g

h

With the idea . . . a5-a4 and . . . llic3, when Black has his share of the play too. Let's follow a sample line to see how things can develop: 26.Ei: l c6 White brings his rook on the other side, before Black blocks the c-file with the knight. 26 ... a4 27.g3 To avoid back-rank mate.

Esben Lund - The Secret Life of Bad Bishops

22

27 ... lt:Jc3 28.lZJgS! White starts an assault on the black king. 28 ... a3 29.Ei:6c7 h6! Black has to be careful: 29 . . . lt:J xa2? 30.lt:Jc6 El:xc7 3 l .Ei:xa8 when he drops his rook due to the mating threats. 30.Ei:xa7 El:xc8 3 l .lt:Jxe6 g6

only two pieces left on the board, the black king is in. no danger of getting mated. 32.lt:Jc7 Ei:d8 32 . . . lt:J xa2?! 33.lZJxdS Wf8 34.lt:Jd3 eyeing the b4-pawn 33.lt:Jxe8 El:xe8 34.lt:Jxg6 Ei:b8! Guarding the b-pawn in advance and preparing to pick up a2 and b3. White is ahead on material and may hold the advantage, but the situation is far from clear, since Black's pawns are the more advanced. All along chis sample line, Black was able to create counter­ chances of his own. Exchange of the DEB to transform the

a

b

c

d

e

f

g

h

This is a good example of a DEB chat looks bad at first glance, but has a very useful role as a defender. Bue the strategy of reducing one of the pieces entirely to the defence only works if one can create counterplay with the remaining army, otherwise Black would simply be passive. Here the attack against a2 by the black knight as well as the far­ advanced a-pawn give Black hope. The concept of reducing one piece to a defensive role (temporary or permanent) is very important in order to understand the nature of the DEB. In case of the DEB, this defensive role can be permanent; and in order not to get a passive position overall, active counterplay has to be initiated elsewhere on the board to compensate for the passivity of the DEB.

Another interesting aspect is chat White over the next few moves will have to get rid of the DEB on e8 to continue his attack - but with

advantage: In order to break through

the opponent's defence, one often

has to exchange off the defender's

DEB. Many players

make such

a

are reluctant to

decision on the false

assumption that the bishop is bad.

But this may be the best or only way to make progress in the position. This concept is in acco rdan ce with -

-

Tarrasch's dictum that what matters is not which pieces

go off the board, but

rather which ones remain.

16.�b3 a4 Perhaps Yates was planning 1 6 . . . �a6 1 7.lZJxaS b4 1 8.Ei:fe l �b5 1 9.lt:Jb7 Ei:xa2 but somehow regretted this line. It is true chat his position is unpleasant after 20 .Wc7 but 20 ... Ei:a7 might hold chances of survival. (20 . . . Ei:xb2? 2 l .lt:J d6+-} 2 1 .Ei:a l �a6 22.Wd6 El:xb7 23.Wxe7 El:xe7 24.Ei:xa6± Black got rid of his DEB, but once again his other issues remain in the position: The target on e6 and worse coordination, which in total makes the black b-pawn the more vulnerable of the two. Another interesting line is 1 6 . . . b4 1 7.lZJcS �a6 - how should White continue?

Chapter 1

-

I ntroducing the Double-edged Bishop

.i�� � Black's l l . . . f5?! was dubious and especially

after 1 3 . . . b5?! he found himself with difficulc problems to solve. There were at least three important factors such as lack of development and coordination, White's control of the c-file and concrete targets to attack on e6 and on the queenside meant the bishop on c8 was only one of chem.

)P> After 2 1 .tll e 5 Black's DEB is definitely

a bad bishop. The discussion if it is a problem or has potential has clearly been stated to its disadvantage. Bue blaming all Black's misery on chis bishop is a fatal simplification.

believe chat the themes presented in chis game give an indication of the dangers for Black to avoid if he wishes to play the Stonewall. I tried whatever I could at the critical stage of chis game to save the honour of Black's bishop, but I must admit chat it was an uphill batcle. At least in a practical game the task is very difficulc for the second player. The above analysis does show some interesting features of the nature of the DEB . in many different positions. In the next game I shall do more justice to the Stonewall sec-up, as the opening is perfeccly playable.

1 .5 More complex examples Hans Tikkanen Silas Lund -

Copenhagen 2009

This game was played before Hans Tikkanen became the strong grandmaster chat he is today. We are teammates in the same club in Sweden (in the city of Lund!) and it has been interesting to follow his development over the past years.

1 .d4 e6 2.c4 c6 6.i.f4!?

5

3.g3 lilf6 4.i.g2 d5 5.�a

A way to sidestep the main theory, but giving extra support to the hole on e5 has its drawbacks too. 6.0-0 i.d6 leads to the modern Stonewall. If compared with the Stonewall-like position in the Alekhine game just analysed, we notice chat: 1 ) The bishop on c8 is not unique, since White has a same-coloured one on g2. Still, it makes sense to treat the c8-bishop as a DEB, for reasons already explained. A future exchange of the g2-bishop for a black knight is not unlikely to happen, and in chis case Black's DEB on c8 would be unique. 2) There are more pieces rema mm g on the board in the modern Stonewall (actually, all of chem) which makes it harder to isolate the DEB from the rest of the army to show its problematic nature.

6 ... i.e7 7.lile5 0-0 S.lild2 \Wes Black simply ignores White's domination of the e5-square. White has three pieces chat would like to occupy chis square, so avoiding exchanges gives White two superfluous pieces. That's why I did not play 8 . . . tll bd7 and instead went for the more ambitious text move.

C h a p te r 1 - Introducing the Double-edged Bishop

9.c!l)d3 c!l) bd7

8 � -j_�"if �� · ,_, '-- - --% r ,� Y,• :--______ : .....�%m �

� �

�

: �.,��. �-lti·---- - �, . %>, . , . , .

�-� " ..,��rd· · · · %tt5 2 8 tf!J �ID 8 tf!J � 3

�

��p

:t· · · %mvm·: · %m%: a

b

c

d

e

f

g

h

10.c5?! But chis move looks suspicious. White's play revolves around the e5-square in particular, and the dark squares in general, but these are only a small part of the overall strategy in the Stonewall. 1 0.0-0 was the normal continuation.

10...b6 1 1 .b4 Aa6! The DEB shows potential . Now it has a nice diagonal a6-fl , whereas the white bishop on g2 momentarily is hitting a wall on d5. Kramnik has pointed out exactly this theme in his survey on the modern Stonewall in Dvoretsky and Yusupov's Positional Play: Why should Black's a6-bishop be any worse than White's g2-bishop now?

1 2.0-0 c!l)e4 I remember being very opt 1 m 1 st 1 c at this stage, but an interesting alternative was 1 2 . . . lt:l h 5!?.

13.c!l):x:e4 fxe4 14.�e5 c!l):x:e5 1 5.Axe5 Wfh5 16J;el

25

s � - R�-%�-�-·�-� 1 � - � r�% • 6 �E {9· � i�m ···-- � 5 ���--�- ·hn·· �i�mRli �� � �� � md' � � � � '#! " .%

�

• • • • •

4 3

2

�

%

.

%••••%

.... %

� �� �.1rFF a

b

c

d

e

f

g

h

1 6 .. JH7

I was looking for ways to get my initiative going, but was not convinced about the alternatives. With the modest text move my initiative slowly fades away. 1 6 . . . bxc5 1 7.bxc5 i.xc5!? was an interesting try. The main idea is a piece sacrifice: l 8.g4 ( l 8.i.xg7 @xg7 l 9.dxc5 E:ab8 gives Black central control at the expense of king security. But it seems to me chat the central control weighs heavier, and White should probably avoid chis line. ) 1 8 . . . W g5 1 9. h4 Wxh4 20 .i.g3 Wxg4 2 l .dxc5 e3 This looks dangerous, but White has defensive resources. 22.Wb3! exf2t 23 .i.xf2 E:f6 (threatening 24 . . .E:g6) 24.Wg3 i.xe2 25 .Wxg4 i.xg4 26.E:ab 1 Frankly I am not sure how to evaluate this unbalanced endgame. ' Unclear' is not far from the truth.

17.£4! Af6 18.e3 Wf:x:dl 1 9J:ie:x:dl bxc5 20.bxc5 Axe5 2 1 .fxe5 gb7 22.i.h3 @f7 23.Afl Kramnik's point is taken: the best White's g2-bishop can do is to exchange itself for the a6-bishop, which was quite annoying on the long diagonal. Since there are still rooks on the board, the a6-bishop can be helpful on either d3 or c4. In cooperation with the remaining pieces, the a6-bishop is still a force to be taken seriously. It is therefore safe to say chat

Esben Lund - The Secret Life of Bad Bishops

26

the a6-bishop has been a good bishop since it arrived on a6 on move 1 1 .

23 ixfl ..•

The last ambitious try was the ugly-looking 23 . . . �c4!? 24.�xc4 dxc4 but White should not be worse in this endgame.

24Jhfl t ©e7 25.:Sfb l gab8 26.:Sb3! Good defence.

26 ... a5 27.ga3 gbl t 28.gxb l gxb l t 29.©£2 gb2t 30.©fl gblt 3 1 .©£2 gb2t 32.©fl gbl t 1/2-1/2

The next encounter is the famous game fragment from the 1 97 1 World Championship semi-final match between Bobby Fischer and Tigran Petrosian.

Robert Fischer - Tigran Petrosian Candidates final, Buenos Aires (7) I 97 1

1 .e4 c5 2.lll O e6 3.d4 cxd4 4.lll xd4 a6 5.i.d3 lll c6 6.lll xc6 bxc6 7.0-0 d5 8.c4 lll f6 9.cxd5 cxd5 10.exd5 exd5 1 1.lll c3 i.e7 1 2.�a4t �d7 13,gel �xa4 14.lll xa4 i.e6 1 5.i.e3 0-0 16 ..ic5 gfeg 17.i.xe7 gxe7 1 8.b4 ©£8 19.lll c5 i.c8 20.8 gea7 2 1 .ge5 i.d7

22.lll xd7t! This is without doubt a strong move Fischer transforms his advantage to something very concrete. Black's DEB might optically look bad (and it has been labelled as such in numerous works) due to the pawns on a6 and d5. Also, the exclamation mark is given for the exchange of the mighty knight on c5 against the supposedly inferior bishop. But it is simply a DEB. And Tarrasch's dictum again tells us to focus on the pieces remaining on the board after the exchange has taken place. Since the DEB might be aiming for the b5square, a natural response for White would be to play 22.a4. This maintains an edge for White, but is a less direct approach, so Fischer's continuation should be preferred in practice. But I have a plan in mind that is no less concrete than Fischer's, and for this reason I believe 22.a4!? is equally strong objectively: 22 . . .�e8 23.:B:ee 1 ! This is the idea: White wants to capture on a6 twice, giving an unbalanced ending with rook vs. two minor pieces. 23 ... lll d7 (23 . . . aS 24.bS gives White a huge protected passed pawn) 24.lll xa6 :B:xa6 2 5 .�xa6 :B:xa6

This famous example has found its way into many books. 8

7

6 5

a

b

c

d

e

f

g

h

26.:B:ed l ! fll f6 27.bS± White controls the entire board and is ready to strengthen his position in many ways. Interested readers can find more about such positions in my previous book for Quality Chess, Rook vs. Two Minor Pieces.

4 3

2

a

b

c

d

e

f

g

h

Chapter 1

-

I ntroducing the Double-edged Bishop

Black can also play 22 ... �c6, but again White replies 23.:B:ee 1 with the same intentions of exchanging on a6. Here 23 . . . �b? looks extremely clumsy. But let's say chat White decides upon a slower approach after 22 . . . �c6 with 23. ©f2 instead of 23.:B:ee 1 , when Black forces the exchange of knights with 23 . . . ltJd? 24.ltJxd?t E:xd7.

s ,i �� ��. 67 �£ � .i. �.i �· ��· 45 ,�%���·��y��� � � � � � � � � -� i!� a

b

c

d

e

f

g

h

This line is by no means forced, and admittedly far from best play by White. Bue I want to illustrate the point chat Black's bishop is a DEB and cannot yet be called 'bad'. Why is chat so? Well, remove all rooks from the board and Black would probably be dead lost after 2 5 .a5. The multiple weaknesses on a6 and d5 plus White's majority on the queenside should secure victory once the white king enters the d4-square. But with rooks on the board the story is different, since now b4 is a target in the white position and things are not so crystal clear.

22 .. J;xd7 23.gcl White has a nice active position with the better pawn structure and a pawn majority on the queenside. The pawns on a6 and d5 are targets in Black's position, whereas White has no such visible weaknesses. These advantages were enough for Fischer to decide upon the transformation of the position with 22.ltJxd?. Again we are reminded ofTarrasch's dictum of

27

the importance of the pieces remaining on the board, not the ones chat are exchanged. I will not go into deep analysis of the remaining 1 1 moves chat it took Fischer to win the game.

23 ... gd6 24.gc7 lLi d7 25.ge2 g6 26.@£2 h5 26 . . . E:b8 27.a3 shows how hard it is for Black to create targets in the white position.

27.f4 h4 28.@f3 f5 29.@e3 d4t 30.@ d2 lll b6 3 1.gee7 lll d5 32.gf7t @es 33.gb7 lll xb4 34 . .ic4 1-0

1 .5.1 Bishop vs. knight Before we turn our attention to some interesting games from grandmaster practice, I would like to delve deeper into the question about the relative strength of bishop to knight. Two sayings about bishops are: 1 ) A bad bishop protects good pawns 2) Even the worst bishop is better than the best knight Both these sayings talk about bad bishops and at the same time say why they are not chat bad after all. Such sayings are not meant to be true, but a good saying usually has a lot of truth in it. And if I were to briefly explain my ambition with chis book, explaining in more detail the logic behind these interesting sayings would possibly be the shortest answer I could give. The short interpretation of the first saying above should come as no surprise to the reader by now: 1 ) The so-called bad bishop chat protects good pawns, is not bad, but a DEB.

28

Esben Lund - The Secret Life o f Bad Bishops

The second saying is more complex, since it indicates that the bishop is simply a better piece than the knight. The difference between these two pieces is one of the most fascinating in chess: how is it possible that two completely different pieces are given approximately the same value?

Today the Nimzo-lndian is one of the most popular Black openings at top level, which speaks for its validity. Even if the bishop is a better piece than the knight, the other positional factors of the position make the picture more complex and somehow outweigh this possible material imbalance.

Both pieces can only protect one colour complex at a time. Albeit in different ways, since the bishop is colour blind and therefore is forever doomed to only work on half of the board at the maximum (that is, 32 squares) , whereas the knight is a chameleon that changes colour with every move - but compared to the bishop it is short-ranged. Could the two pieces be more unalike?

As said: in the end it all comes down to the

In his book Improve Your Chess Now, Jonathan Tisdall devotes Chapter 5 to "The Value of the Pieces". Here he refers to Staunton's 1he Chess Player 's Handbook, which gives the knight the value 3.05 pawns, whereas the bishop is worth slightly more: 3.50. Other authors have also written on the relative values of the knight compared to the bishop, and have no doubt come to different results. My point here is not to argue about the exact numbers, but to point out that the bishop is usually worshipped more than its short­ ranged cousin. Without going deeper into this discussion, do believe there is some hint of truth in the bishop being stronger than the knight. At one end of the scale we have the viewpoint of Hans Berliner, who in his book 1he System believes the Nimzo-lndian to be a dubious opening choice for Black, since with correct play the long-term advantages of the bishop over the knight will prevail: it is simply incorrect for Black to give up the bishop at such an early stage of the game.

other factors of a given position, and weighing up these factors against each other when making decisions at the board. We saw in the gai:ne AJekhine - Yates that Black's problems not only stemmed from his bishop (and more abstractly, the fight bishop vs. knight) , but also from his lack of development and coordination, as well as vulnerable points that were easy to attack. As an indication of the 'validity' of the second

saying, one could consider the following two worst-case scenarios:

Salo Flohr -Jose Capablanca Moscow 1 93 5

Chapter 1 - I ntroducing the Double-edged Bishop

29

Anthony Saidy - Robert Fischer

1 .5.2 Two examples &omJohn Watson

US Championship, New York 1 964

In his book Secrets of Modern Chess Strategy, John Watson writes about good and bad bishops in the chapter entitled "The New Morality of Bad Bishops" . He presents the following instructive example about which he writes:

8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

Position after 26 . . . h S .

With correct play, both these endings should be drawn. Among others, Jacob Aagaard has shown this to be the case with the Capablanca game in his first book, on the Panov-Botvinnik Attack, whereas Mihail Marin devoted a whole chapter in his Learn from the Legends to analysing positions with pawns on d4/d5 and knight vs. bishop (of the same colour as the central d-pawn). I shall indicate the right set-up in Chapter 6, but for now I shall use these two positions to 'prove' the saying: when the worst possible bishop is in battle against the best possible knight the result is still a (theoretical) draw. Of course there are flaws in this reasoning. For instance: that the fight might not only be about bishop vs. knight, since other factors might count too. In the Flohr - Capablanca game, Black had an inferior pawn structure, to point out one.

a

b

c

d

e

f

g

h

"Another fairly obvious qualification is that if the bishop is 'outside' its same-colour pawns (which is to say, it is not trapped behind them) ; then that bishop is still ' bad' , technically speaking, but may be perfectly effective, especially in the middlegame. Here is a stark example of bad bishops of contrasting strengths [see diagram] ." There are some good insights in this example, although I would like to modify the terminology. When he is talking about bad bishops not being so bad, I would simply say DEB. And the phrase 'especially in the middlegame' can be translated to 'with other pieces on the board'. In this example, White's king is likely to come under tremendous pressure in the future, when Black advances his h-pawn and brings his heavy pieces behind it. It is far too simple to just convert the bishop from 'bad' to 'good' , once it gets on the other

Esben Lund - The Secret Life of Bad Bishops

30

side of its own pawn chain. For starters, remove all remaining pieces in the above diagram except the bishops on e3 and e4, and the result will be an obvious draw. Furthermore, the game to follow shortly, Radjabov - Eljanov, will cast more light on this discussion.

. . . the bad white bishop on d3 and static pawn structure are real liabilities, e.g.

10 ... �g6 1 1 . �el 0-0 12.g3 White feels that he must open lines with f4.

1 2 ... ih3 13.�g2 �e8 1 4J;el

I hope that this game and the following example by Watson will make it clear that my objections are not just of an academic nature, but actually make a practical difference in how these strategic positions should be seen and evaluated. Later in his book Watson shows this game of the Nimzo-lndian (Hubner variation) :

Svetozar Gligoric - Jan Timman Bled/Portoroz I 979

l.d4 �f6 2.c4 e6 3.�c3 ib4 4.e3 c5 s.id3 �c6 6.�f'3 ixc3t 7.bxc3 d6 8.e4 e5 9.d5 �e7 8

7

a

b

c

d

14 ... £5! 15.exf5 ixf5

e

f

g

h

Now that White's kingside light squares are weakened, it's okay to rid him of his bad bishop; this also takes a defender away from the weakness on c4.

1 6.f4 hd3 17.Wi'xd3 Wi'd7 1 sJ;b1 �f6 1 9J;b2 gae8 20.gfl Wi'h3 2 1 .£5 �g4 22.�el e4 23.Wf e2 �6e5! 24.Wfxe4 gf7 25.Wfg2 Wi'xg2t 26.i>xg2 �xc4

6 5

4

. . . with a large advantage . . .

3

2

a

b

c

d

e

f

g

h

"

The final position does indeed remind us of the catastrophe in the game Matisons Nimzowitsch, shown earlier in this chapter. Black soon won .

Here Watson comments on castling kingside as follows:

I would like to make the following comments to this example:

"Black's idea is that after

1) On move 1 0: White's pawn structure is compromised, but this is the price he willingly paid to obtain the bishop pair. A very normal

10.0-0

Chapter 1 - I ntroducing the Double-edged Bishop trade-off in the Nimzo-lndian, and it's definitely too early to make a judgement about the position on move 1 0. And we should not forget that the d5-pawn gives White space in the centre. 2) White's bishop on d3 is not bad - it might be considered a DEB. Since Black's only pawn breaks are either . . . b5 or .. .f5 , the bishop looks well placed to me on d3. In fact, when White exchanges this bishop on move 1 6, I would definitely have preferred 1 6.�fl ! to hold on to the bishop pair. 3) I simply do not understand the comment about the weakened light squares around the white king after the exchange of bishops. I don't see a problem with White's move g2-g3; one might instead consider 1 6.f4 as weakening the king position. It seems to me that Watson has already decided that the bishop on d3 is bad, and therefore has to make an argument to back this view up. 4) In a closed position like this one, there is always a risk that the bishops (both of them) end up being inferior to the knights. We discussed this point at the beginning of the introduction when looking for limitations of the usefulness of the term DEB. With White I would really consider which remaining pieces to exchange, and I don't see the point in White's exchange of light-squared bishop, which can only weaken him further on this colour complex. I believe the position after move 26 is enough proof here, with a giant black knight on c4. 5) Watson mentions the weakening of the c4-square when the light-squared bishops are traded off, but since he has already put the predicate 'bad' upon it, he sees no other alternative for White than to gladly exchange it - even if it fatally weakens the c4-square.

31

So there seems to be an inconsistency in his way of reasoning. In the first example he made some good points about the bishop being good or bad, whereas he seemed to fall back on a more traditional and simplified viewpoint in the second example.

1 .5.3 A modem grandmaster game Teimour Radjabov - Pavel Eljanov Astrakhan 20 I 0

1 .d4 llJ f6 2.c4 e6 3.llJf3 d5 4.llJc3 ib4 5.cxd5 exd5 6.ig5 h6 The following concept is a modernizing of the 'older' (we are talking only a few years) solid continuation: 6 . . . ll'i bd7 7.e3 c5 8 .�d3 Wa5 9.Wc2 c4 1 0 .�f5 The bishop is more active here than on e2. 1 0 . . . 0-0 1 1 .0-0 The main continuation goes l l . . .E:e8 l 2. ll'id2 g6 and now either l 3.�xd7 or keeping the tension for a while with l 3.�h3!?. 1 3 .�xd7 ll'ixd7 leads to a complex strategic battle, where Black has a DEB on c8, and White has one on g5 if Black later chooses to exchange on c3 - which he often does in these positions.

7.i.h4 c5 8.e3 c4 Together with . . . h6, this is a very ambitious concept compared to the line 6 . . . ll'i bd7: White is not allowed to play �d3 , and Black himself plans to occupy the diagonal b l -h7.

9.llJd2 This prevents a knight jump to e4. Albeit not the main move here, 9.Wc2 is interesting, since 9 . . . g5 1 0.�g3 ll'i e4? fails to l l .�xc4!.

9 ...g5 10.ig3 i5

Esben Lund - The Secret Life of Bad Bishops

32

Black's compensation for ... i.xc3 is a smooth development and more space in the centre and on the queenside. Because of his active pieces, it is difficult for White to pinpoint the weaknesses in his position at the moment. The greatest weakness is probably the thin air around Black's king position created by the move . . . g5. But since many of these positions see a queen exchange early on, it might not be that big a problem.

l 3.i.d6 The simple retreat l 3.i.g3 is also an option. A possible continuatio,n is 13 ... Wa5 1 4.Wc l 0-0 when Black's full mobilization compensates for his slightly weakened king position as well as White's bishop pair. After l 5.h4!? g4 it is difficult to tell which strategy will triumph; a complex struggle lies ahead.

It is interesting to insert l 1 .h4!? g4.

1 1. ..hc3 1 1 . . .lll b d?? 1 2.i.xf6 loses at least a pawn: 1 2 . . . lll xf6 ( 1 2 . . . Wxf6 1 3.lll xd5) 1 3.Wa4t

12.bxc3 lLlbd7

s i, � �- � 1 1�• m�m � •m 6 ;;,, , /

�

�

-

1 3.Wa4!? was played m the Chinese Championship in 20 1 1 . After 1 3 . . . 0-0 l 4.i.g3 Black has to choose his middlegame strategy, and unlike our main game, he ended up in some trouble after the further l 4 . . . E!e8 1 5 .i.e2

�

�� �md' : �mwtJ' . f�� ' ' "wtJm � m 2 � l8 ,,/rlf , %� l� m ef

� �/

3

��•v•.f•� a

b

c

d

e

f

g

h

The e5-bishop is a unique bishop now, and a DEB. But Black also has a potential DEB in the f5-bishop, not only because of the pawn chain d5-c4, but because of its inability to attack anything concrete in White's position.

s i, � � i. � * 7 ,� i �l.Al � i � 6 � � . � �� 5 � �� �• �..t.r� 4 if �� � � 3 � �,!f J� �� . .If3:US�rid� � �: 2

.

1 5 . . E!e6

: , , , ,

a

b

c

d

e

f

g

h

(This manoeuvre might be too standard here. My analysis engine suggests the aggressive l 5 . . . Wb6 1 6.Wa3 i.d3!? l 7.i.xd3 cxd3. It does make sense to try to exploit the lead in the development. 1 8.0-0 Wc6 Here, as in many similar positions, the DEB is the

Chapter 1 - I ntroducing the Double-edged Bishop interesting strategic feature of the position: unable to attack anything on its own at the moment, but holding a great potential if handled the right way.) 1 6.Wd l We7 1 7.0-0 �a3 1 8 .Wc l Wxc l 1 9.!!fxc l In Zhou Jianchao - Xiu Deshun, Chinese Championship 20 1 1 , Black has achieved the exchange of queens, but under less favourable circumstances than in our main game. Black is unable to expand on the queenside with his pawns due to 1 9 . . . b5 20.a4!.

13 ...�b6 8

7

6 5

4

33

1 4 . .ia3 i s a move we shall see more in the future. The interesting strategic question is if White can organize his forces in such a way chat chis DEB is not just hitting chin air. The move l 4 . .ib4 was tried recently in the game Topalov - Leko, Zug 20 1 3 . White is asking Black a question as to where to place his king. If 1 4 . . . 0-0-0!? White retains the possibility of l 5.a4 crying to open up the queenside for an attack - chis was not possible with the bishop on a3 . In the game Black decided upon short . castling and play on the queenside. On the plus side, the advance of the pawns comes with gain of tempo now: l 4 . . . a5 l 5 . .ia3 Wc6 1 6 . .ie2 b5 1 7.0-0 !!b8 1 8 .f3 b4 1 9 . .ib2 0-0 with a complicated struggle chat White lacer won.

14 ...�b2! 15.�cl �xcl t 16J'hcl

3

2

a

b

c

d

e

f

g

h

14 ..ig3?! 14 ..ia3 avoids the exchange of queens, but Black can then continue with 14 . . . 0-0-0 with a huge lead in development (indication: his rooks are already connected) and his king relatively safe on the queenside. The f5bishop controls the h7-b l diagonal; especially covering the b I -square is of importance. Furthermore, the ambitious . . . g5 now seems more space-grabbing than a weakness. 14 . . . We6?! was played in Werle - Bartel, Bundesliga 20 1 0- 1 1 . After 1 5 .Wa4 a5?! 1 6.Wb5 Wb6? White already had a winning position after the simple trick l 7.llixc4! dxc4 1 8.Wxf5 .

a

b

c

d

e

f

g

h

16 ... bS In chis position, White is not in a position to contest the black queenside expansion, and Black has the more pleasant position. 16 ... 0-0 allows White to stop Black's expansion on the queenside, but after l 7.a4 !!fe8 Black on the other hand prevents White from bringing his king to f2 as in the game.

Esben Lund - The Secret Life of Bad Bishops

34

17.f.3 0---0 1 8.'i:tfl �Ue8 19.a4 a6 20. .ie2

Now White has connection between his rooks, but it is Black to move and he still has a slight lead in development and the better coordination. 20.l:'i:a l is answered with 20 ... tt:l b6 2 1 .axb5 axb5 22.ie2 tt:la4 attacking the c-pawn and avoiding the exchange of rooks for the time being.

a

b

c

d

e

f

g

h

20 .. J:�e6! A multi-purpose move, eyeing the pawn on e3 but also preparing the doubling of rooks on the a-file.

21 .ghel 2 1 .l:'i:al l:'i:ae8! with problems on e3.

21 ...lll b6 22.axb5 axb5 23.gal lll a4 24.ga3

24 ... .ig6! This is the beginning of an amazing concept. Black wants to improve his position by grabbing even more space with . . .f5 , well aware that this will 'bury' his light-squared bishop behind his own pawn chain. Let's follow this plan closely.

