Averbakh - Averbakh's Selected Games
Averbakh - Averbakh's Selected Games
Averbakh - Averbakh's Selected Games
by
Yuri Averbakh
CADOGAN
First published 1998 by Cadogan Books pic, 27-29 Berwick St, London WI V 3RF.
Distributed in North America by The Globe Pequot Press, 6 Business Park Rd, P.O. Box
833, Old Saybrook. Connecticut 06475-0833, USA
All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system or transmitted in any form or by any means, electronic, electrostatic, magnetic
tape, mechanical, photocopying, recording or otherwise, without prior permission in
writing from the publishers.
Foreword 7
My Chess Career 9
Biographical Details 12
Index of Openings 1 92
Foreword
This book should have been published at least quarter of a century ago, when my
generation was still highly placed in the chess world and was creating chess history.
However, journalism, work for the USSR Chess Federation, as well as research on the
endgame and the origins of chess diverted me from this: there simply was not sufficient
time.
Therefore, just before my 75th birthday, I gratefully accepted the offer of Cadogan
Chess to publish a collection of my games.
However, every cloud has a silver lining! When working on games that were played
many years ago, I not only experienced as though anew the events that occurred in them,
but at the same time I gained the opportunity to look at myself from the side, effectively
from a future viewpoint, to look critically, and that means on the whole to evaluate more
objectively my chess playing.
It is considered that present-day chess differs greatly from the chess of the 1950s,
when the peak of my successes occurred. Of course, today chess has become more
competitive, more severe, more professional. Thanks to Informator, the Internet and
computer programs, grandmasters and masters now obtain an incomparably greater
amount of information than in my time. As a consequence of this, opening theory
develops more rapidly and intensively, and the role of opening knowledge is growing.
However, the longer playing sessions, stricter time controls and, finally, the
abolishment of adjournments have left their mark and affected the quality of play. This is
especially noticeable in the concluding phase of the game, the endgame.
One other important factor must be mentioned. The information language of symbols,
which stems from Informator and is now employed in many publications, has led to the
fact that commentaries written with the help of these symbols completely lack the most
important thing that is always present in a chess encounter - the conflict of personalities,
the clash of plans, the battle of ideas. As is said nowadays, the human factor is lost. The
soulless symbols do not, and are unable to reflect anything of this.
And for me the battle of ideas, the clash of plans, and also the various psychological
nuances associated with the play, have always been a significant element in the creative
process at the chess board, sometimes simply determining the result of the game,
irrespective of who had the better or worse position. It is for this reason that in my notes
I have endeavoured to devote the greatest attention to them.
In selecting these games I have aimed to give not only those that I consider my best,
the most memorable, but in the main those that reflect my personality as a grandmaster,
and reflect my understanding of chess. Moreover, in some games I wanted to convey my
experiences at the board.
The knowledgeable reader may be surprised that, although I am considered an expert
on endgame play, here I have given relatively few examples of my play in this field. This
can be explained as follow. Firstly, in my books devoted to the endgame a sufficient
number of such examples can be found, and I did not want to repeat them. Secondly, and
8 Averbakh's Selected Games
perhaps more important, I wanted to show, contrary to the 'top technician ' tag, ascribed
to me by chess journalists for my comparatively good techn ique, that I was on the whole
a player with a style that, although positional, was universal, and that I was quite strong
both in attack and defence, in positional manoeuvring, and in sharp combinational
complications. How well I have succeeded in demonstrating this, the reader can judge.
Only in my later years, when I had stopped being a chess professional and had ceased
working on trying to im prove, when I no longer had ambition and lacked both strength
and nervous energy, I sometimes used to play on techn ique, and incidentally, also not
without success.
I sincerely hope that this collection will not only enable the reader to make the
acquaintance of my games, but will help him to understand better that great game -
CHESS. A game in which, in the words of the ancient sages, victory is gained by
intellect, but I would add that in which, as in life, the factor of chance and luck also plays
a considerable role.
Yuri Averbakh
My Chess Career
I learned to play chess at the age of seven, but I became fascinated by it only much later,
when I was already thirteen.
The first chess book that accidentally came into my hands was My System by Aron
Nimzowitsch. It was hard to think of a worse choice! After all, in chess you must first
learn to attack, and only then to defend, you must gain a mastery of tactics, and only then
strategy. My System is a good book, only not for beginners. It is a textbook on positional
play, and first you must learn to make combinations. As will be seen, with me it all
happened the other way round, and it is not surprising that later, to a significant extent, I
had to relearn.
Late in 193 5 I visited the Moscow Chess Club for the first time, and there I was
fortunate enough to listen to a lecture by the great endgame expert Nikolai Grigoriev. It
made an indelible impression on me. When Grigoriev explained his pawn studies,
moving the pieces on the demonstration board with his thin, artistic fingers, I sensed,
rather than understood, the great depth and beauty of chess, observing with my own eyes
how human thought spiritualises these little wooden pieces, and they, like real actors,
begin performing miraculous spectacles, capable of touching the most sensitive parts of
the human soul. It was this perception of chess as an art that finally linked me with it. I
wanted to understand chess and study it, which is what I earnestly began doing.
My first major chess success was victory in the USSR Cadet Championship (for
juniors up to 1 6 years old). Although among the participants there were probably lads
more talented than me, such as Mark Stolberg who met an untimely death during the
Second World War, my broader understanding of chess bore its fruit.
In 1 93 9 I finished school and at the same time, after passing through the sieve of
elimination events (quarter-final and semi-final), I qualified for the final of the adult
Moscow Championship, where I met some of the acknowledged masters of chess.
Finding myself in such impressive company, I initially lost my head. The responsibility
so pressed, that in almost every game I ended up in time trouble, in the middle of the
tournament I suffered seven successive defeats, and I finished in last but one place. It is
true that there were also objective reasons for this slump - the Championship coincided
with the examination session, so that for preparing for the games and for relaxation there
was no time.
I was naturally very upset by my failure, and, on noticing this, the Chief Arbiter of the
tournament, Nikolai Ryumin, gave me some good advice:
'If you want to become a master', he said, 'you must learn to lose, you must learn to
withstand a blow, so that in the game following a defeat you play as if nothing had
happened!'
I remembered this wise advice all my life and endeavoured to follow it, although, to be
honest, I did not always manage to.
A major role in the forming of my chess outlook was played by two other men - Pyotr
Romanovsky and Benjamin Blumenfeld, both well known masters. Romanovsky taught
10 Averbakh's Selected Games
that in chess, as in any other form of creative activity, there is a battle of ideas, a battle of
different tendencies. Every genuine master should develop his chess philosophy and
follow it. These words of his forced me to think about my play, determine my creative
attitudes, and to pick out those ideas that I intended to carry out in my games.
Blumenfeld drew my attention to the psychological aspect of chess. He emphasised
that chess was played by people, each with their own character and cast of mind, with
different virtues and defects. In the struggle at the chess board one must learn to exploit
all this, one must be able to create those situations in which your skills and talent display
themselves to a greater degree than the skills and talent of the opponent.
After gaining the master title and finishing at the Institute, for five years I tried to
combine engineering work with participation in tournaments of various standards.
However, at the end of the 1940s I had to consider my future. I was at the cross-roads. I
had interesting work as a research engineer, and my candidate dissertation was
progressing. In chess I had become Moscow Champion, and had reached the final of the
USSR Championship, although I did not have any great success there. I could not help
feeling that in the creative sense I was standing still in chess: time for improvement, and
even for preparing for competitions, was practically not available. It was becoming
increasingly obvious that the attempt to sit simultaneously on two chairs did not promise
well: work interfered with chess, and chess interfered with work. I had to make a choice.
And I chose in favour of chess. I decided for a time to give up work in order to try to
become a grandmaster, and to raise my chess playing to a higher standard. At that time it
was much more difficult to gain the grandmaster title than it is now. To do this I first had
to finish in the first four in the USSR Championship Semi-Final. This I did. The next
stage was the 19th USSR Championship, which was simultaneously a zonal qualification
tournament for the World Championship. I managed to finish in the first five, who
gained the right to play in the Interzonal Tournament. The culmination was this
Interzonal Tournament, where I became a Candidate for the World Championshp and
was awarded the grandmaster title. My experiences associated with this tournament are
described in detail in my notes to the games.
The Candidates Tournament was the most significant event in my chess career. I
prepared quite thoroughly for it, but my practical experience before it was slight, and this
told on the final outcome.On the other hand, a couple of months later, with an
exceptionally high score (in these events only Botvinnik had done better) I won the
USSR Championship.
On becoming a grandmaster I began participating regularly in tournaments, but
finished with my scientific work. However, my investigative character forced me to
make a serious study of the endgame, that phase of chess where individual pieces battle
against each other. Initially I published several articles, and then I thought about a book
devoted to various types of endings. Since I was not capable of carrying out this work on
my own, I gathered a small group of masters, and we enthusiastically got down to work.
After a study had been made of endings in which neither side has more than one piece, I
initially had the idea of moving on to endings with a greater number of pieces, but, thank
God, I stopped in time. It became clear to me that such a work would kill chess, as the
game would be transformed into a competitioin in which the one who won would not be
the one who plays better, but the one who knows more.
Averbakh's Selected Games II
I should mention that this protracted work on the endgame had an adverse affect on
my further chess successes. After all, a man's creative energey is not unlimited.
When I stopped studying the endgame, my passion for research forced me to turn to
another interesting topic - the history of chess, which, after all, is full of unexplored
territory. The resolving of many mysteries of its history is a task no less enticing than the
solving of various problems at the chess board. I hope also to write a book about this.
I have spent a long life in chess, and have been not only a player, but also a trainer,
arbiter, journalist and publisher; I have been an administrator of my own country's
federation and internationally. And I am eternally grateful to the royal game, for the fact
that it has brought me so much joy, the joy of creativity. I should like to repeat the words
of Siegbert Tarrasch, that chess, like love and music, can make man happy!
Biographical Details
(added by the Translator)
Author of numerous books, of which English translations have been published of Chess
Endings: Essential Knowledge (first published by Pergamon, 1 966), the 5-volume series
Comprehensive Chess Endings (Pergamon, 1 983-87), and Chess Middlegames: Essential
Knowledge (Cadogan 1 996)
Selected Games 1943-1975
The USSR Championship was not held in I played not less than twenty, and
either 1942 or 1943 . This role was perhaps even as many as thirty games with
essentially fulfil led by the Moscow Cham Simagin. He was a player of original, far
pionship of 1943 /4. This tournament was from routine style, an artist-researcher,
preceded by qualifying events and semi who sought new, untrodden paths. It was
finals, in one of which I took first place, always interesting to play against him.
ahead of Ragozin and Zubarev, who were 1 d4 �f6
already masters, and Simagin, who was 1 c4 c5
soon to become one. 3 d5 e5
4 g3 d6
Game I 5 R.gl R.e7
Averbakh-8imagin 6 e4 �
Moscow Championship Semi-Final 7 �el �bd7
1 943 8 � �e8
Benoni Defence 9 �bc3 'itfh8
10 R.e3
Of the chess players of my generation, one
of the most striking and unusual was
undoubtedly Vladimir Simagin.
A rather pale, freckled face with high
cheek-bones, glasses with a thin metal
frame, light, straw-coloured hair, protru
ding in unruly fashion from his crown -
that is how I remember him from when I
first saw him back in 193 5. During the
war, when he was exempted from service
in the army on health grounds, Vladimir
worked as a fitter in an aircraft factory. He
laboured for twelve hours a day, with little
to eat and with little sleep, but all the same White has played the opening un
he contrived to find time for chess. He pretentiously. 1 0 �hl should have been
went about in a quilted jacket, crude boots considered, to answer 1 0 ...g6 with 1 1 .i.h6
with wooden soles, and a shabby cap with �g7 12 1i'd2.
ear-flaps. 10 R.g5
It so happened that for a long time my The usual plan in this opening is IO ...g6
chess career and that of Vladimir took a followed by Il...�g7 and I2 .. f5 . The
.
parallel course: we used to participate in bishop move is the start of a system which
the same tournaments. We were rivals but at the time was successfully played by
also members of the same team; we played Simagin. He provokes 1 1 f4, in order after
together abroad and often shared a hotel I I . exf4 12 gxf4 .i.f6 to play 1 3 ...g5 and,
. .
11 'ifd2 �xe3 17 �g h 5
12 'ifxe3 Black was hoping for 1 8 'iff3 'ife7 1 9
Strictly speaking, the exchange of g4 e4! 20 'ifg2 �d7 and 2 1 ...:ae8 with a
bishops favours B lack. Firstly, he is complicated game. But White is by no
cramped, and every exchange facil itates means obliged to follow this path. He has a
the manoeuvring of his pieces. Secondly, strong move that essentially refutes the
he rids himself of a bishop which, with its opponent' s plan.
restricted mobility, often proves 'bad ' .
However, these are general arguments, and
everything, of course, depends on the
further development of events.
12 g6
13 f4
18 'ifh6! 'iffS
Because of the threat of 1 9 g4, this
reply is forced.
19 'ifxfS :x fS
Now it is apparent that, for the
successful completion of his original plan,
13 �g7 B lack is short of just one tempo. Were his
The natural continuation was 1 3 ...fS, bishop at d7, he would be able to take on
with a roughly equal game, but Simagin f8 with his other rook.
was not one for following familiar paths. 20 g4 �g7
Here he chooses an original, but risky plan 21 �g3
- he invites the opponent to play f4-f5, so Wh ite is not in a hurry to launch an
as then to try and establish his pieces on offensive. I was afraid that after 2 1 g5
the dark squares. White accepts the �g8 22 lDe4 lDe8 with the threat of
challenge. 23 ... f6 Black would gain counterplay.
14 fS gxfS 21 �d7
B lack consistently carries out his idea. 22 lDge4
1 4 ...g5 IS f6! �h5 1 6 1Ifxg5 �dxf6 1 7 In my notes at that time I suggested that
'ifh4 :g8 18 h3 would have led to an 22 lDce4 lDge8 23 lDxf6 lDxf6 24 g5 lDe8
opening of lines, favourable to White. 25 a4 was stronger, although then 25 ...f6
15 exfS �f6 26 lDe4 fxg5 27lDxg5 lDf6 28 lDe6 :g8 is
16 �h3 :g8 possible. Instead of 25 a4, 25 lDe4 is good,
17 <i>hl since if 25 ... b5 White has the counter 26
Avoiding a little trap - 17 g4? �xf5 ! 1 8 b4! cxb4 27 c5, when B lack ' s position is
'iff3 �xg4! 19 �xg4 1i'h4 20 h3 h5. on the point of col lapse.
Averbakh 's Selected Games 15
29 16
22 �ge8
23 a4 :g8 In the resulting sharp ending White has
24 :gl a6 every chance of winning, thanks to the
I don't want to attach a question mark more active placing of his pieces. He is
to this move, although here it was quite aim ing to decide the game by an attack on
essential to play 24 ... h6. However, as I the kings ide and consistently pursues this
have already mentioned, Vladim ir Simagin course.
very much disliked making natural moves. However, with 29 :a7 he could have
25 g5 �xe4 achieved his aim more simply, for
If 25 ... �h5 26 �e2, threatening 27 example:
�g4. (a) 29 ... :d8 30 :g3 bxc4 3 1 f6 �xh3
26 �xe4 :b8 3 2 :xh3 :f8 33 g6! �g8 34 g7 �xg7 3 5
:g3 etc.;
(b) 29 ... �c8 3 0 :xf7 bxc4 (3 0 ... :g7 3 1
:xg7 �xg7 32 f6+ �g6 3 3 �xc8 :xc8 34
:fl �f7 35 g6+! hxg6 3 6 �g5+ �f8 3 7 f7
etc.) 3 1 g6! hxg6 3 2 fxg6 :g7 3 3 :f8+
:g8 34 :xe8! �xh3 3 5 g7+ �h7 3 6
�f6+, and wins.
29 �xh3
30 :xh3 :g6
Now White's task is greatly simplified.