25 ..ifl lll d7 26.h4 26.e4 This is answered with the strong: 26 . . .f5! 27.exd5 p.exf5 l:'i:xe l 28.'it>xe l l:'i:e8t 29.lt>f2 ixf5 and the opening of the position has favoured Black who has several ways to improve his position and create threats. Notice the difference in concrete targets: c3 is under direct attack and needs protecting, whereas the black pawns on b5 and d5 - the most visible targets - are really hard to get at. One plan could be to transfer the knight to e3, but this is easily countered by Black: 30.ie2? id3! probably forcing the ugly retreat 3 1 .ifl . 27 . . . l:'i:xe l 28. lt>xe l f4 29 .if2 l:'i:e8t 30. lt>d 1

.I �· � � 7 � .I.A) • • 6 • • •-*-fJ 5 �·-�- . 4 :. :� �- �� �� 3� � ��!� "R ;� �-�� .f�" 8

2

a

b

c

d

e

f

g

h

So far, so good. Tactics on c4 are looming, but Black has the strong move: 30 . . . id3! The tactics work well in Black's favour. 3 1 .d6 3 1 .ixd3?! tt:l b2t 32.lt>c2 tt:lxd3 and White loses a piece because of his trapped bishop: 33.tt:le4 tt:lxf2 34.tt:lxf2 l:'i:e2t a

b

c

d

e

f

g

h

35

Chapter 1 - I ntroducing the Double-edged Bishop 3 1 . . .© fl 32.i.xd3 cxd3 33.tt:le4 tt:l b2t 34.c;t> c l d2t! 35.tt:lxd2 3 5 .©xd2 tt:l c4t-+ 35 . . . tt:l d3t This also wins a piece for Black, although the technical task is more difficult. 36.©c2 tt:l xf2 37.E!a5+ The knight on f2 has nowhere to go at the moment! Another point is that without attackable targets, Black enjoys a great freedom of action - and his position will hugely improve once he manoeuvres his knight on d7, which does nothing at the moment. Knights are flexible in a static position like this one.

to defending, while the knight on c3 did the attacking. There was another, more dynamic, approach in the position, though: 28 ... tt:l f6! 29.fxg5 tt:l g4t 30.c;t>gl 30. ©f3 i.h5 3 l .i.f4 tt:lxe3t 32. ©g3 E!ae8+ with too many threats. 30 . . . E!xe3 3 l .Ei:xe3 tt:l xe3 32.i.f4 tt:l d l 33.tt:l b l E!e8 34.g3 Planning to attack d5 with i.g2. If instead 34 .i.e5 then 34 . . . f4-+. 34 ... i.h5 3 5 .i.e5 If instead White plays 3 5 .i.g2 Ei:el t 36.c;t>h2 i.f7 we get to the following position:

26 ... 5 27.hxg5 hxg5 28.f4

a

a

b

c

d

e

f

g

h

28 ...g4 Eljanov follows the strategic plan outlined many moves ago when he played 24 . . . i.g6: space to compensate for his DEB on g6. This bishop does not look good, but it will do a great job defending the pawn on b5 from e8. Meanwhile, Black has many remaining pieces on the board to attack White's weaknesses, and his space advantage gives rise to sacrificial ideas, as we shall see shortly. We saw a similar reasoning in the analysis of the Alekhine Yates game, where the DEB on e8 was reduced

b

c

d

e

f

g

h

Black's rook is very active - his position is probably close to winning. A sample line that illustrates the themes of this book is the following: 37.i.d2 E!e2 38.©gl (Planning to chase the annoying rook away with i.g2-f3) 38 . . . ©g7 39.i.f3

8 7 6 5 4 3 2

a

b

c

d

e

f

g

h

Esben Lund - The Secret Life of Bad Bishops

36

39 . . . l'!xd2! This is a typical transformation of the advantage to something more tangible. Black uses his space advantage on the queenside. Please notice that the white DEB on d2 is eliminated to break down White's defences. 40.'Llxd2 'Ll dxc3 4 1 .l'!al b4-+ Black's position plays itself; the pawns should be unstoppable. We will see more exchange sacrifices in the next chapter. Back to the position before the last two diagrams, after 3 5 .�e5:

29.gbl ga5 30.i.e2 gea6 3 l .i.h4?! 3 1 .l'!ba l ! This was definitely worth a thought. Perhaps Radjabov wanted to avoid the domination of his bishop after: 3 1 . . . 'Ll b8

8 7 6 5 4 3 2

�

,.,,., , , , , ;

a

b

a

c

d

e

f

g

h

35 . . . �f7 36.�h3 Black has many good continuations here, for example: 36 . . . �g6 37.�g2 'Lid 38.'Lld2 �xg5 This simply grabs a pawn with continuing pressure. We return to the main game after 28 . . . g4:

a

b

c

d

e

f

g

h

b

c

d

e

f

g

h

This has the idea 32 .�h4 'Ll c6, when the DEB on h4 is suddenly far off from the defence of the c3-pawn. The bishop sortie to h4 threatens �d8 if Black is not careful. However in the diagram position White has the interesting resource: 32.e4! The point is that 32 . . . fXe4?? 33.f5 drops a piece, whereas 32 . . . dxe4 runs into another tactical shot: 33.'Llxc4+Black can ignore White's tricky move and continue as planned: 32 . . . 'Ll c6 33.exd5 'Ll b6 33 . . . 'Lle?? 34.'Llxc4! bxc4 3 5 .�xc4 loses material.

Chapter I - I ntroducing the Double-edged Bishop Black hopes to take back the pawn with an overwhelming position, but here White can change the nature of the position completely. 34.dxc6!? !!xa3 3 5 .!!b 1 There are a lot of tricky tactics. 35 . . .ie8 Black had better get rid of this passed pawn quickly. After 35 . . . !!3a5? 36.c7 a sacrifice on either b5 or c4 cannot be stopped. 36.!!xb5 ixc6 37.!!xf5 !!xc3 38.!!g5t This position looks pretty unclear to me, with a lot of play for both sides, and certainly it's more fun than the game continuation, where White can only hope for a draw.

3 l ...�ab6 32.gxaS gxaS 33.i.d8 i.e8 34. @el gag

a

b

c

d

e

f

g

h

37

protecting and here the bishop is the best defender. 36.!!c l !!a2 37.id l tt:la4

b

a

c

d

e

f

g

h

The white position looks very passive. But why not then exchange this annoying knight on a4, and at the same time isolate Black's DEB on e8 further? 38.ixa4 bxa4 It is clear that in this position without rooks, White could play ttJ b l -a3 and easily hold the draw. The presence of a pair of rooks makes this impossible - now there are additional targets to take care of for White: c3, e3 and the g-pawn, since these can be attacked by the rook. In this case the presence of one additional piece - the rook on a2 - makes the difference for Black between a draw and a win.

3S ... @f7 36.i.b4 ga2 37.i.dl @e6 38.gcl � b8

Notice the two DEBs on d8 and e8. Both are on the same colour as their pawns (except g2! ), the only difference being that White's bishop is outside the pawn chain in contrast to Black's e8-bishop. I cannot see that this fact makes a real difference in this position, as both DEBs are unable to find any targets in the opponent's position.

35.i.e? Why not exchange the DEB with 3 5 .ixb6 tt:lxb6 then, if it is so useless? Well, it is not useless, especially as the target on c3 needs

a

b

c

d

e

f

g

h

Esben Lund - The Secret Life of Bad Bishops

38

White is doomed to passivity; Black is slowly improving his position. In a static position like this, it is nice to have an extra jumping knight.

43. .ie2 �xb4 44.gxb4 � f6

39J�c2 gal 40.gb2 �a6 4 1.gbl ga2

45.dl galt 46.gb l gxb lt

No, I will not exchange my active rook!

Now the time has come t o exchange rooks.

47.�xb l � e4 48 ..ifl �xg3 49..ig2 � e4

8

A pawn up, Black no longer fears a pure knight vs. bishop (DEB) ending.

7

6

50.e2 d6 5 1 .he4 dxe4

5

4

8

3

7

2

42.g3

White i s defenceless against the penetration of the knight.

6 a

b

c

d

e

f

g

h

White is in a kind of zugzwang. The pawn move is easy to condemn, but the following lines show the dangers of not playing it. 42 ..ie2 allows: 42 ... lt:l xb4 43.l"\xb4 lt:l a4 44.lt:lbl g3! This move prepares the deadly transfer of the knight to d3. (44 . . . lt:l b2 is not nearly as effective, since White can defend: 45 .g3! lt:ld3t? 46 ..ixd3 cxd3 47.lt:l d2 l":i:c2 48 .l":i:b3 l":i:c l t 49.r,t>f2 and it is difficult to improve the position further.) 44 . . . g3 White is lost because of mate on the back rank after for instance: 45.r,t>dl lll b2t 46.r,t> e l lll d3t! 47 ..ixd3 cxd3 48.lt:ld2 l":i:c2 A double threat! 49.l":i:b3 !':\c l mate! After 42.r,t>e2, Black can combine several ideas that we have already seen: 42 . . . g3 43.r,t>e l lt:lxb4 44.l"\xb4 lt:la4! 45 . .ixa4 bxa4-+ and the a-pawn is a winner.

42 ... �d7!

The knight is off to attack the new target on g3!

5

4 3

2

a

b

c

d

e

f

g

h

As I mentioned earlier in the Introduction, the DEB is never completely isolated, since the king will always remain as a piece that has access to the opposite colour complex. Thus, in this endgame Black can still put pressure on the c3-pawn, with a king march to b3 via a4. White can do nothing to stop this, since his king has to guard the advance of the g-pawn. Meanwhile the DEB can easily guard b5 and f5, at the same time stopping the passed d-pawn. Thus, the DEB is reduced to a temporary defender for the near future, whereas the king and pawns do the attacking. This strategy proves sufficient to win the game.

52.�a3 c6 53.�c2 b6 54.fl a5 55.g3 a4 56.d5 .id7 0-1

Chapter I

-

I ntroducing the Double-edged Bishop

1 .6 Chapter recap At the beginning of the chapter we learned a couple of useful definitions: );;> Unique bishop: A bishop that only one

player possesses

);;> Doubled-edged bishop (DEB) : A bishop

that cannot yet be evaluated as either good nor bad at a given moment

Unique bishops are not always DEBs, but it is very useful to spot the unique bishop, since it indicates what colour complex could be a problem in the future - namely that opposite to the unique bishop. Same-coloured bishops (non-unique) can also be a DEB, since the future exchange of bishop against knight makes one of these bishops unique. Usually one has to spot this future DEB in advance, which is why you could call it a DEB of the second degree. The French Advance variation is a case in point. This exchange of one of the bishops could also be the trade-off bishop against a rook - in short: an exchange sacrifice. This is the main subject of the next chapter. We also learned to take many factors into account when trying to evaluate a DEB, and not just label it 'good' or 'bad' prematurely on false grounds. We saw more than once that taking all these other factors into account clearly gave the DEB a better reputation, and in many cases removed the label 'bad' from it, replacing it with the more appropriate 'DEB'. The most important other factors were: );;> The remaining complementary pieces on

the board

);;> Pawn space

39

We also discussed the range of posmons with DEB relevance. If the position is too closed or too open, the concept of DEB loses its practical value. The exaggeration was made that in open positions all bishops are good, whereas in closed positions they are all bad. In the semi-closed or semi-open positions, we estimated 40% of all the positions to be DEB relevant, and a further 20% for endgames with :!'! vs. i. ( 1 0%) and i. vs. lt:J ( 1 0%) by definition, for a total of 60%. On account of a number of hypothetical positions we saw that a bishop that many players would automatically label 'good' , also has a risk for becoming bad. This became clear by removing some of the remaining pieces on the board. The clue to understanding this was the lack of concrete attacking points in the opponent's position, and that the DEB is therefore dependent on the cooperation of the remaining pieces and pawns. An extreme case of these hypothetical positions was the 'worst-case scenario' , where all the remaining pieces were removed. By completely isolating the DEB on its colour complex, we wanted to check j ust how much it was able to achieve on its own. We also learned that a DEB is never completely alone, since the king is always there to support it on the opposite colour complex. In the discussion about bishop vs. knight we agreed that the bishop might as such (material value) be a stronger piece than the knight, but that the positional and dynamic other factors of a position might make a material evaluation too one-sided. A case in point was the Nimzo­ lndian, where Black indicates his willingness to give up his bishop already on move 3, but in

40

Esben Lund - The Secret Life of Bad Bishops

return gets ocher advantages, be it a better pawn structure after the exchange on c3, or a smooth development. We also made the important point chat the bishop protects one colour complex better than a knight does, since the knight changes colour every move it makes. In the games of the top grandmasters we noticed some important features of a DEB: � The DEB can be a good defender, when another part of the remaining army at the same time

has the task of attacking. Otherwise the player with the DEB would simply be passive. This became clear in the analysis of the game Alekhine - Yates when Black played 20 ... lll a3! to preserve chis piece as a future attacker of the white queenside.

� In fact, often the DEB is an important piece, sin�e exchanging it leaves some of the soft spots

in its position difficulc to defend. This was the case in the game Radjabov - Eljanov, when an exchange of the DEB would make the pawn on c3 very vulnerable to attack.

� We also saw the willingness by one of the players to 'bury' his DEB behind its own pawns. This

was done to obtain ocher advantages - and in realization chat the DEB was a good defender of the soft spots in his position, while the remaining forces held the attacking potential.

In general we learned chat the DEB more often than not has some potential chat makes the opponent willing to exchange it.

Chapter 2 The Exchange Sacrifice In the notes to the last game �f the previous chapter, Radjabov - Eljanov, we saw one example of the exchange sacrifice as a means of breaking through (see diagram auhe foot of page 35). This is more standard than one might chink, and not only an accidental occurrence in a random game. Especially in the phase where a player seeks to transform the advantage into something more tangible, the exchange sacrifice becomes an interesting option. In the example mentioned above, the rook is sacrificed for the DEB, but that is not always the case. However, exchange sacrifices are often closely related to the DEB, as they either cry to reveal the hidden potential of it, or try to isolate it by removing remaining pieces that cooperate well with it. fu we know, with every exchange of the remaining pieces the DEB has the risk of ultimately becoming bad. Thus, the rook can be sacrificed for one of these remaining minor pieces. The exchange sacrifice is often the trade-off rook vs. the other bishop, opposite-coloured to the DEB. I shall deal with three cases of the exchange sacrifice: Case 1 : Rook takes DEB - the breakthrough. Case 2: Exchange sacrifice to strengthen the DEB Case 3 : Exchange sacrifice to weaken the DEB The exchange sacrifice is often accompanied by the win of a pawn. The trade-off between a minor piece plus pawn vs. rook often balance each other, so the traditional material 5-3 value of the rook compared to the minor piece is obviously only approximately true. Exchange sacrifices are common in many opening lines, and the logic behind it is often to develop a positional, long-lasting initiative in the early middlegame, since at chis stage the positions are often semi-closed and not ideal for rooks, which like space and many open lines - typical of lacer phases of the game. The risk for the one who sacrifices the exchange is therefore chat the power of the rook will prevail in a later stage of the game, should the initiative evaporate. I shall look more deeply into certain opening lines where exchange sacrifices occur in the last chapter of the book.

Esben Lund - The Secret Life of Bad Bishops

42

2. 1 Case 1 - Rook takes DEB - the breakthrough Nigel Short - Michal Krasenkow Wijk

aan

It can be useful to consider such hypothetical positions. Actually, we do this kind of 'what-if' reasoning more often than we think, a case in point is the 'worst-case scenario' presented in the introduction.

Zee 2008

The unique bishop on c6 is not obviously a DEB here - rather the other bishop on e7 looks like a candidate, buried as it is behind its own pawns. The unique bishop does tell us something important, though: that Black has a firm grip of the light squares. Consider an endgame with only rooks and dark-squared bishops on either side, that is, we remove the minor pieces that are able to control the light squares:

If Black in the above position can bring his king to e6, he would - among other things give extra support to the d6-pawn, when his position looks very solid. We have already discussed this theme in the Stonewall: The white bishop on e3 is hitting the same wall of pa\Yns as Black's on e7, and might therefore not be significantly better. This was Kramnik's comment. Another way of putting it is that the position is so closed that it might fall outside the category of positions useful to analyse with the help of the term DEB, since the bishops are hemmed in due to the closed nature of the position. Notice that White to move in this position will gain nothing from an exchange sacrifice on d6, since the follow-up is dependent on, among other things, the remaining pieces. The fewer pieces on the board, the less the chance of an exchange sacrifice to succeed. Back to the real game:

23.bxa4! The point will be seen on the next move. The immediate 23.!'lxd6 ixd6 24.!'lxd6 axb3 25.cxb3 !'lfd8 26.!'lxd8t !'lxd8 27.ixc5 also looks promising for White, but the text move is even better as White retains a rook.

23 .. J�aS White threatened a4-a5 followed by lli c4 with total domination.

24J�xd6! The breakthrough. Now follows a long forced line.

Chapter 2 - The Exchange Sacrifice

24 ... .ixd6 25J�xd6 gfc8 26.hcS �xa4 27.�bxa4 .ixa4 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

28 . .ib4 28.�xa4! E:xa4 29.�b4 E:aa8 30.lt>b2 E:d8 3 l .E:d5!± is one tempo better for Black compared to the game, but probably this doesn't make a big difference. In the game Black gets an opportunity to fight on (that he doesn't take).

28 ... hS?! Black had a chance to stay in the game with 28 . . .�e8 . It is then harder for White to gain control of the light squares on the queenside and in the centre after the transfer of the black bishop to f7. Furthermore, White has to consider the threat of a rook exchange, since unlike the game continuation, d5 is not an available square for the rook. A few words on positions with opposite­ coloured bishops: Since the bishops operate on opposite colour complexes, the player with the initiative will have the opportunity to practically be a piece up, since he is the one who can create threats with his bishop. It is a general rule of thumb that the initiative counts a lot in positions with opposite-coloured bishops.

43

In this position it turns out that the initiative is not the main factor - rather, White would like to keep the game under control and slowly advance his queenside pawns. Improving the position bit by bit turns out to be more difficult when he lacks light-squared control, as the following lines show: 28 . . . �e8 29.lll d 5 E:d8 30.E:xd8 ( 30.E:b6 E:ab8 3 l .E:e6 avoids the rook exchange, but Black is fine after 3 l . . . �b5. Instead, 3 1 . . .� f7 32.lll e ?t @h8 33.E:c6 also maintains the rook.) 30 . . . E:xd8 3 l .c4 �a4 32. lt>b2 f7 33. lt>c3 E:b8 34.c5 �c6 and it is difficult to gain control of either a4 or c6 . without allowing a lot of counterplay. 3 5 . lt:l b6 E:d8 36.a4 E:d l 37.a5 lt>e6 and if anyone, it is White who is in danger of losing. At the right moment Black's rook can attack White's vulnerable kingside pawns. White is by no means in trouble after he allowed the retreat of the bishop to e8. With this line I merely wanted to show the counterplay Black was able to create once one pair of rooks got off the board.

29.�xa4 gxa4 30.'it>b2 gaa8 3 1 .'it>b3 gds 32.gdS! 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

This is the crucial idea that makes it very unappealing for Black to exchange rooks - the white c- and d-pawns will decide the game. Black is strategically lost here, as he can do

44

Esben Lund - The Secret Life o f Bad Bishops

little to stop White's advancement on the queenside.

2.2 Case 2 - Exchange sacrifice to strengthen the DEB

One can argue that a strong player like Short could easily calculate this line beginning with 23.bxa4 up to this point, and then conclude that it was won. And if it is a mere question of calculation power, there is no need for fancy theories about the breakthrough exchange sacrifice to eliminate Black's DEB. There is some truth in this reasoning, yet still I object:

Levon Aronian - Peter Leko

1) Krasenkow allowed it, and he is also a very strong player.

Nalchik 2009

I .d4 li:)f6 2.c4 e6 3.li:)c3 ib4 4.e3 0-0 5.id3 d5 6.li:)O c5 7.0-0 dxc4 8.hc4 li:) bd7 9.�e2 b6 10.gdl cxd4 l l .exd4 hc3 1 2.bxc3 ib7 13.ib3 �c7 14.c4 gfe8 15.ib2 �f4 16.�e3 �5 17.li:)el

2) Generalizing with rules of thumb makes the idea easier to spot if a similar situation arises in the future - the 'breakthrough idea' will now be in your arsenal of positional ideas for transforming an advantage. 3) There will be positions in which it is not possible to calculate a certain line all the way to a definite conclusion. We shall see this in later complex examples.

32 ... @f7 33.c4 Nothing can stop the white pawns supported by the king, bishop and rook. It does not take a rocket scientist to see that Black's counterplay against the white kingside is almost non­ existent.

a

b

c

d

e

f

g

h

This position is typical of the Nimzo-lndian Defence. Black has given up his dark-squared bishop, but is very solid and has an easy game to compensate for White's bishop pair.

17 ... bS 1 8.cS!?

33 ... @e6 34.a4 gd7 35.aS gc8 36.icS gb8t 37.ib6 gd6 38.@a4 gdxb6 Hopeless, but c4-c5 was on its way.

39.axb6 �b6 40.gd2 g5 41 .cS gbl 42.c6 get 43.c.tibs g4 44.c.tib6 gxf3 45.gxB gb1 t 46.@a7 gal t 47.@b7 gbl t 48.@c8 gb3 49.c7 gxa 50.gc2 Ironically, Black resigns the moment he profits materially from his 'counterplay' on the kingside.

1-0

a

b

c

d

e

f

g

h

Chapter 2

-

The Exchange Sacrifice

45

Aronian has annotated this game thoroughly, giving his thoughts about his reasoning as well as referring to Peter Leko's comments in their post mortem analysis. I will mainly focus on our theme here: How does White handle the DEB on b2 in this position? Right now it does not look too great . . .

18 ... lll dS 1 9.°1Wg3 lll f4 20J;d2 lt:l f6 2 1 .f.3 lll 6 h5 22.'!Wfl i.d5 Aronian suggests 22 . . . a5 as an improvement for Black.

23.i.c2 But as long as White keeps as many pieces on the board as possible, the DEB on b2 does not yet run the risk of being isolated on the dark squares. And although the two white pawns in the centre are blocked at the moment, one should not forget that c5 is a far-advanced, protected passed pawn. Far-advanced pawns (space) usually give rise to several sacrificial ideas - simply because they are closer to the promotion square and therefore demand the opponent's attention.

23 ...°IWgS 24.©hl i.c4 25.g3

a

b

c

d

e

f

g

34.geS! lll xe5 35.gxeS 5 36.i.b3

h

Aronian prefers the immediate 36.2'.xe6.

36 ... lll dS 37.gxe6 The main point of the exchange sacrifice was to win this pawn - thus transforming his hanging pawns in the centre into connected passed pawns.

37 ... ©h8 38.'!Wel lt:lf6 39.°IWeS+-

��

8

y,

7

The knight is driven back from its active outpost.

6

25 ... lt:lg6 26.lll g2 i.d5 27.lll e3 lt:lf6 28.h4 °1Wh5 29.lll xdS lllxd5 30J�el ged8 3 1 .gde2 gab8 32.i.cl h6 33.©g2 lll c3

4

Aronian suggests 33 . . . 2'.bc8 as a better move, but White's idea in the game still works here, albeit not with such a devastating effect. 34.2'.e5! tLi xe5 35. 2'.xe5 f5 36.2'.xe6± Even if Black can exchange some of the remaining pieces on the board with 36 ... 2'.e8 37.Vfie2 ltJc7 38.2'.xe8t Vfixe8 White still enjoys a huge, almost winning advantage in the endgame after 39.ib3t Wh8 40.Vfixe8t 2'.xe8 4 1 . wf2.

2

5

3

a

b

c

d

e

f

g

h

Black is dead lost, since he has no real counterplay to compensate for White's strong passed pawns supported by the bishops. With his last move, White broke the blockade of the d5-square. Notice the DEB on c l : It is beginning to show potential .

Esben Lund - The Secret Life of Bad Bishops

46

39 .. J;es 40.c6 gbc8 41 ."9xb5 "9g6 42.h5 Or simply 42.if4.

42 ... "9xh5 43.if4 The triumph of the DEB - now a strong supporter of the passed pawns.

43...a6 44."9xa6 lDh7 45.c7 lD g5 46.gxeSt "9xe8 47.d5 gas 48.\'Nc4 �h7 49.d6 \'Net 50.\'Nfl "9e8 5 1 ."9d3 "9d7 52."9c4 "9e8 53.hg5 hxg5 54.\'NgSt! A flashy finish - the pawns will promote.

1-0

Carlsen knew he was slightly worse here. Still, there is plenty of play in this position, and Black will have his share of the chances.

1 5."9c2 ghf8 16.f3 e5 17.hxg5 hxg5 18. � gde8 19.id3 "9f7 20.gh6 ia6 2 1 .liJd2 exd4 22.exd4 lD a5 23.gdh l ge7 8

7

6 5

2.3 Case 3 - Exchange sacrifice to weaken the DEB Next we shall analyse a game fragment where Magnus Carlsen exploits the DEB in great style. Once again the dark-squared Nimzo­ Indian bishop is the centre of attention, this time with a different fate:

Etienne Bacrot - Magnus Carlsen Baku 2008

l.d4 llJf6 2.c4 e6 3.liJc3 ib4 4.llJ f3 b6 5.ig5 ib7 6.liJd2 h6 7.ih4 llJc6 8.a3 ixc3 9.bxc3 "9e7 10.e3 g5 1 I.ig3 d6 1 2.h4 � 13.ie2 lDd7 14.liJb3 5� 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

4 3

2

a

b

c

d

e

f

g

h

This is our real starting point. First one has to realize that White's dark-squared bishop on g3 is the bishop in this position that is likely to become a DEB, as it usually is in the Nimzo­ Indian. This position is a rather closed one, so one could argue, as we have already done on several occasions, that this position is beyond the scope of this book; that the term DEB is not useful here as the position is too blocked. If Black plays . . . f5-f4 at some point, the bishop on d3 is hitting no real targets as Black's six pawns are on the opposite colour, so surely this bishop must be a candidate for a DEB, if any? No, a more attentive look at the position reveals that what matters is the central pawn cluster with the doubled c-pawns. Again it is the c4-square that attracts special attention, making the light-squared bishop on d3 the more valuable of White's two bishops at the moment. It demands deep strategic considerations by both players to be alert to the dangers facing White's dark-squared bishop on g3, and Carlsen coped better with this challenge.

Chapter 2

-

The Exchange Sacrifice

24.\Wa4?! Bacrot admitted after the game chat he had missed the idea in the game. Because of chis idea, I have included chis game in the book. 24.�f2 is possible, with a complex position.

24.. J!e3 25J!h7 \Wes 26.\Wc2 li:) f6! On his blog, Carlsen wrote chat chis move deserves an exclamation mark, as it is very easy for White to go wrong.

27J;7h6 27.E!g7! (Carlsen)

47

queenside - where the vulnerable white king is placed. Here White's bishop will be of little help. White's rooks are occupied on the other side of the board, but doing exactly what?! They have no real targets to turn their attention to, and an exchange of one rook (which would be desirable for White as it would reduce Black's initiative) is not possible at the moment. It becomes clear chat after the removal of the d3-bishop, White has a bad coordination between the pieces, whereas the black pieces work wonderfully together.

29 ... ll:)ds

27 .. .a3 �a2t 42.'it>b4 b5 43.'it>a5 a 3 is no better - White simply has nothing to counter Black's several threats: 44.g4 'it>c6--+ and White's king is in trouble.) 4 l . ..�xg2-+ Black can slowly improve and look for a way to win the position.

40.©xa3 gast 41 .©b2 ga2t 42.©cl lll xc3

6

Winning back the exchange, since mate is threatened beginning with 43 . . .i.b3.

5

4

43.i.d2 lll e2t 44.gxe2 i.xe2 45.gg7 ga4 46.i.c3 bs 47.g4 i.a 4s.gs b4 49.i.b2 ga5 50.©c2 i.d5 5 1 .ge7 i.c4 52.i.cl g13 53.i.d2 gri 54.g6 gg2 55.g7 ©c6

3

2 a

39 a3t!

b

c

d

e

f

g

h

•..

If nicely coordinated, �+i.+tt:l can deliver mate! I cannot help noucmg another way to win the position that fits in well with what we have discussed earlier - exchanging the DEB to transform the advantage - in this case a penetration with the rook on the second rank. The bishop on e3 is keeping the rook away from f2 at the moment! I find it amusing that thisDEB that becomes more and more isolated with every exchange of pieces - so isolated that people are j ust waiting for a chance to point fingers at it and label it a 'bad bishop' - this bishop still has a value that makes Black want to give up his strong centralized knight on d5 to get rid of it.

a

b

c

d

e

f

g

h

After some manoeuvring Black has obtained complete coordination and control, and now he can concentrate his effort on the final assault on the white king.

Chapter 2 - The Exchange Sacrifice

59.@b2 ic4 60.@c3 @d5 6 1 .gb7 gg3t 62.@b2 @xd4 0-- 1

16 ...�xc8 17.�xc3 8

The next game is a favourite of mine as it contains an amazingly deep concept in the early middlegame.

7

Suat Atalik - Pavel Eljanov

4

Wij k

aan

49

6 5

3

Zee 2007

1.d4 liJf6 2.c4 e6 3.liJc3 ib4 4.e3 0-0 5.id3 d5 6.a3 ie7 7.ltJ f3 'l!J bd7 8.0--0 c5 9.cxd5 exd5 IO.ic2 b6 l 1 .dxc5 ia6 12.gel ltJxc5 13.l!Jd4 gcs 14.if5

2

a

b

c

d

e

f

g

h

Now it is Black to move, and his initiative has a chance to develop; this is the trade-off White decided on in contrast to a healthy pawn structure. At the same time White has a material plus and potential for the future, should Black's initiative evaporate.

17 ...ih4

a

b

c

d

e

f

g

h

This is our starting point of the game. Both players have a DEB in this position, on cl and a6, and both have the same 'defect' of not attacking any concrete targets in the opponent's position.

This might be the most critical point of the game, putting Black's sacrifice to the test. I believe that this move was a part of Black's whole concept when he sacrificed the exchange: He will force White to play g2-g3 which creates weaknesses on the light squares - the opposite colour of the DEB on c l . Now Black threatens to take immediately on f2 because of a nasty knight fork on e4.

18.g3

14 ... ltJfe4!? 1 5.ixc8 liJxc3 16.�c2

Was White forced to play this weakening pawn move?

1 6.bxc3!? 1Ml'xc8 gives White an extra tempo compared to the game, and thus fights Black's initiative, but at the expense of ruining the pawn structure. If White is less ambitious and does not wish to profit from his extra exchange, the continuation could be l 7.a4 li:J d3 l 8.El:e2 li::l xc l l 9.El:d2! li:J d3 20.El:xd3 ixd3 2 l .1Ml'xd3 with a level, and interesting, position.

l 8.1Ml'c2!? Th i s might be a n improvement upon White's play. Let's analyse: l 8 . . . li:J e4 Not 1 8 . . . li:J d3 1 9.El:e2 since 1 9 . . . li::l xc l ? does not win back the exchange due to 20.El:xc l eyeing c8.

50

Esben Lund - The Secret Life of Bad Bishops 1 9.gxh4 loses to: 1 9 . . . tt:l e4 20.�c2 �g4t 2 1 . ii h 1 i.d3-+

1 9.g3 �h3

f

g

h

20.i.d2!? The crucial difference. It looks dangerous to allow sacrificial ideas, but they do not seem to work in Black's favour. Without chis little bishop move it would be very difficult for White to develop his queenside, and it might be the fear of some sacrifice chat made Acalik reject chis set-up. 20.f3? does not work because of 20 . . .i.xg3!. 20 . . . E:c8 20 . . . lll xf2?! 2 1 .gxh4 i.d3 22.�c7 lll e4 has the double threat of . . . lll xd2 and . . . �g4t, but White defends successfully with 23.�f4! when 23 ... lll xd2 24.E:ad l wins back the material. It is easy to overlook details in chis line. 2 1 .i.d MG 22.f3 lll xc3 22 . . . lllg5 23.�g2! If White can get chis defensive move in, then c2 was definitely the right place for the queen. 23 . . . �h5 24. h l ! i.xd4 25.g4! �h4 26.exd4 and White has an advantage. 23.bxc3 h5 White holds a small advantage here, although Black is in the game.

18 ...�h3 19.gdl White is under pressure, but perhaps defending with too much caution. Ocher options were:

Also possible was the natural-looking: l 9.b4 tt:l e4 l 9 . . . lll d 3? 20.b5 lll xe l 2 1 .�xe l +- with two bishops hanging. 20.�c7 i.f6 Now threatening to cake on f2. 20 . . . lll xf2 2 1 .gxh4! 2 l .E:a2 E:c8 2 l . . . i.c4!? was also an option. 22.�f4 The position is extremely dangerous for White after 22.�xa7?! i.d3 , when it is difficult for him to untangle himself. For instance: 23.f3 lll xg3! Meanwhile, Black can strengthen his attack by advancing the h-pawn or playing . . . lll g5 at the right moment.

a

b

c

d

e

f

g

h

The following lines show chat the game is balanced: 22 . . . h5 22 . . . lll g S!? 23.�f5 �xf5 24.lll xf5 g6 Even with queens off the board, White still needs to be careful. All Black's four pieces are active and cooperating well at the moment, which is not the case for White. 25.lll d4 lll c3 26.E:c2 i.xd4! 27.exd4 i.d3 Black wins back the material. 28.E:d2 lll e2t 29.E:dxe2 i.xe2=

51

Chapter 2 - The Exchange Sacrifice 1 9.�c2 was also possible, but Black can choose to play l 9 . . .�f6 ( l 9 . . . tli e4 is met by 20.�d2!), when he keeps the option open for both knight j umps to d3 or e4, depending on the circumstances. 20.�d2 tli d3 Now after 2 1 .Elfl tlie5, Black wins back the material, and after 2 1 .Eled 1 ? the sacrifice works: 2 1 . . . tLi xf2! 22.'it>xf2 �xh2t 23.'it>f3 �xd4 24.exd4 �e2t with a mating attack.

the rook is pinned to the defence of fl . 22.exd4 Ele2-+) 2 l . . .�xf5 22.tlixf5 tli b3 23.Elb l Elc8 With . . .�e2 on its way. 24.�d2? �d3!

20 J�e8 •.

8

7

6 5

4

3

2

a

a

b

c

d

e

f

g

h

19 ....if6! Very logical, for many reasons. We know that with every exchange White's DEB on cl is being isolated. But with the exchange �f6xtlid4, opposite-coloured bishops will remain on the board and Black will rule on the light squares. For the record, the knight on d4 is defending the light squares (f3 and e2) , so it's logical to exchange it to keep the attack going.

20.�el

b

c

d

e

f

g

h

With White I would probably think along the lines: "What have I done to end up in this position - I made a simple bishop move with 1 4 .�f5 a long time ago, attacking the black rook on c8!?" It is amazing how fast Black's initiative developed after the exchange sacrifice. It is also easy to forget that Black's a6-bishop is a DEB, since as long as Black's initiative is ongoing, the bishop seems only useful and very strong. But again, no direct targets are to be found in White's position that it can attack; all White's pawns are on dark squares. In short, the DEB on a6 is dependent on the remaining forces.

Another cautious defensive move, this time probably the best.

2 1 .b4

20.b4 fails to 20 . . . tli e4 2 1 .�e l tli g5 with problems.

2 1 ...tli e4 22.0

20.�c2 Ele8 2 1 .�f5?! Black's coordination is so good that again a queen exchange will not stop his initiative. (2 l .f3? �xd4! loses outright since

My engine mentions 22.b5 as White's best defence, with the point 22 . . . �xb5 23.tlixb5 �xa l 24.Elxd5, but Black can play the non-

The DEB on cl is sensing a breath of air. 22.�b2? is still too early: 22 . . . tlig5-+

Esben Lund - The Secret Life of Bad Bishops

52

materialistic: 22 . . ..ixd4! 23.Ei:xd4 (Even worse is 23.exd4 .ixb5 24 ..ie3 f5! with a huge attack.) 23 . . . .ixb5 This simply keeps the right pieces on the board, and White remains with the same problems, now with a pawn less. 24.f3 llig5 is similar to the game continuation.

26J!xd4

22... lll gS 23.�fl �hS 24. fl ! 0-0 1 5 .1:%h3 tll b4 1 6.1:%g3

Chapter 3 - From Opening to Early Middlegame transposes, and White has avoided Black's option l 7 . . . i.b4t.

14J�h3!?

65

Al) 17 ...i.b4t! The simple solution.

1 8.'i:t>fl t°i:J e7

14.0-0 gives no advantage.

14 J�ac8 1 5J�g3 'i:t>h8! •.

It is important to move the king to the corner square. 1 5 . . . �h7?! 1 6.�fl ti:J b4 1 7.i.d2 tLl c2 1 8 J''i: b l and now 1 8 . . . ti:J xd4? loses to 1 9 . ti:J xd4 'Wxd4 20.l"i:xg7t!! check. Afcerward!i White cakes on e6 and h6 and picks up the unprotected queen on d4, with a winning position.

Driving the rook back. With the correct white move order, Black is not allowed to play chis. Al)

17 ... t°i:Jb4!?

The more complicated solution.

1 8.'i:t>fl

i'H7

Prophylaxis against sacrifices on h6.

1 6.�fl ! This move order seems to rule out Black's 'simple solution' with l 7 . . . i.b4t (see below) . 1 6 . . . ti:J b4 l 7.tLlxe6 fxe6 1 8.Ei:g6 transposes to l 7 . . . ti:J b4!?.

1 8 . . . ti:J c2 19.i.xh6 gxh6 20.'Wd2 Ei:f7 2 1 .l"i:cl Wxb2 22.�g l ! Ei:c6 23.Ei:xh6t Ei:h7 24.Ei:xh7t �xh7 2 5 . ti:J g5 t i.xg5 26.'Wxg5 White has compensation, but the game is probably j ust a draw. 26 . . . ti:J xd4 27.'We7t=

16 fxe6 17J:;g6

1 9.i.d.2 t°i:Jd3 20.Wfb3 Wfa6 2 1 .'i:t>gl i'k6m

16.t°i:Jxe6

•..

8)

12.@fl!? §°b6 1 3.g3 0--0--0! ?

Another option is short castling. After the text move, Black has to be acquainted with the plan of slow expansion.

14.'i:t>g2 'i:t>b8

a

b

c

d

e

f

g

h

This, I believe, is the most dangerous plan for Black to play against: White exchanges off the DEB on e6 and plays his rook to g6. Black has two ways of dealing with the rook on g6 and the threat to e6: Al) 17 i.b4t! and

Al)

•••

17 t°i:J b4!?. .••

Esben Lund - The Secret Life of Bad Bishops

66

1 5.b3!? !k8 16.a3!? Reinaldo Castineira - Caselas Cabanas, Sanxenxo 2007. The point is seen in the following lines:

16 ... lll aS I7..id2 !k7

4

1 7 ... lll x b3? 1 8 .Ei:b l +-

3

18.°IWel! lll c6 1 8 ... li:Jxb3 1 9.Ei:b l a5 20.a4 with a deadly pin along the b-file. Black did not get his knight to c4.

19.b4 ghc8 20.hS

2

a

b

c

d

e

f

g

h

White has to make a decision here: Cl) 13.�h3 or C2) 13.ci>fl .

White seems to be in control...

8 1

8 7 6 5

i �, r�� �'"J��-J��, ' �� ��

Cl) 13J'�h3 g6! 14.ci>fl

: �fJ�d "•afM;� 'fJ �� � • mttJ� '- ���fj'��

l 4.Ei:g3? g5!

14 0--0-0 1 5. ci>gl .•.

4 3

2

� a

b

• • m� c

d

e

f

g

h

20 ... a6 21.°IWbl lll a7 ... but Black has counterplay down the c-file. C) 1 2.lll f4

After chis move White loses control of g5 as the knight blocks the bishop's access to chis square. Therefore he has to cake a future . . . g5 into account in these lines.

12 ...°1Wb6 Interesting is 1 2 . . . g6!? with the idea of a later ... g6-g5 after having connected the rooks.

16.lll xe6 fxe6 17.hxgS hxgS 1 8.gxh8 l 8.li:Jxg5 1Mfxd4 l 9.1Mf xd4 lLlxd4+

18 gxh8 19.Lgs hg5 20.lll xg5 lll xd4m .•.

Chapter 3 - From Opening to Early Middlegame

C2)

13.�fl

67

16.b3 !k8 1 7.a3 Slow expansion by White.

8

7

17 ... gS! 1 8.hxgS hxg5 19Jhh8 gxhs 20.�xe6 fxe6 2 1 .ixgS .ixg5 22.�xgS llJxd4m

5

The conclusion is that the break . . . g6-g5 usually works with the knight on f4.

6 4 3

2

a

1 3 ... g6!

b

c

d

e

f

g

h

Again this move makes a lot of sense with the knight on f4. If Black instead chooses 1 3 . . . 0-0-0 l 4.h5! 'it>b8 1 5 .�h3 �he8 1 6.�g3 �f8 1 7.'it>g l �c8 White can use the slow expansion with l 8.b3 �c7 l 9.a3 followed by b3-b4. This idea is also seen in line B) 1 2.'it>fl !?. An important point to make here is that without the move h4-h5 I do not recommend Black to castle short because the knight can then land on this square, putting pressure on g7.

14.g3 0--0-0 1s.c.t> g2 c.t>bs

7

6 5

4

3

2

b

c

d

e

3.3 Cbemin's 9 ...ib4!? 9.liJc3 ib4!? This is Chernin's move, although Vasiukov introduced the idea in a similar position two years earlier (in Kholmov - Vasiukov, Moscow 1 984) . Black wants to exchange his dark­ squared bishop for a white knight, thereby reducing the pressure on the d5-pawn. This minor piece exchange isolates both DEBs further on their colour complexes, and is therefore an interesting strategic choice.

10.id2 ixc3

8

a

Having gone through the most important lines and ideas of the position after 8 . . . exf5 , I suggest the novelty 9.�fl !?N for White. The idea is to avoid Chernin's move 9.lt:Jc3 �b4 and at the same time avoid the weakening 9.a3 - at least for the time being. White continues with h2-h4 later; in many lines he does not castle short anyway.

f

g

h

68

Esben Lund - The Secret Life of Bad Bishops

White has two options: I) The solid 1 1 .�xc3 2) The double-edged 1 1 .bxc3!?

3.3. 1 lhe solid 1 1 .ixd

1 9 . �xb4 �a7 20.aS '1Wd8 2 1 .�bS b6 22.gl ge4 23.lbd3 �e8 24.if4 ih5 25.�cl ge2

6

Curiously enough, for many moves now it seems that Black has had a strong initiative goJng. The problem is that this initiative is not going to last.

5

4 3

2

26.lbe5 gaxa2 27.gxa2 gxa2 28.lDxc4 dxc4 29.ie5 a

b

c

d

e

f

g

h

Question 5: How should Black play this position? Find a good plan to follow in the early middlegame.

1 1 . ..ie6 1 2.h4 Actually, this position was reached after the (strange) move order 9. llic3 i.e6 1 O.h4 i.b4 l I .i.d2 i.xc3 l 2.bxc3.

12 ...�d7?! The most natural seems to be 1 2 . . . 0-0N and if White wishes to make his somewhat superficial and time-consuming knight manoeuvre, Black plays .. .f6 without delay. 1 3 .lligl llia5 14. llie2 llic4 1 5 .i.c l (Or l 5.llif4 �e8 1 6.�h3 f6 1 7.exf6 �xf6 1 8.Wfl i.f7 1 9.llid3 and if Black wishes to part with his DEB, he can play 19 . . .i.h5 20.1l*'xh5 lli xd2t 2 1 .Wg l �ac8 with a good position.) 1 5 . . . f6 1 6.exf6 1l*'xf6 and Black's lead in development should get him somewhere. The queen move in the game is a loss of

8 7

6 5

4 3

2

1

a

b

c

d

e

f

g

h

Question 6: Black has won a pawn,

but why is he in trouble here?

The rest of the game can be found together with the answer to the question .

Chapter 3

-

71

From Opening to Early Middlegame

3.4 White allows ...g7-g5

1 9.a4 a6

Nikolai Chadaev - Oleg Chebotarev Sochi 2007

Also possible was: l 9 . . . bxa4 20.E\xa4 f()co

20.axb5 axb5 2 1 .'1Wd2 .if5 22,ga2 f6 23.e6 O! 24.\Wxg5 fxg5 25.lll db4

l .e4 c5 2.lll f3 e6 3.c3 lll e7 4.d4 c:xd4 5.c:xd4 d5 6.e5 lll bc6 7 ..id3 liJ f5 8 . .hfs exf5 9.lll c3 .ie6 10.0-0 .ie7 1 1 .lll e l '1Wb6!? 12.lll c2 0-0 13.b3 gac8 14.lll a4 \Wc7 1 5 . .ia3 We already discussed this position in the theoretical overview.

1 5 ... b5! 16.he7 Let us consider the following move instead: 1 6.lll c S �xc5 1 7.�xcS lll xd4!? 1 8 .�xf8 lll xc2 1 9 .�d6 1!fic3 20.E\c l d4 a

b

c

d

e

f

g

h

25 ... fxg2

a

b

c

d

e

f

g

h

Question 7: A) Why does Black have sufficient compensation - or even more than that - for the exchange? B) How should Black react if White plays 2 1 .'1Wf3?

16 ...\Wxe7 17.lll b 2 1 7.lll c S? is bad due to 17 . . . lll x d4.

17 ... f4 18.lll d3 \Wg5 Black has an active position, and the DEB will soon show up on f5 .

Black secures his bishop on the e4-square. More ambitious was 25 . . . lll xb4 26.lll xb4 �xe6, trying to show that White has some problems with the back rank. The point is 27.Ele l Elc3! and after 28.gxf3 �h3 29.lll xdS Elxb3+ he can try to combine play against the white king with his passed b-pawn.

26.©xg2 lll xb4 27.lll xb4 .ie4t 28.0 gxa The point.

29,gxa g4 30. ©g3 gxf3 3 l .ga7 ge8 32. lll c6 h5 33.e7 h4t 34.©fl ©f7 35,gb7 •!z-1/2

Olli Salmensuu - Kalle Kiik Turku 200 1

l .e4 c5 2.lll f3 e6 3.c3 lll e7 4.d4 c:xd4 5.c:xd4 d5 6.e5 lll bc6 7. .id3 lll f5 8.i.xfS exf5 9.lll c3 .ie6 1 0.0-0 .ie7 l l .a3!? h6 1 2.b4 g5 13.lll a4 b6 14.\Wd3 '1Wd7 1 5.lll c3 f4 16.lll e l gc8 l 7.'1Wd2 0-0 1 8..ib2 f6

Esben Lund - The Secret Life of Bad Bishops

72

From a theoretical viewpoint, this game is not so interesting, but the game shows the dangers for White if Black expands on the kingside and later breaks in the centre.

19.exf6 .ixf6 20.lll e2 i.g7 2 1Jk l ig4 22.f3 1f5 23.lll d3

a

b

c

d

e

f

g

h

1 3.i.e3! looks strong here; the point is l 3 . . . tlia5 I 4.tlic5 i.xc5 l 5.dxc5 tli c4 I 6.i.d4 with a dominant position. White has space and a nice 3 vs. 2 pawn majority, and many ways to improve his position. His DEB on d4 has far more prospects than Black's on e6.

a

b

c

d

e

f

g

h

13 ... bS!? 14.lll c S hcS Again we see this exchange of dark-squared bishop for knight.

Question s:

How should Black proceed here?

3.5 Other positions Yuri Shulman - Magesh Panchanathan Dallas 2006

1 .e4 cS 2.lll f3 e6 3.c3 lll e7 4.d4 c:xd4 5.c:xd4 dS 6.e5 lll bc6 7.i.d3 lll f5 8 . .Lfs exf5 9.0-0 Wfb6?! I prefer only to play this queen sortie when the white knight has moved from c3 to e2, since otherwise Black's queen can be chased after tLi c3-a4.

10.a3 i.e6 1 1 .lll c3 �k8 12.lll a4 Wfa6 13.b4 This gives Black the chance to equalize with a nice counter-idea.

Question 9: White would love to play 1 5.dxcS with the same structural advantage as discussed above after the move 1 3 .i.e3 ! . So why did he refrain from it? Back up your answer with relevant lines.

Chapter 3

-

From Opening to Early Middlegame

15.bxc5 0-0 Black is fine and actually played a good game, outplaying his higher-rated opponent, only to blunder and lose it all. I will focus on a few main points:

16..id2 gbs 17.�c2 �a4 1 S.�xa4 bxa4 19.gfb l h6 20.h4

8 �� m �- •m 7 � m m 1a wr

6 5

4 3

2

'a�la1W ' • · � ·-- -�-' ' n' ' • �:tj m mqj� '• �� �-��M- m � a

20 ... f4!

b

c

d

e

f

g

8 7

6 5

4 3

2

a

21 .ixf4 gb3! 22.ie3 i5 23.gel gfbsi Black has more than adequate compensation. White sacrifices a pawn to get some play.

24.if4 24.lll h2 Ei'.d3 25 .:B:ed l Ei'.xd l t 26.Ei'.xd l Ei'.b3 27.Ei'.a l Wf8 This is a very unpleasant ending for White. Black can win back the pawn whenever he wishes - for now he simply improves the position of his king.

b

c

d

e

f

Question 10: Argue why the exchange sacrifice 26 . . . Ei'.xf3 ! is strong here.

The rest of the game can be found in the answer to the question.

Willy Hendriks Igor Glek

h

An interesting pawn sacrifice to improve the DEB on e6.

73

-

Dutch Championship 1 996

1 .e4 e6 2.d4 d5 3.e5 c5 4.c3 lLic6 5.lll f'3 llJ h6 6.id3 cxd4 7.cxd4 ttJ e; S.i.xfs ex5 9.lll c3 ie6 10.lll e2 .ie7 1 1 .h4 h6 12.llJf4 �b6 1 3.@fl 0--0--0 14.h5 @bs 1s.gh3 gheS 16.gg3 if8 17.@gl gcs 1 S.llJe2 gc7 19.lLic3 gecS 20.lLia4 �b5 2 1 .if4 b6 22.gbl @as 23.lll c3 �a6 24.lLiel lll e7 25.�d2

24 ... gSb5 25.e6 fxe6 26.id6

a

b

c

d

e

f

g

h

74

Esben Lund - The Secret Life o f Bad Bishops Question 1 1 : Argue why 25 gS! is a strong move in this position. •..

The moves until the next diagram are together with the answer to the question.

8 7 6 5 4

8 7 6 5 4 3

2

a

1 2 ... cj;i d7!?

3

2

a

b

c

d

e

f

g

32.�4! h3?! Instead 32 . . . lll xf4 33.'1Wxf4 h3 34.gxh3 l'!h8 looks simple and strong. The DEB on e6 is brilliant, defending both d5 and f7, and also within reach of the white king.

33JU6 ghs 34.'!Wg7 gcc8 35.gxe6? Best was 35.l'!xg6! fxg6 36.lll e 3 hxg2 37.lt>xg2 when the game is far from over.

35 ... fxe6 36.'!Wxg6 gcg8 37.'!Wxe6 hxg2-+ 38.'!WxdSt '!Wb7 39.'!WxgSt gxg8 40.lll e4 '!Wh7 0-1

Praveen Thipsay - Christian Bauer Bled (ol) 2002

b

c

d

e

f

g

h

Many players seem to take a liking to this double-edged king move. Admittedly, you don't often get a chance to place the king in the centre like this, but I believe Black has safer ways of playing the position.

13.hS 1 3 .b4 g5! is Black's idea - and perhaps the concrete argument as to why the double­ edged king move works in this position: it has connected the heavy pieces on the back rank!

13 ...'!Wb6 14.'!Wd3 1 4.b4?! a5! and White is in no position to play l 5.b5; hence the queen move in the game.

14 ... lll aS Just in time to get the knight to c4.

15.b4 lll c4 16.lll g3 g6 Bauer must have calculated this pawn sacrifice when he decided upon the king move to d7. He gets decent compensation in the following play.

17.gb l

l.e4 cS 2.lll f3 e6 3.c3 lll e7 4.d4 cxd4 5.cxd4 dS 6.eS lll bc6 7.i.d3 ll:\£5 8.Lf5 exf5 9.a3 �e6 10.lll c3 �e7 1 1 .h4 h6 1 2.lll e2

1 7.hxg6 fxg6 1 8 .l'!xh6 l'!xh6 1 9.�xh6 a5 wins back the pawn on the queenside soon.

We have seen this kind of position before, but here White has spent time on a2-a3.

17 ... '!Wa6 1 8.gb3 gac8 19.hxg6 fxg6 20.gxh6 gxh6 2 1 .hl6 lll xa3 22.'!Wxa6 bxa6

Chapter 3

-

From Opening

to

Early Middlegame

75

1 2 . . . �e? 1 3 .�d2! is the point: Black has no time to bring his king to g7 to defend the h6pawn.

1 3.0-0 .ie7 14.�d2! �a5 1 5.b3 .ib4 An extremely clumsy way to defend the h-pawn.

16.�d3 � c6

How would you evaluate this position?

Why did Black not play 1 6 . . . Wf8? The answer might be l 7.�d2 �e7 l 8.h5 g5 l 9.'Llg3 f4 20. 'Ll f5 'Ll c6 2 1 .:!"!ac l when White's position looks dominant. It is not a good idea to part with the DEB on e6: 2 l . . .�xf5? 22.�xf5 Wg7 23.e6± with problems on the light squares.

Ara Minasian - Valerian Gaprindashvili

17.a3 .ie7 1 8.b4!

a

b

c

d

e

.

f

g

h

Question 1 2 :

Linares I 999

1.e4 e6 2.d4 d5 3.e5 c5 4.c3 � c6 5.�8 �ge7 6 ..id3 cxd4 7.cxd4 � f5 8.Lf5 exf5 9.�c3 .ie6 10.h4 h6 1 1 .�e2 g6?!

White changes plan: he allows the black monarch to reach g7, but instead expands on the queenside.

1 8 ...�d8 Better was: 1 8 . . . �f8 1 9.'Llc3 Wg7 20.'Ll a4t

19.�d2! hh4 20.hh6 .ie7 2 I ..ig5! Exchanging the DEB and isolating Black's on e6.

2 1 . .. i.xg5 22.�xg5±

8

7

6 a

b

c

d

e

f

g

h

This was played a couple of times before the strong move . . .

1 2 ..ie3! . . . was introduced.

1 2 ...�b6

5

4 3

2

a

b

c

d

e

f

g

h

76

Esben Lund - The Secret Life of Bad Bishops Question 13: Explain why chis position

is better for White.

22 ... ©d7 23.f4 '!Wb6 24.gab l a6 25.�0 '!Wa7 26.�cl �e7 27.�d3 b6 28.a4 ghc8 29.aS! Securing the c5-square for a knight.

29 ... gc4 30.axb6 '!Wxb6 3 1 .�cSt ©e8

1 4.hS! Preventing . . . g5, for now and almost for ever!

14 ... ©c7 1 5.gg3 gg8 1 6.a3 a

b

c

d

e

f

g

h

Question 14: Find a way for White to

build up an initiative.

Jiri Nun Gennadi Zaichik -

Polanica Zdroj 1 989

1.e4 cS 2.c3 e6 3.d4 dS 4.eS � c6 5.�f'3 � h6 6.id3 cxd4 7.cxd4 �e; 8.ix5 ex5 9.�c3 i.e6 10.�e2 ie7 1 1 .h4 h6 12.�f4 �k8?!

In chis position I shall suggest a major improvement and at the same time a typical DEB idea: 1 6.Wfl W b8 1 7.�g l tt:J a5 1 8 .Wd3 Wb6 l 9.tt:Jxe6 fxe6 20.:l'!g6 Wc6 2 l ..id2! tt:Jc4 22 . .ic3± 7

6 4

12 . . . Wb6 is the normal move. Neither Shore nor Psakhis has any comments to chis move. I believe it is very imprecise, since now Black cannot cascle queenside - chat is, get his king to safety on chis side of the board. As long as White hasn't played h4-h 5, it is less attractive for Black to cascle kingside instead - since it allows a knight jump to chis square.

a

b

Question 1 5 :

c

d

e

f

g

h

Why i s White on top in chis position?

1 6... ©bs 1 6 . . . tt:Ja5!?

Chapter 3

-

From Opening to Early Middlegame

17.b4 �b6 1S.gbl ltJa5 19.g2 h5 3 l .ge2 d4 32.i.f4 i:ha3 33.!'!b2 gb3 34.ga2

26 .. Jhf)! Nice idea - I would label it under the theme "Exchange sacrifice to weaken the DEB". In the resulting semi-closed position White's bishop on d6 is doing little, and furthermore it is difficult for White to achieve a relieving exchange of remaining pieces, since Black's DEB on f5 does a great job controlling the b l -square. Black's cooperation between the remaining army of rook, knight, DEB and pawns is very good.

27.gxf) lllxd4 28.!'!edl

a

b

c

d

e

f

g

h

34 ... gxf)!? Fancy, but exchanging the rook weakens his piece coordination. I prefer 34 . . . e5+.

35.@xf3 i.g4t 36.i>e4 i.xdl 37.i.cl i.b3 38.ge2 .idl 39.gd2 i.b3 40.ge2 i.c4 4 1 .gel i.a2 42.i.b2 i.d5t 43.i>d3 e5 44.gal g5! 45.!'!gl 45.hxg5 h4 46.We2 h3 47.f3 e4 decides.

45 ... g4 46.!'!al e4t This makes the position loose. After 46 . . . .ib3 47.We4 ©f7 Black brings in the king: 48.Wf5 d3 49 . .ic3 .ic2 with too many threats. 28 . . . tt'lxf3t!! was possible, and I believe Black avoided it because of White's counterplay with the c-pawn. But Black's rook, bishop and knight (supported by the pawns) threaten to mate the white king! 29.Wg2 (29.Wh l .ie4! or 29.Wfl E!:b2! 30.c6?? .ih3 mate) 29 . . . tt'l xh4t

47.i>d2 i.c4?? Oh, no!

48.i.xd4! lll xd4 49.i>c3 lll b3 50.i>xc4! The c-pawn queens if Black captures the rook.

1-0

Chapter 3 - From Opening to Early Middlegame

Question 1 1

Question 12

8 7

6 5

4 3

2

a

b

c

d

e

83

f

g

h

25 ... g5! This is an interesting way to create counterplay: at the cost of weakening the pawn structure, Black has achieved the pawn push against all odds (White's h4-h5 normally prevents it). Please again notice that the configuration of d5-�e6-f5-f7 makes the black position very solid as the pawns are well guarded by the DEB on e6. Black does not mind that this DEB on e6 is a defender at the moment as his other pieces can become active. Still, the pawn push .. .f5-f4 is in the air, which will liberate the DEB to join the action on the kingside against the white king.

26.hxg6 lll xg6 Black is eyeing f4 and has future pawn pushes with the f- and h-pawn in mind. For tactical reasons, the pawn on h6 is not hanging, as will soon be clear.

27.lll c2 27.�xh6 �xh6 28 .Wxh6 lll xe 5! 29.dxeS d4 is Black's point. 30.tll d l ? Wxa2-+ loses the rook on b l .

27 ... h5 28 ..ih6 h4 29J�f3 .ixh6 30.�xh6 f4 3 U�el �k4 Black has taken over the initiative.

This endgame is complex. Black's pawns on the a-file look ugly, but more important is the activity of the pieces. Black's DEB on e6 is not a problem for now, since Black can play actively with his other bishop, rook and knight. His king is also actively placed in the centre of the board. Talking about concrete weaknesses, Black can easily put pressure on White's pawns on b4 and d4. And because of White's uncoordinated pieces on the kingside, Black has good play on the queenside. Especially the knight on g3 looks rather clumsy and needs to be relocated. So Black's pawn weaknesses are made up for. The position is approximately balanced.

23.iidl lll c4 24..ig5 ghs 25.lll fl ghl 26.he7 gxflt 27.iie2 gal 28 ..ig5 f4!

8 7

6 5

4 3

2

a

b

c

d

e

f

g

h

Black activates his DEB on e6.

29.lll d2 29 .�xf4 �g4 with the idea of . . . Ei:g 1 .

29 ....ig4t 30.f3 ggl 112-1/2

84

Esben Lund - The Secret Life of Bad Bishops

Question 13

•.•

This position is a disaster for Black. It shows chat he has handled his DEB on e6, and the position in general, poorly. After the exchange of dark-squared bishops Black has lost control of the g5-square and with it his possible kingside counterplay. Even with additional pieces he cannot really stop White's play - both on the dark squares in general and on the queenside in particular. It is still too early to call the DEB on e6 'bad' as ocher pieces still remain on the board. But Black's problems here are not mainly due to the DEB, but his lack of real counterplay and the missing cooperation between his pieces. The removal of the h-pawns really made a crucial difference! With pawns on h6 and h4 Black would rule out any knight j ump to g5, and he would have a potential pawn break with . . . g6-g5.

Question 14

32.%'fel! Aiming fo r h 4 and the poorly defended dark squares on the kingside.

32 xh7 25.lll g St .ixg5 26.'!tfxg5 lll xd4 27.'!tfe7t=) 19 ..td2 l'ild3 20.'l'Nb3 'l'Na6 21.@gl gc6m 82422) 12.l'ilf4 'l'Nb6 ( 1 2 ... g6!?) 82422 1) 13.@fl

82422 1 1) 13 ... 0--0--0 14.hS! @b8 ISJ;h3 ghe8 16.gg3 .if8 17.@gl gc8 18.b3 gc7 19.a3 82422 12) 13 ...g6! I4.g3 0--0--0 1s.@g2 @b8 I6.b3 gc8 I7.a3 g5! I8.hxg5 hxg5 I9.gxh8 gxh8 20.l'ilxe6 fxe6 21 .hgS hgs 22.l'ilxgS l'ilxd4m 824222) 13.gh3 13 ... g6! 14.@fl ( 1 4.Ei:g3? g5!) 14 ... 0--0--0 15.@gl g5! 16.l'ilxe6 fxe6 17.hxgS hxg5 18.i'!xh8 ( 1 8.lll xg5 ?! '!tfxd4 1 9.'!tfxd4 lll xd4) 18 ... gxh8 19 ..ixgs hgS 20.l'ilxg5 l'ilxd4m 82423) 12.@fl!? 'l'Nb6 13.g3 0--0--0 I 4.@g2 @b8 15.b3! gc8 I6.a3! Filas I 7.J.d2 &7 ( 1 7 ... lll x b3? 1 8.Ei:b l ) 18.'l'Nel! l'ilc6 ( 1 8 ... lll xb3 1 9.Ei:b l a5 20.a4) 19.b4 ghc8 20.hS a6 21.'l'Nb l l'ila7 22.'l'Nd3 'l'Nb5 23.'l'NxbS l'ilxb5 24.a4 l'ila7 25.ghb l gc2 26.l'ilf4 b6 27.bS axb5 28.axbS @b7

88

Esben Lund - The Secret Life of Bad Bishops 3.8

Chapter recap

In chis chapter everything about the DEB we had encountered so far was put to the test through 1 6 questions while reading through the text. The framework was given by the Advance French, using the specific variation arising after 8 ... exf5. Not everyone plays chis particular opening line, but my examination showed a way to use the concept of the DEB actively in the opening preparation, which also included the transformation into the early middlegame. We looked at several types of positions arising from the critical continuations, each position demanding different treatment. A recurring theme was the solidity of the configuration d5-�e6f5-f7. The DEB is blocked in by its own pawns for the time being, but the bishop is doing a great job defending the central structure. This gave Black a free hand in many positions to launch an attack elsewhere on the board, knowing chat his possible weaknesses were well guarded. And his bishop could often enter the game later after either . . . d5-d4 or . . . f5-f4. However, in the position after the double-edged 1 1 .bxc3!? we saw chat Black had to play ... f7-f6 to break in the centre, thereby making his DEB slighcly loose and giving White's DEB a nice future square on e5. In return for chis break, Black freed his position and had active pieces. In other words, the other factors such as better coordination and development were more important. One should not be too rigid, since different positions demand different solutions. In the positions after 8 ... exf5, ideas with exchange sacrifices were not too surprising; both players have a DEB in a semi-closed position. The most common of these exchange sacrifices were the ones to weaken the opponent's DEB (Case 3).