After the more critical, but also sharper
30 ... bxc4, I would have had a choice of
two continuations:
27 :a3 ! (a) 3 1 :g4 :g6 (3 1 ...:xb2 32 g6! fxg6
Intending 2 8 f6 with a very strong 33 :xh7+! �xh7 34 :h4 mate) 32 :gh4
attack. Black's desire to obtain at least h6 33 Lh6+ :xh6 34 :xh6+ �g8 3 5 g6
some counterplay is therefore quite under fxg6 (3 5 ... :xb2 36 :h8+! �xh8 3 7 gxf7,
standable. or 3 5 ... :b7 36 �xd6! �xd6 37 g7) 36
27 b5 :xg6+ �f7! (3 6... �h7 3 7 f7) 3 7 :h6 (3 7
28 axb5 axb5 :g2 c3! 3 8 bxc3 :bl + 3 9 :gl :xgl + 40
16 Averbakh IS Selected Games
�xgl c4! and Black can resist) 3 7...�g8 34 ...�h7, trying to break free with the
3 8 lDg5 lDxf6 39 :g6+ �h8 40 IIxf6 king, would have been met by the crushing
IIxb2, and White still has to win this 3 5 :'g3 hxg5 36lDxg5+.
ending. During the game this did not 35 :'g3
appeal to me, and I was intending to play:
(b) 3 1 g6! :xg6 3 2 :xg6 fxg6 33 f7
lDc7 (3 3 ...lDg7 34 lDf6 h5 3 5 :'g3 ! :'fS 36
:'xg6 :'xf7 3 7 :h6 mate) 34 lDg5 h5
(34...�g7 35 :'xh7+ �f6 3 6 lDe4+ �fS
[36...�e7 3 7 fS1I'+!] 3 7 lDxd6+ �f6 3 8
fS1I'+ :'xfS 3 9 :'xc7 etc.) 3 5 :'0 �g7 3 6
h4! (not immediately 3 6 lDe6+ lDxe6 3 7
dxe6 �fS 3 8 :'f6 g 5 3 9 :'g6 �e7, when if
40 :'xg5 Black has 40 ... c3! 4 1 bxc3 �xe6,
after which ... White loses) 36...e4 3 7 IIf4
e3 3 8 �gl , and Black has no useful
moves. Thus 3 8 ... lDxd5 is met by 3 9
lDe6+ �h7 40 fS 1I' :'xfS 4 1 IIxfS c 3 42 Now if 3 5 ...�h7 White has the decisive
bxc3 lDxc3 43 :'f7+ �h8 44 :'e7, and 36 g6+.
White must gradually be able to realise his 35 hxg5
material advantage. 36 hxg5 :c8
37 :h3+ �g8
38 �g2 lDc7
39 lIal
Aiming to conclude the game with
mate.
39 lDe6
40 dxe6 fxe6
41 g6
Black resigns
holding my own in the games with grand the side of the latter. Perhaps this was
masters Botvinn ik, Smyslov and Kotov, I because Botvinnik was considered the
took sixth place and achieved the master favourite, although the match ended in a
norm. An important competitive barrier, draw.
which I had approached closely before the Incidentally, it was then that I ended up
war, had been overcome. The tournament for the first time in the National Hotel. It
winner was Botvinn ik, playing hors happened like this: a group of young chess
concours, and the Moscow Champion players was walking along Gorky Street
became the second-placed Smyslov. (now it is cal led, as it was in the past,
Tverskaya). Sudden ly Boris Vaksberg (I
Game 2 defin itely remember that it was he)
Averbakh-Botvinnik suggested: 'Lads! Levenfish is staying in
Moscow Championship 194314 the National. What if we call in and show
French Defence him our analyses in the Sicil ian?'
I should explain that at that time a
I do not recal l for what reason, but my variation, that subsequently even became
game with Botvinnik took place in his known as the Moscow Variation, was
room in the National Hotel, where he was being intensively analysed in Moscow - 1
then staying. e4 c5 2 lLlf3 d6 3 d4 cxd4 4 lLlxd4 lLlf6 5
I should rem ind the reader that in the f3 e5 6 �b5+.
days of my youth Botvinnik was the idol of It was no sooner said than done! After
our young generation. We passionately making over way, not without some
supported him, intently followed his difficulty, past the heavily-moustached
successes, and learned from his games. doormen in uniform at the entrance to the
Botvinnik ' s outstanding chess successes hotel, we went up in the lift, and Vaksberg
had brought him enormous authority and cautiously knocked on the door of the
extraordinary popularity. He was known required room. I recall that the door was
throughout the country. Botvinnik was opened by Yakov Rokhlin.
admired, and he was followed and 'We need to see Grigory Yakov
imitated. levich!' , said Vaksberg boldly, and we
And here I was playing against hesitantly followed him into the room.
Botvinn ik, one of the strongest grand That which we saw threw us into utter
masters in the world, who had beaten confus ion. On a sofa, lean ing over a chess
Alekhine, and Capablanca, and Lasker and board, sat Levenfish and ... Botvinn ik!
Euwe. We said hello, and then there was a
When I sat down at the board and wrote pause. The first to find his tongue was
on my scoresheet: White - Averbakh, again Vaksberg. 'Grigory Yakovlevich!' ,
Black - Botvinnik, I wondered if I was he said to Levenfish, 'we would like to
dream ing. I even secretly pinched myself invite you to visit the Stad ium of Young
on the hand. No, it was not a dream. Pioneers! And you too, Mikhail Mois
Before me sat Botvinnik himselfl eev ich!', he added. And with that we
I remember that when in 1 93 7 proudly trooped out.
Botvinnik played a match for the title of But let us return to my game with Bot
USSR Champion with Grigory Levenfish, vinnik. I played, as you will see, extremely
the sympathy of the Moscow lads was on cautiously and, to my surprise, I gained a
18 Averbakh 's Selected Games
15 c3
A draw in this game was the limit of
my dreams, and so I consistently aim to
simplify the position. However, I S :dl
would have been much stronger, in order
to recapture with the rook. There is nothing
for the king to do at d3 .
'A new move, that has not been 15 lDxd3
considered by theory', wrote Baturinsky, 16 �xd3 �d7
the co-author of the three-volume work on 17 �g5 f6
Botvinnik that was published in the 1960s. 18 �e3
Now, many years later, it can be said A timely change of m ind. Initially I was
that it does not in fact deserve particular evidently intending to play 1 8 �h4, but
consideration. The best that Black can then I rejected this because of 1 8 ... lDfS 1 9
hope for here is a draw. �g3 �cS, when it i s White who may have
7 lDxd4 1Ife7+ difficulties.
8 1Ife2 1Ifxe2+ 18 :ac8
9 �xe2 �d7 19 :ad l l:the8
10 lDlfJ a6 20 g3 :c4
11 �d3 �d6 11 lDd4 �e5
The attempt to gain the advantage of 22 lDc2 l:tec8
the two bishops by 1 1...lDb4 could have At last Black shows his teeth. He
been answered by 12 lDeS. threatens 23 ... �xc3 .
12 lDf5 �xf5 23 �e2 a5
13 �xf5 lDge7 Black prevents the knight from going to
14 �d3 b4.
White should have considered 14 �h3 , 24 l:td3
with the idea of playing the bishop to g2, 24 :d2! was much more subtle. Then if
aiming at the dS pawn. 24 . . . bS there could have followed 25 lDa3!
Averbakh 's Selected Games 19
24 b5
25 a3
Now 25 tiJa3 l:ta4 26 tiJxb5 l:txa2
would have led to an advantage for B lack.
25 �d6
26 �d4 l:t4c6
27 l:thd l l:tb8
28 l:t3d2 b4
Game 3 7 "'xc4 e5
�ikenas-)lverbakh In this way Black tries immediately to
Moscow Championship 194314 solve the problem of the centre.
Catalan Opening 8 tiJgfJ
Nowadays theory considers 8 dxeS
1 d4 tiJf6 tiJxeS 9 "'c3 to be more accurate, aiming
2 c4 e6 to exploit the knight's insecure position.
3 g3 Now, however, it is White who has to be
At the time when this game was played, concerned about the insecure position of
the Catalan was a comparatively new his knight.
opening. It was introduced into tournament 8 tiJb6
play by Saviely Tartakower at the inter 9 "'d3 exd4
national tournament in the Catalonian 10 0-0 .i.e7
capital of Barcelona in 1 929. I did not want to defend the pawn by
3 d5 1 0 ...cS , as this would have opened the
4 .i.g2 dxc4 diagonal for the white bishop and delayed
5 "'a4+ tiJbd7 the development of my pieces.
6 tiJd2 11 tiJxd4 0-0
If 6 .xc4 Black usually used to 12 tLJ2b3 :e8
continue 6 ... a6 followed by 7 ... bS. The
idea of taking on c4 with the knight was
successfully employed in the late 1 93 0s by
the Latvian master Vladimir Petrov. If, for
example, 6...a6 7 tiJxc4 bS , then 8 "'c2
:b8 9 tLJas .i.b4+ 1 0 .i.d2 with the better
chances for White.
22 CiJ t'S?
The knight has to move, as 22 l:tad 1
.i.xd4 23 l:txd4 CiJb5 is completely bad.
But it should have moved so as to cover
the approaches to the white position.
Therefore 22 CiJe2! was the only correct
The s ituation has become extremely continuation, and if 22 .. .CiJb5 23 'it'c2! It is
sharp. Just as in an aerial battle the oppos true that in this case by 23 . ...i.xf2+ 24
ing fighter planes rush headlong towards l:txf2 'it'xa5 Black would have achieved a
each other, so here the adversaries, without good game, but that is all. Now, however,
deviating, go in for a sharpening of the his attack becomes threatening.
struggle. 22 CiJbS
Mikenas was a player of active, com 23 'it'c2 .i.xt'S
binational style. With the following blow, 24 ext'S CiJd4
prepared in advance, he was hoping to The black pieces rush forward, and the
refute the opponent's plan. white queen still has to worry about the
19 .i.aS defence of its knight, while in the mean
Black appears to be in a critical time the main events are developing on the
position. If, for example, 1 9 ... axb3 20 opposite wing.
22 Averbakh 's Selected Games
25 .xa4 28
There is nothing better. The culm ination of Black's attack! His
25 lLle2+ queen approaches the enemy fortress with
26 � h2 �xtl decisive effect. Mate in three moves is
27 "'c2 threatened, and White can defend against it
only at the cost of heavy loss of material.
29 :xtl .xg3+
30 �hl .h4+ !
31 �h3 .xtl
The threat of 3 1 ...lLlg3 mate can be
parried only by giving up the queen.
Therefore White resigned.
Perhaps the most interesting feature of
this game is that the black knight, which
created the deadly threat of mate at g3 ,
made a truly heroic journey: b8-d7-b6-d5-
c7-b5-d4-e2, whereas its opposite number,
on reaching as , was stuck there to the end
If 27 ...•e3 White was intending 2 8 of the game .
•e4, including h i s queen i n the defence of
the kingside. Analysis shows that in this Game 4
case too, by replying 28 ... "'xg3 + 29 �hl Averbakh-Zak
h5! with the threat of 30 ...lLlg4!, Black 2nd Match Game
retains winning chances, although with 3 0 Moscow 1947
.xe2 lLlg4 3 1 .xe8+ �h7 3 2 "'b8! White Ruy Lopez
gains a chance to prolong the resistance:
32 ... "'xb8 33 hxg4 .g3 34 :xf2 .xf2 3 5 In 1 946 the Al l-Union Chess and Draughts
b4 "'b2 3 6 :n .xa3 3 7 lLlxb7 "'xb4 3 8 Section granted the right to a number of
lLld8. However, Black has a stronger reply. candidate masters, who had gained good
27 lLlg4+! results, to play qualification matches for
28 hxg4 the master title. It was mainly young
If 28 �h 1 lLlxg3 mate. masters who were chosen as examiners -
Petros ian, Taimanov, and myself. Not one
of the hopefuls managed to become a
master, which in general was not sur
prising: we were al l progressing, rapidly
developing in the chess sense, and within a
few years were to become grandmasters.
It fell to me to play a match with
Vladimir Zak, an experienced player and
well-known trainer, who was later to coach
Spassky and Korchnoi.
There was an interesting prelude to this
match. Towards the end of 1 946 I took part
in the Sem i-Final of the 1 5th USSR
Averbakh's Selected Games 23
Championship, which was held in Lenin therefore the innovation I employed was
grad. In the first round I had White against for him a strong psychological blow,
Semyon Furman, and in reply to 1 e4 he which to some extent influenced the final
chose the Open Variation of the Ruy outcome.
Lopez. Sitting at the board and considering 1 e4 eS
my next move, I noticed that the same 2 lDfJ lDc6
position had been reached in the Kopaev 3 �bS a6
Levenfish game. I had heard that Leven 4 �a4 lDf6
fish had prepared for this tournament 5 0-0 lDxe4
together with Furman, and the idea 6 d4 bS
occurred to me of delaying my move a 7 �b3 dS
l ittle, in order to see what plan the two 8 dxeS �e6
Leningrad players had prepared. And so I 9 c3 �e7
began waiting. And my clock ticked. 10 �e3
Kopaev meanwhile was sunk in thought. Th is old move, which had been
Ten minutes passed, then twenty, and still employed by Alekhine, used to occur com
Kopaev did not make a move. He final ly paratively rarely, and did not bring White
did so only after 40 m inutes (II), and more good results.
over he deviated from the main theoretical 10 0-0
continuation and won the game. But I 11 lDbd2 lDxd2
followed the theoretical path, fell into a 12 .xd2 lDaS
prepared variation, and as a result lost, 13 �c2 lDc4
overstepping the time limit on the 3 8th 14 .d3 g6
move. And in subsequent games, when in IS �h6 lDxb2
the Ruy Lopez 4 ...lDf6 was played against A theoretical position, given in many
me, I repl ied 5 d3 , not allowing the capture books, has been reached.
on e4.
Zak was present at the tournament and
it was there that he challenged me to a
match. He undoubtedly made a careful
study of my games and, of course, would
have noticed how I had lost helplessly to
Furman and how I had subsequently
avoided the Open Variation.
Before the encounter with Zak I
decided to look at my recent games, in
order to imagine to myself what sort of
impression my play would make on my
future opponent. Of course, it immediately
became clear that in the first instance I 16 .e3 ! ?
needed to improve my opening preparation The opening books considered only 16
in the Open Variation of the Ruy Lopez. .e2, which threatens to regain the pawn,
And this was done. but diverts the queen from participating in
Zak clearly did not expect me to the attack on the kings ide. The plan with
prepare seriously for this match, and 16 .e3 was specially prepared by me for
24 Averbakh's Selected Games
the forthcoming match. I do not think that I would have had to reply 1 9 lLlxe6 fxe6 20
it leads to an advantage for White, but f5, again with sharp play.
certainly it seriously sharpens the play, 19 .g3 c5
which in combination with the surprise
factor plays a considerable role.
16 :'e8
If 1 6...lLlc4 1 7 "'f4 :'e8 I was pro
posing to play 1 8 lLlg5.
17 lLld4
White has created two threats. One is
obvious - to exchange the opponent's
dark-square bishop by 1 8 lLlc6, while the
other is a latent one - to begin a pawn
attack on the kingside with 1 8 f4.
If Black disregards the first threat and
plays 1 7...c5, there follows 1 8 lLlc6 "'d7
1 9 lLlxe7+ .xe7 20 �g5 .fS 2 1 �f6 White's attack can no longer he halted.
followed by 22 f4. During the game I con Now comes a forcing combination.
sidered the strongest reply to be 1 7 ...�fS, 20 f5 cxd4
when I was intending 1 8 �xfS :'xfS 1 9 11 fxg6! hxg6
.h6, and if 1 9 •d7 20 f4 f5 (if 20 ...c5
. . . Or 2 1 ...fxg6 22 �xg6 �g4 23 �f5.
2 1 f5!, threatening 22 f6) 2 1 g4 :'ae8 22 22 �xg6 <it>h8
gxf5 �xf5 23 lLlxf5 gxf5 24 �h I with an Black was hoping for 23 �x.f7 :'g8 24
attack on the g-file. However, Black could �xg8 :'xg8, but White had prepared in
reply 1 9 ...c5!, when after 20 �xg6 hxg6 advance the concluding blow!
2 1 lLlxe6 fxe6 22 .xg6+ <it>h8 it is doubt
ful if White has more than perpetual check.
But Black's genuinely strongest reply,
which I noticed only when preparing this
book, was 1 7 ...�d7, when White still has
to demonstrate that his initiative is worth
the sacrificed pawn.
17 .d7
Guided by the spectacular fin ish, the
commentators considered this natural
move to be virtually the decisive mistake.
In fact even after it Black has quite
adequate resources for parrying the attack.
18 f4 lLlc4? 23 �g7+! �xg7
Only now, when Black helps the queen 24 �xt7+ �h8
to switch with gain of tempo to an attack Or 24 ...�g4 25 e6.
ing position, does White's attack become 25 "'g6 �f8
threatening. It was essential to play 1 8 ...c5, If 25 ...�g5 White has a mate in four
and if 19 f5 cxd4 20 cxd4 �xf5 2 1 �xf5 moves with 26 "'h5+ �g7 27 "'xg5+.