Chapter 4 Endgame Section In this chapter we shall remove all the remaining pieces from the board and discuss positions where one side has a bishop. That is, only one side has a bishop. Positions with same-coloured and opposite-coloured bishops do not fulfil the aim of this chapter. Instead I wish to show the difference in nature between the bishop and the other two pieces: rook and knight. This gives us two worst-case scenarios (no remaining pieces for the bishop side) worth discussing: I ) Rook vs. bishop. 2) Bishop vs. knight.

Later in the chapter I shall briefly discuss the idea of the worst-case scenario and why it is useful. The two types of endgames I shall present here have a striking similarity: the presence of an isolated bishop can give positions that are very static and yet can become extremely sharp. 'Static' means when the bishop side fails to gain control of the colour complex opposite to the bishop. This is due to the nature of the bishop. And on the other hand the positions usually get very sharp once the pawn races start. The most extreme case is rook vs. bishop (with no pawns!) and this is where we begin.

4. 1 Rook vs. bishop From a material point of view, the rook is a stronger piece than the bishop. The relative value of 5 versus 3 pawns is perhaps only an approximate estimate for both pieces, but 5 is clearly more than 3! If we compare the two long-range pieces, the advantage of the rook over the bishop is that it has access to both colour complexes at all times, and also with every move. This might seem like a trivial truth, but it is worth stating anyway since it affects the position in two important ways: I ) The rook can, at the same time, both stop as well as attack a passed pawn, in contrast to the bishop. This comparison also holds for rook vs. knight.

Esben Lund - The Secret Life of Bad Bishops

90

2) Because of its colour blindness, the bishop is much more dependent on the remaining army than the rook is - in chis case the king and pawn(s) . So, the colour opposite to the bishop should be given serious attention, as well as the coordination of the few remaining pieces. Usually the side with the bishop will have to have some material compensation (that is, extra pawns) to match the rook, but lee's first take a look at positions with equal and reduced material. The bishop side hopes to create a fortress.

Horwitz & Kling 1 85 1

a

b

c

d

e

f

g

h

4J�g2! Now the bishop is forced onto the e-file. Or even further away, but chat doesn't help him either.

4 ...ie5 Boch 4 . . . �h4 5 .lt> h 5 t and 4 . . . �f4 5.lt>f5t drop the bishop to a sneaky discovered check.

5J3e2 id6 6J�e8t ifB 7J3d8 With mate on the next move. Here the zugzwang on the 8 ch rank worked since Black's king was in the wrong corner.

l...igl! Black tries to hide his bishop behind the white king. With the bishop on the f-, g- or h-file, White cannot threaten the bishop and threaten back rank mate simulcaneously. However, the shared effort of rook and king chases the bishop back out of the bush.

2J3fl ih2! 3J3£2 ig3 3 . . . �g l 4.Ei:g2

No fortress chis time, but under normal circumstances, the pure rook vs. bishop is a draw. The easiest way to draw is to bring the king to the corner opposite to the colour of the bishop. In the diagram position above, Black's king was trapped in the wrong corner. Another atypical situation is if the king and bishop are separated under unfavourable circumstances, and here many different ideas have found their way into the world of endgame studies. We shall see some of these ideas as we go through chis chapter. Notice chat in our diagram position above the rook was not only controlling the seventh rank, but also blocking the f-file. Thus it prevented the black king from escaping out of the corner to the other side of the board via f8 .

Chapter 4 - Endgame Section The next posmon is a study by a Danish composer, and is very relevant for practical play. One pawn has been added to either side, and at first glance the winning task does not look easy for White, since the extra pawn is a rook's pawn.

Jens Enevoldsen

91

Uk7t 'i:td8 First we notice chat l . . .© b8 brings the king to the wrong corner, and White wins similar to the introductory diagram. 2.1%c3 Zugzwang - Black loses his pawn or gets mated. 2 ... ie2 brings the bishop out into the open, and again White can combine threats to bishop with mating threats. 3.1%e3 ig4 4.1%e8t ic8 5.1%f8 Zugzwang and mate next move.

1 949

Thus, the black king has been forced to take the first step away from the a6-pawn.

8

7

2.'i:tb7

6

Black is in a kind of zugzwang again, since he will have to move his bishop, allowing White to check on the d-file.

5

4

2 ...ie2 3.gc2

3

The threat to the bishop.

2

3 ... ib5 4.gd2t 'i:te7 s.'i:tc7 'i:te6 6.gd4! a

b

c

d

e

f

g

h

White wins by following a plan very typical of these positions: I) Combine zugzwang and threats to the

bishop to check the black king as far away from the a-pawn as needed. The presence of an a-pawn allows Black to make a draw after the immediate l%a7xa6. 2) Return with the king to attack the a6-pawn. 3) Then, return with the rook to capture the pawn such chat in the pawn ending, the black king is not in time to reach one of the vital squares c8 or c7. (Defending ag1inst a rook's pawn there is chis additional possibility instead of trying j ust to control the promotion square.) This is how it is done:

A waiting move. After 6.1%d6t ©e5 the white king cannot move to c5.

6 ...ifl 6 . . . 'tt> e 5 7.1%d6 and because of zugzwang, Black will have to allow the white king to reach the c5-square.

7.gd6t 'i:te5 8.'i:tc6 ie2 9.'i:tc5 ib5 8

7

6 5

4

3

2

a

b

c

d

e

f

g

h

Esben Lund - The Secret Life of Bad Bishops

92

IOJ�b6! Not even on b5 does the bishop have a stronghold; White threatens to take it and promote the pawn. Again the bishop is forced out into the open.

10 ...i.e2 l l J�b2! The threat to the bishop.

4. 1 . 1 Fottress Positions

Even with extra pawns for both sides, the player with the bishop sometimes succeeds in building a fortress. Here follows the most imponant fortress positions; in some of the cases the rook side even has the extra pawn.

Ercole del Rio

l l ...i.d3 12J!b3! i.fl l3.ge3t @f4 14.@d4 The black king has been forced further away from the a-pawn.

14...i.bS

8 7 6 5 4 3

1 750

8 7 6 5 4 3 2

1

2

a

b

c

d

e

f

g

a

b

h

c

d

1 5 ... i.a4 l6,ge6 And now the threat to the pawn instead! And with the same effect.

16 ... i.bS l 7,gf6t @gs 1 8,gfll The black king has been chased far enough away for White to turn his attention to the pawn.

18 ... @g6 19.@cs @g7 20.!:!a8 @f7 2 t .@b6 @e7 22.!:!xa6+-

f

g

h

f

g

h

Carlo Cozio

15.geS! Again this well-known threat to the bishop.

e

1 766

8 7 6 5 4 3 2

Black is not in time to reach c8 or c7 after:

22 ... i.xa6 23.@xa6 @d7 24.@b7 White promotes the pawn.

a

b

c

d

e

Chapter 4 - Endgame Section

Yuri Averbakh (473)

93

8

7

8

6

7

5

6

4

5

3

4

2

3

2

a a

b

c

d

e

f

g

h

g

h

Yuri Averbakh (477) 8

7

6 5

4 3

2

a

b

c

d

e

f

These four positions are worth remembering. Especially Averbakh (473) where Black draws as follows. To improve his position, White will have to play b2-b4 at some point. Black exchanges on b4, and because he keeps a pawn on b6 securing the important c5-square, White's king cannot approach the black king in the corner. Here the coordination between the light-squared bishop, king and pawn on b6 is sufficient for a fortress. Let's imagine that White could play b2-b4 before Black had a chance to play . . . a5:

b

c

d

e

f

g

h

Lb4 i.f3

Black can try l . . .ic2!? to stop White from playing a2-a4. After 2.�g2? ia4 3.�g5 a6 Black prevents White from advancing his pawns, and the pawn on b6 and bishop on the a4-e8 diagonal keep the white king off the queenside pawns. This position is a fortress. Please note that the position with the a-pawn on a3 is well-known to endgame theory since the game Olafsson - Larsen, Las Palmas 1 974. Dvoretsky, amongst others, has mentioned the position in Dvoretsky's Endgame Manual. With the pawn on a2 White can chase the black bishop with Wb2-a3, but with precise defence it is nevertheless a draw. It is important for Black to be able to deliver a check on the g8-a2 diagonal once the white king goes to b3 in order to support the pawn advance a2-a4. Instead White should reply to 1 . . .ic2!? with 2.b5. This move also provides an alternative win to the main continuation: White wants to play a4-a5, and since Black is forced to take (otherwise a5-a6) he will get access to the important c5-square. 2 . . ..id l 3.�f7 .ic2 4.�f4 .ig6 5.a4 Wb7 6.�f6 ie4 7.a5! bxa5 8.Wc5 a4 9.�f7t Wb8 1 0.�f4 .ig2 (Black should cover the c6-square since 1 O . . .ic2 1 1 . Wc6+- loses right away.) l l .�xa4 This is not a fortress: a7 is a point to attack and White will play b5-b6 at the right moment; the black king is in the wrong corner.

Esben Lund - The Secret Life of Bad Bishops

94

2.a4 .ie4 3.a5!

fu we just saw, this is not the only path to victory. Also possible was 3.b5 followed by a4-a5 .

3 ... bxa5 3 . . . �d3 4.axb6 axb6 5.'it>c6 gives White's king access to the queenside. 5 ... b5 6.'it>b6 Wc8 7.1%c7t ©d8 (7 ... ©b8 8.1%d7 �f5 9.1%d8t �c8 1 0.1%e8 Zugzwang. 10 ... 'it>a8 l l .1%xc8 mate! This should be a familiar theme by now.) 8.1%c5 White takes on b5 and wins the pawn ending. If instead 3 ... �f3 4.a6! White mates on the back rank.

4.bxa5 a6 4 ...�d3 This is a tougher defence, trying to keep the dark square b6 under control. 5.'it>c6 �e4t 6.Wc5 �d3 A waiting move on the diagonal, claiming that a5-a6 would place the pawn in danger, is refuted by: 6 . . . �f3 7.a6! �e2

b

c

d

e

f

g

h

8.1%b7t! White sacrifices the pawn and plays for mate, since the black king is in the wrong corner. 8 . . . ©a8 (8 . . . 'it>c8 9.1%xa7+-) 9.'it>c6! �xa6 1 0.1%b4 Threatening mate beginning with ©c6-c7. 1 O . . .�fl l l .1%f4 �g2t 1 2. '1t>c7 a6 1 3. ©b6!+-

c

d

e

f

g

h

7.1%g3! ·Wherever Black moves his bishop, he cannot prevent both the threats of a5-a6 and 'it>c6. 7 . . .�e2 8.'it>c6 a6 Forced to avoid mate. 9.'1t>b6 '1t>c8 1 0.1%g7 +We have the study by Enevoldsen in essence. Back to the main continuation after 4 . . . a6:

a a

b

a

b

c

d

e

f

g

h

s.©cs .td.3 6.©b6 ©cs Again we have the study by Enevoldsen. White wins. Let's return to the four fortress positions once again. You might wonder why the position with a bishop's and a rook's pawn on the sixth rank is a draw, as in the first two diagrams, and the position with a knight's pawn is not. The answer can be found in the exercise section,

Chapter 4

-

Endgame Section

but at this point I can reveal that it is important for the defending king to be close to the right corner (opposite to the colour of the bishop) . Another important position that is not a fortress - albeit it looks as one at first glance is the following:

Kholmov 1 973 8

7

6

95

s.©g2 hxg3 6.©xg3 !!d4 7.©h4 ©xf4! 8.g6 (8.©h5 !!d3! 9.©h6 !!h3t i o.©g6 !!g3-+) 8 . . . !!d8-+ and because of the mating threat, White has no time to support his passed pawn. 5 . . . 'it>xf4 Black has sacrificed a pawn (temporarily) but more important is that he gained a passed pawn himself, and that he obtained access to the f4-square, which in essence is the c5-square with colours reversed. 6.�c8 6.'it>g2 !!d2t 7.'it>h3 !!d3t s.©g2 !!g3t 9.'it>h2 Wg4 1 0.�c8 !!h3t l l .'it>g2 !!xh4 1 2.g6 !!h6-+ 6 . . . 'it>g3 7.'it>fl f4 8.h5 f3-+

5

4 3

2

a

b

c

d

e

f

g

h

White to play

In contrast to the diagram above with only two pawns each on the kingside, the presence of f-pawns makes the difference between a draw and a win. But in order to win, Black has to sacrifice two pawns.

l ..ic6 gd2 2. ©gl gd6 Black prepares the breakthrough. This is an ideal square for the rook: from here it keeps an eye on the kingside pawns and forces the bishop close to the corner.

3 ..ib7 g5! 4.fxg5! The best defence according to Averbakh. The alternative was: 4.hxg5 h4 5.gxh4

a

b

c

d

e

f

g

h

4 ... f4 5.gxf4 ©x:f4 Again Black has gained access to the important f4-square.

6.©£2 ©g4 7.ie4 ©xh4 8.g6 gd7 9.©a ©g5 1 0.©g3 White is unable to improve the position further, and can only wait for Black to break through.

10 ... ge7 1 1 .ib l h4t 12.©h3 ge3t 13.©h2 gg3 14.ic2 ©f4 1 5 ..ib l gb3! This is the clearest way for Black to victory. The bishop is trapped and White will have to sacrifice his g-pawn to save it.

96

Esben Lund - The Secret Life of Bad Bishops

l 5 . . . h3!? This is more complicated, but also wins. Averbakh's mainline goes: 1 6.if5 Typical are the following lines: 1 6.g7 E:xg7 l 7.Wxh3 :B:g3t! An important trick ensuring chat the king will not run away from the corner. 1 8.Wh2 ( 1 8.©h4 :B:gl-+ threatens both the bishop and mate on h 1 .) 1 8 . . . '\t>f3-+ and the white king is trapped in the wrong corner. Black wins as in the first diagram of the chapter. 1 6 . . . Wf3 l 7.ic8! 17 .ixh3 :B:xg6 followed by . . . :B:g3 again traps the king in the corner. 17 . . . :B:xg6 1 8 .ib7t ©g4 1 9.ic8t Wh4 20.ixh3 E:c6! In chis line the white king is actually allowed to escape to fl , but the unfortunate separated position of king and bishop secures victory for Black after all. 2 1 .id7 2 1 .if5 E:f6 22.id7 E:f2t 23.Wgl ©g3�+ 2 1 . . .E:c2t 22.©g 1 ©g3 23.'it>fl

27.ie8 E:e7! You cannot hide behind the king! 28.ic6 E:c7 29.ia4 :B:c l t 30.id l E:b l Zugzwang, and mate in 7 according to my engine. Back to the game after the better l 5 . . . E:b3!:

a

b

c

d

e

f

g

h

1 6.g? gg3 17.ic2 gxg7 1 8.i>h3 Otherwise Black plays . . . :B:g3 and brings the king to f2.

1 8 ... gg3t! 1 9.i>xh4 gg2-+ Black wins the bishop due to the mating threat on h2. Already chis study gives an idea of how these positions can suddenly become very sharp, and chat finding the win often means chat you should dare to enter complications. a

b

c

d

e

f

g

h

23 . . . ©f3! There are no bishop checks. 24.©e l ©e3! 25.'it>fl 25.Wd l E:d2t drops the bishop. 25 . . . E:f2t 26.©e l 26.Wgl Wf3-+ traps the king in the corner. 26 . . . E:f7 Black will harass the bishop until he can deliver the deadly check on the first rank.

A very important lesson from chis study is chat when relying on the position as being a fortress, you should be absolutely sure chat it is in fact a fortress. This is why the above handful of fortress positions are important to know by heart, so as to have something to compare with. At chis point I would like to introduce a useful term:

Chapter 4

-

97

Endgame Sec tio n

Mikhail Botvinnik - Coenraad Zuidema

4. 1 .2 Fortress-like position

I shall abbreviate fortress-like position as FLP for short. The definition of an FLP is:

Amsterdam 1 966

Fortress-like position (FLP): A position of rook vs. bishop that has come to a standstill, where the bishop side cannot improve his position further and only the rook side can try to win it. The characteristic of such a position is that it is not yet clear if the position is a fortress or not.

Analysis will show if the FLP is a fortress or not. And here the knowledge of fortress positions comes in handy. If the position is not a fortress then the bishop side is lost. We shall see many FLPs in the pages to come, and some of them actually turn out to be fortresses. This is mainly due to the fact that the bishop side has material compensation in the form of at least one pawn. The following three positions are taken from Averbakh's famous 6-book endgame work:

Grigory Levenfisb - Sergey Freiman

White to play

Emanuel Lasker - Vyacbeslav Ragozin Moscow 1 936 8

7

6 5

Leningrad 1 934

4

8

3

7

2

6

1

5

4

a

b

c

d

e

f

g

h

Black to play

3

The reader can verify for himself that these three positions are in fact winning for the rook side, e.g. in the first diagram (Averbakh 485):

2

a

b

c

d

e

f

White to play

g

h

I .gel! 'i:td7 2.gb l .ic6 3,gb6 i.xa4 4.gb7t 'i:te8 5,ga7 .id? 6.gxaS .ie6

98

Esben Lund - The Secret Life of Bad Bishops is a borderline example, and almost a fortress for Black! White has to work hard to win the game, and in the good old days with adjourned positions, Botvinnik had time to find the winning plan and successfully implement it over the board. 4. 1 .3 Fire on Board

a

b

c

d

e

f

g

h

We shall use our knowledge of FLPs in the following example from Alexei Shirov's Fire on Board, the section with selected endgames.

Yuri Balasbov - Alexei Shirov

After the exchange of a-pawns, Black 1s doomed to passivity. Thus we have an FLP.

Klaipeda 1 988

As mentioned above, this is the kind of position

that Black should a void, unless he is absolutely certain that it is a fortress. In this case it is not - among other things, White can easily pick out at least two weaknesses to attack: d5 and the pawns on the kingside. Meanwhile, Black has no counterplay. This does not mean that the position is easy to win over the board. And even if White has everything under control and has a winning position, there usually comes a point where one has to allow a certain amount of counterplay to transform the advantage to something more concrete. That is not easy either. Having said this, these three positions are not very interesting to analyse from the perspective of this chapter. We realize that they will at some point come to a standstill where the bishop side can only wait - in short, an FLP - and unless it is a fortress, the bishop side is simply lost. The game Botvinnik - Zuidema is the most interesting of the three games: The position has come to a standstill, but Black has an extra pawn and two far-advanced passed pawns that restrict the white king and rook slightly. This

a

b

c

d

e

f

g

h

We enter after White's 5 l .!!b6, threatening the b-pawn. The game continuation was:

5 I ...@c4 Shirov gives this move two question marks.

52.@d2 1-0 Black resigned here because of 52 ... �a2 53 .©c2 �b3t 54.'it>cl �a2 5 5 .!!d6 ©b3 56.!!d8 'it>c4 57.©c2 �b3t 5 8 . © b l and zugzwang (Shirov) . Black loses a pawn after 58 . . . b4 59.axb4 cxb4 60.!!d4t etc.

Chapter 4

-

99

Endgame Section

Instead Shirov gives 5 1 . . .i.c4 in the diagram above as 'quite an easy draw'. It turns out that this is a hasty conclusion, since: 1) There are an equal number of pawns and so far only a few FLPs are in fact fortresses with a limited amount of pawns. The above study by Kholmov was a case in point. 2) At some point the position will come to a standstill and only White can try to win it.

8

7

6 5

4 3

2

a

Shirov's drawing line goes:

5 I ...i.c4 52J�b8 'it>e5 53J�e8t 'it>d5 54.'it>f4 i.a2 55J�dst 'it>c4 56.'it>e5 'it>b3 57J!d2 b4= 8

7

6

b

c

d

e

f

g

h

My plan in this FLP is the following: 1 ) Place the rook on the fourth rank to stop Black's king from going to c4-b3. This stops counterplay against b2 and the pawn push . . . b5-b4. White has achieved the desired standstill. 2) Bring the white king on a long march around the board to the queenside. The king marches along the 8th rank to attack the black b- and c-pawn from behind.

5

4 3

2

The execution of this plan takes a while! a

b

c

d

e

f

g

h

' . . . when the draw is obvious'. After a few pawn exchanges this conclusion of the final position is confirmed by Nalimov's tablebase. However, in this line both sides seemed to be in a hurry to achieve their goals, but in reality this benefits only Black, who would do anything to avoid a standstill. White had no reason to force the black king towards the b2-pawn and should instead have avoided it. Instead of 53.Ei:e8t White can play more strongly:

53JU8 .ia2 54JU4

Meanwhile Black has no counterplay. Here are some sample lines to illustrate the point:

54 .. .i.e6 55J�h4 i.f7 55 . . . ©d5 56.©f4! gets the king on the other side, since 56 . . . 'tt> c 4 is answered with 57.'tt> e 5t winning.

56J!g4 i.c4 57J!g5t 'it>d6 58.'it>e4 i.f7 59.!!gl i.d5t 60.@5 i.e6t 6 l .'it>f6 i.b3 62.!!g5 Stopping any counterplay with . . . 'tt> d 5-c4.

62 ...i.d5 63.'it>g7 i.e6 64.'it>ra i.d7 65.'it>f7 White is slowly but surely approaching the queenside.

1 00

Esben Lund - The Secret Life of Bad Bishops

65 ...J.e6t 65 ... .ic6 66.�g6t! @d7 (66 ... Wd5 67.@e7) 67.�g l with a check on the d-file to follow.

8

7

6

66.@es ih3 67.@ds Black is again in zugzwang and must allow the white king to get closer to the pawns.

5

4 3

67... @c6

2

8

7

a

6

78.c4t

b

c

d

e

f

g

h

White is winning.

5

4

78 ... J.b3 79.c5 @b2 80.c6 a3 8 l .c7 a2 82.c8=Ylli al=Ylli 83.Yllf c3t

3

And wins.

2

a

b

c

d

e

f

g

h

So far so good. The time has come to take concrete action which involves allowing some counterplay.

6sJ;g6t! @ds 69.gg3! With the rook on the third rank, there is a neat tactical point when the black king tries to approach the b2-pawn.

69.. .ie6 70.@c7 @c4 7l .@b6 White is much better placed for Black's counterplay now, and the following lines confirm the win for White:

71 ...J.d7 7 1 . ..Wb3 72.c4t! Wxc4 73.�c3t @d4 74.�xc5+- also wins. Equally hopeless is 72 ... Wxb2 73.Wxb5.

72.gg7 J.e8 73.ge7 J.g6 7 4.ge5 @b3 75.@xc5 @xb2 76.@xbS @xa3 77.ge3 J.c2

Actually, I have not yet seen a fortress with three pawns each (doubled or tripled pawns excluded, of course) . And I am not sure if it is possible to construct one such. The closest was possibly the Kholmov study, but as we saw earlier, Black won with energetic play. Let's turn our attention to more complex examples, armed with the useful knowledge of the FLP. 4. 1.4 Complex endgame examples

In the following three examples the outcome of the battle is extremely hard to predict from just looking at the starting positions. And play will be extremely sharp. These examples show that even with few pieces on the board (or because of few pieces on the board!) the coordination among the remaining army is of crucial importance. This goes for both parties, but especially the bishop side since, as already pointed out, this colour-blind minor piece is dependent on

Chapter 4 - Endgame Section the king and pawns as i t has no access to the other colour complex. And inadequate control of the opposite colour complex can lead to a dangerous FLP. Lev Belov - Anatoly Utiatsky

101

the queenside for the time being, and start pushing his own pawns. Now White instead gets a passed pawn.

2 ... he5 2 . . . dxe5?? 3.c5+-

3.c5 dxc5 4Jhc5 ©f6 5Jha5

Kislovodsk I 960 8

With active play White has won the a-pawn, but still Black has three connected passed pawns.

6

5 ...i.d4 6J:�b5 ©g6 7.a5 f5 8J:�b4

4

8.a6 is equally good - the final result is the same.

7

White forces the black bishop to g 1 .

5

3

2

8 ... i.gl 9.a6 g4 I OJ:�b7 a

b

c

d

e

f

g

h

White to play. What is the verdictfor this position?

The material distribution is rook vs. dark­ squared bishop plus two pawns. The position definitely has not come to a standstill, as Black can get his kingside pawns rolling. In other words, he can fight for light-squared control. White hopes to create a passed pawn on the other side of the board and here the pawn on a5 looks suspiciously vulnerable. White needs to get at this a5-pawn as soon as possible to counter Black's play, and the position quickly becomes sharp.

l .c4! i.d4 1 . . ..if6 2.e5! is similar to the game continuation.

2.e5! Without this move, White would probably be worse, since Black could consolidate

White will win the bishop, but Black builds a fortress with his remaining three pawns.

IO ... f4 l l.a7 i.xa7 1 2Jha7 White has won the bishop, but Black keeps the balance in study-like fashion. 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

1 2 ... ©f6! Black should not allow the rook to attack the pawns from behind: 1 2 . . . e5? 1 3.:B:e? \t>f6 l 4.:B:e8 g3 1 5. \t>f3 Zugzwang - White wins.

1 02

Esben Lund - The Secret Life of Bad Bishops

13,gas

1/2-l/2

©f7!

the queenside - had it been on, say, c4 in this position, the position would be lost for Black.

The white rook is not allowed behind the pawns. Black places his pawns on e5, f4 and g3 - we have a well-known theoretical draw. One can argue that this position could in fact have been an exercise, and all the moves could have been calculated at the beginning of the first diagram. It might be so, but still it shows the sharp nature of these positions. The last two examples in this section are even more complex.

Alexander Alekhine - Ludwig Rellstab Sr Kemeri 1 937 8

7

6 5

4 3

2

a

b

c

d

e

f

g

1 . ©c6! ..

Active play is the only saving chance for Black. This position deserves a diagram, since now White has to make a crucial decision, but we j ust had one, so I will save some ink.

2.@aS?! Another important move was: 2.b5t! Averbakh gives this move an exclamation mark followed by the remark that Black would have had more trouble than in the game to obtain a draw. As we shall see shortly, this is not the case since White actually has a (narrow) path to the win from the diagram position above. 2 . . . d5 3.l%d3t! e4 Black has a couple of worse alternatives: 3 . . .�d4 is answered by 4.l%g3 (Averbakh) . 3 . . . ci>e6?! 4.h4 ( 4.h3!?) 4 . . .�f6 5.l%g3 \ti f7 6.l%h3 Temporary passivity; the a7-pawn will come under fire shortly. 4.l%d7 �xh2 5 .l%xa7 g5 6.b6 Black has more difficulties here than in the game, to obtain a draw (Averbakh) . In fact, the position is lost.

h

Black to play

In this position Black has a dark-squared bishop plus pawn to compensate for White's rook. To avoid a standstill and a dangerous FLP, he builds his hopes on his kingside pawn majority for survival. Under more normal circumstances White would have an almost decisive advantage in such positions, but here the rook is slightly misplaced on h3, guarding the pawn on h2 that is hanging in numerous lines. Likewise, the white king is on the rim of the board on

a

b

c

d

e

f

g

h

6 . . . h4 6 . . . g4 7.l%h7 �b8 8.b5 f4 9.l%h8! shows Black's problem: the b8-h2 diagonal is too short, and the rook and white king dominate

Chapter 4 - Endgame Section the bishop. 9 . . . .id6 I O.'iflc6 .ie5 l 1 .l!e8 f3 1 2.gxf3 t gxf3 1 3.b7+- etc. 7.l!e7t ©d5 7 ... 'it>f4 8.b7 ©g4 9.Wb5 .ib8 I O.l!e8 .if4 I I .l!f8 .ig3 1 2.©c6 .ie5 1 3.l!d8 White blocks the h2-b8 diagonal and wins. 1 3 ... 'it>g3 1 4.l!d6 ©xg2 1 5 .b8='.W+8.Wb5 .if4 9.l!h7 Again Black has problems advancing the pawns without the bishop standing in the way, and White wins.

White opts for a static position (FLP) where Black has no real counterplay, and only then plans to undertake active play. For that purpose the king is needed in the centre and on the kingside. In the pages to come I shall examine a lot of continuations that lead to an FLP, and then see if these FLP's are in fact fortresses or simply lost positions.

� �� � � . , ;B. B B i � � n � · •�• " .% n�� n� � � � � �� � � a �� � �,� "· · ·

� 5 4 6

�

a

a

b

c

d

e

f

g

h

9 ... .ig3 1 0.b7 .ib8 1 0 . . . g4 I 1 .l!xh4 l I .l!h8 .ic7 l 2.l!d8t! Notice chat 1 2.b8='.W? .ixb8 1 3 .l!xb8 Wd4! leads only to a draw for White. 12 ... ©e4 1 3.Wc6 .if4 1 4.l!d6 White queens the pawn. There was another early option for White, though: 2.©b3! This move is part of a totally different plan from the game continuation, and is mentioned neither by Averbakh nor Alekhine. The idea is simply to avoid all pawn races, where the bishop supported by the king will prove almost as strong as the rook when we take into consideration that the white king is too far away on the queenside. It is interesting how the relative value of the pieces can differ depending on the cooperation between chem.

1 03

b

�

c

d

�

e

f

g

h

2 . . . ©d5 3 .l!d3t ©e4 3 . . . ©e6 is too passive: 4.h4 with the same idea as in the line given by Averbakh: 4 . . ..if6 5 .l!g3 'iflf7 6.l!h3 Black's counterplay is under control and White can focus his attention on the a7-pawn and his own b-pawn. 4.l!d7 .ixh2 If instead Black plays 4 ... .id4 5.'it>c4 .igl 6.l!g7 f4 7.l!xg6 .ixh2 8.l!h6 White wins. Please note chat the position after 8 ... f3 9.gxf3t Wxf3 1 0.l!xh5 .ig l is not a fortress, as the black king needs to be back home at b7!

1 04

Esben Lund - The Secret Life of Bad Bishops

5.1%xa7 The best move. Instead White does not win after 5.1%g7? f4 6.1%xg6 f3 7.gxf3t @xf3 , the extra h-pawn securing the draw. 5 ... h4! 5 ... f4 6.@c4 g5 7.1%e7t @f5 8.b5 g4 9.b6 f3 1 0.gxf3 gxf3 l l .@d5 is similar to a line we already looked at; White is winning. 6.b5 f4 7.b6 h3 8.1%a2! 8.gxh3 f3 gives enough counterplay for a draw. 8 ... ig3 9.gxh3 f3 1 0.b7 @e3

2 ... ©d5 3.©a6 After White's mistaken move 2. @a5 , the win has already disappeared! This really shows how sharp these positions are; every tempo counts and one imprecise king move can cost you half a point.

Another option for White was: 3 .1%d3t @e4 4.1%d7

a

b

a

c

d

e

f

g

h

This position is in fact a study: White to move and win! (Lund 20 1 3). The reader can find the study in Chapter 6 as a (difficult) bonus exercise at the end. To summarize the outcome of my move 2.@b3!: the position is won for White, but only after very accurate play. 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

b

c

d

e

f

g

h

4 . . . g5! Black's counterplay comes in time. Besides having wasted a move on @a4-a5, the king is now even further away from the kingside than it was before. And since the white rook is the one picking up the a7-pawn, the white king really went astray on the queenside. 5 .1%xa7 5.h3? is j ust a waste of tempo, since Black's counterplay is based upon the f-pawn. After 5 . . . f4 6.1%xa7 g4 7.hxg4 hxg4 8.b5 f3 Black draws easily. 5 . . . f4 5 . . . g4 leads to the same thing, but White gets an extra option. 6.b5 (6.g3!? is possible, although 6 ... f4 7.gxf4 ixf4 seems to draw fairly easily.) 6 . . . f4 when we reach the game continuation. 6.b5 g4 7.b6 f3 8 .gxf3t gxf3 9.1%e7 The continuation 9.1%f7 ixh2 1 0. b7 @e3 also leads to a draw. 9 . . . f2 1 0.1%xe5t @xe5 l l .b7 fl =� 1 2.b8=�t With a drawn queen endgame.