"'a7, gaining time for the defence. Perhaps 26 .g8 mate
Averbakh 's Selected Games 25
board, out of the comer of my eye I saw 9 'it'c2 a6 1 0 b3, and it is far from easy for
that the last few seconds on my clock were Black to equal ise.
ticking away, and I feverishly made the 8 a6
first move that came to hand. It turned out 9 l:td l 'it'e7
that the typist had missed out one move on 10 i.dl dxc4
the scoresheets! The move I made was a 11 i.xc4 b5
bad one, and as a result I lost a pawn, 12 i.d3 i.b7
although I was able to save the draw. 13 a3
Game S
Taimanov-Averbakh
Trades Union Team Championship
Leningrad 1947
Nimzo-Indian Defence
1 d4 lDf6
1 c4 e6
3 lDc3 i.b4
4 e3 0-0
5 lDf3 d5
6 i.d3 c5
7 0-0 lDbd7 13 cxd4
This move was first employed in the Hardly a good move: now Black gets
game Ravinsky-Ragozin, Moscow 1 942. into difficulties. He should have retreated
The idea of it is to retain both the dark his bishop to as, and if 14 lDdS lDxdS IS
square bishop, and the tension in the i.xaS cxd4 with roughly equal chances.
centre. The game Krogius-Bondarevsky from
8 'it'el the 17th USSR Championship Semi-Final
26 Averbakh 's Selected Games
(Leningrad 1 948) went 1 3 ....i.aS 1 4 :lacl Having convinced himself that the
.i.b6 I S .i.b 1 :lac8 1 6 lbeS cxd4 1 7 exd4 direct attack on the king by 2 1 "'c2 lbxeS
.i.xd4 1 8lbxd7lbxd7 1 9 .i.xh7+ �xh7 20 22 .i.xf6 lbg6 23 .i.eS lbxe5 24 dxeS
'ifd3 + �g8 2 1 'ifxd4 lbcs with advantage .i.xeS 25 "'h7+ �fS does not promise
to Black. anything, White tries to clear the situation .
14 exd4 .i.d6 However, the move played is a mistake. 2 1
IS .i.gS :lcd 1 , maintaining the tension, was correct.
A very unpleasant pin. 1 6 lbe4 is
threatened, as well as 1 6 lbeS , and in
addition White can carry out the standard
plan with 1 6 .i.c2 and 1 7 "'d3 . Black must
now play very precisely.
IS h6
16 .i.h4 'ife8
17 lbeS 'ifb8
21 .i.xe4
This secures Black a positional advan
tage. After 2 1 ....i.xeS 22 dxeS lbxe4 23
.i.xe4 "'xeS 24 .i.h7+ �h8 (24 ...�xh7 25
"'d3 +) 25 'ifxeS lbxeS 26 :lxeS �xh7 he
would have won a pawn, but the opposite
colour bishops would have seriously
hampered its realisation.
By intensifying the pressure on eS , 22 .i.xe4 .i.xeS
Black provokes 1 8 f4, after which the play 23 dxeS lbxe4
becomes markedly sharper, for example: 24 "'xe4 %:tc7
1 8 ...:le8 1 9 .i.c2 .i.e7!, and if 20 'ifd3 I 25 f4 :lxcl
would be able to reply 20 ...lbfS. However, 26 :lxc l %:tc8
by continuing 20 fS White would retain a 27 :ld l
dangerous initiative. Black's chances are also better after 27
Taimanov prefers a less committing :lxc8+ 'ifxc8, as queen and knight are
plan. nonnally stronger than queen and bishop.
18 :leI :a7 Here the bishop is restricted by its own
19 :lac l pawns, whereas the knight has a number of
After 1 9 .i.c2 Black was intending to strong-points (dS , fS). In addition the
play 1 9 ...:le8, and if 20 "'d3 lbxeS 2 1 invasion of the queen on the c-file is
dxeS .i.xeS 22 .i.xf6 .i.xf6 23 "'h7+ �fS. threatened.
19 .i.a8 27 "'c7
20 .i.b l :e8 28 h3 lbfS
21 lbe4 29 :ld6
Averbakh 's Selected Games 27
31 'iffl.!
The murderous 32 ... l:c 1 is threatened,
so that the opponent is forced to exchange In this position the game was ad
queens. I should mention that the next part journed. Analysis showed that, despite
of the game, right up to move 40, took Black's extra pawn, White's chances are
place in a severe time scramble. better, since the knight is restricted and
32 'ifd3 'ifxd3 White is threatening to win it. However, I
33 :'xd3 l:c2 was able to find a clever trap, and much
This allows the opponent good counter depended on whether my opponent had
chances. Correct was 33 ... h5!, aiming to seen it.
restrict the bishop sti ll further, for The reader will no doubt have noticed
example: 34 h4 tlJe7!, and after 3 5 l:d6 how greatly the character of the position
l:c2 3 6 b4 tlJf5 3 7 l:xa6 tlJe3 , in view of has changed during the last ten moves.
the threat of 3 8 ... tlJfl +, Black has a clear Whereas earlier the knight was stronger
advantage. than the bishop, now, on the contrary, the
34 :'d8+ �h7 bishop has become stronger.
35 l:d7 �g8 42 �tl
36 l:d8+ �h7 The sealed move.
37 :'d7 l:xb2 42 l:a4!
Throwing caution to the winds. Switching the rook to a more
38 l:xf7 l:b3 favourable position with gain of tempo.
39 h4! �g8? 43 �g3
Here I was obliged to play 39 ... h5 40 fS If 43 g3 , then 43 . . . l:a2 44 �gl b4.
exfS 4 1 :'xfS l:xa3 42 l:xh5+ �g8, when 43 b4
Black's chances are at any rate no worse. 44 �c5
Now, however, the advantage passes to This move slightly raised my hopes.
White. After 44 l:b7! b3 45 l:xb3 l:c4 46 l:b7 a5
28 Averbakh 's Selected Games
82 �e3
82 . . . �f3 was simpler.
83 :gl :xh5
84 :g3+ �d4
85 :a3 :h2
86 �f5 :fl+
87 �g4 �c4
The time trouble mist, in which the two 88 �g3 :c2
players have been wandering about, has 89 :a8 �b4
finally lifted. The black king has been 90 �f4 :e2
forced to return, and after 73 �d4! this White resigns
Averbakh's Selected Games 31
Despite the numerous mistakes, or per active pieces and transposing into an
haps because of them, this tense game, endgame.
interesting in all its phases, caused great
anxiety both to the participants themselves,
and to the members of their teams. The
rook ending that arose turned out to be full
of various subtleties, which the players,
particularly in time trouble, were unable to
fathom.
Game 6
Averbakh-Veresov
Semi-Final 1 6th USSR Championship,
Moscow 1 94 7
Sicilian Defence
11 �5 e6
1 e4 c5 12 ll)de3 'iWh4
2 ll)f3 �6 Dreaming of an attack, Black switches
3 d4 cxd4 his queen to the kingside. Of course, it
4 ll)xd4 g6 could have been driven away immediately
5 c4 by 1 3 g3, when 1 3 . . . 1Ifh3 is bad because of
At that time the opening books asserted 1 4 Jtg4, but I considered that at h4 the
that allowing c2-c4 gave Black a bad queen did not present any danger.
game, as he had no way of opposing 13 l:tbl
White's strong centre. In the present game Preparing 14 b3 and 1 5 Jtb2 with
my opponent carries out an unusual plan, exchanges to follow.
devised by the Swedish player G6sta Stoltz 13 ll)cd4
(who was later to become a grandmaster), Falling in with White's intentions.
involving attacking the enemy centre with 14 ll)xd4 ll)xd4
. . . f7-fS . 15 b3 h5
5 Jtg7
6 ll)cl
For me Black' s idea was a new one,
and so I decided to move the knight out of
harm ' s way.
6 d6
7 Jte2 ll)h6
8 ll)c3 0-0
9 0-0 fS
10 exfS ll)xfS
On encountering an innovation, I have
played cautiously, and my opponent has
obtained an active position. However, his
pawn structure has been weakened, and so Against the immediate 1 5 . . . Jte5 White
I devised a plan aimed at exchanging his had the reply 1 6 ll)g4. Continuing to think
32 Averbakh 's Selected Games
39 bS
Analysis showed that Black would also
have failed to save the game after 39 . . . .i.a8
40 .i.d7 .i.b7 4 1 b4 axb4 42 axb4 .i.a8 43
c5 bxc5 44 bxc5, and now: At first sight it is not clear how White
(a) 44 . . . .i.d5 45 .i.e8 �f5 46 c6 ! (the can win here. After all, as soon as he
bishop sacrifice 46 . . . .i.xc6 47 .i.xc6 also advances his passed pawn, Black will give
does not save Black, e.g. 47 . . . g5 48 up his bishop for it, which leads to a
.i.xe4+ �g4 49 hxg5 �xg5 50 �n h4 5 1 theoretically drawn ending. This would be
g4, or 47 . . . �g4 48 �f2 e3+ 49 �g2, and so, if the black pawn were at h6. However,
49 . . . g5 is not possible because of 50 .i.d7 thanks to the fact that it is on a square of
mate) 46 . . . �g4 47 c7 .i.e6 48 .i.xg6 �xg3 the colour of his bishop, White can win.
49 .i.xh5 �xh4 50 .i.f7! .i.c8 5 1 �xe4 The diagram position virtually constitutes
�g5 52 �e5 .i.a6 53 .i.e6 .i.b7 54 .i.h3 a study!
i.a6 55 �d6 �f6 56 �c6 �e7 57 �b6 46 .i. a4
�d6 58 .i.g4 and wins; If 46 ... .i.e8 47 i.n �f5 48 .i.e2 �e5
(b) 44 . . . �d5 45 .i.e8 g5 (if 45 . . . �xc5 49 .i.d3 �d5 50 �f4, and after 50 . . . �d4
46 .i.xg6 �d6 47 .i.xh5 �e5 48 .i.g6 .i.c6 White can even play 5 1 .i.e2 <ifi'c3 52
49 g4 and wins) 46 hxg5 �xc5 47 .i.g6 ! .i.xh5 ! .i.xh5 53 b5 .i.e8 54 b6 .i.c6 55 h5
(47 g6? �d6 with a draw) 47 . . . i.d5 48 and wins.
i.xe4 .i.g8 49 �f4 �d6 50 �f5 �e7 5 1 47 .i.g6 .i.d l
�g6 and White wins. 48 bS �dS
40 exbS .i.xbS 49 �f4 �eS
41 .i.e8 .i.e6 SO �gS .i. e2
42 b4 axb4 If 50 ... �xb5 White wins by 5 1 i.xh5
43 axb4 .i.bS i.c2 52 .i.e8+ �c5 53 h5 �d6 54 �f6 .
44 .i.b7 ! gS! SI .i.e8!
34 Averbakh 's Selected Games
It was not yet too late to go wrong. continuation is 7 �e3 . At that time I did
After 5 I �xh5? �xb5 52 �g4 i.e8 53 not yet have a reliable antidote to the
�fS �d6 54 �g6 �e7 Black would have King's Indian Defence, and I tried the most
saved the game. varied plans.
51 �b6
Regrettable though it is, the black king
has to move even further away from the h-
pawn.
52 � xh5 � xb5
53 i.g4 �e8
54 �fS �c7
55 � g6 �d8
7 �b8
8 �e2
8 �d3 , taking control of e4, is more
active.
8 e6
9 0.-.. exd5
10 exd5 :le8
11 � e3 �bd7
56 �f6! 12 %tel � c5
The final finesse. 56 h5 or -5 6 �xe8 13 �d4 �fe4
would have led only to a draw.
Black resigns.
Game 7
Averbakh-Lilienthal
Moscow Championship 1948
King's Indian Defence
1 d4 �f6
2 c4 d6
3 �f3 g6
4 �c3 �g7
5 e4 0.-..
6 h3 �c6 White has played the opening most
7 d5 unpretentiously, and, apart from a slight
This continuation is justified if Black superiority in space, he has not gained any
replies 7 �e5, as after 8 �xe5 dxe5
... advantage.
White's position is preferable. The usual 14 �xe4 �xe4
Averbakh 's Selected Games 35
29 d5?
By playing his queen to h5, Black has Only now does White gain real winning
created the threat of . . . J.xb3 . However, chances: with the weakness at c6 the
the queen is not altogether well placed superiority of the knight over the bishop
there, and White could have tried to exploit increases.
this by 22 l:xe2 ! and if 22 . . . J.xb3 23 30 c5 �d7
J.xg7. For example, if 23 . . . J.xg2 24 �xg2 31 �d4 f6
.g4+, then 25 �h2 .xfl 26 J.c3 .h5+ 32 l:e2 J.f7
27 �g3 f5 28 fl and wins, or 23 ...•g4 24 White's plan is to restrict the bishop as
�e l �xg7 25 l:e3 and wins. Final ly, if much as possible and then launch an
23 . . .�xg7, then 24 .c3+ and 25 gxh3 . offensive on the queens ide.
But now the game goes by force into an 33 f4 l:g8
ending. 34 g3
II J.xd4 I was intending to reply to 34 . . . g5 with
Of course, not 22 . . . l:e8 23 .xe8+ 35 f5, and if 35 . . . g4 36 h4, not allowing
J.xe8 24 l:xe8+ J.f8 25 J.f6! .f5 26 J.e7 the opponent to open lines .
•d7 27 l:xf8+ �g7 28 �g5 and wins. 34 h5
23 �xd4 Wxe2 35 l:e3 l:e8
36 AverbaJch 's Selected Games
Game 8
Kotov-Averbakh
1 6th USSR Championship
38 rJilc7 Moscow 1 948
Here Black should have played 3 8 ... g5 ! , Queen 's Gambit
when after 39 fxg5 fxg5 4 0 lil o rJile6 ! 4 1
lLlxg5+ rJile5 he has good drawing chances. 1 c4 lilf6
I was planning to reply 39 £5, but in this 2 lilc3 e6
case too Black could have put up a tena 3 lilfJ cS
cious defence. Now, however, he is lost. 4 e3 dS
39 rJilg3 �d7 S cxdS lilxdS
If 39 . . . g5 there follows 40 fxg5 fxg5 4 1 6 d4
lile6+, winning a pawn. The game has transposed into a position
40 gS fxgS from the Queen's Gambit, which is
White also wins easily after 40 . . . f5 4 1 normally reached after 1 d4 lilf6 2 c4 e6 3
rJilo as 42 rJile3 rJilb7 43 rJild3 rJila6 44 a4 lilc3 d5 4 lilo c5 5 cxd5 lilxd5 6 e3 . This
rJilb7 45 lilO rJilc7 46 rJild4 rJild8 47 rJile5 move order has enabled my opponent to
rJile7 48 lild4 �e8 49 lLlc2 followed by b3- avoid the Nimzo-Indian Defence, which I
b4. often employed at that time.
41 fxgS �c8 6 �e7
42 rJilf4 7 �c4
Here the game was adjourned. A In this way White forces the opponent
straightforward analysis showed that Black to declare his intentions in the centre.
is unable to prevent the enemy king from Nowadays 7 �d3 is more usual.
breaking into his position. 7 lilxc3
For example: 42 . . . a5 43 rJile5 �g4 (or 8 bxc3 0-0
43 . . . �a6 44 rJilf6 �d3 45 rJile7 followed by 9 0-0 lild7
46 lile6+ and 47 rJild6) 44 rJilf6 �h5 45 10 e4 "c7
lile6+ rJild7 46 lilf4. 11 �d3
Averbakh IS Selected Games 37
39 :al
40 :d l
41 �hl
42 :gl
Here I could not restrain myself and I A rarely played move, examined in
declared: 'It is mate in five moves ! ' Sovremenny Debyut* ( 1 940) with the
45 "'r.H indication that after 1 0 d4 exd4 1 1 cxd4 a4
46 l:tg2 lb a+ 1 2 .tc2 lbb4 Black gains the advantage of
47 �gl l:te1+ the two bishops, and that White should
White resigns. After 48 .tfl lbh3+ 49 therefore play either 1 0 a4 or 1 0 d3 .
"'xh3 there follows mate by 49 . . . "'xfl or 10 d4 exd4
49 . . . l:txfl . 11 lbxd4
After signing the scoresheet, my In this way the above variation is
opponent commented sarcastically: avoided, but Black' s game develops easily
'You and your mate! There was a much and freely.
simpler one, without any tricks and in three 11 lbxd4
moves ! ' Indeed - 45 . . . lbf2+ 46 �g2 0+ 12 cxd4 .tb7
47 �fl "'e l mate. 13 .tel
After this episode I never again After 1 3 lbc3 b4 1 4 lbd5 lbxd5 1 5
announced mate aloud! .txd5 .txd5 1 6 exd5 the position would
have become too dull, but now Black
Game 9 launches an immediate attack on the
Averbakh-Bondarevsky opponent's pawn centre.