105

Chapter 4 - Endgame Section O n e really wonders what the white king was doing in all these lines - taking a nap on a5 !??

3 f4! 4.Eia3 ...

4.lt>xa7 is answered with 4 . . .\t>c4!. The move that might pose most problems for Black is 4.b5!? and now:

the b-pawn is the more advanced. White will win Black's bishop, and with his king on the way to defend the kingside, the win is not far off. If Black undertakes nothing, the white king simply goes for the g6-pawn. b) 4 . . . \t>c4! Again this king move is correct.

5

4 3

2

a

b

c

d

e

f

g

h

a) 4 . . .id4? This looks very natural, but in fact loses. 5 .l:!f3 lt>e4 6.h4!

a

b

c

d

e

f

g

b

a

d

c

f

e

g

h

5 .l:!a3 g5 6.l:!a5!? A last trick found by my engine; White wants to collect Black's pawns on the kingside. 6 . . . id4 7.b6 axb6 8.l:!xg5 h4 9.h3 9.l:!g4 h3! 1 0.gxh3 f3 is a draw, one sharp and entertaining line being: 1 1 .l:!f4 f2 1 2.h4 ©d5 1 3.h5 ie3 1 4.l:!f8 lt>e5 1 5 .h6 if4 1 6.h? fl =�t 1 7.lt>a? �a l t 1 8.lt>b? �h l t with perpetual check. 9 . . . @d3 1 0.l:!g4 if2 1 1 .l:!xf4 ©e2

h

6 ... ie3 7.©b7 It is difficult for Black to undertake anything. The only active plan is: 7 ... itif5 8.g3! lt>g4 9.l:!xf4t! lt>xg3 9 ... ixf4 1 0.gxf4 and Black loses the pawn race. 1 O.l:!a4 if2 1 1 . ©c6 Black is lost, since if he wins the h4-pawn, his own on a7 will drop too. And in this case

a

b

c

d

e

f

g

h

Black has counterplay against g2. Again, the white king is doing nothing far, far away on the other side of the board.

Esben Lund - The Secret Life of Bad Bishops

1 06

1 2.lt>b5 ©fl 1 3.1%g4 i.g3 1 4.'it>xb6 'it>xg2 1 5 .Wc5 'it>xh3 With the white king closer, this might have been a dangerous situation for Black with his own king in the corner, but now it's a dead draw.

Vladimir Alatortsev Vitaly Chekhover -

Leningrad 1 9 36 8

7

6

4 ... g5 5.©xa7 g4 6.b5 ©e4

5

4 3

2

a

b

c

d

e

f

g

h

Black to play

a

b

c

d

e

f

g

h

Black has an optimal cooperation between his forces: the king supports the .. .f4-f3 pawn push, and the bishop stops the enemy pawn and can also be used actively on the long diagonal.

7.b6 f.3 8.gxf.3t gxf.3 9J:fa4t ©d3 lOJ!a3t ©e2 I U :fa5 .if4 1 2.l:ta2t ©el 13.l:ial t 1/2-1/2

Please note that it was very convenient for Black that his a7-pawn was on the same colour as his bishop. In that way Black was able to defend it from a distance. It can be very useful to place pawns on the same colour as the bishop. We shall see this in the .coming section on bishop vs. knight endgames, as a way of displaying good technique. The last example is my favourite rook vs. bishop endgame:

The posmon contains a lot of instructive material to show how cooperation between few pieces can and should be obtained.

l ...g5t!! I give this move two exclamation marks; Averbakh was more restrained and gave it only one. The underlying strategic idea is very deep. When I give this position as one to solve, almost everyone chooses: 1 . . . 'it>e6!? After 2.1%e3t 'it>d6 3.1%d3 Black has a draw. Even strong club players fail to understand that it is Black who is playing for a win in this position! Another option is: l ... c4 2.bxc4 bxc4 Aiming to activate the king via the c5-square. 3 .1%c3 3.l%a3?! looks active, but after 3 . . . Wc5! and White has a choice: a) 4.1%xa5t Wd4 It turns out that the a-pawn is less important than bringing the white king into the defence against the far-advanced c-pawn; White is in deep trouble here.

Chapter 4

-

Endgame Section

b) If White instead plays 4.We5 c;f;ib4 he has simply lost a tempo with his 'active' rook move to a3. c) Even worse is the embarrassing 4.c;f;ie3?? Wb4 when the rook is trapped. 3 ... Wc5 4.We3

a

b

c

d

e

f

g

h

4 . . . g5! This move fixes White's pawn on g4. The pawns on the kingside will soon be liquidated, and only pawns on the other side will remain. Black's play on the queenside is likely to come to a standstill, but Black has and extra material to prove that this FLP is a fortress. 5 .hxg5 hxg5 6.a3 �e6 7.:B:c l �xg4 8.:B:gl �e6 9 .:B:xg5t �d5

a

b

c

d

e

f

g

h

This is a fortress for Black: the two extra c-pawns give good protection and shelter to Black's king and bishop, and the c4-pawn threatens to advance, should the white king

1 07

move too far away in an ambitious winning try.

2. hxgS hxgS t

a

b

c

d

e

f

g

h

White now has a difficult choice. Out of three possible moves, only one of them draws.

3.@f5? With this move White is playing for an FLP where he hopes that Black will soon run out of moves and thus be forced to let the white king into e5. It is also bad to play: 3 .Wxg5? c4 4.bxc4 bxc4 5 .:B:e3 ( 5 .:B:a3 We5! 6.:B:xa5 c3 7.:B:c5 Wd4 wins) 5 . . . c;f;ic5 6.c;f;if4 Wd4 The black coordination is ideal: the king is supporting the c4-pawn, at the same time body-checking the white king. The bishop is defending the g8-square from its outpost in the centre of the board, and is even protected by the c6-pawn. This bishop can also be used for attacking purposes if necessary. Meanwhile White has problems with his coordination, as he would love to have his king on the queenside, defending against the march of the c-pawn. This reminds us of the previous game Alekhine - Rellstab where the king was a somewhat superfluous supporter of the passed pawn.

108

Esben Lund - The Secret Life of Bad Bishops In the game there is no salvation. 8

7

6 5

a

b

c

d

e

f

g

h

7.1"i:e l c3 8.1"i:d l t ©c4 9.©e3 c2 1 0.1"i:c l ©c3 The white king is still being shut out by Black's king. l I .g5 ©b2 1 2.©d2 i.e4-+ Defending everything on the b l -h7 diagonal. White can do liccle about the march of the c6-pawn to c3! Bocvinnik lacer gave the right defensive method for White, and backed it up with some variations. White should not try to stop the pawns with the rook alone; for chat purpose the king is needed. Therefore the third option is the right one: 3.©e3! ©e5 (Bocvinnik also gives: 3 ... b4 4.1"i:d l a4 5 .bxa4 ©c7 6.a3 ©b6 7.1"i:b l ©a5 8.axb4t cxb4 9.©d4 ©xa4 1 0.©c5 and White is fine. Now I O ... b3?? is not possible in view of I l .1"i:a l mate!) 4.1"i:c3 c4 5 .bxc4 i.xc4 6.1"i:c2 ©d5 7.1"i:f2 ©c5 8.a3 i.e6 9.©d2! ©c4 I 0.1"i:f6 i.d5 l I .1"i:f5 ©b3 l 2.1"i:xg5 ©xa3

8

�• �• �• 7 �• • • • � � · � .i.�� �� %- �- - �Y.� 3 � �- �- �. 2 , ,;,. �� �- �� �� , /,� �� 6 5

4

a

b

c

d

e

f

g

h

1 3.1"i:xd5! cxd5 1 4.g5 with a drawn queen ending. An impressive piece of analysis by Bocvinnik.

4 3

2

a

b

c

d

e

f

g

h

3 ... a4 3 ... c4? 4.bxc4 bxc4 5 .1"i:e3 'tt> c 5 6.©e5! ©b4 7.a3t and White cannot lose (Averbakh). As we saw earlier, it is unlikely that White can win, even if the g5-pawn is an additional target. The far-advanced c4-pawn secures the f draw for Black.

4.@f6 4.bxa4 c4 5 .1"i:d l b4! is given by Averbakh with no further comments. Since my engine is not immediately impressed by Black's position, I will cake the line a bit further: 6.a5 c3 7.a6 'tt> c7 8.©e5 ©b6 9.©d4 'tt> xa6 I O.'tt> d 3 ( I O.a3 is refuted by 1 0 . . . c2! I l .1"i:e l b3 1 2.©c3 c5 1 3.'tt> b 2 c4 and White has no time to activate his rook with 1 4.1"i:e5 since 1 4 . . . c3t 1 5 .'tt> c l i.f7 I 6.1"i:xg5 b2t 1 7.©xc2 i.a2 wins.) 1 0 . . . 'tt> b 5 l l .a3 c5! 1 2.axb4 cxb4 The presence of g-pawns gives Black good winning chances. Without them, 1 3 .1"i:c l would draw immediately.

4... b4 s.@f5 White is sticking co his waiting strategy, and it seems as if Black is slowly running out of useful moves. Bue Black can continue to improve his position, as the game continuation makes clear. Thus we do not have an FLP.

1 09

Chapter 4 - Endgame Section

Already at this stage, my engine claims a winning advantage for Black, and he is indeed winning quite easily, since he can play his king to b5 and then continue ... c5-c4. In fact I think that 5 . . . ©e7 could even be awarded a '?', since bringing the king to chat side of the board is missing a crucial point: why let two pieces do the same job? Now that is bad cooperation! a

b

c

d

e

f

g

h

5 ©e7?! Averbakh wrongly awards this move an exclamation mark. It turns out that the black king is not needed to defend against White's g-pawn; it is far more useful on the queenside. ••.

The best move is � .cj;; c 7! - which constitutes a far more useful distribution of the remaining forces: the black bishop stops the g-pawn whereas the black king supports the pawns. This is important since the king can help to gain control of the dark squares on the queenside. On the kingside, dark-squared control is not important; only the g8-square needs to be under control. 6.©e5 (6.©xg5 .ixb3! winning, as in the game) . In contrast to the lines beginning with 5 . . . cj;; e7?! Black's king can be improved and take active part in the game: 6 ... ©b6!-+

a

b

c

d

e

f

g

h

a

b

c

d

e

f

g

h

6.©xg5? The correct king move is: 6. cj;; e 5! This seems to draw; the priority of the king is not to be on the kingside . 6 . . . c4 The alternatives are: 6 . . . .ixb3? 7.axb3 a3 8.:gd l c4 9.bxc4 b3 10. cj;; d4 b2 1 1 . cj;; c 3 and White wins. Also bad is 6 . . . ©d7?? trying to get the king back to the queenside: 7.:gxd5t cxd5 8. ©xd5 and White wins the pawn ending. Instead 6 . . . axb3 7.axb3 .ie6 is Black's safest choice. 7.bxc4 .ixc4 8.:gd4 .ixa2 9.:gxb4 .ib3! Drawing. The FLP that arises after 9 . . . .ie6!? 1 0.:gxa4 .id7 also seems to draw.

1 10

Esben Lund - The Secret Life of Bad Bishops 8

7

6 5

4 3

a

b

c

d

e

f

g

h

Black has a counterattack against the g4pawn, and if White tries to free the rook by playing the king to f3, Black will initiate counterplay with the c-pawn, supported by the king. Again this FLP never really did come to a standstill since White had to care too much about defence. Back to the position after 9 . . . �b3!.

2

a

6 ....bb3!!

b

c

d

e

f

g

h

This fantastic blow is the reason why Black could still improve his position.

7.axb3 a3 8.'tti f4 More resistance was offered by 8.:ge3t 'tt> d7 9.'tt> f6!. 8

7

6 4 3

2

b

d

f

g

h

1 0.:gb?t In this position only White has winning chances. The question is whether Black can make a fortress from this FLP. 1 0 ... 'tt> d8 1 1 .'tt> fS c5 1 2.'tt> xgS c4 1 3.c;tif4 a3 1 4.'tt> e 3 a2 1 5 .:ga? c3 1 6.gS c;tie8 1 7.'tt> d 3 c2 1 8 .'tt> d 2 c;tifs 1 9 .g6 c;tigs 20.. c 1 'tt> fs 2 i . b 2 'tti g8 Drawing - White cannot improve his position.

Freeing the path for the g-pawn and intending to transfer the rook to a7 in some lines. Black has to avoid one pitfall here: 9 . . . a2?? 1 0.:ge?t c;tid6 1 1 .:ga? c4 1 2.:gxa2 and White wins. Instead the win is secured with 9 . . . c4 1 0.:ge?t c;tid6 1 1 .gS a2 1 2.:ga? cxb3 l 3.g6 b2 and Black comes first.

Back to the game after the unfortunate 6.c;tixgS?.

Averbakh points out that more easily.

a

c

e

8 ... a2 9.gdl c4 1 0.g5 cxb3 1 1.g6 b2 1 2.g? a l ='*' l 2 . . . c;ti f7

wins

Chapter 4

-

Endgame Section

111

4.2. 1 Zugzwang

8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

Once again one wonders what exaccly the white king is doing on f4!?

13Jhal bxa l =lW 14.gS=lW lWd4t 1 5 .�g3?? allows 1 5 . . . Wg l t, and otherwise the queens will be liquidated by I 5 . . . Wd5t.

Zugzwang describes a situation where the side chat has to move can only worsen his position. Zugzwang positions usually come about in the endgame with few pieces; in lace middlegame positions it is only occasionally seen. In pure bishop vs. knight endgames the zugzwang most often appears when one side makes a waiting move chat places the opponent in zugzwang. Here the difference in nature between the two minor pieces becomes apparent: the bishop can easily make a waiting move and still control the same squares on a diagonal, whereas the knight cannot, as it changes colour with every move. Thus, it is often the knight side chat ends up in zugzwang. Here are two examples to illustrate the difference,

0-1

I chink the reader will agree with m e chat these endgames are extremely sharp. 4.2 Bishop vs. knight

In chis section we shall have a deeper look at pure bishop versus knight endgames. This section will be structured in the same way as the previous one with rook vs. bishop: first examples with few pawns and then we slowly move on to more complex positions. My main idea with chis section is not to give a full description of bishop vs. knight endings. And as in the previous section, there are positions where the winning side has to display good technique in order to win; but to show chis technique is not my ambition in itself only if it in some way helps to understand the difference in nature between the bishop and the knight. This is my main focus here.

a

1 .�cl

b

c

d

e

f

g

h

White to play

This is a position of reciprocal zugzwang. If White were to move, he would lose immediately, as it would allow the black king to escape out of the corner. Bue unfortunately for Black, it is his move. The white king only has the squares c l and c2 if he wishes to keep the black king in the corner to prevent the promotion of the a-pawn. The knight cannot make a waiting move, nor

1 12

Esben Lund - The Secret Life of Bad Bishops

simultaneously control both cl and c2, so the game ends in a draw. Lee's try replacing the knight with a bishop in the next diagram. The square c2 is the focus of attention. To show the superiority of the bishop in this situation, Black even has an additional 5 pawns (!) . The example is taken from a book I wrote for schoolchildren.

Silas Lund 2008

10.©cl if5 Zugzwang.

1 1 .cS='IW hes 1 2.©c2 .if5t 1 3.©cl

� �� � � : �.{;%��.%��.%� � � 5 � '°./. � .� � � : ��.� . . �.� � % % % � � - m 8

l. ..if5 The white king cannot move without allowing the black king to escape from the corner, so he has to make a pawn move.

2.g4 Lg4 3. ©c2 icS! A waiting move.

4.©cl if5! White is in zugzwang again.

5.e6 i.xe6 6. ©c2 icS! Again this waiting move.

7.©cl if5 Zugzwang.

3

2

a

b

c

d

e

f

g

h

1 3 ...ih7! Zugzwang. This is exaccly what the bishop is good at: to make a waiting move on the diagonal (in this case b l -h7) and still control the same squares.

1 4.f5 ix5

Zugzwang. White has to let the black king out of the corner.

Chapter 4 - Endgame Section

1 5.@d2 @b2-+ The a-pawn queens. Zugzwang is a very important cool in these endgames to fight for colour control: the bishop side is fighting for control of the opposite colour complex, and by forcing the knight to move this can be achieved. Without the zugzwang - that is, without being able to pass the move on to the opponent - the bishop would lose a lot of strength compared to the knight! Zugzwang and space are closely connected as the following two examples show.

This move is the first indication of j ust how tricky these bishop vs. knight endings can be. We shall see more as we go through the chapter.

4.@e6! 4.�xf7?? is stalemate.

4... �dst s.@d6 Black is in zugzwang and will have to allow the white king access to the e7-square. Then White will control the promotion square. s

..

. @g7 6.@e7 +-

In the next diagram the pawn is moved one row back.

Averbakh (4)

Averbakh (13)

8

7

8

6

7

5

6

4

5

3

4

2

3

a

b

c

d

e

f

g

h

White to play

Black controls the promotion square d8. White's bishop has no access to this square, but still White wins thanks to zugzwang.

I .i.e4 �f7 2.i.f3 A waiting move.

2 �d8 3.i.dS •••

White dominates the knight - or so it seems.

3 tll f7! •••

1 13

2

a

b

c

d

e

f

g

h

No matter who is to move, the position is a draw. White cannot simulcaneously control all six squares the knight has access to from its blockading outpost on d7: b8, b6, c5, e5, f6 and f8. If White were to move in the diagram, l .�a7 would be close to chis goal, but f8 is still not covered. So, more space for the knight gave it additional squares chat saved the game for Black.

Esben Lund - The Secret Life of Bad Bishops

1 14

We could even add one pawn extra for White and the following position is still a draw.

Averbakh (95)

The other factors are mainly: I ) A rook's pawn.

2) King and pawn space.

7

The pawn structure is of course important as well and can make a crucial difference. But the above two themes are especially important for bishop vs. knight endings.

5

4.2.2. 1 The rook's pawn

3

The . presence of a rook's pawn introduces limitations to both the bishop and the knight; we shall have a brief look at some examples.

8

6 4 2

a

b

c

d

e

f

g

h

Again the knight has too many squares to jump to and no zugzwang will appear. 4.2.2 The other factors

In the following we shall have a look at several other factors that influence the evaluation of the bishop vs. the knight in these endings. If the influence of these other factors is too dominant, one of the minor pieces will appear to be inferior to the other - and the interesting question, of course, is whether the worse position is due to the minor piece itself or the influence of the other factors. In many cases it is difficult to clearly separate these two aspects. I shall dare to try, and in this process I will show some common misconceptions as to what degree the minor piece is to blame for an inferior position, and to what degree the other factors are more important to focus on. The same logic was applied to the treatment of the DEB in the middlegame: often the blame for all evils in a bad position is placed on the (bad) bishop alone.

To illustrate the strength of the bishop over the knight in open positions, Averbakh gives the following example:

Averbakh (172) 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

White to play l .a7

©g4 2. ©£2 ©£4 3. ©e2 ©f5

Black cannot try to win the bishop for the h-pawn: 3 . . . ©g3 4.©e3 @h2 5.i.b7 @gl 6.@d4 h2 7.@c5+- Here White has a direct threat to the knight, but even if White's king were only on d4, he would still win: a rook's pawn on the seventh rank is the only

115

Chapter 4 - Endgame Section case where a knight cannot fight successfully against king and pawn.

4.©e3 ©es s.©d3 ©e6 6.©c3 ©d6 7.©b4 ©d7 8.©bs ©c7 9.©a6+White wins.

The bishop also has a few issues with a rook's pawn. It is well known that bishop plus rook's pawn does not win if the bishop has the opposite colour to the promotion corner square. The next couple of examples show some limitations of both the knight and bishop.

Ernst Holm

9 ... h2 IO.i.g2 Black is in zugzwang and will have to allow the a-pawn to queen. Imagine that it was Black to move in the diagram:

l . .. llic8 This move easily holds the draw, since now Black is able to win the bishop for the h-pawn. A sample line goes:

2.©fl h2 3.©e2 lli a7 4.©e3 llic8 s.i.b7 llia7 6.©d4 ©fl 7.©cS ©gl 8.©b6 llic8t! 9.©c7 llia7 10.©bs llibS I l .i.c6 h l =Wf 12.hhl ©xhl With an easy draw. The knight j ust has to stay in the ring consisting of the squares a7 c8-d6-b5. Black's king is not needed in the defence against the a6-pawn. In short: the real importance of this example is whether the a-pawn is blocked on a6 or a7 - another important factor for the difference between a draw and a win. Thus it has less to do with the difference in strength between the bishop and knight. But this example does show us the long­ range qualities of the bishop. Here it could both defend against the h-pawn and support the a-pawn. Therefore I don't believe that this example gives a true impression of the difference in strength between bishop and knight in an open position. But it does tell us that rook's pawns can be especially troublesome for knights.

1910 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

White to play

The situation looks grim for the first player. The h-pawn is far advanced and there seems to be no reasonable way to stop it.

I .i.gl llJ f3 2.i.h2!! This is the star move. 2. 'tt> e 2? has the same idea in mind, but fails to: 2 . . . lll xg l t 3. 'tt> f2 lll e2! Black can approach with the king and if necessary, play . . . lll f4. With the rook's pawn on the sixth rank, the position is not a fortress.

2 ... lLixh2 3.©e2 ©f6 4.©fl Threatening 5 . 'tt> g 3 eliminating the h-pawn.

4 ... lL!g4t s.©g3! h2 6.©g2=

Esben Lund - The Secret Life of Bad Bishops

1 16

l ... @b5!!

8

7

If Black moves the bishop, for instance l . . .�b6, White responds with 2.lt:lb8! a5 3.lt:lc6 followed by 4.lt:lxa5 and a draw.

5

2.lDxa7t @c5!

6 4

This diagonal distance between king and knight gives a perfect domination.

3

2

3.lll c8 a5 4.lDe7 a4 5.lll f5 a3 6.lDe3 a

b

c

d

e

f

g

h

This is a fortress well worth remembering. The white king simply moves between h 1 and g2, and if the black king approaches, it will be stalemate.

The white knight is able to stop the black pawn, but unfortunately for him the pawn will reac;:h the a2-square.

6 ... a2 7.lDc2 @c4 White loses after the further moves:

s.@h2 @c3 9.lll a l @b2 If the pawn could be stopped on a3 instead (with the knight on a2) , the position would have been drawn.

Bela Sandor - Pal Benko Budapesc 1 949

0-1

8

Because of its short range, the knight is often dominated by other pieces. But let's take a brief excursion and have a look at the opposite case: when the knight dominates.

7

6 5

4.2.2.2 The knight dominates

4 3

2 a

b

c

d

e

f

g

h

White to play

White is two pawns down in this ending, but since the bishop has the opposite colour to the h l -square, White can secure the draw ifhe can sacrifice the knight for Black's a-pawn.

1 .lll c6 1 .Wh2?? loses to l .. .�b8.

Displaying good technique with the bishop often involves cooperation with the other pieces (king and pawns). We already saw it in the previous section when opposed to a rook. It is also valid in bishop vs. knight endgames. Handling the knight is a slightly different matter. Cooperation between the pieces is important, of course, but the knight itself can move in mysterious ways and it can achieve a lot on its own. The self-dependence can probably be explained by the fact that the knight has access to both colour complexes, albeit only one colour every move, and not all the time. One could say that bishops and knights are a bit like dogs and cats in that sense.

Chapter 4

-

Endgame Section

The following example shows how a knight can dominate virtually the whole board from the centre:

1.oltan Vecsey (Yuri Averbakh 1956) 1 934

1 17

only plans with a knight move to b4 or c5 make sense. White will have co abandon the blockade of the cl-pawn in order co win the bishop and push his own b-pawn, which leads co a sharpening of the position. The knight move co c6 thus has a clear plan in mind. A similar idea is to transfer the knight to c5 in order co win the bishop: 4.li:Je6 (or 4.li:J b3) 4 . . . ©h7 The king move played in the main continuation. 5.li:Jc5? {It is best co return the knight with 5.li:Jd4 and follow the main continuation.) 5 . . . d4 6.li::i xa6 d3 In contrast co the main continuation, this is not a win for White. 8

7 a

b

c

d

e

f

g

h

White to play

The first few moves are forced:

l.b4 i.h3 Or I . . .i.f5.

2.b5 i.fl 3.b6 i.a6 At this point we notice that the shortest diagonal including the b7-square is only three squares long (a6-c8) .

4.�c6! This position is actually a kind of zugzwang since the diagonal a6-c8 is coo short for the bishop to make waiting moves. At the same time the black king will have to step carefully not to · lose the cl-pawn to an unfortunate knight fork, as the coming variations will show. White's threat is li:J b4 when the black bishop would be in trouble. Nalimov's tablebase also indicates several other winning moves for White, but co a human,

6 5

4 3

2

a

b

c

d

e

f

g

h

4... @h7 This king move looks strange, but is actually the only one that keeps the pawn for the time being! 4 . . . ©f5 5 . li:J b4 d4 6.li::i xa6! d3 7.li:Jb4 d2 8.li:Jd5! d l =� (8 . . . ©e4 9.li:Jc3t ©d3 1 0. li:J d l © c 2 1 1 . li:J f2 does not help Black either) 9.li:Je3t and wins. 4 . . . © f6 5 . li:J b4 d4 6.li::i xa6 d3 7.li:Jc5 d2 8.li:Je4t again loses the pawn co a knight fork. Or 4 . . . @f7 5 . li:J b4 d4 6.li:Jc6! d3 7.li:Je5t ©e6 8.li:Jxd3 followed by li:J b4.

1 18

Esben Lund - The Secret Life of Bad Bishops

Staying on the g-file is no improvement: 4 ... ©g7 5.lLie7 d4 6 . lLi f5t 4 . . . ©g5 5 . lLi b4 d4 6.lLixa6 d3 7.lLic5 d2 8.lLie4t The black king does not even escape the radius of the almighty knight on all squares of the h-file: 4 ... ©h6 5.lLie7 d4 6 . lLi f5t 4 ... Wh5 is a safe square for now, but White plays 5. i>b8!.

8

7

6

5

4 3

2

a

b

c

d

e

f

g

h

s.©bs! After Black's previous king move, Wh ite has to come up with a new idea to win the game and chis idea is ©b8-a7. With the black king on h7, White cannot capture the bishop on a6 with the knight, so instead he aims to chase away the bishop using the king. The point will follow shortly. a

b

c

d

e

f

g

h

le turns out chat Black is in zugzwang and

chat all five king moves will drop the pawn to a knight fork:

a) 5 ... ©g4 6.lLib4 d4 7.lLixa6 d3 8.lLic5 d2 9.lLie4 d i =� 1 0. lLi f2t b) 5 ... i>h4 6.lLie7 d4 7.lLif5t c) 5 ... i>g6 6.i>a7 .ic8 7.lLie7t d) 5 ... i>g5 6.lLib4 d4 7.lLixa6 d3 8.lLic5 d2 9.lLie4t e) 5 ... i>h6 6.lLie7 d4 7.lLif5t The way the knight controls the black king is called 'building a barrier' and chis barrier consists of squares chat are controlled both directly and indirectly (through tactics).

5 ... ©h8 The only move. White wins after both 5 . . . i>g7 6.lLie7 d4 7.lLif5t and 5 . . . ©h6 6.lLie7 d4 7.lLif5t. Going to g6 or g8 loses the pawn to an immediate knight check on e7.

6.©a7 .ic8 7.�e7 d4 8.�xc8! d3 9.b7 d2 l0.b8=Wf d l =Wf 8

7

6

5

4 3

2

1

a

b

c

d

e

f

g

h

Chapter 4

-

1 19

Endgame Section

The point of White's play was to capture the bishop on c8, and not a6. From c8, the knight is closer to Black's king. Amazingly enough, this position is won for White, mainly because of the unprotected black queen on d l .

1 1 .tll e7t @g7 1 1 . . .c;t;>h? leads to mate in two: 1 2.�g8t c;t;>h6 1 3.�g6#

12.�gst @f6 13.tll d 5t! @es 1 3 . . . cj;>f5 l 4.lll e3t

14.Wfg7t @d6 White also wins after 14 . . . c;t;>e4 l 5.lll c 3t, or 14 . . . c;t;>e6 1 5 .�e7t cj;>f5 1 6.lll e3t or 1 4 . . . c;t;>xd5 1 5 .�d7t.

cooperation among the remaining army is not optimal. King and pawns both influence the positions. An active king is always an asset, and since the board is 8x8 and the kings cannot come into direct contact, when they approach each other one side will always have the advantage of having more king space laterally or vertically. When I use the term 'king space' , this is not identical to 'king activity'. This will be apparent in the examples to come. Here are three positions from Averbakh's 6-volume endgame work: Averbakh (278)

Kazimierz Plater Mikhail BoMnnik -

1 5.Wfe7t @c6 1 6.Wfc7t+Black loses the queen: l 6 . . . c;t;>b5 l 7. lll c 3t or l 6 . . . c;t;>xd5 l 7.�d7t

Moscow 1 947 8

We now return to the other factor that influences the evaluation of pure bishop vs. knight endgames.

7

4.2.2.3 King and pawn space

5

Until now I hope to have given a brief overview of the difference in nature between the bishop and knight. Still, these 'pure' bishop vs. knight endings are not entirely pure, since the king is always there to support either side. And pawns on either side will also have an influence on the evaluation and outcome of a position. So, compared to the pure bishop vs. knight endgame, the king and pawns constitute the other factors in the position. Pawn space is an asset that has to be taken into consideration as another factor. More space usually means that the pawns are closer to promotion and that the opponent has less space. I say 'usually' mainly because too-far­ advanced pawns can also be vulnerable if the

6 4 3

2 a

b

c

d

e

f

White to play

g

h

1 20

Esben Lund - The Secret Life of Bad Bishops

Gosta Stoltz - Isaac Kashdan

pawns and the inferior pawn structure overall, does not make his chances of survival better. We shall see the continuation on page 1 22.

The Hague 1 928 8

In the second diagram Black is to move, and because of chis he will reach the d5-square:

6

l. .. ©ffi 2.©fl ©e7 3.©e2 ©d6 4.©d3 ©ds

7

5

4 3

2

1

5

a

b

c

d

e

f

g

h

Black to play

Vitaly Chekhover - Emanuel Lasker Moscow 1 93 5

4 3

2

1

a

b

c

d

e

f

g

h

Here Black holds the advantage because he has more king space. If White were to move in the starting position, he would bring the king to c4 when the game would most likely be drawn. Let's see a few more moves to get an idea of the problems that lie ahead for White:

5.h4 ic8! Black wishes to drive the king on d3 away with a check from a6.