16th USSR Championship 13 c5
Moscow 1 948 14 lbd2 cxd4
Ruy Lopez
* The first volume of a planned three
1 e4 e5 volume work on opening theory, edited by
2 lbo lbc6 Levenfish. The other two volumes were
3 .tb5 a6 never published (Translator's note).
Averbakh 's Selected Games 41
15 liJa
If 1 5 b3 Black had the reply 1 5 . . . liJd7 !
1 6 .ib2 .if6 with the threat of . . . d4-d3 .
15 d5
16 e5 liJe4
17 liJxd4
29 �h2 �xg2
30 :gl
Why didn't White take the d4 pawn
with his queen? Then there could have
followed 30 . . . �xh3 ! 3 1 �xh3 :h6+ 32
�h4 "'f3+ 33 �h2 "'f4+ 34 1It'xf4 :xf4,
and Black has every chance of winning.
23 �xe4 �xe4
24 �g5!
This is the point! After 24 . . ....xg5 25
"'xe4 White has nothing to fear. Now,
however, Black could have sacrificed his
queen with 24 . . . �xg2, when the game
could have concluded in perpetual check:
25 �xe7 �f3+ 26 �h2 ! :g2+ 27 �h l 30 �xh3 !
:g6+ 28 �h2 (28 "'xf3 :xf3 29 �h2 31 � xh3 :h6+
:xf2+ 30 �h l :gg2) 28 . . . :g2+. In severe time trouble B'ondarevsky
24 "'f1 forces matters.
Bondarevsky tries to maintain his fad 32 1It'xh6
ing initiative, and thanks to White's next If 32 �h4 "'d7+ 33 �g3 (3 3 :g4
move he succeeds in this. :f3+) 33 . . ....f5 34 "'g5 :g6 and wins, or
25 e6? 34 "'xd4 "'f3+ 35 �h2 1It'f4+, and Black
An obvious oversight: I forgot that the remains a pawn up.
pawn could be taken by the rook. Mean 32 "'c8+
while, by continuing 25 "'xe4 ! "'xf2+ 26 33 �h2 gxh6
�h l l:lxg5 27 :e2 (also good is 27 :fl 34 �d6+ �f1
"'xfl + 28 :xfl :xfl + 29 �h2, and in this 35 �xfB "'xfB
ending White has every chance of success) 36 :g3
27 . . ....f4 28 "'d5+! �h8 29 "'xb5 Whj�e In this ending White can offer a resis
would have won a pawn. tance only if he succeeds in coordinating
25 :xe6 his forces. This is not an easy task.
26 � h4 �d5 36 �e6
27 "'d2 37 :el+ �d5
If 27 "'xb5 :g6 28 �g3 Black has the 38 �gl
reply 28 . . . :xg3 ! Now White loses another pawn, but
27 :g6 even after 38 :e2 "'f4 39 �g2 h5 the
28 �g3 "'b7 defence would have been extremely
Black's attack bears fruit - the g2 pawn difficult.
cannot be defended. 38 "'b4
Averbakh 's Selected Games 43
theoretical disputes, where often the game transpires that it is difficult for Black to
is decided not by who plays better, but by develop his queenside play, whereas
who knows more. Therefore I preferred not White's threats on the kings ide become
to allow Black' s central freeing advance. very real . With the move played I prevent
13 d5 1 8 . . . b4, which would be answered not by
1 9 Wxc4 .i.a6, but by 1 9 .i.d3 lLlb6 20 c4,
when after blocking the position on the
queens ide White can conduct his attack on
the kingside without distraction.
18 lLla5
1 8 . . .lLlb6 came into consideration, when
White was intending to play 1 9 a3 (but not
1 9 .i.d3 c4 20 .i.c2 a3) 1 9 . . . .i.a6 20 .i.d3 .
19 :gl b4!
has no defence. If, for example, and switch completely to the realisation of
3 1 . . . 1i'xf2+ 32 :g2 1i'f4+ 33 1i'xf4 exf4, his extra pawn, which in this complicated
both 34 �f6 and 34 :ag l are possible. position, where Black has the two bishops,
Of course, all these variations do not is not an easy matter.
exhaust the possibilities in the position, but
they vividly demonstrate just how strong
White's attack could have been.
20 .i.xa4 bxc3
21 bxc3 .i.d7
Szabo reckons that after 22 .i.xd7
li)xd7 23 .i.e3 �b6 he will seize the
initiative on the queenside. And yet this
move is not the best. By continuing
2 1 . . .�d7 22 �f5 .i.f8 23 .i.c2 .i.a6 24
'iWd l �b6 he could have achieved a
position in which White's extra pawn is of
no great significance.
22 .i.c2 �b3 26 g6
27 �d2 :al
This eases White's task. Since b3-b4 is
not yet threatened (the c3 pawn is
hanging), Black should have played
27 . . . .i.e8, and if 28 1i'c4 �d7, when by the
threat of 29 . . . �b6 he again prevents the
advance of the b-pawn.
28 b4 Wc8
29 1i'c4 :a4
Game 1 1
Averbakh-Goldberg
Semi-Final 18th USSR Championship
Tula 1 950
Dutch Defence
1 d4 e6
2 lD a f5
47 b5! 3 g3 lDf6
Less good is 47 i.xd4 .xb4 48 i.xe5+ 4 i.g2 c5
dxe5 when, in view of the insecure In combination with . . . t7-f5 this
position of the white king, Black retains advance can hardly be recommended, as it
chances of a successful defence. merely leads to a weakening of the pawn
47 lDc2 structure.
On 47 . . ..b6 White wins a second pawn 5 c4 cxd4
by 48 l:ta4 ! , as if 48 . . .. xb5 49 l:txd4. But 6 lDxd4 i.e7
now the game concludes very quickly. 7 lDc3 a6
48 :a7 .b8
49 b6 lDxe3
50 .xe3 :c2
8 � jlc7 12 d6
9 e4! 13 -*.gS �d8
The natural reaction to the opponent' s 13 . . . �d7 is met by 14 llxe7+, while if
slow plan. White aims to open u p the 13 . . . �f1 or 13 . . . 0--{) , then 14 �d5 is
position. If 9 . . . fxe4 he has a choice decisive.
between the solid 1 0 jle2, and the sharp 1 0
�xe4, sacrificing a pawn for the initiative.
9 g6?
The decisive mistake. Black tries to
prevent the opening of lines, but he
achieves directly the opposite result.
10 ext'S gxt'S
1 0 . . . e5 1 1 �c2 gxf5 1 2 �e3 is also
unfavourable.
14 llxe7! jlxe7
15 �dS � xd S
16 -*.xe7+ �xe7
17 jlxd6+ �d 7
18 lld l !
11 �xt'S! ext'S
12 l:el
This quiet move is the point of White ' s
play. How is Black to defend against the
numerous threats?
1 2 . . . �d8 can be met by 1 3 llxe7 ! �xe7
1 4 -*.g5 when he is unable to avoid great
loss of material. If instead 1 2 . . . h6, averting
the threat of -*.g5 , then 1 3 -*.f4 jld8
( 1 3 . . .1Ifxc4 1 4 llc l ) 1 4 �d5 �xd5 I S An amusing position has arisen. There
1Ifh5+ �f8 1 6 -*.xd5. Also no better is is no defence against 1 9 -*.xb7, since
12 . . .�c6 1 3 -*.f4 jla5 ( 1 3 . . . d6 14 -*.xd6 1 8 . . . lla7 is met by 1 9 j1b6+.
jld7 I S -*.xc6 1Ifxc6 1 6 -*.xe7) 1 4 �d5 18 llg8
�d5 I S cxd5 �d8 1 6 jlh5+ �f1 1 7 d6 19 -*.xb7 l:g6
etc. 20 jld4 �c6
My opponent finds a way to give up his 21 jlb6+ �e8
queen for three pieces, but this merely 22 -*.xc6
delays the end somewhat. Black resigns
Averbakh 's Selected Games 51
21 lLle2 :'d6
22 :b6 :'h6
At the cost of a pawn my opponent
aims to gain time for the attack. To
22 . . . i.d7 I was intending to reply 23 lLlg3 ,
and if 23 . . . :h6 24 i.f5 ! 'ii'h4 25 h3 .
23 :'xc6 lLlg6
If 23 . . . :ee6, then 24 :xe6 fxe6 25
.c7 ! is good enough, covering h2. But
All this has been played many times. now Black has two threats - 24 . . . 'ii'h4 and
White plans a m inority pawn attack on the 24 . . . lLlh4.
queenside. Modem theory considers 1 1 24 lLlgl !
i.xf6 i.xf6 1 2 b4 to be more effective.
11 i.g4
12 lLleS i.hS
13 b4 a6
14 a4 lLlg4
15 i.xe7 .xe7
16 lLlxg4 i.xg4
17 bS axbS
18 axbS .gS
19 �hl :'ad8
20 bxc6 bxc6
The two players have consistently
carried out their opening plans. By opening
52 Averbakh 's Selected Games
Game 1 3
Averbakh-�o�v
Moscow Championship 1 950
Sicilian Defence
1 e4 cS
2 �e2 e6
3 d4 cxd4
4 �xd4 �f6
29 i.b3 ! 5 �c3 d6
To 29 g3 Black had the reply 29 . . . i.g6. 6 J.. e2 J.. e 7
29 :xe3 7 0-0 0-0
30 i.xdS .i.g6 8 f4 a6
31 :f4 �f5 9 �hl .c7
32 h4 �e7 10 .el �c6
After 32 . . . :d3 33 �e2 :d2 34 h5 ! 11 J.. e3 J.d7
.i.xh5 (34 . . . :xe2 3 5 hxg6 fxg6 36 g4) 3 5 12 :d l bS
lhfS Le2 36 .i.xf7 White also wins. 13 a3 :ac8
33 �h2 �d6 14 .g3
Averbalch 's Selected Games 53
Game 1 4
Golovk�Averbakh
Moscow Championship 1950
Catalan Opening
1 d4 lDf6
2 c4 e6
3 g3 Jlb4+
4 Jld2 Jle7
The idea of this manoeuvre, which used
to be employed by the Moscow master
Nikolai Ryumin in the Dutch Defence, is
to force the white bishop to occupy the
27 "a7 poor square d2.
By invading the opponent's position, 5 Jlg2 d5
White exploits the lack of coordination of 6 lDf3 lDbd7
his pieces. 7 0-0 0-0
27 Jlc8
28 "b8 �g8
After 28 . . . lDxf4 29 "xf4 �g8 30 Jlf3 !
the rook has no good retreat square.
29 Jlf3 l::tc 4
30 Jld6 lDxc2
This accelerates Black' s inevitable
defeat. The pawn, which for many moves
has been en prise, is also immune now.
31 JlxfS lDxfS
32 "b3 lDe3
8 "c2
Here 8 lDc3 would have lost a pawn.
8 c6
9 a4
Yevgeny Zagoryansky, an expert on
this opening, recommended 9 :d 1 here,
followed by 1 0 Jle 1 and 1 1 lDbd2.
9 a5
10 lDc3 b6
11 cxd5 lDxd5
Not fearing 1 2 e4 because of 1 2 . . . lDb4
33 lDd l ! l::td and 1 3 . . . Jla6.
34 "xe3 :al 12 lDal
35 h3 White defends against the threat of the
Black resigns knight move to b4 and prepares e2-e4, but
Averbakh 's Selected Games 55
now Black succeeds i n playing . . . c6-c5 defence would probably have only led to a
and firmly seizes the initiative. draw.
21 b3 -*.xb3
22 1Wxb3 :&3
23 1Wc4 :xf3
24 1Wg4
12 -*.a6
13 :fel c5
14 e4 lDb4
15 lDxb4 axb4
16 d5 exd5 It was on this - that my opponent
17 exd5 -*.16 would be tempted by the win of a piece -
18 :ad l b5 that I was counting. After 24 -*.xb4 cxb4
19 axb5 -*.xb5 25 1Wg4 lLle5 26 :xe5 -*.xe5 27 ..xn or
even 24 -*.xd7 1Wxd7 25 "xc5 it is
doubtful whether Black' s insignificant
advantage would have sufficed for a win.
24 :d3
25 "xd7 :xd5!
This is the whole point! The ill-starred
bishop at d2 comes under a mortal pin, and
even the threat of mate at e8 does not save
White.
26 1Wxd8 :fxd8
27 -*.g2
If 27 �fl I was intending to continue
not 27 ... -*.c3 28 Jlxc3 :xd 1 29 :xd 1
Black's pieces are very active. The :xd 1 + 30 �e2 :h l 3 1 Jle5, when White
immediate threat is 20 . . :a2 2 1 -*.c l -*.a4.
. can still put of a tough resistance, but 27 . . .
20 -*.h3 b3 ! 28 �e2 b2 . If now 29 -*.g2, then
By attacking the knight and threatening 29 . . . :xd2+! 30 :xd2 :e8+ 3 1 �fl
to win the c5 pawn, my opponent tries to :xe 1 +, while on 29 :fl there follows
stem Black's offensive. 29 . . . c4 ! 30 Jlg2 :e8+ 3 1 Jle3 :xd l 32
20 Jla4 :xd l c3, when White has no defence. It
Black begins a complicated should be mentioned that if he had played
combination, which against accurate 27 Jlg4, aiming to escape from the pin,
56 Averbakh 's Selected Games
of the pawns, Black wins) 45 . . .:a3 46 This game was awarded a special prize,
j.b5 f4 47 j.c6 g4 ! 48 j.d7 h5 49 j.e8 as the best in the tournament.
0+ 50 �d2 :a2+ 5 1 �e l :a l + 52 �d2
:fl 53 �e3 :e l + 54 �d3 �fS 55 j.xh5 Game 1 5
�g5 56 j.fl :e2 and Black wins. Averbakh-Suetin
38 h5 18th USSR Championship
39 j.d3 �g7 Moscow 1950
40 �e3 h4! Sicilian Defence
41 gxh4 :xh4
42 j.n �f6 1 e4 e5
43 �a :f4+ 2 fila file6
44 �g3 �e5 3 d4 exd4
45 j.g2 �d4 4 filxd4 filf6
5 lile3 d6
6 .i.g5 e6
7 "'d2 a6
8 0-8-0 h6
9 j.f4 j.d7
10 .i.g3
14
Black returns the compliment. He
hopes to intensify the pressure on the e4
pawn with 1 5 . . .... b7, but underestimates
the knight sacrifice that occurs in the
game.
15 "'d6 ! "'b7
If 1 5 . . ....xd6 1 6 ':xd6 j,xe4 there
would have followed 1 7 l£lxe4 l£lxe4 1 8
':d4 f5 1 9 f.t
19 j,fJ l:te8
20 "'g3 !
After 20 ':xe8+ l£lxe8 2 1 "'f8 Black
can gain counterplay by 2 1 . . . ':c8, and if
22 d6 j,fS ! 23 j,xb7 ':xc2+ 24 �b l ':d2+
with perpetual check.
20 g5
Now White wins very simply. Things
would have been more difficult for him
after 20 . . . ':xe l 2 1 ':xe l ':c8 22 "'xg7
16 l£ld5! "'b6.
Spectacular! The knight, which has
been standing en prise for three moves,
again offers to sacrifice itself.
16 exd5
The knight can be taken in any of three
ways, but none of them brings Black relief.
Thus if 1 6 . . . j,xd5 1 7 exd5 l£lxd5 I was
proposing to play 1 8 j,f3 , while if
16 . . . l£lxd5 1 7 exd5 j,xd5 1 8 ':he l , and
now 1 8 . . . j,xg2 1 9 j,h5 ! or 1 8 . . . ':d8 1 9
j,xa6.
17 exd5 j,d7
In reply to 1 7 . . . j,xd5 White has the
decisive 1 8 j,f3 j,xf3 1 9 gxf3 , while if 21 "'d6! l£lg8
1 7 . . . l£lxd5 he wins by 1 8 ':he l , when There is nothing better. If 2 1 . . .':xe 1 22
1 8 . . . l£le7 is met by the crushing 1 9 j,xa6! "'xf6+ ':e7 23 d6.