6.ll:)f3 ia6t

In all three positions, the player who plays for a win also has an advantage in king space, or will shortly obtain an advantage in king space. In the first diagram, Black's king is already very active and ready to attack White's vulnerable queenside. The face chat White has doubled

Such a check chat forces the king to choose a side is displaying good technique. Now White has an unpalatable choice: If he goes co e3, the black king can enter the position via the queenside (the b4-square) whereas the game continuation leaves the kingside vulnerable; the black king has access to the e4-square.

7.©c3 h6 s.ll:)d4 g6 9.lDc2 ©e4 10.lDe3 f5

121

Chapter 4 - Endgame Section 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

a

b

c

d

e

f

g

h

White is in trouble and Black went on to wm :

2 1 . .. b5!

l 1 .©d2 f4 1 2.tll g4 h5 13.liJf6t ©f5 14.liJd? ic8 1 5.tll ffi g5 1 6.g3 gxh4 17.gxh4 ©g4 18.liJg6 if5 19.liJe? ie6 20.b4 ©xh4 2 1 .©d3 ©g4 22.©e4 h4 23.liJc6 if5t 24.©d5 f3 25.b5 h3 26.liJxa? h2 0-1

22.©el ib2 23.a4 bxa4 24.bxa4 ©c6 25. ©d2 ©c5 26.liJc3

White will lose a pawn on the queenside.

Or 26.c2 id4! 27.f3 c4 and the a4-pawn is doomed within the next couple of moves.

26 ... ©b4 27.liJb5 a5 28.liJd6 White tries a desperate counterattack.

In the third diagram Black holds the advantage and White is to move.

28 ... ©xa4 29.©c2 29.li:Jxf7 b3 and the a-pawn is a winner.

Averbakh points out that with the pawn on a4 White could play 2 1 .li:J c l c6 22.li:Jd3 when he blocks the queenside and holds the draw. But since 2 1 . liJ c l is met with 2 1 .. .ib2 picking up the a3-pawn, this is not possible. Thus Black uses his greater king space and breaks through on the queenside. 2 1 .a4 is too slow: 2 1 . . .c6 22. li:J c l c5 23.li:Jd3t Black can play 23 ... d4 and break through. Instead, the game continuation was:

2 1 .©fl

29 ... ie5 30.liJxf7 ixh2 Black has a winning position. Thus, chis game was not very interesting and in fact did not tell us much about the difference between bishop and knight; what counted the most here was the difference in activity of the kings.

3 1 .liJdS e5 32.tll c6 igl 33.f.3 ic5 34.liJbS ©b5 35.g4 ie7 36.g5 fxg5 37.liJd? id6 38.liJf6 ©c4 0-1 My point arising from these three positions is chat they do not really show us the difference in nature between bishop and knight, simply

1 22

Esben Lund - The Secret Life of Bad Bishops

because the other factors were too dominating: A major factor was the king's position, but in the first diagram White furthermore had a vulnerable pawn structure with doubled pawns. These three positions are not unique in the sense of king space; I found many positions in Averbakh's endgame work where one side having an advantage also enjoyed more king space. Such positions can be useful if one wants to be familiar with the technique of winning a won endgame, but in the context of my purpose of exploring the difference between bishop and knight they are less interesting.

Kazimierz Plater Mikhail Botvinnik -

Moscow 1 947 8

7 6 5

4 3

2 a

In the next section we shall have a closer look at important tools to display good technique in bishop vs. knight endings. 4.2.3 Good technique

I will consider three methods of displaying good technique: 1 ) Pawns on the same colour as the bishop. 2) Bishop check that forces the king to choose a side. 3) Counterplay is allowed.

d

e

f

g

h

l .�d4 The counterattack l .f5 exf5 2.lll f4 �e8 3.tll d5 is answered with 3 . . . b5! 4 .tll e3 'tt> e 5 (Averbakh) .

l . ..g6 2.g3 e 5 3.fxe5t fxe5 4.�c2 i.e4 5.�el 'tt> c5 The black king breaks through on the queenside.

6.@e3 i.f5 7.�f3 @b4 8.�d2 i.c2 9.@f3 i.xb3 l 0.@e4 i.xc4 l l.@xe5 i.d3

7

Pawns and bishop on the same colour

4

This is done for defensive purposes only, and it can be quite effective. Once again we shall return to the first diagram position from above: Averbakh (278)

c

White to play

The first two points concern the bishop. The third point is useful especially for the knight's side in these endings. We also note that points 1 and 3 were covered in the previous section on rook vs. bishop. 4.2.3. l

b

8 6 5

3

2 a

b

c

d

e

f

g

h

Chapter 4 - Endgame Section

1 23

Black defends the pawn on g6 from a distance. In this case it is to his advantage that the kingside pawns are on the same colour squares as his bishop, since it is on the queenside that Black is doing the attacking. He just wishes to keep the kingside under control and here the bishop does a great job from afar. In this sense the theme is not solely for defensive purposes, since the solidity of defence gives greater freedom in the attack.

draw is obvious; White has to be careful not to allow that.

1 2.@d4 if5 13.llic4 b5 14.llid2 a4 15.i>dS ih3 0-1

2 . . .'it>f6 3.'it>f4 is zugzwang. Black will have to allow the white king to the e5-square. If the knight moves, the white a-pawn is free to move forward.

4.2.3.2

Bishop check that forces the king to choose a side

We briefly touched upon this subject earlier, but let's see another example of this useful endgame technique.

City of Liverpool - City of Glasgow

The position is won for White; this is just mentioned in fairness to Black's knight.

Lie2 i>e6 2.ic4t! Here it is: the bishop check that forces the black king to choose a side.

2 ... i>d6

3.i>f5 llid5 Black starts a desperate counterattack.

4.i>g6 llie3 s.ie2 c4 6.i>xh6 c3 7.id3 From d3, the bishop stops the c-pawn and also supports the passed a-pawn.

7 ... llig4t 7 . . . c2 8.�xc2 lli xc2 9.Wg6 lli d4 I O.'it>f6! Again this domination between king and knight. 1 0 . . . lli e6 l l .a6! wins.

Correspondence I 905-06 8

7

The bishop's side has to be ready for a transformation of the position as well, in order to win it. The sacrifice of the bishop is a typical theme that uses the fact that the knight is short-ranged. Transformation of an advantageous position is the theme of the next section.

6 5

4 3

2

s.i>g5 lli e5 9.ic2 lli c4 a

b

c

d

e

f

g

h

White to play

We notice that White has king space and that three rook's pawns are present. If Black can sacrifice his knight for White's a-pawn, the

More stubborn was 9 . . . 'it>c5. Here White has to find the move 1 0.�a4! which cuts off the black king from the a5-pawn . After 1 0 . . . tt.J f7t ( 1 0 . . . Wb4 l l .a6 'it>xa4 1 2 .a7 c2 1 3 .a8=1Mft+­ promotes with check.) l l .Wf4 lli h6 White has to show patience and win slowly by eliminating the annoying c3-pawn: 1 2.'it>e4!+- With the

Esben Lund - The Secret Life of Bad Bishops

1 24

c-pawn gone, White can decisively reactivate his bishop.

1 . . .i.d l 2.�d5t! ci>b3 3.f6 i.h5 4.�f4 8

10.a6

7

Black resigned. 4.2.3.3 Counterplay is allowed

Even in positions where the knight side has full control, there usually comes a point where concrete action is needed. The bishop will awake and the position comes alive; perhaps a pawn race will take place and accurate calculation is needed. In order to win such positions, the knight's side often has to take up the challenge and psychologically one has to get out of the comfort zone.

6 5

4 3

2

1·

a

b

c

d

e

f

g

h

4... ci> xb2!

7

We notice that the shortest diagonal including f7 is h5-e8 (four squares). Since knight and king can control two squares each on a diagonal - four in total - this diagonal is too short: 4 . . .�e8 5.'it>f8! The white king covers e8 and f7, the knight g6 and h 5. 5 . . .�b 5 6.'it>e7 �c4 7.CUe6 White queens the pawn and wins.

5

Instead, Black goes for active counterplay as the white knight is far away from the a-pawn.

3

5.�xh5 a4 6.�f4!

G. Fedotov 1 9 54 8

6 4 2

a

b

c

d

e

f

g

h

White to play

This study shows some useful features of bishop vs. knight endings when the position becomes sharp.

1.tll f4 Black's bishop has to stop the passed f-pawn on f7, but it is now dominated by the white knight which controls both d5 and h5.

6.f7? a3 7.f8 ='.W a2 and although my engine claims a '+ 1 1 ' advantage, this endgame is drawn. We know that queen vs. rook's pawn on the seventh rank is usually a draw, and it turns out that the presence of a knight on h 5 does not make a difference. Talk about the disadvantages of a short-ranged piece!

6 ... a3 7.�d3t ci>c3 8.�b4! This is a trick worth remembering: White stops the pawn and by placing his knight en prise, he lures the king to a mined square, when the promotion of the f-pawn follows with check.

Chapter 4 - Endgame Section An alternative win is: 8.f7! a2 9.f8 =� a l =� 1 0.�f6t with a deadly skewer.

1 25

7.lDd7 gl 8.d8 ig6 9.e7 if5 10.ll:)cs ic8! White threatened l 1 .lll e 6.

8 ... xb4 9.f7 a2 1 0.f'S=�t Check! White wins as Black is not in time to gain control of the promotion square:

l l.ll:)d7 hl 1 2.dS ia6 13.c? ib5 14.ll:)eS ie8!

10 ... b3 l l .�f6 c2 1 2.�al +-

1/2-1/2

I would like to follow up on this game with a short intermezzo:

Back to positions that suddenly become sharp; we end this section with a complex example from the hand of the author.

And so on.

Averbakh (47)

Nikolai Korolev Silas Lund -

8

Taascrup 2002

7

8

6

7

5

6

4

5

3

4

2

3

a

b

c

d

e

f

g

h

White to play

Black has to stop the d-pawn on d7. Two diagonals emanate from this square: a4-e8 and h3-c8. The shorter is five squares long (i.e. more than the four the king and knight can control) , which means that Black can draw this position without the help of the king.

l. e6 ib5 2. e7 ic6 3. d8 ib5 4. c7 gl A waiting move, but that is the only help the bishop gets!

5.lDd3 hl 6.lDeS ie8! White threatened to block the diagonal with 7.tt:lc6.

2

a

b

c

d

e

f

g

h

Black to play

With Black I have been trying for a while to find a breakthrough into White's position. Black has a clear advantage, and it might be won with correct play, but to this day I have not been able to find a win. Black has the better pawn structure and a lot of king space, since his king has reached White's fourth rank. Still, it's not so easy to make progress from here. White has all the pawns on the same colour as the bishop, and this might be convenient as they can then be easily defended. Again we see that it is not always

1 26

Esben Lund - The Secret Life of Bad Bishops

such a bad idea to have the pawns on the same colour as the bishop. In the game I tried the breakthrough:

52 ... lll f5 53.@f"2

57 . . . lll c2 A very critical position:

8 7 6 5

2 a

a

b

c

d

e

f

g

h

53 @d3 .••

Now the position becomes very sharp with opposite pawn races where every tempo counts.

[email protected] White starts a counterattack against g6.

54 lll e3 55.@f4 � dSt .••

After this move White has an easier task making a draw than after the alternatives. The most testing move was definitely: 55 . . . lll c4! This has the idea of a capture on a3. I missed this idea over the board. 56.'ifig5 lll xa3! 57.'ifixg6! It is amazing how relative the value of pieces becomes in very sharp positions. Black sacrificed the knight on a3, but White doesn't stop to take it; he prefers to create a passed pawn as quickly as possible. 57.ixa3? 'ifixc3 58.'it>xg6 b4 59.ic l 'it>c2 60.ig5 b3 6 1 .if6 a3 62.'it>xh5 b2 etc. loses in simple fashion.

b

c

d

e

f

g

h

58.id6!! Other bishop moves lose, as will soon be apparent. Or 5 8.g4? hxg4 59.h5 g3 60.id6 g2 6 l .ih2 and Black's a-pawn is faster than the h-pawn: 6 1 . . .a3 62.h6 a2 63.h7 a l ='IW 64.h8='1W g l ='IWt 65.ixg l '.Wxg l t-+ 5 8 . . . a3 White's point is seen in the variation 58 . . . b4 59.cxb4 a3 60.b5 a2 6 1 .b6 (6 1 .ie5 lll d4-+) 6 1 . .. a l ='IW 62.b7 and there is no queen check on a6 with the bishop on d6! Instead Black can try the sneaky 6 1 . . . lll d4. Here White has drawing resources too: 62.b7 lll c6 63.b8='1W! lll xb8 64.ie5

a

b

c

d

e

f

g

h

This final phase of the game is very sharp. White takes on h5 and has two connected pawns, but they are not far advanced

1 27

Chapter 4 - Endgame Section compared to Black's pawn on al. Black wishes to block the bishop's diagonal with a knight j ump to b2 in order to promote the pawn. With correct play the game ends in a draw: 64 . . . li:l d7 65.�a l lll c5 66.©xh5 li:l b3 67.�b2 wc2 68.�f6 li:l d2 69.©g6! li:l c4 70.h5 li:l b2 7 1 .h6 a l =W 72.h7 The promotion cannot be prevented. 72 ... \Wa3 73.h8=W \Wxg3t draws. After 58 . . . a3 Black wins the bishop immediately: 59.�xa3 li:l xa3 60.Wxh5 ©xc3

I had succeeded in finding an extremely narrow path to the draw beginning with the move 6 1 .Wg5!? (the g6-square is poisoned since Black then queens with check), but once you spot 6 I .g4! you see how easily the game is drawn. Thanks to Christian Eriksson for testing my rigid thinking. Candidate moves! Back to the game after 55 . . . li:l d5t: 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

The knight usually has an easier task facing a g-pawn than a rook's pawn. For this reason I mainly considered king moves in this position to get the h-pawn going. However, this is not always the case; from this game alone I can extract lines where the only drawing move is either with the g-pawn or the h-pawn. One of my pupils showed the right approach in this actual position: 6 l .g4! li:l c4 6 1 . . .b4 62.g5 li:l b5 63.g6 li:l c7 64.g7 li:l e8 65.Wg6 leads to the same result. 62.g5 li:l d6 62 . . . li:l e5 63.g6 li:l xg6 64.Wxg6 b4 65.h5 b3 66.h6 b2 67.h7 b l =Wt 68.Wg7draws. With a rook's pawn on the seventh rank the position is drawn. 63.g6 b4 64.g7 li:l e8 65.©g6 li:lxg7 66.©xg7 b3 67.h5 b2 68.h6 b I =W 69.h7 Draw.

a

b

c

d

e

f

g

h

56.@gs lDxb4 This is a draw right away. Slightly more testing was: 56 . . . li:l xd 57.©xg6 li:l d5 5 8.�d6 b4

a

b

c

d

e

f

g

h

Here White has to find the only move that draws. 59 .�xb4! Yet another minor piece sacrifice.

1 28

Esben Lund - The Secret Life of B ad Bishops

59 . . . li:Jxb4 60.Wxh5 li:Jc2 6 1 .�g6! Bad is 6 l .g4? lll xa3 62.g5 lll b 5 63.g6 li:J d6! 64.g? lll e 8! Tricky knight. 65.g8 =1Mf lll f6t winning. Black also wins after: 65.�g6 instead of promotion - 65 . . . lll xg? 66.�xg? a3 67.h5 a2 68.h6 a l ='Mft With the h-pawn only on the 6th rank, Black wins. 6 1 . . .lll xa3 62.h5 lll c4 63.h6 In chis position there is no mined square chat the black knight can sacrifice itself on. 63 ... a3 64.h? lll e 5t 65.�g? a2 66.h8=W a l =W 67.Wh?t Draw. 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

57.cxb4 @c3 58.@xg6 @b3 59.@xh5 @xa3 60.g4 @xb4 6 1 .gS a3 62.g6 a2 63.g? a l='ilY 64.g8='ilY 'ilYeSt 65.'ilYgS Draw agreed.

l/z-1/z

4.2.4 Worst-case scenarios Already in the first chapter of chis book I introduced the idea of a worst-case scenario, and we have discussed it further since. The idea of the WCS is to see what happens if all the additional pieces are removed from the board. This leaves the bishop side with only the king and pawns to fight for control of the squares of the opposite colour to the bishop.

Thinking in terms of WCS is a useful tool when deciding upon which pieces to exchange, or if trying to avoid a specific piece exchange. Additional pieces are one of the other major factors in complex middlegame positions and are, for instance, the reason why many exchange sacrifices work in a given position: there are ocher pieces left to cooperate with the minor piece and thus compensate for the material loss of the sacrifice. I believe chat many players fear to sacrifice the exchange (even if the sacrifice is sound) because of chis abstract way of chinking: Oh no, if the remaining pieces go off the board then I am j ust going to lose the endgame! This 'What-if' should be a useful tool, not a ghost chat scares you and thereby guides you wrongly. My advice: use the idea of the WCS and then back it up with concrete lines to show if any exchanges unfavourable to you are about to cake place in the near future. Lee's return to the position from the Introduction; in the analysis to the game Kramnik - Gelfand we had a rook vs. bishop endgame. I will now remove the light-squared bishops: 8

7

6 5

4 3

2 1

a

b

c

d

e

f

g

h

This is of course to Black's advantage, as the bishop is now completely isolated.

1 29

Chapter 4 - Endgame Section

37.i.cS ga2t 38.'i!;>gl 'i!;>f7 39.b6 gb2 We have an FLP that is very unlikely to be a fortress. Black can bring his king over c6b5-c4 and to the kingside, where he will find additional targets, or he can push the h-pawn to create more weaknesses. After this short deviation from our path, we shall return to positions with bishop vs. knight. Here we already saw several positions where the bishop side made it to a draw even though it looked grim. The last example was the game fragment Korolev - Lund where I have been unable to find a win for Black. Let's have a look at a further two positions from Averbakh's six-volume endgame work:

Yuri Averbakh Vassily Panov -

Moscow 1 9 50

Other than j ust comparing the knight on e4 with Black's bishop we notice the following other factors: 1 ) White can create a passed pawn on the kingside. 2) White will break through to the f5-square and will therefore be able to obtain king space. 3) The pawn on d5 already constitutes pawn space for White. This means that Black's bishop has less room to manoeuvre. 4) d6 is a concrete target to attack.

64.gS 'i!;>g7 6S.'i!;>f3 'i!;>f7 66.'i!;>g4 ie7 67.5 i.f'8 68.�f6 h6 69.gxh6 hh6 70.�e4 i.f'8 7 1 .h6 i.xh6 After 7 1 . . .�g8 the simplest way for White is: 72.g6 ie7 73.h?t h8 74. f7 ii.dB 75.li:lxd6 xh7 76.li:le4+- etc.

72.llJxd6t 'i!;>e7 73.�e4 ie3 74.d6t 'i!;>d7 75.'i!;>xeS A straightforward win that does little justice to the bishop compared with the knight.

1-0

Nikolai Zubarev - Alexandrov Moscow 1 9 1 5

a

b

c

d

e

f

g

h

White to play

Averbakh's position is without the b-pawns and the black king on g8 instead of h8. I am not sure if Averbakh's memory or the ChessBase Mega Database is wrong - I have chosen the position from the latter. In either case the winning method is the same.

Esben Lund - The Secret Life of Bad Bishops

1 30

i.@f'2 @e7 2.@e3 @ds 3.@d4 @c7 4.@cs White has conquered king space. Besides king space we notice the other factors relevant to the position: Pawn space - White's e-pawn on the fifth rank. Concrete targets on a6 and e6.

4.. .i.c8 5.�b4 ib7 6.g3 ic8 7.�d3 id? 8.�f4! The mobile knight creates more weaknesses on the kingside. White can combine the threats and has a won position as Averbakh shows.

1-0

4.2.4.l Creative Chess Strategy One game fragment chat I saw about six years ago really impressed me. It is taken from the interesting book Creative Chess Strategy by Alfonso Romero Holmes.

refers to the observation chat in order to transform the advantage into something more tangible and to win the game, one often has to use creative and alternative means as the straightforward try often doesn't prove enough to win the game. This example from the hand of the future world champion is a case in point. We enter the game at its most critical point.

1 8.icS!! Romero Holmes awards this move two exclaqlation marks. Instead of this excellent move White could have opted for an endgame with knight vs. Black's bishop by exchanging on b6 and further liquidation of the rooks on the d-file. Would Black's DEB not turn into a bad bishop then, after the remaining pieces to support it have departed? Romero Holmes considers the following worst-case scenario (without using this term - instead he calls it a 'target position'):

Tigran Petrosian - Anatoly Bannik USSR Championship, Riga 1 9 58 8

7

6 5

a

4 3

2 a

b

c

d

e

f

g

h

Position after l 7 . . . Ei:cd8.

Romero Holmes spends eight pages in the introduction on exactly chis game. The tide

b

c

d

e

f

g

h

Romero Holmes writes: "As Petrosian remarks, it would be very complicated to convert the theoretical advantage into a full point." We have seen several cases of these WCS chat look good for the side with the knight but are only drawn with the correct defence. One can say chat Black's DEB in the above example is a bad bishop by now but if it is not enough to win the game, then who cares which name we attach co it?

Chapter 4 - Endgame Section

1 8 .. Jhdl t 1 9.B:xdl Lc5 20.c!Dxc5 B:e8 2 1 .lll e4 With the absence of bishops the pawn on f6 is suddenly under attack. White possesses the d-file and has space and good squares for the knight on the kingside. Petrosian went on to win the game confidently on move 6 1 . I can only recommend the study of chis game with the comments by GM Romero Holmes. The last example shows the stage before the worst-case scenario. In the following game of mine, the rooks were still present on the board. I was not sure if the position could be won after the exchange of rooks, so instead I used this additional piece to create havoc in White's position - before the exchange:

131

This little dance with the rook has made White's f-pawn a target. 8

7

6 5

4 3

2 a

b

c

d

e

f

g

4 1 .lll b2 B:xa3t 42.@xa3 i.xf'3

h

This position is winning, but it is instructive to see the final breakthrough.

Efraim Laksman Silas Lund -

Copenhagen 2006

43.@b3 c5 44.@c2 e4 45.dxe4 Le4t 46.@d2 c4 47.c!Ddl f3 48.h4 id3!

8

The final preparation for the breakthrough.

7

49.g5

6

49.h5 does not make a fundamental difference. 49 . . . c3t 50. © c l c2 5 1 .lll f2 We3 52.lll h3 i.fl 53.h6 (53.Wxc2 does not work either: 53 . . .i.xh3 54.h6 i.xg4 5 5 . h? i.f5t followed by . . .i.xh7) 53 . . .i.d3! 54.g5 i.f5 winning.

5

4 3

2 a

b

c

d

e

f

g

h

Position after 38.lll xa4.

38 ... B:g8! Black can exchange all the remaining pieces: 38 .. .i'!a8 39.Wb3 i'!xa4 40.i'!xa4t i.xa4t 4 l .Wxa4 Wxd3 42.g4 but the pawn ending is not winning. a

39.g4 B:a8! 40.@b3 f4

b

c

d

e

f

g

h

White is in zugzwang and will either lose

1 32

Esben Lund - The Secret Life of B ad Bishops

Boris Grachev - Igor Lysyj

his kingside pawns or allow one of Black's pawns to queen. The bishop is perfect on f5 : it supports the c-pawn and stops White's pawns. 5 5.W b2 Wd2 or 5 5 . tt:l g l f2. The game went along similar lines.

49 ... c3t! 50.@cl 50.tt:lxd fl 5 I .tt:le2t We5 52.tt:lg3 Wf4-+ and the pawn queens.

50 ... c2 5 1 .liJfl @e3 52.liJg4t @f4 53.lll fl i.5 White resigned without waiting for 54.Wd2 Wg3.

Taganrog 20 1 1 8

7

6 5

4 3

2

0-1

The next section is devoted to the bishop pair, or in other words: bishop vs. knight with an additional bishop on either side.

4.2.5 The bishop pair vs. bishop and knight fu

a contrast to the worst-case scenario, the first half of the book has dealt with many pieces on the board and thereby many possible transformations of the position. In this section I will investigate a case with few additional pieces and therefore fewer transformations. It is generally considered advantageous to possess the bishop pair, and my examples will all confirm this. The reason for this is not that the bishop pair is good by definition, but the insight that the bishop usually protects one colour complex better than the knight. And with two bishops working together, they can work wonders. These considerations are based on the assumption that the other factors of the position are not in the favour of one of the players. I wish to show how the possibilities to play for a win increases with additional pieces (actually one piece) to support the DEB. The following examples are therefore an extension of the pure bishop vs. knight insights.

a

b

c

d

e

f

g

h

Position after 24 . . . tt:l xe?.

25.�f3 With j ust one additional piece added to the DEB, the number of possible transformations increases dramatically compared to the pure bishop vs. knight. In the above diagram the possible transformations are: 1 ) The exchange of the light-squared bishop for the knight leaves same-coloured bishops. 2) The exchange of the dark-squared bishop for the knight leaves opposite-coloured bishops. 3) The exchange of bishops would give a pure knight vs. bishop ending. Also possible is of course a pure pawn ending; or a minor piece sacrifice for pawns - but this leads to a different material imbalance. The likeliness that such a piece sacrifice will work increases with the number of pieces left on the board. The problem for Black in the diagram is that there is no recipe for how to place the pawns - since all of the above transformations are possible.

Chapter 4 - Endgame Section

25 ... b6 26.@e2 I must mention at this point that White enjoys the advantage of an important "other factor" : the more active king. However, the next example will show that it is easier for the side with the bishop pair to conquer king space.

1 33

8

7

6 5

4 3

26 ... i.d6 27.hS @fS 28.g4 White takes space on the kingside. With the present pawn formation on the kingside, White should be aware of the light-squared bishop as a possible DEB.

28 ...@e8 29.i.e4 @d7 30.f4 f6 3 l .@d3 @e6 32.i.f3 i.c5 33.i.dl @d6 34.i.b3 b5 35.@e4

2

a

b

c

d

e

f

g

h

Black is in a kind of zugzwang now and has to let White take over the diagonal a3-f8.

39 ...i.b6

39 . . . id6 40.c4! No knight j ump to d6 here! 40 . . . bxc4 4 1 .ixc4 and g4-g5 is already a threat. 39 . . . iid8 40.c4! lll c8 4 1 .cxb5 ! A little combination. 4 1 . . .lll d 6t 42.iid5 With the king on d8 the knight is no longer protected. 42 . . . ib4 43.a3 is good for White.

40.i.a3 White seizes the long diagonal and takes d6 under control. a

b

c

d

e

f

g

h

A common theme when facing a bishop and a knight is that the king can more easily move on the colour complex opposite of the bishop. I have mentioned this before: a bishop is better than the knight to protect one colour complex (it is the only thing it can do!) . So even if White had a slight advantage of king space at the beginning of the game fragment, the side with the bishop pair will often gain more king space in any case with his two powerful bishops.

35 ... aS 36.i.d2 i.b6 37.i.f7 @d7 38.i.cl i.c5 39.i.b2

40.c4?! is less effective now due to 40 . . . lll c8!. This is why Black did not place his bishop on the d6-square.

40 ...i.c? 40 . . . lll c8 4 1 .ig6 lll e7 (4 1 .. . lll d 6t 42.iid5) 42.if5t e8 43.ie6 and White has made progress.

41 .i.cS i.d6 Instead 4 1 . . .lll c8 42.ig6! (aiming for f5) 42 ... lll e? 43.if5 t ! followed by g4-g5. Notice that exchanging on f5 gives Black a hopeless same-coloured bishop ending mainly because

1 34

Esben Lund - The Secret Life of Bad Bishops

of the static pawns placed on dark squares: f6g7-h6. 8

7

6 5

grab this space at the beginning of the game fragment. Black can try the desperate: 43 . . . f5t 44.gxf5 li.J g8 4 5 .ie5 w f7 But White wins by focusing on the weak queenside; a possible continuation is: 46. Wd5 li.J e?t 47.lt>c5 li.Jxf5 48.lt>xb5 g6 49. hxg6t Wxg6 50.c4 h5 5 l .c5 h4 52.c6 h3 53.c7 li.J e7 54.f5t Wxf5 5 5 .ih2

4 3

2 a

b

c

d

e

f

g

h

42.ieSt! @c7 43.id4 Black resigned here as he can do little against the coming g4-g5 when White will obtain a deadly passed pawn.

1-0

Let's instead go back to the diagram position and see what happens if Black takes the bishop on e8: 42 . . . Wxe8 43.ixd6

a

b

c

d

e

f

g

h

White threatens to exchange on e7 to enter a won pawn ending. Notice how most transformations of the position are to White's advantage due to the fact that he has more pawn and king space. And it was due to the bishop pair that he was able to

a

b

c

d

e

f

g

h

This position is won for White, but because of the a-pawn and wrong-coloured bishop compared to the a8-corner, Black can fight on: 55 ... We6 56.Wxa5 Wd7 57.Wb6 Wc8 58.a4 li.Jd5t 59.Wc6 li.J b4t 60.Wc5 li.J c6 6 1 .Wb6 li.J e7 62.a5 li.Jd5t 63.Wc6 li.J b4t 64.Wc5 li.J a6t 65.Wb6 li.J b4 66.a6 li.Jd5t 67.Wc6 li.J b4t 68.lt>b5 li.Jd5

135

Chapter 4 - Endgame Section

Alexander Grischuk - Michael Roiz Ningbo 20 1 1 8

7

6 a

b

c

d

e

f

g

h

This is a position well-known to endgame theory. White transfers his bishop to a5 and king to a6 when Black will succumb to zugzwang and ultimately allow the white king to b7. 72 . . . tfJ c7 73.if4 tiJ a8 74.id2 tfJ c7 75 .id tiJa8 76.ia5 tfJc7 77.Wb6 tiJ a8 t 78.Wa6 Zugzwang. 78 . . . ©d7 79.©b7+Next is another recent example from grandmaster practice. Please notice an important method to improve the position for the bishop pair side, which we also saw in the previous example: by grabbing space, the pawns become positioned on the same colour as the DEB. We already devoted a section to this theme earlier in the chapter and it works well as a defensive method: by placing them on the same colour as the DEB, they can easily be defended. By doing this, the side with the bishop pair runs the risk of ending up with a bad bishop vs. knight ending, should the additional bishops be removed. But only with careless play does the bishop pair side allow the exchange of the additional bishop. Let's follow this method of improvement more closely in the next game.

5

4 3

2

a

b

c

d

e

f

g

h

Position after 25 . . .ixd7.

26.f.3 c!De6 27.©fl ic6 28.©e3 id5 For now the white king is not able to move forward.

29.a4 f5 30.g3 ©fl 3 l .b4 a6 32.ie5 g6 33.g4 ib3 34.gxf5 gxf5 35.a5 Black's problem is that if he had tried earlier to take space on the queenside, this space would most likely become a liability, creating weak pawns that White's bishops could attack. We saw the same in the previous game: the side with the bishop pair can more safely grab pawn space. Please notice that White places his queenside pawns on b4 and a5, the same colour as the DEB.

35.. .J.d5 36.id3 ©g6 37.h4 With the idea of h4-h5 to create an imbalance in the pawn structure.

37... h5 38.f4! With his last two moves Grischuk has placed his remaining pawns on the same colour as the DEB.

Esben Lund - The Secret Life of Bad Bishops

1 36

38...ic6 39.@d2 ib5 40.ic2 Of course White should wait with the exchange of bishops. The extra piece on the board gives additional transformation possibilities. Let's consider the exchange of bishops with 40.ixb5? axb5 .

b

a

c

d

e

f

g

We have come to a critical point in the game. White shows a great willingness to transform the position.

43 .. .i.f'7 Black is not saved by: 43 . . . tt:l xe5 44.fxe5 ic6 4 5 .'tt> d 4! 'tt> g7 45 . . . f4 46.ie4t! ixe4 47.'tt> xe4 f3 48.'tt> x f3 'tt> f5 49.e6 'tt> xe6 50.'tt> e4! 'tt> f6 5 J .'ifif4+- A deadly opposition. 46.'ific5! f4 46 . . .ixd5 47.'tt> x d5 @f7 48.e6t 'ifie7 4Q.'tt> e 5+- etc. 4 7 .ixc6 bxc6

8 7 6

h

It is clear now that White's DEB on e5 has been completely isolated, and if anyone can play for a win here it is Black, who can start by bringing his king to d5. This is only to remind us how fragile the DEB can be: at first a strong part of the bishop pair, and after the exchange of one pair of bishops - an isolated DEB, a bad bishop.

40 ... �d8 4I.i.b3 i.e8 42.@c3 � c6 43.idS! 8

7

6 5

4

�

4 3 2 a

b

c

d

e

f

g

h

48.'it>d4!+White changes plan and goes for the f-pawn, with a similar win as already shown: 48 . . . 'tt> g6 49.'ifie4 f3 50.'tt> xf3 'tt>f5 5 l .e6 'tt> xe6 52.'tt> e 4! 'tt> d 6 53.'ifid4! 53.'ifif5 ? 'tt>d 5 and the resulting pawn race is not a win. 53 . . . 'ifid? 54.'it>e5 'tt> e7 5 5 .'tt> f5 +Now White aims for the h 5-pawn and wins easily.

44.iO b7 and h5 are the points to attack in Black's camp, so the bishop is ideally placed on f3 .

3

2

44... ie6 a

b

c

d

e

f

g

h

Chapter 4 - Endgame Section 8

7

6 5

4 3

2

a

b

c

d

e

f

g

h

45 .ic7! 1-0 .

The bishop move protects a5 and prepares b4-b 5. Actually, White's threat is 46.b5 axb5 47.a6! when the knight on c6 is loose. Roiz resigned the game here, although he could still put up some resistance with 45 . . . CiJ a7 (threatening . . . CiJ b 5 t ) . But after 46.�d2 CiJ b 5 47.�e5 �c8 (b7 had to be defended.) 48.�d3 White's king will invade via c4-c5-b6. The last example showed us that additional pieces on the board (in this example, one additional bishop besides the DEB) yield extra possibilities and transformations and make it more difficult for the defender to simplify the position or to build a fortress. As we saw in this last example (on move 40) Black was only one bishop exchange away from obtaining an advantageous knight vs. bishop ending.

4.2.6 Chapter recap 4.2.6.1 Rook vs. bishop We learned that the pure rook vs. bishop ending is usually a draw. Exceptions are if the king is trapped in the wrong corner (the same colour as the bishop) or if the king and bishop are separated under unfavourable conditions. From the study by Jens Enevoldsen ( 1 949) we also learned the typical winning method for

1 37

positions with one pawn each with no passed pawns. By combining threats to the enemy's bishop (or pawn) the coordinated effort of king and rook forces the black king as far away as needed. Then the rook returns to capture the pawn ensuring that the remaining king and pawn vs. king ending is won. In fact there seem to be no fortress positions in endgame theory with three pawns each (doubled and tripled pawns excluded) . The Kholmov study was probably the closest. We found a win for White in the position from Fire on Board. We used this knowledge and moved on to the FLP (Fortress-like position) . Such a position is characterized by the absence of counterplay for the bishop side. Analysis still needs to show if it is in fact a fortress. It is advisable that the bishop side is absolutely sure that it is a fortress when he enters an FLP, otherwise it is simply lost. Well, in theory - the rook side still has to show good technique to win the position, and good technique also means the willingness to allow a certain amount of counterplay on the way to transforming the advantage. We learned a handful of useful fortresses. These can be used as a comparison if there is doubt as to whether the FLP is in fact a fortress. In the second half of this section we turned our attention to more complex examples. Special focus here was on the cooperation between the few remaining pieces, but we also used the knowledge about FLPs. The positions were very sharp and entertaining! We also discussed the usefulness of placing own pawns on the same colour as the bishop. This was done mainly for defensive purposes, on the side of the board where the bishop side was only intending to defend.

1 38

Esben Lund - The Secret Life of Bad Bishops

4.2.6.2 Bishop vs. knight We saw that usually zugzwang appears for the knight side due to the nature of this piece; it changes square-colour with every move it makes. The bishop can easily make a waiting move and maintain control of the squares on a diagonal. On the other hand, zugzwang can appear for the bishop side as well: this happens if the diagonal, from where the bishop tries to stop a far-advanced pawn, is too short and therefore leaves no room for waiting moves. We also saw how a knight can dominate the whole board from a central position in the study by Averbakh/Vecsey. In this way the knight makes a 'barrier' (Dvoretsky) ; not only are the squares that the knight protects inaccessible to the opponent, but so also are squares that are indirectly mined. Thus a wall of squares can appear to which the. opponent has no access. We also saw how the knight could be very tricky in the critical phase with pawn races. In Korolev - Lund it displayed an amazing repertoire of tricks reminiscent of the Averbakh/Vecsey study. Special attention should be given to the knight move to a mined square, which is a sacrifice simply to gain time for the pawn race. We discussed the other factors of the position: � Rook's pawn; � King and pawn space.

There were several limitations to both bishop and knight with the presence of rook's pawns. With king space I meant the fact that one of the kings will always be able to take up more space than the opponent's; that is, the king with more space is closer to the opponent's first rank and therefore closer to the promotion squares of his own pawns. Taking these other factors into account diminished the superiority of the bishop over the knight. I showed that many positions from Averbakh were falsely said to have a decisive advantage due to the bishop's superiority over the knight, when it was in fact the other factors that played a crucial role. In this way I hope to have contributed to the mystery as to why these two minor pieces, which are so different in nature, have approximately the same value. When more complex examples were introduced we discussed what constitutes good technique in these positions: I ) Bishop check that forces the king to choose sides. 2) Pawns on the same colour as the bishop. 3) Counterpla y is allowed.

Chapter 4 - Endgame Section

1 39

The bishop check is a good technical cool to know if the bishop side has king space. By checking the opponent's king away to one side, the attacking king is enabled to advance further on the other side. Once again we saw the advantage of placing the pawns on the same colour as the bishop if done for defensive purposes. Especially in the critical phase of the game a certain amount of counterplay has to be allowed when trying to realize an advantage. We learnt something about the relative value of the pieces: Many sacrifices of the minor piece were made to gain time for the pawn race. In general the king also has a high value in these endgames with only one minor piece each. The king can protect both dark and light squares at the same time, which .is more than either of the minor pieces can do. This is one of the reasons why king space is a very important factor. We had a closer look at worst-case scenarios. Many bishop vs. knight endgames are in fact theoretically drawn for the bishop side in positions chat look grim, although in a practical game the winning chances have to be estimated more highly. A case in point is the game Korolev Lund, where Black also had a considerable advantage in both pawn and king space. We moved on to consider the stage before the worst-case scenario, with extra pieces on the board. The example from Creative Chess Strategy showed in an instructive manner why the seemingly favourable transformation into a pure bishop vs. knight ending was not the best winning try. I concluded the chapter with a couple of examples with the bishop pair vs. bishop and knight. A certain method for improvement can be observed in these positions: Because the bishop is better than the knight at controlling a one-colour complex, the side possessing the bishop pair is able to gain king space. The bishop pair side can place his pawns on the same colour as his unique bishop (making it a DEB) thus forcing the opponent to place his pawns on the same colour as his bishop. In chis case the side with the bishop pair should be wary of exchanging his ocher bishop. On the ocher hand, ocher transformations are to his advantage: same-coloured bishop endings are advantageous due to the pawn structure, and many bishop vs. knight positions are as well, if the attacking side was able to grab more space during the phase when he possessed both bishops.

Chapter 5 Exercises Over the following pages the reader will have a chance to challenge his or her understanding of the themes presented in this book. It is my strong recommendation that you read through the earlier chapters first before turning to the exercises in this chapter. The concept of the DEB and other themes may be new to you, so you should get accustomed to them first. For every exercise there are one or two questions. From some of the questions I pose, it may be possible to derive which theme of the book you should pay special attention to. In other exercises this is not the case - the clue to the exercise might j ust be ' Black to move and win' . Besides specific questions, the reader will find my recommended time consumption for solving the exercise. I urge you to take the trouble to set up the position on a chessboard and set the suggested time accordingly on a timer. Often the time recommendation is closely connected to the level of difficulty. Every exercise is given a level from 1 to 5, with 5 being the most challenging. The exercises are ordered in increasing order of difficulty and, as I mentioned, are independent of a specific theme. The level is based upon the demands regarding calculation, judgement and spotting creative or surprising ideas. When I ask you to evaluate a position in an exercise, it means that you should include both words and calculations to the extent that you find it relevant. My point is that I do not want to influence your choice or thinking process; instead the ideal situation is to consider the exercise as just another over-the-board position. If I want a clear answer to a question, I will have formulated the question accordingly. The next chapter gives solutions, and there will be no doubt how the exercise is correctly solved. If, for instance, an exchange sacrifice seems the natural first choice for you in an exercise, and you believe this to be the intended solution, you probably can't j ust go on this idea alone. In the solution I will pinpoint the one or two critical lines to give the idea proper support. That is, you cannot just make your decision based solely on ideas. In a practical game you are not responsible towards anyone else, and this is where the over-the-board situation differs from the solving of

1 42

Esben Lund - The Secret Life of Bad Bishops

these exercises: Here I indicate which critical lines you should consider in order to claim that you solved the exercises! Many exercises are extracts from my own games. This makes a lot of sense, since I know my own games very well and chis is a guarantee for thoroughly worked-through exercises. But otherwise I have tried to include as many modern grandmaster games as possible. In any case, a single exercise usually presents one or two important themes, and a game between two amateurs can thus easily lay the foundation of a Level 5 exercise chat could challenge a grandmaster. I wish you a lot of joy with these 30 exercises, and hopefully you will experience the feeling that they are making you a better chess player. If you wish, you can use the Guiding questions I presented at the beginning of Chapter 3 as a helping tool.

Chapter 5

Kramnik - Anand, Wijk aan Zee 2007

Q) s

-

Exercises

1 43

Gelfand - Karjakin, Monaco (rapid) 20 1 1 8

7

7

6

6

4

4

2

2

5

3

a

b

c

d

e

f

g

h

White to play and win. Please display good technique!

White to play. What is your evaluation of this position?

Recommended time: 5 minutes. Level: 1

Recommended time: 8 minutes. Level: 2

Avrukh - Guseinov, Gothenburg 2005

� �� �-��! / �,d , /�r;; � 6 �al; �f� - - f� �� s I- - ---%� ti%� 4 ,.�8j.�.�� 3 � -tzJ � 2 �.�. ,.r�

@�

1

"//�

a

�� �� �.� ...

b

c

d

e

f

g

h

Radjabov - Bacrot, Khamy-Mansiysk 20 1 1

7

6 4 3

2

a

b

c

d

e

f

g

h

Black to play. In the game Black grabbed a pawn with 36 l!Jxa4. Why was this a mistake?

How do you evaluate this endgame? Build your reasoning on the insights from Chapter 4 .

Recommended time: 8 minutes. Level: 2

Recommended time: 8 minutes. Level: 2

...

Esben Lund - The Secret Life of Bad Bishops

1 44

R. Akesson - Lund, Klaksvik 2008

Lavrik - Lund, Vladimir 2007 8

7

6 5

4

a

b

c

d

e

f

g

h

White to play. I had prepared chis position with Black. Try to explain the reasoning behind Black's strategy. How do you estimate Black's chances?

a

b

c

d

e

f

g

h

White to play. 1 ) Please list some strategic considerations about Black's bishop on f6. 2) What is the evaluation of the position?

Recommended time: 10 minutes. Level: 3

Recommended time: 8 minutes. Level: 2

Averbakh ( 1 1 9)

Johannesson - Lund, Reykjavik 2007

6 4 3

2 a

b

c

d

e

f

g

h

White to play. This position is a draw without the extra pawn on d3. Give the reason why as well as the win for White here.

Recommended time: 8 minutes. Level: 2

a

b

c

d

e

f

White to play and draw.

Recommended time: 1 0 minutes. Level: 3

g

h

Chapter 5 - Exercises

Dominguez - Topalov, Sofia 2009

1 45

Komev - Morozevich, Taganrog 20 1 1

®

a

b

c

d

e

f

g

h

Black to play. He sacrificed an exchange:

24 Jk7!? 25.i.b6W!fb8 (25 . . . l:!xc2 26.W!f a5! +-) 26.i.xc7 W!fxc7 What is the reasoning?

a

b

c

d

e

f

g

h

Black to play. Find the strongest continuation.

.•

Recommended time: 1 0 minutes. Level: 3

Recommended time: 1 0 minutes. Level: 3

Lund - J. Akesson, Sweden 2009

White to play. Find the strongest continuation.

Recommended time: 1 0 minutes. Level: 3

Trifunovic - Pirc, Birmingham 1 9 5 1

White to play. How does he proceed from here?

Recommended time: 1 0 minutes. Level: 3

Esben Lund - The Secret Life of Bad Bishops

1 46

Volokitin - Moiseenko, Kiev 20 1 1

Sozio - Alatortsev, Moscow 1 93 1

7

6 5

4 3

... a

b

c

d

e

f

g

h

2

1

a

b

c

d

e

f

g

h

Black to play. This is a fairly typical Sveshnikov endgame. Should Black fear the exchange of rooks on the d-file?

White has just played the move 44J�e2-el with the idea :gh l -h7 to attack the b7-pawn. How should Black react?

Recommended time: 1 0 minutes. Level: 3

Recommended time: 1 0 minutes. Level: 3

Horwitz & Kling, 1 8 5 1

Sam Loyd, 1 860

4 3

2

a

b

c

d

e

f

g

h

White to play and win. Show that a knight's pawn on the sixth rank does not enable a fortress to be set up by the defender.

Recommended time: 1 0 minutes. Level: 3

Recommended time: 1 0 minutes. Level: 3

Chapter 5

Henri Rinck, 1 923

-

Exercises

1 47

Mikkelsen - Semcesen, Horsens 20 1 3

®8 7

6 4 3

2 a

b

c

d

e

f

g

h

White to play and draw.

Recommended time: 1 0 minutes. Level: 3

a

b

c

d

e

f

g

h

White played 2 1.be? �he? 22.ll:)c4 with the idea of securing the knight on c4 and playing against Black's g7-bishop. Please evaluate this plan. Recommended time: 1 0 minutes. Level: 3

S. Bekker-Jensen - Glud, Horsens 20 1 3

Genrikh Kasparian, 1 9 58

@8

7

7

6

6

5

4

4

2

2

3

a

b

c

d

e

f

White to play and draw.

Recommended time: 1 0 minutes. Level: 3

g

h

1

a

b

c

d

e

f

g

h

What do you think of White's plan 26.f4 i.h6 27 ..ixf6 to play against Black's DEB on g6? Explain your evaluation ofthe position. Recommended time: 1 0 minutes. Level: 3

Esben Lund - The Secret Life of Bad Bishops

1 48

Dominguez - Polgar, Khamy-Mansiysk 20 1 1

a

b

c

d

e

f

g

Grischuk - Morozevich, Khamy-Mansiysk 20 1 1

a

h

Position after 78 J:�b2. White to play and draw. •.

Recommended time: 1 0 minutes. Level: 3

b

c

d

e

f

g

h

Morozevich played the non-standard move 14 ...i.bS, which involves a pawn sacrifice. Try to explain the reasoning behind this move. How do you evaluate the resulting position? Recommended time: 1 5 minutes. Level: 4

Najer - Vol.k.ov, Taganrog 20 1 1

Ponkratov - Morozevich, Taganrog 20 1 1 8

7

i. �

�

6 4 3

2 a

b

c

d

e

f

g

h

Black to play here unleashed the exchange sacrifice 2 1 ..Jhf3. What is Black's j ustifica­ tion? What is the evaluation thereafter?

Recommended time: 1 5 minutes. Level: 4

a

b

c

d

e

f

g

h

Black to play. White has temporarily sacrificed a pawn for attacking chances. How should Black react to White's latest move: 3 1 .fxg6?

Recommended time: 1 5 minutes. Level: 4

Chapter 5 - Exercises

Paaske Lund, Copenhagen 2003 -

1 49

Cheparinov - Melkumyan, Helsingor 20 1 3

6 4 3

2

a

b

c

d

e

f

g

h

Black to play. White intends to play 24.tll a S attacking both the bishop and the pawn on b7. How should Black react to this?

-

A.

b

c

d

e

f

g

h

White to play. Explain why 26 .ixdS! is a strong move. •

Recommended time: 1 5 minutes. Level: 5

Recommended time: 1 5 minutes. Level: 4

Lund

a

Smith, Borup 2009

Gashimov - Ponomariov, Khamy-Man. 20 1 1 8

7

6

5

4 3

2

a

b

c

d

e

f

Black to play -find the win.

Recommended time: 1 5 minutes. Level: 4

g

h

a

b

c

d

e

f

g

h

Black to play. With the white king one step further away from the corner, Black is winning. Explain why and support it with some lines.

Recommended time: 20 minutes. Level: 5

Esben Lund - The Secret Life of Bad Bishops

1 50

Kanli - Lund, Rogaska Slatina 20 1 1

Silas Lund, 20 1 3

@ 4 3

2 a

b

c

d

e

f

g

h

Black has a theoretically won position, but it is not easy to prove in a practical game. Show a winning plan for Black. Recommended time: 20 minutes. Level: 5

.1

-=='----�����....__.

a

b

c

d

e

f

White to play and win.

Recommended time: 20 minutes. Level: 5

g

h

Chapter 6 Solutions 1. Vladimir Kramnik - Viswanathan Anand Wijk

aan

c

d

Zee 2007

8

7

6 5

4 3

2

a

b

e

f

g

h

53 .icSt! .

Good technique - the bishop check forces Black to choose a side. The white monarch will move to the opposite flank.

1-0

Anand resigned at this point. The reason is: 53 ... �f7 54.�e5 and White will soon break the blockade on b5. Otherwise, 53 . . . �d? 54.�f6 picks up the h-pawn. Note that it is convenient for White that his h-pawn simply stays on h2 - it is easily protected by the bishop from a distance. We also noticed from the beginning that h8 is the same colour as the bishop. Therefore Black cannot sacrifice his knight on b5 (if allowed) and then hide the king in the h8-corner. But if White grabs space on the kingside the following scenario can become reality if White is not careful: 53.h4 �f7 54.h5 �e7 5 5 .i.c5t �f7 56.h6?! li::l c3 57.�e5? (Pushing the h-pawn was a dubious plan, but this throws away the win. White could still play for a win with for instance 57.i.d4, only now he cannot allow Black to sacrifice his knight for the b-pawn.) 57 . . . li:J a2!

Esben Lund - The Secret Life of Bad Bishops

1 52

2. Boris Avrukh - Gadir Guseinov

58.b5 lll c 3 59.b6 lll a4! 60.i.d4 (The point of Black's manoeuvre was 60.b? lll xc5 6 1 .b8=� lll d 7t winning the queen.) 60 . . . lll xb6 6 1 .i.xb6 Wg8=

Gothenburg 2005 8

7

6 5

4 3

a

b

c

d

e

f

g

h

This position is a fortress, another limitation with a rook's pawn. If White tries to approach the h7-pawn, the black king is stalemated.

2

a

b

c

d

e

f

g

h

If you wondered what this exercise was about, the answer is that even with reduced material, mating patterns occur:

36 c!Dxa4? 37.i.c6! .•.

Forcing the black rook to abandon the back rank.

37 .. J�a6 38.idS Threatening mate on g8.

38 ... hS 39.tll h4! ©h7 40.c!Dg6 Black loses material if he wants to prevent the mating attack by White's rook, bishop and knight. Therefore he resigned. csr In positions with opposite-coloured bishops (here both DEBs) one often has to be on guard for an attack against the king. The cooperation of these three pieces can be very effective.

Chapter 6 - Solutions

3. Boris Gelfand - Sergey Karjakin Monaco (rapid) 20 1 1

1 53

29 id7 30J�e5 '!Wd6 3 1 .g5e4 •••

8

7

6 5

4 3

2

a a

b

c

d

e

f

g

h

At first glance it is easy to point fingers at Black's bishop on c8, but this DEB does a great job defending the numerous weaknesses in his camp. Meanwhile the remaining two rooks and queen can make White's life uneasy by attacking the pawns on b4 and d4. So because of the remaining pieces on the board, all talk of a bad black bishop is premature. The position is simply unclear. One sample line goes: 26.fi:c l id? 27.tLlc5 ic8 In this way Black can keep both a6 and c6 defended.

26.h3 id7 27.ll:)cs ic8 28.@hl '!Wb8 Attacking b4.

29.lDd3 29.fi:b 1 ? fi:xa5 benefits from the pin along the b-file.

b

c

d

e

f

g

h

After Black's next move the position opens up and becomes sharp. Both players play well in this rapid game until an unfortunate blunder by Karjakin on move 39 that loses the game on the spot.

31 f5!? 32.g4e2 gxd4 •.•

The weak pawn on d4 drops.

33.liJcS gbxb4 34.ll:)xe6 In order to break through Black's defences, White will have to exchange the DEB another point in favour of the view that the bishop is not a bad piece.

34 gbc4 35.'!Wb2 ixe6 36Jhe6 '!Wf4 37.'!Wb7 ge4 38.g6xe4 fxe4 39.'!Wxa6 '!Wxf'2?? •.•

Black must have noticed that the rook on c4 can now be captured with check and therefore resigned before Gelfand could actually make the move.

1-0

Esben Lund - The Secret Life of Bad Bishops

1 54

4. Teimour Radjabov - Etienne Bacrot Khamy-Mansiysk 20 1 1 8

7

The game continuation speaks for itself: Black has no counterplay since W'hite dominates the light squares and can easily improve his position. Eventually, the black position succumbs.

30 ... @g6 3 1 .ges hS 32.@e2 @f5 33.g3 g6 34.@f3 gS 35.g4t hxg4t 36.hxg4t @f6 37.ggs ie7 38.@e4 @e6 39,ghs icS 40.gh6t f6

6 5

4

8

3

7

2

a

b

c

d

e

f

g

h

In this rook vs. bishop endgame we have what I called an FLP (fortress-like position) in Chapter 4. It is not among the fortresses known to endgame theory (which usually contain a maximum of two pawns) , so for a start we simply note that it is not so easy for W'hite to make progress, since Black can easily protect his weaknesses on c7 and the kingside to begin with. On the other hand, on the basis of the knowledge presented in Chapter 4, the suspicion is that this position is lost since the likelihood that this FLP actually turns into a fortress is low with so many pawns remaining on the board.

6 5

4 3

2

a

b

c

d

e

f

g

h

4 Lf4 gxf4 42.gS ie7 43.@xf4 id6t 44.@g4 ieS 45.@hs id4 46.g6 f5 47.gh? f4 48.gf7 1-0 Th e answer to the question is that Black's position is lost already when entering the rook vs. bishop endgame.

Chapter 6 - Solutions

5. Ralf A.kesson Silas Lund

23.h3 gd4 24.i.g6 @f8?!

-

More solid is: 24 . . .�d6 2 5.�f3! �d7 (25 . . . b6? 26.�a8t is bad for Black) 26.b3 lll h7!? The idea is to transfer the knight to f8 to secure both the eighth rank and the light squares around the king. (26 .. . @f8?! 27.ie4! b6 28 .ig6! The threat is again � a8t and after 28 . . . �c8::!:: Black is slightly tied up.) Now 27.ixh7t is slightly better for White, but nothing really alarming for Black, while something like 27.@h2 lll f8 28.ie4 b6 seems okay for Black.

Klaksvik 2008 8

7

6 5

4 3

2

a

b

c

d

e

f

1 55

g

h

6

Black's concept in the opening was extremely ambitious: ultimately I was hoping for an ending where I could prove that White's DEB on c2 would be more a problem than a strength. To achieve as many exchanges out of the opening as possible I even accepted an inferior pawn structure. Fortunately Black's weaknesses are fairly easy to cover, and I have a stronghold on d4; the plan is . . . E!:d4, . . . �d6, . . . b6 and an exchange of rooks and/or queens if possible. Objectively speaking Black is slightly worse as he has more visible weaknesses than White, including the worse king position. White should keep the additional pieces - queen and rook - on the board for the time being, as otherwise the black king would feel safer. A good plan (which my grandmaster opponent did not play in the game) would have been the space-grabbing h4-h5 to pinpoint the light­ squared weaknesses around the black king. So my opening concept was too ambitious.

8 7 6 5 4 3 2

a

b

c

d

e

f

g

h

I like the idea of temporarily declaring the knight on f8 a defensive piece. Later on this knight can come alive and meanwhile Black can play with his queen and rook. Throughout the book we have seen the DEB declared a defensive piece, but the difference here is of course that the knight does not have the inherent risk of becoming a bad piece in the same way as a DEB has.

25.b3 @e7?! Very risky and I cannot remember what went through my mind when going for this plan. White is better but the game was eventually drawn in 67 moves.

Esben Lund - The Secret Life of Bad Bishops

1 56

6. Averbakh (119)

. a a

b

c

d

e

f

g

h

Liie7! The move order is important: l .i.a7? Wd8!= is only a draw. If you saw the zugzwang pattern, but failed to acknowledge the importance of the move order, the exercise was not solved.

l. .. ltlb8 2. .ia7! lll c6t 3.iie8 lll e 5 4.d4! ltld7 s.iie7+-

b

c

d

e

f

g

h

Black is in zugzwang as the pawn on d4 protects the e5-square. If White does not have the extra d-pawn and instead places the bishop on d4 to cover e5, the b8-square is available to the knight. That is, under normal conditions a blockaded pawn on the sixth rank is not enoug h to bring the defender into zugzwang, . but with two d-pawns it works.

Chapter 6 - Solutions

7. Dmitry Lavrik - Silas Lund Vladimir 2007

1 57

2) I slightly pref er Black due to the better pawn structure with less visible targets. The pawn on g7 is safely guarded by the DEB on f6. And once the white knight is driven away from f5 , no minor piece can attack g7. If White plays �g5 to exchange the DEB on f6, Black can always recapture . . . gxf6. Evaluation: + In the game White decided upon a dubious plan:

27.id2 tf::i e7 28.tf::i e3 �HB 29.c!Dg4?! �Hh8 30.c!Dxf6? gxf6 3 U 'Ihl gh4 32.f3 d5 a

b

c

d

e

f

g

h

Answers: 1 ) The bishop on f6 is placed on the same colour as the pawns, but it is a good defender of the g7-square, and Black plans to chase away the annoying knight with . . . l/J c6-e7. Notice that the black bishop is hitting the same wall of pawns as its white counterpart on e3, and in such a closed position it may not be seen as inferior. It is arguable that this position is irrelevant in respect of the DEB concept because of the closed nature of the position, but I find it meaningful to call Black's bishop on f6 a DEB as Black has to consider its future prospects carefully. White also has some weaknesses to take care of: the doubled pawns on the b-file and the isolated pawn on the h-file. These are factors in the position that weigh more heavily than Black's DEB. With an additional two rooks and a knight for each player, White's pawn structure is a real liability.

a

b

c

d

e

f

g

h

Perhaps he overestimated the strength of his h-pawn , which is nicely blockaded by Black on h4. Also, Black's knight is more mobile than White's bishop here. Black has a tangible advantage.

33.exd5t tf::i xd5 34.©fl tf::i e7 35.©g2 gds 36.icl tf::i e; 37.gg4 gh5 38.gc4 c5 39.b4 tf::i d6 40.gg4 cxb4 4l .cxb4 gc8 42.h4 gc2t 43.©h3 a4 44.b5 c!Dxb5 45.gxa4 tf::i d4 46.ga6 'if.le; 47.gxb6 gn 48.ga6 gxat 49.©g2 ghs 50.ga3 ggst 5 t .©h2 gnt 52.©h3 tf::i f3 53.ie3 gxb2 54.h5 gbg2 55.ig5 g2xg5 56.gxf3t ©e4 0-1

Esben Lund - The Secret Life of Bad Bishops

1 58 8.

Ingvar Johannesson - Silas Lund

8

7

Reykjavik 2007

6

8

7

6 2

4 3

.a

2

b

c

d

e

f

g

h

The threat is . . .�g4 blocking the g-file. a

b

c

d

e

f

g

h

The game continuation was 8 l .E:f8t?? (my opponent was in time trouble) 8 1 .. .�fS 0- 1 . Now 82.E:g8 �g4t is with check!

81.©el!= A study-like draw. White saves himself with the help of stalemate.

8 1...ia

82.gg4t! 82.E:g3! also draws.

82 ... ©es 82 .. �xg4 is stalemate.

83J!g5t ©d4 84,gg4t! ©cs 8SJ�gSt i.dS 86.©e2 ©d4 87.gg4t i.e4 88.©el ©es 89.ggSt ©f4 90.gg8 Back to the start of the exercise.

90 ...i.0 9 IJ!g4t!= And so on.

Chapter 6 - Solutions

1 59

9. Leinier Dominguez Perez - Veselin Topalov

10. Silas Lund - Joel Akesson

Sofia 2009

Sweden 2008-09

24.. J:k7!? 25.ib6 Wfb8 26.ixc7 Wfxc7

7

8

6 5

m m m•m m•��lil·� f • �....,� • mlj)m m

4 :i W1L� W � 3 ,, n � m

2 �.r.�.!. a

a

b

c

d

e

f

g

h

White has a DEB on g4 that has been isolated after he parted with his dark-squared bishop. Black has sacrificed the exchange but there are still many pieces left on the board. It is difficult for White to undertake anything; the most obvious plan is to push pawns on the queenside (a2-a4 or c2-c4) , but this just endangers the white king as long as the queens are still on the board. And heavy piece exchanges are not easy to bring about. The presence of opposite-colour bishops also promises Black play on the dark squares. The game continuation showed nothing tangible for White and eventually Black initiated counterplay against White's king and forced the draw:

b

m • m c

d

e

f

g

h

L

The solution is a transformation of the advantage:

32.LdS! tll x d5 33.gcst 'itih7 34.gds In return for the light-squared bishop, White has driven the black king to the edge of the board and now hopes to break the blockade on d5 and get his d-pawn going.