"'d7 20 ':xe7+ "'xe7 2 1 "'xc6+. 22 l:txe8+ �xe8
18 l:thel �d8 23 l:te1+ �d8
Here we can take stock. For the 24 "'18+ �c7
sacrificed piece White has two pawns and 25 d6+ �b6
a powerful attack. 26 "'xa8 "'xa8
60 Averbakh 's Selected Games
1 d4 dS
2 c4 e6
3 �c3 c6
4 �f3 dxc4
seen i n the variation 1 2 lDxgs? hxgs 1 3 The dangers awaiting Black are also
�xgs 'fIc7, and i f 1 4 f4 b4, while i f 1 2 illustrated by the following variation:
exf6 gxh4 1 3 lDes there follows 1 3 . . . "'xf6 2 1 . . .lDc3 22 lDxgs lDxe2+ 23 :xe2 hxgs
with a good game for Black. 24 Aae 1 and there is no defence. Thus if
12 � g3 lDdS 24 . . . 'fIxd4 there follows 25 Axe6 ! Axe6
13 lDe4 "'b6 26 'fIg8+, while if 24 . . . es 25 dxes fxes 26
The immediate 1 3 . . . cs was also AxeS ! :xes 27 1I'g8+.
possible, for example: 1 4 lDd6+ �xd6 1 5 However, when I began examining this
exd6 'ifb6 1 6 dxcs "'xcs 1 7 lDd4 0-0 . game for the present book, I came to the
14 lDd6+ � xd6 conclusion that the king should go not to
15 exd6 f6 d8, but to fB . In this case Black is able to
This move weakens Black' s position parry all the threats by the queen man
and creates the grounds for various tactical oeuvre . . . 'fId8-e8. For example: 1 8 . . . �fB
blows. Therefore I s . . . cs was more (instead of 1 8 . . . <i>d8) 1 9 lDes lDxes
accurate. ( 1 9 . . . fxes 20 �hs, and there is no defence
16 b4 ! ? against the mate) 20 dxes 'fId8 2 1 �hs
'fId7 22 exf6 'fIh7 23 �es 'fIxg6 24 �xg6
Ag8 25 �h7 <i>f7 26 �xg8+ :xg8 and in
this sharp ending Black has at least equal
chances. Or 1 9 lDh4 'fId8 20 �hs 'fIe8 2 1
'fIxe8+ Axe8 22 lDg6+ �g7 23 lDxh8
Axh8 24 :fe l b4 (24 . . . lDfB 25 �e8 lDb6
26 as) 25 l::txe6 c3 and Black's passed
pawns are extremely dangerous.
All these variations are so complicated
and difficult that, from the practical point
of view, perhaps I was right to decline the
pawn sacrifice.
16 cxb3
Petros ian offers a second pawn This exchange is forced, otherwise the
sacrifice, adding fuel to the fire. When I bishop will be shut out of the game.
annotated this game many years ago, I 17 'fIxb3 <i>f1
assumed that its capture would lead to 1 7 . . .g4 1 8 lDh4 'fIxd4 19 :fd 1 would
unclear complications and extremely sharp have led to an unnecessary sharpening of
play. For example: 1 6 . . . lDxb4 1 7 'fIb l lDds the play. I rejected queenside castling
(also after 1 7 . . . lDd3 1 8 �xd3 cxd3 1 9 because of 1 8 :fc 1 , and decided to leave
'fIxd3 0-0-0 2 0 Ilfc 1 it i s not easy for my king in the centre.
Black to disentangle his group of pieces) 18 l::tfe1 l::tae8
1 8 'fIg6+ <i>d8 1 9 'fIg7 l::te 8 20 as 'fIa7 2 1 1 8 . . . hs was possible, not allowing any
Afe l . tactics, but I provoked Petros ian into
Here Black has a material advantage, making the following sacrifice, thinking
but his pieces are extremely cramped and that, by giving up the exchange, I would
uncoordinated. If 2 1 . . .hs even 22 lDxgs obtain a position in which the knight at ds
fxgs 23 �xhs AfB 24 :xe6 is possible, would be worth a rook.
with a winning attack. 19 lDeS+
62 Averbakh 's Selected Games
21 .ta4!
Black plans 22 . . . l:ac8 and then
23 . . . bS . At first sight this looks risky
because of 22 bS, cutting off the bishop' s
retreat, but it i s easy to see that after
22 . . . �7 23 .te3 (23 .tb4 �b6 ! )
2 3 . . . �cS Black has a fine game.
22 "'bl
In defending against the threat of
22 . . . bS, on which there would follow 23
.td3 , White concedes the bS square
13 b4 without a fight. However, also after 22
A dubious move. My opponent does not .te3 bS 23 .tb3 l:xc I 24 .txc I .txb3 2S
want to allow the black bishop onto the a7- 1i'xb3 1i'c8 Black' s position would have
g I diagonal, from where it will threaten the been preferable.
12 square, but this allows Black to begin 22 l:ae8
active play on the queens ide. 23 .td3 .tb5
13 a5 24 l:xe7 l:xc 7
14 :bl axb4 25 .te2
15 axb4 1i'e7 If 2S l:c I , trying by exchanges to
16 1i'b3 h6 simplify the defence, very strong is
17 �b5 :re8 2S . . . l:xc I + 26 .txc l .txd3 27 1i'xd3 "'a4,
18 .td2 1i'e8 when White must lose one of his pawns.
19 �xd6 25 1i'e8
1 9 1i'd3 was probably more accurate. 26 .tb3
After this exchange the second black rook A serious mistake. In defending against
also comes into play. the opponent' s threats on the queenside,
19 exd6 White has placed his pieces so badly that
20 1i'd3 this allows Black instantly to create an
66 Averbakh 's Selected Games
attack on the kingside. After 26 .i.d I or �e l iDg2 mate) 30 . . . iDh3+ 3 1 �g2 iDdxf2
even 26 .i.d3 Black would have still have 32 .i.e2.
had to demonstrate that his positional Now, however, Black's attack wins by
advantage, consisting of his more active force.
pieces, could bring anything significant.
29 "'h3 !
26 iDh5 30 .i.a4
The black cavalry, which up to here has Desperation ! There was no defence
merely been observing the events taking against the threat of 30 . . . iDdf4.
place on the opposite wing, now comes 30 iDdf4
into play with great effect. 31 gxf4 "'g4+
27 g3 32 �hl "'xf3+
White tries to set up a pawn barrier, but 33 � gl iDh4
the knights easily overcome it. Could he White resigns
have managed without this weakening
move? If, for example, 27 .i.d I iDhf4 28 Game 1 8
.i.xf4 (or 28 :e3 iDxg2 ! 29 �xg2 iDf4+ Novotelnov-Averbakh
and 29 . . ...h3) 28 . . . iDxf4 29 h3 , then 19th USSR Championship
29 . . . :c l 30 "b3 iDd3 3 1 :fl iDb2 32 Moscow 1951
:e l (32 "xb2 .i.xfl ) 32 . . . iDxd l 33 :xc l Catalan Opening
"c4.
27 :e3 iDhf4 28 iDe 1 was the best 1 d4 iDf6
defence, but even in this case the invasion 2 c4 e6
of the knights by 28 . . . iDe2+ 29 �h l iDgf4 3 iD f3 d5
looks very dangerous. 4 g3 .i.e7
27 iDhf4 5 .i. g2 0-0
28 .i.d l iDd3 6 0-0 iDbd7
29 :0 7 "'c2 c6
The sudden attack has caught my 8 .i.f4 b6
opponent unawares, and he fails to offer a Here White does not achieve anything
worthy resistance. He could have put up a with 9 cxdS cxdS 1 0 .i.c7 "'e8 I I :c l .
tougher defence with 29 :e3 ! iDgf4 ! 30 There follows I l . . . .i.a6 1 2 e3 :c8, and
iDe l (3 0 gxf4 "'g4+ 3 1 �fl iDxf4+ 32 1 3 . . . iDcs is already threatened.
Averbakh 's Selected Games 67
19 "bl
By placing his queen at b I , White was
hoping to neutralise Black' s pressure on
the d-file. If 1 9 "a4 Black could have
continued 1 9 . . . l:ta5 20 "d l J.xg2 2 1
�g2 l:txa2 ! 2 2 l:txa2 "d5+ 2 3 �f3
"xa2. Although White can regain his
pawn with 24 "a i , by 24 . . . "b3 25 "xa7
J.c5 Black retains an active position.
19 J.xg2 26 "e2?
20 ci>xgl l:td5 26 f3 was probably the only defence,
21 l:td l J.f6 when I was intending to play 26 . . . �d4 27
Preventing 22 �, when there would J.xd4 (27 "£2 �b3 ) 27 . . . J.xd4+ 28 ci>h l
have followed 22 . . . b5 . (28 ci>f1 "f6 29 "d3 l:te3 ! ) 28 . . . ..f6 29
22 ci>gl l:te8 l:tf1 l:td7 with a strong attack. Now,
23 84 however, Black's knight comes into play
with great effect.
26 �4
27 "0
Perhaps in his preliminary calculations
White had been intending to play 27 "g4
here, but then saw that there would have
followed 27 . . . l:te4 ! 28 "xe4 �2+ 29
ci>f1 l:txd 1 + 30 ci>xe2 l:txa l .
27 "c8?
Short of time on the clock, Black in
tum goes wrong. 27 . . . �f3+ 28 ci>h l l:txd l
29 l:txd l "a8 ! would have won
immediately. I of course saw this
Averbakh 's Selected Games 69
13 a3 �c6
14 d5
1 4 �g3 is more flexible, as yet not
fixing the position in the centre.
14 �e5
15 �xe5 dxe5
19 g6
The commentators considered this to be
a poor move, but in my opinion it is not at
all bad. Smyslov's aim of initiating play on
the kings ide is positionally well-founded:
White has delayed the development of his
A problematic position has arisen. pieces. After the recommended 19 ... :n
White has a protected passed pawn and the there could have followed 20 b3 , and if
possibility of an attack on the kingside 20 . . . �f8 2 1 fxes fxes 22 �b2.
with f2.-f4. Black has three pawns against 20 �b6+ �b8
two on the queenside. 21 f5
16 � g3
Bronstein suggested 1 6 �e3 here, and
if 1 6 . . . �e4 1 7 �c4, but Keres showed
that after 1 7 . . ...fS 1 8 f3 �d6 1 9 l:xes
"g6 ! 20 l:xe7 �xc4 2 1 l:xb7 l:fd8 Black
has a reasonable game.
16 �e8!
17 f4 � d6!
Trying to force the opponent to declare
his intentions. 1 7 . . . f6 would have created a
new weakness in Black's position - the es
pawn.
18 �f5 f6
Of course, not 1 8 . . . exf4 1 9 �xd6 "xd6 A committing advance, which during
20 es "h6 2 1 ..g4 with the initiative for the game I thought to be obligatory, as
White. otherwise Black would play . . . f6-fS . How
19 g3 ever, after 2 1 "f3 the move 2 1 . . . fS would
This idea of maintaining the pawn have been premature because of 22 exfS !
tension leads to a loss of time. 1 9 "f3 was (22 fxes i.xes 23 exfs i.d4+) 22 . . . exf4 23
more accurate, when if 1 9 . . . exf4 20 �xf4 fxg6 ! , and if 23 . . . fxg3 , then 24 �n+ �g8
i.xf4 2 1 "xf4 White has a good game. 25 l:xe8! :axe8 26 gxh7+ is possible, with
74 Averbalch 's Selected Games
a strong attack, while if 2 1 . . ....g7, then 22 f-file, on which he is counting, does not
fS gS 23 "'hS is good. It has to be admitted prove possible.
that 2 1 "'0 would have been more in the
spirit of the position than 2 1 fS .
21 c4
22 h4
White has no option other than to
continue in the same vein. The idea of the
plan with 2 1 fS and 22 h4 is to gain control
of fS after 23 fxg6 hxg6 24 hS . However,
against accurate defence White is not able
to carry this out, and therefore 22 :fl
came into consideration, and if 22 . . . �cS+
23 �g2 lDd6 24 "'e2 .
22 �c5+
23 �g2 27 �h6 :f7
28 exfS :xfS
29 "'c2 !
Apparently this simple reply was not
anticipated by Smyslov. Unexpectedly his
position becomes critical .
29 :xn
30 :xn :g8
23 gxfS
Smyslov is let down by his sense of
position. By 23 . . . lDd6 ! he could have set
White serious problems. If 24 fxg6 hxg6
2S hS, then after 2S . . . gS he would have to
reply either 26 g4 "'h7 27 lDfS lDxfS 28
gxfS, or 26 lDg4 "'h7 27 "'e2, but in both
cases Black has fair prospects. 31 "'fS!
24 lDxfS lDd6 Having taken the initiative, White
2S lDxd6 .*.xd6 consistently exploits it. After 3 1 "'xc4 :g4
26 :n fS 32 "'d3 (or 32 "'e2) 32 . . . e4 Black would
After consistently pursuing his strategy have gained strong counterplay.
of opening lines on the kingside, Black 31 "'xfS
fails to take account of the fact that after 32 :xfS :g6
the exchange of knights the situation has Hardly better was 32 . . . e4 33 .*.f4 .*.xf4
changed sharply, and that the seizure of the 34 Lf4 :d8 3 S :fS, when White should
Averbakh 's Selected Games 75
38 �c5 12 c5!
This loses immediately. 38 . . . �e7 39 Excellently played! It transpires that the
l:txa6 �f7 would have prolonged the acceptance of the pawn sacrifice by 1 3
resistance. .i.xa6 .i.xa6 1 4 dxcS bxcS 1 5 "'xcS leads
39 d6 �h8 to difficulties for White after I S . . . 'ii'b 7 !
40 �d5 13 "'a4
Black resigns On encountering for the first time the
variation played by Black, I chose an
Game 2 1 incorrect order of moves. 1 3 dxcS bxcS
Averbakh-Kholmov and only then 1 4 ...a4 was more accurate,
Semi-Final 20th USSR Championship although even in this case Black has a
Minsk 1 952 good game.
Queen's Gambit 13 c4!
In my preliminary calculations I had
1 c4 e6 been intending to answer this with 1 4 b3 ,
2 lDc3 d5 but only now did I notice that after
3 d4 lDf6 1 4 . . . bS ! 1 5 "'xbS "'a3 White faces serious
76 Averbakh 's Selected Games
difficulties. It has to be conceded that the positional mistake, allowing White again
queen move to a4 was an empty threat, to take the initiative and to develop his
leading to the complete loss of the offensive on the kingside. After the correct
initiative. 1 8 ......d6 1 9 g4 the bishop could have
14 .i.e2 &iJc7 retreated to h7, and if 20 f4 Black could
15 "'c2 bS have replied 20 . . . f6, driving back the
16 0-0 "'f6 knight, while 1 9 "'as could have been
answered by 1 9 . f6 20 &iJf3 "'b6.
. .
19 g4!
This is the point! The bishop has to
retreat to c8, and this means that playing it
to fS was merely a waste of time.
19 .i.c8
20 f4 "'d6
21 f5
White consistently carries out his pawn
offensive on the kingside, the goal of
which is His Majesty the enemy king.
Black must try to defend his kingside
while continuing his attack on the
Now the position can be assessed. queenside.
Black has good possibilities on the queen 21 f6
side, where he has an extra pawn. White 22 &iJg6 l:d8
must seek chances on the kingside and in With the intention of playing his bishop
the centre. With his last move Kholmov to f7, Kholmov leaves the e8 square free
aims to develop his bishop at fS, in order to for it.
eliminate White's possible play with e3-e4. 23 .i.f3 .i.d7
17 &fJes .i.f5 24 &iJf4 .i.e8
18 "'d2 as
25 :tl !
Kholmov made this natural move, White places his rooks on the second
depriving the white queen of the as square, rank, so that they defend the queenside and
almost without thinking, but in fact it is a help the offensive on the kingside.
Averbakh 's Selected Games 77
instantly, it was not easy to find the correct calmly. It is clear that the situation has
continuation of the attack, as the position become extremely sharp, both kings lack
was still very sharp. Not seeing anything pawn protection, and a piece attack is
forcing, I simply decided to protect once possible from both sides.
more my bishop at h5 . 41 lDe6+ � e7 !
Subsequent analysis showed that White Mate would have followed after
would have retained a strong attack after 4 1 . . . .txe6 42 fxe6+ lDf6 43 "'xf6+ ! "'xf6
39 :7g6 ! .txg6 40 lDxg6+ :xg6 4 1 fxg6, 44 :g8+ and 45 :e8.
for example: 42 .txt7 e2 !
(a) 4 1 . . ....xe3+ 42 :f2+ �g8 (42 . . . �g7 Kholmov defends resourcefully. If
43 .td l ! ) 43 .tf3 ! lDc7 44 "'f6 with the 42 . . . �xf7 there would have followed 43
threat of 45 .txd5+; l:th2 c2 ! 44 :xh6 ! c l "'+ 45 �h2, when
(b) 4 1 . . .:xe3 42 g7+! lDxg7 43 :f2+ Black has no defence, while after 42 . . . :h4
�g8 44 .tf7+ �h7 45 .tg6+! and wins. 43 .th5 :xh5 44 :xh5 "'xh5 45 "'e l l
The move played by me is a mistake, White retains every chance of winning.
after which the roles are reversed, and it is 43 "'xe2 "'h4
now White who has to find a way to save 44 :5g3 :e1+
the game, as the initiative again passes to 45 �a "'hl
my opponent.
39 :xe3
Black returns the compliment.
Although he had quite sufficient time for
thought, he in tum does not make the
strongest move. 39 . . . lDf6 was correct,
tying up even more tightly the clump of
white pieces on the kingside, after which
the c3 pawn could have played the decisive
role. Now, however, I gain the opportunity
to disentangle my pieces.
40 "'a :e4
46 :g4
During the game I considered this to be
the best move, as I assumed that after 46
"'d3 :b6 47 .txe8 :b2+ 48 �f3 :fl + 49
�g4 :xg2 Black would have definite
counter-chances. But in subsequent analy
sis it was discovered that with 50 "'a3+
White gives mate in six moves: 50 . . . �xe8
5 1 "'f8+ �d7 52 "'d8+ �c6 53 "'c7+
�b5 54 "'b7+ �c4 55 "'b3 . And if Black
tries to bring his knight into play -
46 . . . lDf6, then 47 :g7 ! and if 47 . . . lDe4+
Here White's time trouble ended, and at 48 "'xe4 ! , when taking the queen allows
last it was possible to evaluate the position 49 -*.g6+ with mate next move.
Averbakh 's Selected Games 79
54 :h2 ! lbe8
After 54 . . . lbxf5, which Black initially
planned, White wins a piece by 5 5 lbe5 .
55 :h8+ �e7
56 :gg8 :d8
80 Averbakh 's Selected Games
19 lDb3
If 1 9 cs there would have followed
1 9 . . . dxcs 20 bxcs lDxcs 2 1 "xcs "xd2,
and therefore White's desire to remove his
knight from the attack by the queen is quite
understandable. However, due to this loss
of an important tempo . Black now has
counterplay sufficient for equality.
Unzicker rejected 1 9 a4 because of the
14 lDb6 following variation : 1 9 . . . cs 20 dxc6 ':'xc6,
A serious positional mistake, after and if 2 1 as lDxc4 22 lDxc4 bs 23 axb6
which the opponent' s advantage becomes lDxb6 24 �b3 lDxc4 25 �xc4 %tfc8, when
obvious. Since White is intending an Black regains his piece. However, in a
offensive on the queenside, Black should joint analysis after the game we established
aim for counterplay on the kingside, and that by 26 ':'xa6 ! ':'xa6 27 �xf7+ �xf7 28
this must be done energetically. 1 4 . . . h6 and "xc8 White could nevertheless have
I s .....gs was probably better, or even gained a decisive advantage. Therefore
1 4 ...""4. Black's desire to develop his Black would probably have had to answer
second knight is quite understandable, but 2 1 as with 2 1 . . . lDc8, going totally onto
in so doing he puts it on a bad square, the defensive.
where it subsequently comes under attack. 19 15!
15 �c2 lD8d7 Black' s counterplay is based on tactics.
16 b4 If 20 exfS "xc4, while after 20 cs I was
Averbakh 's Selected Games 81
Game 23
Averbakh-Stahlberg
Interzonal Tournament
SaltsjobadeniStockholm 1952
French Defence
1 e4 e6
2S lLlxb7 2 d4 dS
Now a fierce tactical battle commences. 3 lLld2 cS
25 . . . l:lxb7 would be met by 26 c6 l:lbb8 27 4 exdS WxdS
cxd7. S lLlgB cxd4
2S lLlf6 6 .i.c4 Wd8
S2 Averbakh 's Selected Games
16 "'eS!
1 6 �e5 was possible, for the moment
keeping the queens on, but going into the
endgame is also good.
16 'ii'xeS
17 �xeS �g6
By also exchanging the knights, Black
tries to simplify the defence. However,
11 �bxd4! J.xd4 1 7 . . . b5 I S J.b3 �g6 was probably more
After 1 1 . . .�xd4 1 2 �xd4 "'a5 the accurate (but not I S . . . J.b7 1 9 ':'d7 �c6 20
reply 1 3 �b5 ! ? is not so clear: 1 3 . . . axb5 �xfl, winning a pawn).
Averbakh 's Selected Games 83
(b) 42 . . . .i.h3 43 �e4 ! .i.g2+ 44 �e5 me that Black had no defence against 48 a7
.i.xd5 45 :xd5 :e8 46 :d7. followed by moving away the rook with
Now, however, the play becomes check.
slightly sharper. What was I to do? Choose the new
path, discovered at the board, or that which
had been considered in the analysis? After
much hesitation, I nevertheless chose the
second path. This decision was correct, if
only because in the last variation, by
playing 47 . . . �e5 48 a7 :e7 ! , Black would
have parried the threat of the a-pawn' s
advance.
However, the consideration of the
variations associated with 44 b5, and
especially my doubts, cost a great deal of
time, and I ended up in time trouble.
44 bxe6
40 �g7 4S :b6 :d7!
41 :g8+
The game was adjourned in this
position and Black sealed his next move.
He had to choose between playing his king
to h7 or g6, as 4 1 . . .�f6 would have lost to
42 :£8+! �e5 43 :f7, when he is
completely helpless, e.g. 43 . . . g4+ 44 �f2
g5 45 :g7 �f6 46 :h7 �e5 47 :f7 ! �d4
48 .i.e6 etc.
Analysis showed that White's
advantage was enough for a win, but that
the way to it lay via a sharp rook ending
that demanded accurate play.
41 � h6 46 :xa6?
42 :h8+ �g7 This is what time trouble means! White
43 :b8 .i.e6 captures the wrong pawn. 46 :xc6 would
44 .i.xe6 have won, for example:
I examined this move in my night-time (a) 46 . . .:d2 47 e5 ! :xh2 48 :c7+ �h6
analysis. However, during the adjournment (48 . . . �g8 49 b5 ! ) 49 :f7 ! , and Black has
session the following morning, I noticed no defence against the advance of the a
the possibility of the pawn breakthrough and c-pawns;
44 b5 and feverishly began calculating (b) 46 . . . :d3+ is no better: 47 �g4 ! f.3
variations: 44 . . . .i.xd5 45 exd5 :xc5 46 d6 (47 . . . �h6 48 :xa6) 48 �xg5 f2 49 :xg6+
:d5 47 :xb7+ �f6 48 d7 �e7 49 bxa6 and 50 :f6.
:xa5 50 a7 and wins, or 44 . . . axb5 45 A justification for my mistake was that
.i.xc6 bxc6 (45 . . . :xc6 46 :xb7+ �h6 47 in my night-time analysis I did not
:xb5) 46 a6 �f6 47 :a8 and it seemed to establish which pawn it was better to take,
Averbakh 's Selected Games 85
61 cxbS'" Game 24
White resigns Gligoric-Averbakh
Interzonal Tournament
After signing the scoresheet, I shook Saltsj6badeniStockhoim 1 952
my opponent's hand and set off for the Nimzo-Indian Defence
hotel. It was just coming up to lunch time.
However, on entering the restaurant, I felt 1 d4 lDf6
that I was incapable not only of eating, but 2 c4 e6
even of sitting down at the table. 3 lDc3 .i.b4
We were playing and living in Salts 4 a3
jobaden, a suburb of Stockholm. Going out Undoubtedly the most critical
into the fresh air, although there was a continuation . White shows that he does not
strong wind and a light Autumn rain was consider the exchange on c3 to be a threat.
falling, I wandered around a forest, not On the contrary, in order to gain the two
following any paths, until I found myself bishops he is ready to waste a tempo.
on the banks of a small lake, edged with 4 .i.xc3+
boulders. I sat down on a bench that 5 bxc3 0--{)
opportunely appeared, and began gazing at
the water. I sat there for at least half an
hour . . .
Strangely enough, this calmed my
nerves that had been so angered by the
vexing defeat. Incidentally, realising that
water had a pacifying effect on me, on
more than one subsequent occasion I got
myself back into a normal state after an
especially nervy encounter, by observing
the surface of water.
' What are you getting so hysterical
about?' , I persuaded myself. 'After all, the
tournament is not over, and your chances 6 f3
have not yet been lost. There are still nine This desire to advance e2-e4 In one
rounds to go. ' move is perfectly understandable.
On returning to the hotel, I set up a plan 6 d5
for the coming battles. In order to finish in In other games played at that time,
the first five, which would give the right to Black tried to manage without this move.
participate in the Candidates Tournament Thus Szabo against Bronstein (Budapest
and also the grandmaster title, I needed to 1 950) played 6 . . . lDh5 7 lDh3 f5 here, and
score at least 7 points out of 9. And I slightly later Taimanov with Black against
succeeded in fulfilling this difficult task. I Szabo (Szczawno Zdroj 1 950) chose 6 . . . b6
will not describe what effort and nervous 7 e4 lDe8 followed by . . . lDc6-a5 and
energy this cost me. As the unforgettable . . ..i.a6, attacking the c4 pawn.
Grigory Levenfish once said: ' Major 7 cxd5
successes do not happen without a major If 7 e3 c5 8 cxd5 Black has the
expenditure of nervous energy! ' . satisfactory reply 8 . . . lDxd5, as played in
Averbakh 's Selected Games 87
11 c5
Black should not have been in a hurry
to make this move. The correct contin
uation was 1 1 . . . lLld7, and if 1 2 �f2 lLle6
This move marks the start of a new 1 3 �d3 lLldfS with a good position.
plan, which I prepared specially for the 12 �fl lLlg6
Interzonal Tournament. 13 h5
9 'ifcl l:te8 An inaccuracy in reply. After 1 3 �d3
10 g4 lLlc6 14 lLle2 �e6 1 5 g5 ! it is bad to play
After prolonged thought Gligoric 1 5 . . . l:tc8 because of 1 6 h5 lLlfS 1 7 g6 ! with
decides on the immediate advance of his g an attack for White (Gheorghiu-Fischer,
pawn. The whole point is that the usual Havana Olympiad 1 966), although
plan of development in such positions, 1 5 . . . lLlfS is quite possible, with equal
involving kingside castling and the chances.
preparation of e3-e4, is difficult to carry
out here. Thus if 1 0 �d3 there can follow
1 0 . . . 'ifh4+ 1 1 'iff2 lLlf4! 1 2 �f1 'iff6, and
Black successfully regroups his forces.
Therefore White takes the risk of
beginning an immediate pawn offensive on
the kingside.
10 lLlf4
11 h4
This position clearly deserves a
diagram . In the course of the first eleven
moves White has succeeded in moving all
his pawns ! And yet, because the position is
a closed one, this has not caused him any 13 lLlfS
serious difficulties. After all, the white 14 lLlel lLlc6
pawns constitute a kind of barricade, 15 lLlf4
through which for the moment the black Black also has a good game after 1 5
pieces are unable to break. lLlg3 i.. e6 1 6 i.. d3 I:tc8.
88 Averbakh 's Selected Games
35 J.xh6 �xf5
36 J.d �xd4
37 :exe4 �xf3
38 :xe7 :xe7
39 �a �e5
The knight, which has completed a
lengthy manoeuvre: b8-c6-a5-c4-b6-a4-b2-
c4-d6-fS-d4-f3-e5, has not done badly! The
win for Black, who is two pawns up, is not
far off.
40 :d4 :d7
41 �e3
30 �b2 !
31 �fl
3 1 J.e l �c4 32 dxc5 was hardly any
better. After 32 . . . dxe4 3 3 :hxe4 l:xe4 34
fxe4 (34 l:xe4 l:xe4 35 fxe4 �xa3)
34 . . . l:c8 Black would have gained the
bette r ending.
31 �c4
32 J.d
6 lDxe4
7 d4 �d7
8 l:lel lD f6
9 dxe5
The game Geller-Lipnitsky (Kiev 1 9S0)
continued 9 . . . lDxeS 10 lDxeS dxeS 1 1
l:lxeS+ �e7 1 2 �gS ( 1 2 �xd7+ 'ii'xd7 1 3
'ii'e2 O--O-O ! ) 1 2 . . .�xa4 1 3 'ii'xa4+ 'ii'd7 1 4 After examining these variations I came
'ii'xd7+ �xd7 I S lDd2 h 6 1 6 �h4 l:lae8 1 7 to the sad conclusion that I did not even
lDo �d8 ! with roughly equal chances. have my one bird in the hand, and, gritting
In an earlier game Ryum in-Ragozin my teeth, I prepared to defend desperately.
(Moscow 1 93 8) White first played 9 �xc6 On my way to the grandmaster title I had
�xc6 and only then 1 0 dxeS dxeS 1 1 encountered so many barriers, and now,
'ii'xd8+ l:lxd8 1 2 lDxeS .i.e4 1 3 iDd2 �e7 when the goal was within reach,
1 4 lDxe4 lDxe4 I S �f4 iDes with a quick everything had collapsed like a house of
draw. cards !
Before my game with Pilnik I asked 10 �xa4
Geller what he thought about this 11 'ii'xa4+ .d7
variation, and he replied very evasively. It 12 'ii'd 4!
was only five years later that I discovered
why. In the game Geller-Weltmander
(Gorky 1 9S4) there followed I S �h6 ! with
advantage to White.
I, of course, remembered the Geller
Lipnitsky game, and without thinking I
played:
9 lDxe5
10 �f4!
This move looks much stronger than 1 0
lDxeS . While developing, White intensifies
the pressure in the centre. If 1 0 . . . i.e7 there
now follows 1 1 lDxeS dxeS 1 2 i.xeS,
92 Averbakh 's Selected Games
Avoiding a subtle trap. Against the lLlaS l:d7 1 9 "'a8+ .i.b8 20 l:ae l , when if
natural 1 2 "'b3 I had prepared the reply 20 . . ....c2 2 1 lLlxc6 bxc6 22 "'xc6+ l:c7 23
1 2 . . . 0-0-0 ! , and if 1 3 .i.xeS dxeS 1 4 "'xa6+ <it>d8 24 lLld4 with a powerful
lLlxeS "'f5 ! , when after I S lLlxf7 .i.cs attack, or 20 . . . l:c7 2 1 lLld4 l:d8 (2 1 . . . cS 22
Black has a strong attack. l:xcS ! is strong) and White has the
Here I was again obliged to think. At convincing 22 l:Se3 ! "'g6 (22 . . ....d2 23
first I was intending to play 1 2 . . . 0-0-0, but lLlc4 ! ) 23 lLldxc6.
I rejected this because of 1 3 .i.xeS dxeS 1 4 It should be mentioned that including
"'a7, after examining these variations: the bishop in the defence of the queenside
also does not help: I S . . . .i.d6 1 6 lLlc4 .i.b8
is met by 1 7 "'cS ! ( 1 7 lLlb6+ �c7 1 8
lLlxd7 .i.xa7 is not dangerous) 1 7 . . . .i.c7 1 8
lLlfxeS "'fS 1 9 "'a7 ! with numerous
threats.
It must also be added that Black cannot
answer 1 2 "'d4 with 1 2 . . . cS 1 3 .i.xeS !
cxd4 because of 1 4 .i.xf6+.
However, in reply to 1 2 . . . 0-0-0 1 3
.i.xeS it would seem possible to interpose
1 3 . . . cS, for example: 1 4 "'c4 bS ( 1 4 . . . dxeS
I S lLlxeS "'e8 1 6 lLla3 ) I S "'b3 , and then
not I S . . . c4 1 6 "'a3 , but simply I S . . . dxeS
1 4 . . . c6 ! (evidently the best reply; 1 6 lLlxeS "'fS with counterplay.
1 4 . . ....dS is bad because of I S "'a8+ �d7 The most difficult thing in chess is
1 6 lLlxeS+ "'xeS 1 7 "'xd8+ and 1 8 :xeS) choosing, out of two evils, the one which
I S lLlbd2 ! (this is much stronger than I S causes less damage. I rejected queenside
lLlxeS "'dS, when both 1 6 . . ....xeS and castling, and played instead:
1 6 . . . .tcS are threatened). 12 .i.e7
Now if I S . . ....c7 there follows ' I 6 lLlc4 13 .i.xe5 dxe5
"'b8 1 7 "'xb8+ �xb8 1 8 lLlfxeS and the f7 14 "'xe5
pawn is lost, while if I S . . . �c7 1 6 lLlc4
"'dS 1 7 "'b6+ �c8 ( 1 7 . . . �b8 1 8 lLlaS :d7
1 9 lLlxeS :c7 20 lLlexc6+ �c8 2 1 "'a7 !
bxc6 22 "'a8+ �d7 23 l:tad l ) 1 8 "'b3 !