34 .. ,gas Black's best chance of survival might have been 34 . . . lLi xe3!? 3 5 . fxd E:b4 36.E:b8 E:xb2t 37.@f3 b5 38.d5 although it does look unpleasant with the king cut off on h7.

35.id2! gb5 36.'itid3 lll f6 37.ic3

27.ie2 Wfd8 28,ggl Wfb6 29.Wff4 if'2 30,gg2 ie3 3 1 .Wfe4 ge7 32.Wfb4 gc7 33.h4 ic5 34.Wfe4 WfaS 35.gg3 if'2 36,gh3 h5 37.gxh6 f5 38.Wff4 gxc2 39.'itixc2 Wfxa2t 40.'iticl Wf alt 41 .'itic2 Wf a2t 1/z-V2

a

b

c

d

e

f

g

h

Esben Lund - The Secret Life of Bad Bishops

1 60

The main idea for White was to declare the bishop on c3 to have a defensive role protecting the pawns, thus giving greater freedom to the rook, king and passed d-pawn. The outside position of Black's king on h7 was an important additional factor when taking chis decision.

1 1. Alexei Komev - Alexander Morozevich Taganrog 20 1 1

37 .. J:U5

37 . . . g5!? was a better try.

38.f3 gh5 39.gffi gxh4 40.gxf7 gh2 4l .g4 gh3 42.g5! .!Dh5 43. 'itie4+8

7

6

22 ... gxe3!

4

22 . . .ixc4 23.\Wxf6 gxf6 24.ixc5 dxc5 25 .!'i:a? is unclear.

Ocher options:

5

3

2 a

b

c

d

e

f

g

h

With the superb coordination between king, rook, bishop and d-pawn, White is winning.

43... tll g3t 44.'itif4 .!Dh5t 45.'itie5 gg3 46.d5! gxg5t 47.'itie6 The potential of the bishop shows. White is winning with his active pieces and deadly d-pawn.

47 ... ggl 48.d6 gdl 49.d7 g5 50.'itie7 'itig6 5 1.gffi 1-0

The straightforward 22 . . . \Wxd4 23.ixd4 ixc4 24.!'i:dc l !'i:b4 gives White a lot of play for the pawn after: 2 5 .ic3 !'i:b5 26.id4 id3 27.!'i:c3 ic2 28.!'i:a? lll d3 29.f5 ! (This is better than 29lkxc7 lll xf4! 30.!'i:xf7? lll e2t 3 l .©h2 lll xd4 32.!'i:xg?t �h8, when the bishop covers the h7-square and Black remains a piece up.)

23.\Wxe3 �xb2 After the exchange sacrifice the important dark-squared bishop vanished from the board. Thus the white DEB on g2 has been isolated. Other factors that benefit Black are the strong (untouchable) knight on c5 and passed b-pawn. White also has an open king position. Black is better.

Chapter 6 - Solutions

161

-

12. Petar Trifunovic Vasja Pirc

Since the position has stabilized so far and Black can try to improve his position from here in order to transform the advantage, you can claim the exercise solved at this point.

Birmingham 1 9 5 1

m m �� ­

8

•. : ��m·�·00'""'� 1 "'� '�t····"'� ·00/, ... � : . iu'.f ! ,i 3 8 m vm 8 �@

24.ifl 'Wf6 25.E!dbl g6 26Jfa5 b2 27J'!a3 h5 28.ie2 i.c8 29.'Wd2 i.d7 30.i.dl lD e4 3 1 .'We3 ge8 32.c;!;>g2

,

.

2

� 8 m m :: B a

- m m �/,� b

c

d

e

f

g

h

6

68.gxe5! Le5 69.gxe5 ges

a

b

c

d

e

f

g

h

32 ...'Wh4!? Black transforms his advantage. By giving up the passed pawn he wishes to make White's king even more vulnerable.

33.gxb2 33.gxh 5? i.xh3t! and here 34.'Wxh3 leads to forced mate: 34 . . . �f2t 35.i>h l ll'i g3t 36.2:!xg3 2:!e 1 t-+

33 ... hxg4 34.hxg4 i.xg4 35.hg4 'Wxg4t A very unpleasant position for White. The king will never find shelter as long as Black keeps the queens on the board. Black went on to win the game.

This move is not the first choice of my computer, but a human being should definitely consider it as one of the first candidates: Black wishes to exchange off as many of the additional pieces (queen and rook) as he can, and try to blockade the d-pawn. If you did not consider this move after the exchange sacrifice on e5 and find an adequate answer to it, you did not solve the exercise. After 69 . . . 2:%d6 White keeps the extra rook on the board and enjoys a great freedom of choice. One sample line goes: 70.'We3 'Wg7 7 l .2:!e 7 'Wal Threatening mate on h 1 . 72.i.e6 'Wf6 Forcing the exchange of queens, but White gets an endgame with good winning chances. (72 . . . 'Wh l t? 73.©g4) 73.2:!c7 'Wxf3 74.'Wxf3 2:!xf3 75.2:!xc5 2:!f2 76.2:!b5 2:!xa2 77.2:!xb4± Remove one pair of rooks from the board, and Black would have great chances of survival due to his dominance on the dark squares. Unfortunately for the second player, White still has a rook remaining to support the light-squared DEB.

Esben Lund - The Secret Life of Bad Bishops

1 62

13. Andrei Volokitin - Alexander Moiseenko

8

7

Ukrainian Championship, Kiev 20 1 1

6 5

4

3

2

a

b

c

d

e

f

g

h

70.�e4!? Another option is 70.l'!xe8t �xe8 7 1 .d6. Unfortunately for Black, he cannot both exchange a pair of rooks as well as blockade the d-pawn on d6. Now this pawn becomes a real danger. Black also has concerns about his king and the c5-pawn. 7 1 . . .�e l 72.�f5 ! �c3 (72 . . . �h l t 73.Wg4) 73.\We4 (Threatening 74.�e7) 73 .. . l'!f8 74.d7 �f6 75.�d5 ©g7 76.\Wxc5+-

70 .. JheS 71.�xeS �f'8 72..ie6! Preventing the queen sortie to d6 which would give Black hopes of saving the game.

72 ... i>g7 73.f4 gxf4 74.gxf4 i>h7 75.i>g4 �g7t 76.i>f3 gm 77..if5t @gs 78.�e6t i>hs 79.d6 �c3t so.i>g4 ggst s i ..ig6 �b2 82.�e5t 1-0

25.gxd8 i>xd8

�� �� � ���� � �� �� � 1 � m m•�• 6 • m �:;::.r%m��,�m �� � 8

� �--:3 ��. m m �:-

2

:rf.::o;:: lf{ :rf.::o N

A WJ .(� f:.2.J ff':'. � m o ,,L.J;:: a

/ /,m /,m/, %� b

c

d

e

f

g

h

Looking superficially at the position, it looks as if Black's bishop might be a problem. Well, it is a DEB, but other factors have to be taken into account: Black has space on the queenside (b4) and in the centre. Also, his king will be easier to centralize, thus fighting for king space. White cannot prevent Black from

Chapter 6 - Solutions playing . . . e5-e4 preparing the king march to e5. Black's pawns are not easy for White to attack either. So, in compensation for his DEB, Black has pawn space and will soon bring his king to e5. This last point of the king march to e5 is important in order to claim the exercise solved.

1 63

38 ... ©xc4

26.g4 The best winning attempt. Solid play with ©fl -e2 is not dangerous for Black. After 26.lll e 3 e4 27.b3 i.d4 28.lll f5 ©d7 29.g4 ©e6 the king is going to e5 and Black is absolutely safe.

26 ... e4 27.b3 ©e7 28.f4!? White wishes to activate his king and fight for king space.

28 ... erl3 29.©f'2 ©e6 30.©rl3 ©es 3 1 .h4 �f6 32.gS �d8 33.©g4 f6! Black gets more air for his bishop and an attacking point on g5. This is very logical.

34.lDe3 fxg5 35.hxgS ©d4 36.ttJdS a5 37.©f5 a4 38.bxa4 38.ttJf6?! axb3 39.axb3 ©c3 40.tlJxh? i.xg5 4 1 .tlJxg5 ©xb3 42.©e4 is also drawn, but here Black can try a few more moves. However, the game saw a bit of a drama towards the end that could have questioned the most likely result.

a

39.liJf6??

b

c

d

e

f

g

h

39.©e4 i.xg5 ?? (39 . . . b3 draws) 40.lll b6t! ©c3 4 1 .a5 i.d8 42.tlJd5t (Or 42.tlJa4t, in both cases with check. This is the point: Otherwise a5-a6 is not winning.) 42 . . . ©b2 43.a6+-

39 ... b3? Black misses his chance. 39 . . . ©d4! 40.tlJxh? c4 4 1 .g6 b3 42.g? b2 43.g8=� b l =�t 44.©e6 �e4t and now either mate or exchange of queens will follow, when the c-pawn will decide the game. 45.©f7 (45.©d7 �e7t 46.©c6 �c7t 47.©b5 �b6#) 45 . . . �d5t-+

40.axb3t ©xb3 4I.ttJxh7 hg5 42.ttJxgS 1/2-112

Esben Lund - The Secret Life of Bad Bishops

1 64

14. Horwitz & Kling

2.g8=�t? Wxg8 3.@g6 Wf8! The black king escapes from the wrong corner. White has no cricks chat win the bishop, so the position is a draw.

1851 8

2 . J.e5

7

.

.

2 . . .i.xg7 3 .:gc7 Zugzwang. 3 . . . @g8 4.Wg6 i.f8 5 .!k8 and mate next move: 5 . . . @h8 6.:gxf8#

6 4 3

2 a

b

c

d

e

f

g

h

1 .g7 i>h7 l . ..i.xg7 2.@g6+-

2J:U7! White prepares the sacrifice of the g-pawn, and there is licde Black can do about ic It is important that the black king is kept in the wrong corner (same colour square as the bishop) .

4 3

2 a

b

c

d

e

f

g

h

3.gS=�!t i>xgs 4.i>g6+The first position (in essence) of the endgame chapter has been reached.

Chapter 6 - Solutions

15. Venyamin Sozio - Vladimir Alatortsev USSR Championship, Moscow 1 93 1 8

7

1 65

Black has a winning advantage in the diagram position, but in the game he played too safe with: 44 . . . d4? This simplifies the position and gives away the advantage. 45.cxd4 ixd4 46.Ei:e700 Suddenly the position is j ust unclear; later it was drawn on move 67.

6

45.ghl

4

45 ... 'itie4! 46,gh7 5 47,gxb7 f4 48J�b6

45.iif2 d4! 46.cxd4 ixd4t 47.e4 5 .'ific2 �g5 6.lll c 3t e3 7.lll b l =.

5.©c2 ©e2 6.c!Df4t ©el 8

7

6 4 3

2

1

a

7 ll:) g2t

b

c

d

e

f

g

h

,

Another way to achieve the draw is: 7.lll d 3t e2 8.lll c l t 'ifie3 9.lll b3=

a

b

c

d

e

f

g

h

7 ... ©e2 8.c!Df4t 1/z-1/2

168

Esben Lund - The Secret Life of Bad Bishops

l . .. ©e3 2.©b6 ©d2 3.©c5 i.g8 4.©d6 e4 s.©e5 e3

18. Genrikh Kasparian 1 958 8

7

6 5 4 3

2

a

b

c

d

e

f

g

h

l.liJcl 1 .lll d2t? loses to: l .. .We3! (But not l . ..Wd3? 2.lll fl when the access to the g3-square secures the knight a route behind the pawn. 2 . . . We2 3.lll g 3t Wf3 4.lll f5 e4 5.Wb6=) 2.lll fl t (or 2.lll b l Wd3) 2 . . . Wf2 Now 3 .lll h2 ie6 traps the knight, whereas after 3.lll d2 We2 4.lll b l Wd3 the king again outmanoeuvres the knight. Another way to achieve the draw is 1 .lll c 5t ©d4 2.©b6 ©c4 3.lll d7 e4 4.lll e 5t ©d4 but here White has to show accuracy: 5 .lll g6! (5.lll g 4? ie6 6.lll h 2 ih3! 7.Wc6 Wd3 8.Wc5 e3 9.lll f3 e2-+) 5 ... e3 6.lll f4 ic4 7.lll h 3! The idea of the knight transfer to g 1 has to be spotted from the start of the exercise to claim it solved if you chose 1 .lll c 5t as your first move. 7 ... e2 8.lll g l =

a

b

c

d

e

f

g

h

6.©e4! White avoids 6.Wf4? id5 , when he is in zugzwang and loses. The same goes for 6.Wd4? id5! . It is essential to spot the mutual zugzwang i n order to claim the exercise solved.

6 ...�c4 6 . . .id5 t 7.Wd4! and Black is forced to give up the control of either b3 or e4.

7.©a �dst

7 . . .if7 is met by 8 .We4!, again avoiding the zugzwang.

8.©f4! Black cannot make progress.

1/2-1/2

Chapter 6

19. Nikolai Mikkelsen - Daniel Semcesen Horsens 20 1 3

21..be7 !h'.e7 22.lll c4 8

-

Solutions

1 69

on b2, d5 and f4 . In comparison, Black has targets on a4 and c7. White's knight may look great, but it is not protected; it is tied to the defence of the b-pawn and it is a long way from attacking Black's targets. 2) If the queens and rooks were exchanged, White could start manoeuvring his knight to attack Black's targets and in general play against the bishop on g7. The problem for White is that the heavy pieces are still on the board, and thus White has all his weaknesses to take care of

7

6 5 4 3

2

a

b

c

d

e

f

g

22 ... hdS! 23.exdS �he l! 24.Wfxel

h

24.Ei:xe 1 a3! effectively breaks the dark­ squared blockade on the queenside.

24...Wf5

I am reluctant to call Black's bishop a DEB in this position. His pawn chain in the centre calls for it, but White's pawns on the queenside are on the other hand placed on the same colour and can become vulnerable later. DEB or not, the concept of trying to prove the (supposed) superior mobility of the knight in such a position is still a valid way to play for an advantage. In this game it backfired, though.

25.gdl Wfxf4 If you evaluated the position correctly and saw this far, you have solved the exercise.

26.Wfe2 hS 27.a3 Wf gS 28.Wfe4 h4 29.gel .if6 30.ge2 @g7 3 1 .llie3 ieS 32.Wfg4

Because of the numerous threats to b l , b2, d5 and f4, White loses a pawn without adequate compensation. I do not like the idea played by White in the game, for several reasons: 1) White has targets in his position that are far more vulnerable than Black's: the pawns

a

b

c

d

e

f

g

h

1 70

Esben Lund - The Secret Life of Bad Bishops

Now Black transforms the advantage into a material gain with the following tactical operation:

20. Simon Bekker-Jensen - Jakob Vang Glud

32 ...Wixg4! 33.'Llxg4 if4

26.f4 ih6 27.i.xf6

Threatening . . . ic l .

34.llJ� c4 Preventing 'Lld3 and introducing the idea of . . . :B:b5 attacking d5.

35.ge4 gxb2 36.gxf4 gbl t 37.liJdl gxd l t 38.c3 h5 50.i>b2 ti'h6 5 1 .ti'fl �U6 52.ti'gl @fl 53J�eel i>e6 54.tt:\e3 @d7 55.i>c3 ge6 56.ti'fl ti'h8 57.tt:\g2 if6 58.ti'd2 i>c8 59.h4 ixg2 60.ti'xg2 ixd4t 61 .@b3 @b8 62.gf5 i>a7 63.ti'd2 ial 64.ti'cl ie5 65.ti'e3 ge8 66.gf7 gf8 67.gefl gxf7 68.gxf7 ial 69.ti'fl ti'e5 70.ti'e3 ti'b2t 71 .i>a4 ti'a2t 72.ti'a3 ti'c2t 73.ti'b3 ti'cl 74.gfl id4 75.gc2 ti'fl 76.ti'a2 ie5 77 .gfl ti'b5t 78.i>a3 e3 79.gg2 ti'd3t 80.i>a4 d4 8 1 .ti°e2 ti'e4 0-1

3

2

a

b

c

d

e

f

g

h

Before we come to the solution of this exercise, we go some way back in the game as it shows a common transformation into a rook and bishop vs. rook endgame. The defending side finds his position slightly disrupted after he succeeds in exchanging the opponent's last pawn.

74.ghs Objectively, White is lost, but it is more difficult to prove after 74.E:i:g l . Both players were down to the additional 30 seconds each per move.

74 ... gxh8 Here Black had the nice trick 7 4 . . . c3 77.ga4 gh2t? 77 . . . E:i:bS-+ is a simple solution: Black just needs to get rid of the b-pawn. Such a position with the king on the sixth rank is won for the rook and bishop side, unless the defender can bring his rook into the second-rank defence.

Chapter 6 - Solutions

1 73

22. Pavel Ponkratov - Alexander Morozevich Taganrog 20 1 1

a

b

c

d

e

.f

g

h

78. a3 gc2 74.i>b3 gd2 75.i>a3 i>e2

As we discussed in the endgame chapter, it is

quite common in positions like these, chat in order to transform the advantage into a win, a certain amount of counterplay should be allowed and Black should be ready for such a development of events.

The idea of parting with one's own remaining pawn(s) to get a favourable version of the endgame rook and bishop vs. rook and mate the opponent instead, is worth remembering.

76... i>d3 After White's slip, the position is again won.

77.gbs @ci 78.i>a2 icS 79.gb3 gdl so.gbs gb1 A curious zugzwang position.

8

8 l .b4

6 5

8 l . ..axb4 82.cxb4 id4

8 1 .8'.b3 8'.al t-+

7

Mate follows.

0-1

4 3

2

a

b

c

d

e

f

g

h

Esben Lund - The Secret Life of Bad Bishops

186

30.Lund

3...n 3...g5 4.E:e6t cj".ifl 5.E:g6 f2 6.E:xg5 c;t>g2 leads more or less to the main line, with the rook on g5 and not g6. Also losing is 6...�f4t ?.c;t>dl �xg5 8.b8=W i>g2 9.Wb7t i>gl. The bishop blocks the g-file from checks, but White wins with the 'staircase manoeuvre' typical of these endings. 10.Wb6 i>g2 11.Wc6t i>gl 12.Wc5+- threatening the bishop.

2013

I.©c2! The white king takes an active part in the defence.

i.c;t>c4 �b8 (1...£2 2.E:a3t c;t>d2=) 2.c;t>d5 f2 3.E:al c;t>e2 4.c;t>e6 fl=W 5.E:xfl c;t>xfl 6.c;t>f6 i>g2= 1. ..©e2

Not 1...£2 4.bS=W+-.

2.E:a3t i>e2

3.E:xg3

fl=W

a

b

c

d

e

f

g

h

6J'ff6! 6.E:xg3t? i>xg3 7.b8=Wt i>xh3= In this endgame, the white king is too far away, so Black is saved by stalemate tricks: 8.Wf4 c;t>g2 9.Wg4t c;t>h2 10.Wf3 c;t>g1 11.Wg3t c;t>h1! etc. 6...i.e5 7,gxflt ©xfl 8.h4 ©e3 9.h5 ©d4 IO.h6 ©cs 11.h7 Black is one tempo short. 11...©c6 12.h8=� Or 12.b8=W. 12...i.xhs I3.h8=� White wins. 1-0

a

b

c

d

e

f

g

h

Games Index Chapter 1 Hermanis Matisons - Aron Nimzowitsch, Carlsbad 1929 Sergey Karjakin - Vladimir Kramnik, Moscow 2011 Alexander Alekhine - Frederick Yates, London 1922 Hans Tikkanen - Silas Lund, Copenhagen 2009 Robert Fischer - Tigran Petrosian, Candidates final, Buenos Aires (7) 1971 Salo Flohr - Jose Capablanca, Moscow 1935 Anthony Saidy - Robert Fischer, US Championship, New York 1964 Svetozar Gligoric - Jan Timman, Bled/Portoroz 1979 Teimour Radjabov - Pavel Eljanov, Astrakhan 2010

14 16 17 24 26 28 29 30 31

Chapter2 Nigel Short - Michal Krasenkow, Wijk aan Zee 2008 Levon Aronian - Peter Leko, Nalchik 2009 Etienne Bacrot - Magnus Carlsen, Baku 2008 Suat Atalik - Pavel Eljanov, Wijk aan Zee 2007 Vladimir Kramnik - Boris Gelfand, Wijk aan Zee 2008

42 44 46 49 54

Chapter3 Evgeny Sveshnikov - Alexander Chemin, Sochi 1986 Sergey Tiviakov - Julen Luis Arizmendi Martinez, Istanbul 2003 Aljosa Grosar - lztok Jelen, Ljubljana 1992 Nikolai Chadaev - Oleg Chebotarev, Sochi 2007 Olli Salmensuu - Kalle Kiik, Turku 2001 Yuri Shulman - Magesh Panchanathan, Dallas 2006 Willy Hendriks - Igor Glek, Dutch Championship 1996 Praveen Thipsay - Christian Bauer, Bled (ol) 2002 Ara Minasian - Valerian Gaprindashvili, Linares 1999 Jiri Nun - Gennadi Zaichik, Polanica Zdroj 1989

68 68 70 71 71 72 73 74 75 76

Chapter4 Grigory Levenfish - Sergey Freiman, Leningrad 1934 Mikhail Botvinnik - Coenraad Zuidema, Amsterdam 1966 Emanuel Lasker - Vyacheslav Ragozin, Moscow 1936 Yuri Balashov - Alexei Shirov, Klaipeda 1988 Lev Belov - Anatoly Utiatsky, Kislovodsk 1960 Alexander Alekhine - Ludwig Rellstab Sr, Kemeri 1937 Vladimir Alatortsev - Vitaly Chekhover, Leningrad 1936 Bela Sandor - Pal Benko, Budapest 1949

97 97 97 98 101 102 106 116

1 88

Esben Lund - The Secret Life of Bad Bishops 1 19 1 20 1 20 1 22 1 23 1 25 1 29 1 29 1 30 131 1 32 1 35

Kazimierz Plater - Mikhail Botvinnik, Moscow 1 947 Gosta Stoltz - Isaac Kashdan, The Hague 1 928 Vitaly Chekhover - Emanuel Lasker, Moscow 1 93 5 Kazimierz Plater - Mikhail Borvinnik, Moscow 1 947 City of Liverpool - City of Glasgow, Correspondence 1 905/6 Nikolai Korolev - Silas Lund, Taastrup 2002 Yuri Averbakh - Vassily Panov, Moscow 1 9 50 Nikolai Zubarev - Alexandrov, Moscow 1 9 1 5 Tigran Petrosian - Anatoly Bannik, USSR Championship, Riga 1 9 58 Efraim Laksman - Silas Lund, Copenhagen 2006 Boris Grachev - Igor Lysyj , Taganrog 20 1 1 Alexander Grischuk - Michael Roiz, Ningbo 20 1 1

Chapters 5 & 6 Vladimir Kramnik - Viswanathan Anand, Wijk aan Zee 2007 Boris Avrukh - Gadir Guseinov, Gothenburg 2005 Boris Gelfand - Sergey Karjakin, Monaco (rapid) 20 1 1 Teimour Radjabov - Etienne Bacrot, Khanty-Mansiysk 20 1 1 Ralf Akesson - Silas Lund, Klaksvik 2008 Dmitry Lavrik - Silas Lund, Vladimir 2007 lngvar Johannesson - Silas Lund, Reykjavik 2007 Lenier Dominguez - Veselin Topalov, Sofia 2009 Silas Lund - Joel Akesson, Sweden 2008-09 Alexei Kornev - Alexander Morozevich, Taganrog 20 1 1 Petar TrifunoviC - Vasja Pirc, Birmingham 1 9 5 1 Andrei Volokitin - Alexander Moiseenko, Ukrainian Championship, Kiev 20 1 1 Venyamin Sozin - Vladimir Alatortsev, USSR Championship, Moscow 1 93 1 Nikolai Mikkelsen - Daniel Semcesen, Horsens 20 1 3 Simon Bekker-Jensen - Jakob Vang Glud, Horsens 20 1 3 Lenier Dominguez - Judit Polgar, Khanty-Mansiysk 20 1 1 Pavel Ponkratov - Alex. Morozevich, Taganrog 20 1 1 Alex. Grischuk - Alex. Morozevich, Khanty-Mansiysk 20 1 1 Evgeny Najer - Sergey Volkov, Taganrog 20 1 1 Asger Paaske - Silas Lund, Copenhagen 2003 Silas Lund - Axel Smith, Borup 2009 Ivan Cheparinov - Hrant Melkumyan, Helsingor 20 1 3 Vugar Gashimov - Ruslan Ponomariov, Khanty-Mansiysk 20 1 1 Kaan Kanli - Silas Lund, Rogaska Slatina 20 1 1

1 43, 1 43 , 1 43, 1 43, 1 44, 1 44, 1 44, 1 44, 1 44, 1 44, 1 45 , 145, 145, 145, 1 46, 1 46, 1 46, 1 46, 1 47, 1 47, 1 47, 1 47, 1 48, 1 48,

151 1 52 1 53 1 54 1 55 1 57 1 58 1 59 1 59 1 60 161 1 62 1 65 1 69 1 70 1 72 1 73 1 75 1 76 1 78 1 79 1 80 1 82 1 84

Index of Studies and Positions Chapter 4 Bernard Horwitz & Joseph Kling Chess Studies 1 8 5 1 Jens Enevoldsen, 1 949 Ercole del Rio Sopra if Giuoco degfi Scacchi 1 750 Carlo Cozio If Giuoco degfi Scacchi 1 766 Yuri Averbakh No.473 in Turm gegen Leichtjigur see Bibliography Yuri Averbakh No.477 in Turm gegen Leichtjigur see Bibliography Kholmov 1 973, quoted in Averbakh's Turm gegen Leichtjigur see Bibliography Silas Lund Skofeskak 1 2008 see Bibliography Yuri Averbakh No.4 in Springer gegen Laufer see Bibliography Yuri Averbakh No. 1 3 in Springer gegen Laufer see Bibliography Yuri Averbakh No.95 in Springer gegen Laufer see Bibliography Yuri Averbakh No. 1 72 in Springer gegen Laufer see Bibliography Ernst Holm Sydsvenska Dagb!adet Snaflposten 1 9 1 0 Zoltan Vecsey Honourable mention, Magyar Sakkvi!dg 1 934 (version Y.Averbakh 1 9 56) G. Fedotov Shakhmaty v SSSR 1 9 54 Yuri Averbakh No.47 in Springer gegen Laufer see Bibliography -

-

-

-

-

-

-

-

-

90 91 92 92 93 93 95 1 12 1 13 1 13 1 14 1 44 1 15 1 17 1 24 1 25

Chapters 5 & 6 Yuri Averbakh No. 1 1 9 in Springer gegen Laufer see Bibliography Bernard Horwitz & Joseph Kling Chess Studies 1 8 5 1 Sam Loyd Chess Monthly, February 1 860 Variation from Henri Rinck, 2nd Hon. Mention, Schweizerische Shachzeitung 1 9 23 Genrikh Kasparian 1 9 58 Silas Lund 20 1 3 (original compositions) -

1 44, 1 46, 1 46, 1 47, 1 47, 1 50,

1 56 1 64 1 66 1 67 168 1 86

Name Index Chemin

A Aagaard

8,29,178 144,145,155,159

Akesson Alatortsev

106,146,165

Alekhine

4,17,18,24,28,35, 40,102,103,107,186

Alexandrov

129 143,151

Anand Arizmendi Martinez Aronian Atalik Averbakh

143,152

Balashov Bannik

16,46,47,59,143,154 98 130 33

Bauer

63,74 147,170

Belov

101

Benko

187

Berliner Botvinnik

28 29,97,98,108,119,122

Capablanca Carlsen

Fedotov Fischer

66

Castineira

66

Chadaev

71

Chebotarev

71

Chekhover

106,120

Cheparinov

149,180

127

124 26,27,29 28,29 97

G Gaprindashvili Gashimov Gelfand Glek Gligoric Glud Grachev Grischuk Guseinov

28,29 16,46,47,48,59

91,94,137

F

Grosar

c Cabanas

Eriksson

Freiman

Bartel Bekker-Jensen

4,30,31,35,40,41,49,52,59

Flohr

B Bacrot

7,25,93,138

E

49,50,59

125,129,130,138,144,156

92 145,148,159,172

Dvoretsky

Enevoldsen

114,117,119,121,122, Avrukh

Del Rio Dominguez

Eljanov

106,108,109,110,113,

92

D

68 44,45,59 93,95,96,97,102,103,

4,12,62,64,67,68,69

Cozio

75 149,182 54,55,57,59,128,143,153 73 30 147,170,171 132 135,148, 175 70 143,152

H Hagen Hendriks Holm Horwitz

7,8 73 115 90,146,164

191

N

J Jelen

70

Johannesson

144,158

148,176,177 99,117

Nilsson

K

7,8 14,30

Nimzowitsch

Kanli

150,184

Karjakin

16,41,143,153 120

Kashdan Kasparian

147,168

Kasparov

11

Kholmov

67,95,99,100,137 71

Kiik Kling

90,146,164

Kornev

145,160

Korolev

125, 129,138,139

Kramnik

11,16,17,25,42,54,

Krasenkow

59,128,143,151 42,44,59

L Laksman

131

Larsen

93

Lasker

97,120 144,157

Lavrik Leko

33,44,45,59

Levenfish

97 146,166

Loyd Lund

Najer Nalimov

8,24,104,112,125,129,131, 138,139,144,145,149,150,155, 157,158,159,178,179,184,186

Lysyj

132

M Marin Matisons

29,54

Melkumyan

14,30 149,180,181

Mikkelsen

7,8,147,169

Minasian Moiseenko Morozevich

75 146,162 145,148,160,173,174,175

Nun

76

0 Olafsson

93

p Paaske

149,178 72 29,129

Panchanathan Panov Petrosian

26,130,131

Pirc

145,161

Plater

119,122

Polgar Ponkratov

148,172 148,173

Ponomariov

149,182

R Radjabov

4,30,31,36,40,41,143,154

Ragozin Rellstab Rinck

97 102,107,186 147,167

Roiz

135,137

Romero Holmes

130,131

Rowson

61

s Saidy

29

Salmensuu

71

Sandor Semcesen Shirov Short Shulman Smith Sozin Stoltz Sveshnikov

116 147,169 98,99 42,44,59,76 72,81 149,179 146,165 120 68,146

Esben Lund - The Secret Life of Bad Bishops

1 92 T

22, 26, 27, 55 6 3 , 74 24 30 28 68 33, 1 45 , 1 59 1 45, 1 6 1

Tarrasch Thipsay Tikkanen Timman Tisdall Tiviakov Topalov Trifunovic u

101

Utiatsky v

1 1 7, 1 38 1 48, 1 76 1 46, 1 62

Vecsey Volkov Volokitin w

4, 29, 30, 3 1 33

Watson Werle y

Yates Yusupov

i 1 7, 1 8, 1 9, 22, 23, 28, 3 5 , 40

25, 54

z

Zaichik Zhou Jianchao Zubarev Zuidema

76 33 1 29 97, 98