�c7 ( 1 8 . . . .i.cS 1 9 :xeS) 1 9 :ad l "'bS 20
%txd8 �xd8 2 1 lLlcxeS .
Since 1 6 lLlc4 is a threat, Black can
play I S . . ....dS . This did not appeal to me
because of 1 6 :xeS "'d3 1 7 lLlb3 ,
threatening 1 8 lLlaS . Forty-five years later I
subjected this position to a thorough
analysis, but was still unable to find any
tolerable defence. For example: 1 7 . . ....c2
1 8 lLlaS "'xb2 1 9 :ee I ! and there is no Black has a difficult position. Castling
defence against 20 l:eb 1 , or 1 7 . . . .td6 1 8 is impossible, and the initiative is
Averbakh 's Selected Games 93
21 .b4+
At first sight 2 1 l:e4 h4 22 l%c4 looks
dangerous here, but this is not so. Black
replies 22 . . . •d8! 23 h3 c5 24 b4 l:h5 ! ,
retaining possibilities of a defence.
21 �g8
22 .a4 b5
Apparently Black has to play this. After
the game I thought that 22 . . .c�h7 would
have been more accurate, as if 23 lDe5
Black has the good reply 23 . . . l:e8 !
18 lDxf6 (weaker is 23 . . . .i.xe5 24 l:xe5 g6 25 .f4 !
This exchange favours Black, as it .f8 2 6 l:e l c 5 27 .f6) 2 4 .d l �g8, and
eases his defensive problems. After 1 8 .f4 25 .xh5 is bad because of 25 . . . g6 ! ,
lDxe4 1 9 l:xe4 he would not have replied winning a piece.
1 9 . . . .i.f6 20 l:xe8+ �xe8 2 1 lDe5 .i.xe5 However, 23 .c4 (instead of 23 lDe5)
(2 1 . . .•e6 22 lDg4 .i.d8 23 .d4) 22 is stronger, and only after 23 . . . l:f8 - 24
. •xe5+ �f8 23 .d4 with a decisive lDe5 .
"
advantage, but 1 9 . . . .i.d6, although after 20
lhe8+ .xe8 he would have to reckon with
the exchange sacrifice 2 1 l:xd6.
18 .i.xf6
19 .f4 l:xe1+
20 l:xel
The situation has clarified - whereas
the white forces are fully mobilised, Black
still requires time to bring his rook into
play.
20 h5!
After 20 . . . h6 White had the very strong
2 1 h4, and if 2 1 . . . �g8 22 g4, in order to
answer 22 . . . �h7 with 23 g5 . 23 .f4 l:h6
94 Averbakh 's Selected Games
24 h3 :g6 32 �h7
25 'ife4 �f8 33 Wxc7 Wel+
26 Wd5! h4 34 �h2 Wxfl
27 :e4 :h6 35 b3 'ifxal
36 'ifx17 Wc2
37 'ifh5+ �g8
38 Wd5+ �h7
39 c4 bxc4
It was probably simpler to take on b3 ,
but the last few seconds were ticking away
on my clock.
40 bxc4
1 d4 tDf6
2 e4 eS
3 dS e6
4 tDe3 exdS
S exdS d6
6 e4 g6
7 .i. e2 .i. g7
8 tDf3 0-0
9 0-0 %te8
10 1i'e2
14 tD fd7
This move was later condemned by the
theorists, but the more active 1 4 . . . h6 1 5
.i.g3 g5 1 6 f4 ! , as occurred in the game
Malich-Browne (Amsterdam 1 972), also
leads to an advantage for White. It would
seem that the plan chosen by Black
altogether fails to equalise.
IS .i.e3 f6
16 f4 tDr7
At that time the theory of this opening, 17 .i.tl 15?
then called the Indian Defence, was only
just beginning to develop. Nowadays 1 0
00 i s considered the strongest.
10 86
At one time I O . . . tDa6 was very popular.
Thus the game Averbakh-Tal (25th USSR
Championship, Riga 1 95 8) went 1 1 .i.f4
tDb4 1 2 1i'b 1 tDxe4 1 3 tDxe4 .i.fS 1 4 tDfd2
tDxd5 1 5 .i.xd6? tDf6, and Black emerged
a pawn up. However, by continuing 1 5
J.. g3 ! White would have retained the
advantage. 1 5 . . . .i.h6 1 6 .i.b5 ! is good for
him, while if 1 5 . . . 1i'e7 1 6 .tf3 %tad8 1 7
:e 1 , as suggested by Kholmov. My opponent tries to seize the initiative
11 84 tDbd7 by forcible means, for which, it has to be
12 .i.f4 1i'e7 said, he had purely psychological grounds.
The alternative is 1 2 . . . 1i'e7. When I was still a master, I played Alex
13 tDd2 tDes ander Tolush four times, with a dismal
14 h3 outcome for me -one draw and three
White intends to retreat his bishop to e3 defeats. The point was that I simply could
and then play f2-f4. not adapt to his exceptionally aggressive,
Averbakh 's Selected Games 97
forceful style, and even reckless, risky play This allows a spectacular finish. How
brought him success. But on this occasion ever, even after the more tenacious
he met his match ! 23 ...•d8 24 lLle4 lLlh8 25 'ii'h 5 �g8 26
18 exfS j,xc3 j,h4 .fS 27 :e3 Black has no defence
This is the point of Black's idea - he against the combined attack of the enemy
wins a piece. pieces.
19 fxg6 hxg6 24 :e8+! �xe8
20 bxc3 25 .g8+
Of course, not 20 Wxg6+ j,g7 2 1 j,d3 Black resigns
lLlfS, when Black has everything in order. After 25 . . . �e7 26 j,h4+ he is mated in
20 lbe2 two moves.
21 Wxg6+ �f8
Game 27
Keres-Averbakh
Candidates Tournament
NeuhausenlZurich 1 953
Nimzo-Indian Defence
1 d4 lLlf6
2 c4 e6
3 lLlc3 j, b4
4 e3 0-0
5 j, d3 d5
6 lLlo c5
7 0-0 lLlc6
22 Itael ! 8 a3 j,xc3
The entry of the second rook into the 9 bxc3
battle refutes the opponent's combination.
If 22 . . .ltxd2 there follows 23 j,h4 lLlb6 24
:e8+ with the same mate as in the game.
22 :xe1
23 lbe1 lLlb6
12 exdS
Keres decides to relieve the situation in
the centre, but in so doing he loses one of
his trumps - the advantage of the two
bishops.
12 "'xdS
13 "'e1 exd4
14 exd4 llJb3
15 :bl llJxd
16 %lbxd :ae8
17 "'b1 11 gS
If White's bishop were at b2, his central 11 :a "'xd4
pawns would be very strong. But here, 13 "'xd4
even after c3-c4, they are likely to become There is no choice. If 23 "'e2 %le7 24
weaknesses, which need to be defended. %le I :de8, and Black wins.
Averbakh 's Selected Games 99
Game 28
Averbakh-8zabo
Candidates Tournament
NeuhausenlZurich 1953
Ruy Lopez
1 e4 e5
2 lbfJ lbc6
3 J.b5 a6
4 J.a4 lbf6
Probably more tenacious was 3 1 J.g6 5 � lbxe4
J.xc4 32 :xc4 :xc4 33 :xc4 :xg6 34 6 d4 b5
:xh4, when Black would still have to win 7 J.b3 d5
1 00 Averbakh 's Selected Games
noticed the dangerous pawn sacrifice Having convinced himself that his
1 9 . . . c4! , and after 20 "d4 there could attack is threatening to peter out, with a
follow 20 . . . .i.fS ! 2 1 .i.xfS :xf5 22 :d l rook sacrifice my opponent forces
lDe6, and if 23 "xd5 1i'b6+ 24 �fl :dS. perpetual check.
Unsettled by this discovery, I decided 24 :xeS
to risk a sharp continuation, in which 25 "xeS .g3+
Black has nothing better than to force a 26 �hl .xfJ+
draw. 27 �gl
Draw agreed
Game 29
Euwe-Averbakh
Candidates Tournament
NeuhausenlZurich 1 953
Nimzo-Indian Defence
1 d4 lDf6
2 c4 e6
3 lDc3 .i.b4
4 e3 0-0
5 .i.d3 dS
19 lDh3+ 6 lDfJ cS
20 gxh3 .xh4 7 0-0 lDc6
21 :a .i.xh3 8 a3 .i.xc3
22 : h2 :ae8 9 bxc3 b6
23 "xd5+ �h8 10 cxdS exdS
24 .i.d2 11 lDd2
It appears that White has managed to The former World Champion demon
stabilise the position. If 24 . . . :e6 he had strates his preparation for the tournament.
prepared 25 .i.xf4, while after 24 .....g5+ He intends to play his rook to e I , transfer
25 �h l :XeS he has 26 "d3 ! .i.fS 27 his knight via fl to g3 and, after supporting
"d6, winning. his d-pawn by moving his bishop to b2,
1 02 Averbakh 's Selected Games
17 1Vd2
One has to admire Euwe's composure.
Here he could have played 1 7 lIc 1 , so that
after I S . . . b3 his rook would not be shut in.
However, he did not want to waste time on
the rook move - he thought that within a
few moves all the same it would come into
play.
17 b3 And this move, which I had planned far
18 .tbl a4 in advance, is in fact a serious mistake,
Thus Black has achieved his aim by after which White should have developed a
creating a strong protected passed pawn on strong initiative. 22 . . . 1VdS ! was correct,
the queens ide, but now he has to withstand and if 23 1Vg5 ltJg6.
his opponent's offensive in the centre, and 23 1Vg5 %tt7
later his attack on the kingside. 24 exfS .txfS
Averbakh 's Selected Games 1 03
32 g4
Could the former World Champion
have saved this ending? In order to win,
1 04 Averbakh 's Selected Games
Game 30 15 'ilh3
Averbakh-Taimanov If I S 'irf2 Black can sacrifice a piece:
Candidates Tournament I S . . . 0-0! 1 6 g4 lDxb3 1 7 axb3 'ilxeS 1 8
NeuhausenlZurich 1953 gxhS 'irxhS with an attack.
Sicilian Defence
1 e4 c5
2 lDf3 lDc6
3 d4 cxd4
4 lDxd4 lDf6
5 lDc3 d6
6 .i.c4 e6
7 0-0 a6
8 .i.e3 'ilc7
9 .i.b3 .i.e7
10 f4 lDa5
11 'irf3 b5
15 'ilxe5?
It was this capture that I was counting
on ! In the later game Bannik-Taimanov
(23rd USSR Championship, 1 9S4)
Taimanov showed what Black should play
- I s . . . lDxb3 1 6 lDxb3 'ilxeS 1 7 lDaS b4 1 8
lDc4 'irc7 1 9 'irxhS g6 20 'ile2 bxc3 with
the better chances.
16 .i.xe6! fxe6
If Black declines the sacrifice by
1 6 . 0-0 , White has the decisive 1 7 %Irs
..
li)g4, when, after placing his knight at e3 , considers here the variation I S . . . ...c6 1 6
White has a solid enough position. li)b4 "'b7, although after 1 7 �e3 li)c4 1 8
c6 the situation would have become more
complicated.
16 llbl llec8
17 h4
In this way White prepares 1 8 �f4, as
on the immediate bishop move he was
afraid of 1 7 . . . gS .
17 d5
18 �f4 f6
With the obvious threat of exchanging
on cS and playing . . . e6-eS .
12 d6
13 li)d3
Hardly better was 1 3 li)c6 "'d7 1 4
li)xe7+ ( 1 4 dS �f6) 1 4 ......xe7, when the
pattern remains the same - the game
reduces to a battle of knight against bishop.
13 li)a5
14 c5 "'e8
15 "'xe8 llfxe8
19 li)b4
A clever attempt to complicate matters.
Now after 1 9 . . . bxcS there would have
followed 20 li)a6, and if 20 . . . cxd4 2 1
li)xc7, e.g. 2 1 . . . 11ab8 22 llxb8 llxb8 23
li)xdS or 21 . . . eS 22 li)xdS with unclear
complications. However, Black has no
reason to go in for such play.
19 a6!
20 cxb6
If 20 li)d3 , then good is 20 . . . bxcS 2 1
The result of White's pseudo-active li)xcs e S 22 �e3 �xcS 2 3 dxcS llab8 24
play is that there are weaknesses in his llfd 1 llxb 1 2S llxb 1 q;f7 with a
position, and the initiative is completely in favourable ending.
the hands of his opponent. Of course, 20 cxb6
Najdorf could have retained the queens by 21 �d2
playing IS 'ii'c 2, but it is doubtful whether All the commentators at the time
this would have improved his chances of a condemned this move, suggesting instead
successful defence. Euwe, for example, 2 1 li)d3 , but after 2 1 . . . bS 22 llfc l �a3 23
l OS Averbakh 's Selected Games
1 d4 �f6
2 c4 e6
3 �c3 i.b4
4 e3 c5
5 �f3 0-0
6 i.d3 d5
7 0-0 �c6
8 a3 i.xc3
9 bxc3
16 c4!
A pawn sacrifice specially prepared for
this tournament by the former World
Champion. It creates considerable prob
lems for White, and they have to solved
directly at the board.
17 i.xc4
Declining the sacrifice by 1 7 i.e2
would have given Black a comfortable
game after 1 7 . . . i.c6. But now White must
play very accurately to avoid losing.
Exactly this position was reached in the 17 �g4
game between the same opponents from 18 f4
the first cycle of the tournament, only with 1 8 g3 "hS 1 9 f4 (but not 1 9 h4 �eS 20
reversed colours. I played 9 . . . b6. Euwe i.e2 i.g4) 1 9 . . . ..cS would have come to
prefers the more popular plan, involving roughly the same thing.
. . . e6-eS . 18 "c5
9 dxc4 19 "d3
10 i.xc4 "c7 White is obliged to calculate exactly at
11 i.d3 e5 every step. Thus if 1 9 i.d3 there could
12 "c2 "e7 have followed 19 . . . �xe3 20 "f2 l:fe8,
Here 1 2 . . . l:te8 is usually played, and if when 21 i.c l is bad because of 2 1 . . ...xc3 ,
1 3 e4 c4. while 2 1 l:e2 "b6 22 l:ae 1 �dS is good
13 dxe5 �xe5 for Black.
1 10 Averbakh 's Selected Games
25 :e7
25 :ed2 also promises little because of
25 . . . :c8.
25 �e6
26 :xd8 :xd8
27 �xe6 fxe6
28 :xb7
19 :ad8
20 :ad l !
After 2 0 .d4 Black has the unpleasant
20 •h5 2 1 h3 �c6, while if 20 �d5
. . .
23 :e2 !
This modest rook move, which I
anticipated back on move 1 7 when I took
the c4 pawn, is the whole point: because of
24 :a2 Black is unable to take on a3 , and
he is forced to exchange queens.
23 .b6
24 .xb6 axb6
As a result of the interesting tactical
complications, White has a favourable
ending, but here this is not enough to win. 35
Averbakh 's Selected Games III
defended, White's king goes to the support fxe4 1 0 tLlxe4 tLlxe4 1 1 1%xe4 'ii'xe4 on
of his passed pawn. account of 1 2 tLlh4, when the queen has no
38 �d2 retreat square.
38 �e7 8 tLle4
39 �cl �d6 9 'ii'c2 'ii'g6
40 �b2 :a4 10 .i.e3
41 g3 �c6 1 0 tLlxe4? fxe4 1 1 tLld2 would have lost
42 �b3 :a1 to 1 l . e3 ! Now, however, White is
. .
1 1 . . .�f6 1 2 :ad 1 lDc6 1 3 'it'b3 a5 1 4 c5 should first have played 1 5 . . . l:b8, and after
with a slight advantage for White, but 1 2 1 6 l:ec 1 �d7 1 7 lDe 1 �c6 he would have
b4 ! , employed later by Furman, i s even had a solid game, although I was intending
better. 1 8 lDd3 and considered the resulting
position to be favourable for White.
However, Ragozin was already in an
optimistic mood, which explains his
following move, after which Black's
position collapses like a house of cards.
15 b6?
16 �g5! �xg5
1 6 . . . �b7 loses a piece to 1 7 �xe7
lDxe7 1 8 lDh4.
17 lDxg5 �b7
After 1 7 . . .• xg5 1 8 �xd5 exd5 1 9
.xd5+ �h8 20 .xa8 Black has no
compensation for the lost material.
12 c5 If now the knight were obliged to
This move is justified: White has retreat, Black would gain a reasonable
completed his development and can game, but White has a new possibility.
embark on active play. At the same time he
is obliged to hurry, otherwise after
1 2 . . .�f6 and 1 3 . . . e5 Black will have a
good game.
12 lDf6
13 cxd6 cxd6
14 'it'b3 lDd5
15 �d2
18 e4! fxe4
19 lDxe4
During the last four moves the position
has changed radically. My opponent has
managed to develop his bishop, but at what
a price ! His central pawns are weak, and
the knight at d5 is insecure. Passive
defence was not to Ragozin' s taste. He
It may seem that Black has successfully initiates a tactical skirmish, but apparently
solved his opening problems. His knight at miscalculates.
dS is well placed, and it only remains to 19 lDe7
develop his queen' s bishop. To do this he 20 lDxd6 �xg2
Averbakh 's Selected Games 1 13
Game 34
Averbakh-Suetin
21 st USSR Championship
Kiev 1954
Slav Defence
1 d4 dS
2 c4 c6
3 lOc3 lOf6
4 1Of3 dxc4
5 a4 1.rs
6 e3 e6
7 1.xc4 1.b4
21 "xe6+! 8 0-0 0-0
Perhaps Black had been counting on 2 1 9 .e2 lObd7
:xe6 1.dS, when White loses a piece.
21 .xe6
22 l:xe6 1.h3
23 l:xe7 l:ad8
24 lOe4 l:xd4
25 l:el h6
In time trouble, Black gives up a
second pawn in order to retain his bishop
and activate his second rook. If 2S . . . aS I
was intending to play 26 lOgS 1.rs 27 1Oe6
1.xe6 28 l:7xe6 :d2 29 l: l e2.
26 l:xa7 l:c8
27 f3 :c2
Black still overlooks the impending 10 e4 1.g4
danger. 11 :d l .e7
12 h3 1.xf3
13 "xf3 eS
14 dS lOb6
15 1.b3 cxdS
16 as d4
17 axb6 dxd
18 bxd 1.cS
19 bxa7
At that time this variation of the Slav
Defence had been extensively analysed,
and, had Black played 1 9 . . . :xa7, after 20
:xa7 1.xa7 a position could have been
reached from the game Lilienthal-Aronin
28 lOf6+ ( 1 6th USSR Championship, 1 948), where
Black resigns after 2 1 1.gS :d8 22 1.dS White gained a
1 14 Averbakh 's Selected Games
slight advantage. Alexei Suetin was always This intermediate move is the whole
well prepared theoretically, and he makes a point.
new move, preventing the bishop from 24 f5
going to gS . 25 :xd5
2S exdS would have led to unclear play.
25 fxe4
19 h6
20 .i.e3 b6
If 20 . . . bS there would have followed 2 1 26 .i.e3
l:aS, winning a pawn. I made this move, defending the f2
21 .i.d5 l:ba7 pawn, without particular thought. But now,
By first playing 2 1 . . .�xdS Black would studying this position many years later, I
have maintained material equality, but by noticed the strong reply 26 "'g4 ! with the
replying 22 exdS White would have unpleasant threat of 27 l:d7. If Black
created a dangerous passed pawn. replies 26 . . . .i.xf2+ 27 �h2 l:n, then 28
22 :xa7 "'xa7 l:d8+ �h7 29 J.xg7 is decisive, while if
he plays 26 . . . ...n, then 27 cJa>h2 "'f6 28
.i.e3 with the better chances for White.
26 "'al+
27 cJa>h2 .i.xe3
27 . . ....xc3 28 J.xcS "xg3+ 29 cJa>xg3
bxcS 30 l:xeS would have led to a lost
rook ending, e.g. 30 . . . c4 3 1 l:xe4 :c8 32
l:e2 c3 33 l:c2, and the king goes across to
the c3 pawn.
28 "'xe3
29 l:d2
30 l:d7
Black seeks salvation in a queen
23 .i.xh6! ending, but in so doing he gives up a pawn.
This combination refutes the Much better was 30 . . :.e6 ! 3 1 l:b7, and
opponent's innovation. here not 3 1 . . :.f6 32 "'xe4 "'xf2 33 "'xeS,
23 . � xd 5 but 3 1 . . . ... f5! 32 "'g3 "'f6, when White's
24 "'g3 ! advantage is insufficient for a win.
Averbakh 's Selected Games 1 15
41 'ife2
In such positions, where there are no
passed pawns, White's plan is to penetrate
with his king into the opponent's position
and weave a mating net.
42 g4
Opening the way for the king.
42 'ifd2
43 �g3 'ifc3+
44 �h4 'ifd4
45 .f5+
31 l:txf7 �xf7
32 .xe4 'ifxc3
33 'ifb7+ �g8
34 .xb6 'ifcl
35 .d8+ �f7
36 'ifd7+ �g8
37 .e6+ �h7
38 .f7 .e4
39 g3
39 g4 was possible, but I did not hurry,
aiming to begin the realisation of my
advantage only after adjourning the game 45 g6
and analysing it at home. This makes things slightly easier for
39 �h6 White. Had the black king retreated to the
40 �g1 �h7 back rank, after f2-f.3 and �h5 White
41 �h2 would have played g5-g6 and again set up
a mating net, but in this case Black would
have acquired stalemating chances, and the
win would have been not at all simple.
Now, however, White's king quickly
invades the enemy position.
46 .f7+ �h6
47 .f6! �h7
48 �g5 .d2+
49 f4! exf4
Going into the pawn ending is no better
- 49 . . ..xf4+ 50 .xf4 exf4 5 1 �xf4 �h6
52 �e5 �g5 53 �e6 �h4 54 �f6 and
White wins.
In this position the game was 50 .f7+ �h8
adjourned, and Black had to seal his next 51 �h6
move. Black resigns
1 16 Averbakh 's Selected Games
Game 3 S
Averbakh-Bannik
21st USSR Championship
Kiev 1954
English Opening
1 c4 eS
2 lDc3 lDc6
3 g3 g6
4 R.g2 R.g7
5 e3
10 R. c8
In his notes to the game with Resh
evsky, Botvinnik attaches an exclamation
mark to this move, whereas Konstantin
opolsky, in his notes to the present game in
the tournament book, thinks that the retreat
to d7 is more natural . However, then White
is not obliged to play d4-dS .
11 dS lDeS
12 b3
In the same notes Botvinnik asserts that
In the 1 93 0s this was Botvinnik' s the continuation chosen by me also
favourite continuation. However, in his last achieves nothing, although he does not
book of collected games, published in the give any variations to demonstrate this.
1 980s, he himself points out that S d3 or The question arises: if this variation is
even S e4 sets Black more difficult so unpromising, why did I choose it?
problems. Firstly, I did not consider it to be
5 lDge7 unpromising. Secondly, and this is the
6 lDge2 d6 main thing, the variation employed gives
7 � � White a fairly solid position, and at this
8 d4 exd4 moment, two rounds before the end of the
9 exd4 R.g4 tournament, I was leading my nearest rival
The well-known game Botvinnik-Resh by a whole point, so that I did not need to
evsky (AVRO Tournament 1 93 8) went aim for the maximum . My main task was
9 . . . lDfS 1 0 dS lDeS 1 1 b3 as 1 2 R.b2 lDd7 not to allow the opponent active play.
1 3 a3 lDcs 1 4 b4 with a positional advan 12 a6
tage. Botvinnik thought that by the bishop This move was recommended by
move to g4 Black could ful ly equalise. Botvinnik, reckoning on 1 3 R.b2 bS ! with
10 f3 good counterplay for Black, but my plans
If 1 0 h3 Botvinnik gives 1 0 . . . R.xe2 1 1 included a different way of developing the
lDxe2 lDfS 1 2 dS lDeS 1 3 'ii'c2 :e8 with an bishop. Black could have considerably
acceptable game for Black. sharpened the play by 1 2 . . . c6, but Bannik,
Averbakh 's Selected Games 1 17
a player of strictly positional style, did not Being cramped, Black is glad of every
make such moves. exchange, but this does not improve his
position.
22 %:txel .f8
23 ci>a
Anticipating the transition into a minor
piece ending, White improves the position
of his king.
23 %:te8
24 :xe8 j,xe8
25 �e2 �d8
13 -*.g5! h6
14 -*.e3 �f5
15 -*.a %:te8
16 .d2 %:tb8
17 �d4 �xd4
18 -*.xd4 -*.d7
White has a spatial advantage, and also
a slight lead in development, so that Black
has to play very cautiously. The more
active l B . . . bS also has its drawbacks. Tarrasch in his time said that if one
White replies 1 9 cxbS axbS 20 f4 �g4 2 1 piece stands badly, then the entire position
j,xg7 ci>xg7 22 %:tfe 1 with a good game. becomes bad. In our case the black knight
is poorly placed, and this later plays the
decisive role. Whereas White gradually
strengthens his position, Black is unable to
find any possibilities for counterplay.
26 .e3 q;n
27 b4!
Beginning to cramp the opponent on
the queens ide.
27 -*.d7
28 1.0 -*.xd4
29 �xd4
White was threatening c4-cS, so Black
decides to exchange the dark-square
19 h3 f5 bishops, but now the white knight has
Defending against the threat of winning taken up a strong position.
the knight. 29 j,c8
20 f4 �n 30 j,e2 j,d7
21 %:tfel :xe1+ 31 j,dl .g7
1 18 Averbakh 's Selected Games
By removing his control of c5, Black knight. However, this was the last move
allows the opponent to make an important before the time control, and he had only
pawn advance. seconds left on his clock. Now his position
32 cS! "'f6 becomes hopeless.
33 -*. b3 �g7
34 cxd6 cxd6
41 h4!
The only move that leads to a win.
With the queens on the board it is easier White no longer allows the knight
for Black to defend, so White decides to manoeuvre.
exchange them, after which he acquires 41 �h8
new possibilities for strengthening his This move was sealed by my opponent,
position - the centralisation of his king and but it is already too late.
the advance of his queenside pawns. 42 hS! �f7
3S "'c3! gS 43 as �d8
This move was condemned by the 43 . . . �5 would have offered some
commentators, as it weakens the f5 pawn, practical chances, but after 44 fxe5 dxe5
but on the other hand it could have allowed 45 b5 -*.c8 46 �4 ! axb5 47 -*.xb5 -*.d7 48
Black to markedly improve the position of -*.xd7 �xd7 49 �c5+ �d6 50 �b7+
his knight. �xd5 5 1 a6 �c6 52 �5+ �b6 53 �4+
36 �e2 ! "'xc3 �xa6 54 �xe5 White wins.
37 �xc3 gxf4
Black should not have been in a hurry
to make this exchange. After 3 7 . ./ijiJ
followed by 3 8 . . . �h8 and 39 . . . �g6 he
could have hoped to save the game.
However, the move played is not a fatal
mistake.
38 gxf4 �f6
39 -*.c4 �f7
40 a4 �e7
The last opportunity for a successful
resistance was 40 . . . �h8, when B lack
succeeds in improving the position of his
Averbakh 's Selected Games 1 19
44 b5!
White's plan is crystal-clear. He needs
to open lines on the queens ide and then try
to penetrate with his king into the enemy
position.
44 Jlc8
45 �e3 �e8
46 �d4 �e7
47 Jld3 �f7
48 �a4 axb5
Because of the threat of 49 �b6, this
exchange is forced.
49 Jlxb5 Jld7
This move too has to be made, after 59 �d4 �8
which the win for White becomes a matter 60 �b5+ �d7
of simple technique. 61 �b6 �e7
50 Jlxd7 �xd7 62 �c7
Black resigns
By my victory in this game I won the
title of USSR Champion with a round to
spare.
Game 36
Korchnoi-Averbakh
21st USSR Championship
Kiev 1954
Sicilian Defence
But here to my aid came. . . Korchnoi And here 49 �f6 was more accurate,
himselfl Instead of adjourning the game, and only if 49 . . . �a4 50 g4.
and then at home, in a calm atmosphere, 49 :a3 !
clarifying the way to win, he decided to 50 �xe6
finish me off without adjourning, seeing as That's it. The win is no longer there!
we still had time on our clocks. And so the After 50 h4 ! l:g3 5 1 �f4 l:h3 52 h5 l:h l
battle continued with its former intensity. 53 l:e2 ! l:a l 54 g5 White would still have
retained winning chances.
50 l:h3 !
But now, as often occurs in rook
endings, the active placing of the black
pieces prevents White from realising his
extra pawn.
51 �f5 �a4
52 g5 hxg5
53 �xg5 l:c3
54 :e2 l:c5+
55 �g4 l:c4+
41 �f6 :d8
42 �e5 �xc5
43 :d4 a4
44 bxa4 :a8
45 �e4+
Of course, simpler was 45 �xd5 exd5
46 :xd5+ �c4 47 :d2 :xa4 48 �f6 etc.
45 Jtxe4
46 :xe4 :xa4
47 :e2 :a6
48 :c2+ �b5
49 g4 Here the game was finally adjourned
and declared a draw without being
resumed. After 56 �g3 l:c3+ 57 �g2 :c4
5 8 h3 �a3 59 �g3 Black plays 59 . . . l:c l
60 h4 :a l 6 1 h5 :xa2 with an easy draw.
Game 3 7
Averbakh-Panno
USSR v. Argentina
Buenos Aires 1954
King's Indian Defelice
1 d4 �f6
2 c4 g6
Averbakh 's Selected Games 1 23
17 �e2 !
Solid barriers have been set up in the
middle of the board, and so here White's
king feels completely safe. To continue the
offensive it only remains for him to switch
his heavy pieces to the h-file and then to
open it.
17 :tg7
Averbakh 's Selected Games 1 2S
23 l:th l !
Now it i s difficult for Black to defend
against the destructive bishop sacrifice at
f4. For example: 23 . . . i.d8 24 i.xf4 exf4
2S l:th4 �e7 26 :xf4 with a decisive
attack.
23
With the desperate 23 . . . bS Black could
have tried to reduce the tempo of the
attack, for example 24 axbS axbS 2S cxbS
li)c7 26 i.xf4 ! li)xbS, but after 27 i.xeS
dxeS 28 li)xeS+ �e8 29 li)xbS 1Wa6 30
1I'xg7 1I'xbS+ 3 1 'i>f3 White has an easy
win. 29 li)h4!
24 i.xf4! 1I'c7 White's mating attack develops very
As 24 . . . exf4 loses to 2S :h4, Black has smoothly.
to defend against the threat of 2S i.xeS ! If, 29 i.d8
for example, 24 . . . li)d7, then 2S 1Wh3 li)b6 30 li)xg6+ � g7
26 i.xeS ! , winning material. 30 . . . :xg6 would have allowed mate in
two moves - 3 1 :h8+ 'i>g7 32 1I'g8 .
31 li)xe5
Black resigns
Game 3 8
Averbakh-Aronin
USSR Team Championship
Riga 1954
King's Indian Defence
1 d4 li)f6
2 c4 g6
3 li)c3 i.g7
25 1Wh2 ! 4 e4 d6
Again threatening to capture on eS . 5 i.e2 0-0
25 li)d7 6 i.g5 c5
26 1Wh3 7 d5 h6
The triumph of White's plan. The At that time the theory of this variation,
queen succeeds in invading the weak which later became known as the
squares in the enemy position. Averbakh Variation, was only just being
26 li)fS developed, and my opponents tried various
27 :xfS+! plans against me.
The only defender of the e6 square has 8 i.f4 li)bd7
to be eliminated. 9 li)f3 li)g4
27 'i>xfS Black immediately tries to begin active
28 1I'e6 :g8 play in the centre.
1 26 Averbakh 's Selected Games
22 lbt7+!
Black resigns
After 22 . . . rlilxfl 23 1i'hS+ rlilg7 24
"'g6+ �f8 2S lDxe6 he is mated.
cxd4 tDxd4 1 9 tDxf7 tDf3+ 20 gxf3 tj;xf7. and probably I should have prevented him
14 c4 from connecting his rooks by 22 "'f4, and
Averbakh 's Selected Games 1 29
48 &£jc7 !
Black defends very accurately. After
48 . . .&£jxf6 49 a6 his position could again
have become critical.
49 IIb6 IIxb6
Averbakh IS Selected Games 131
The Dutch was always one of Nothing good comes of 1 8 �xb4 1i'aS
Botvinnik's favourite defences as Black. In 1 9 a3 "'xa3 20 �d3 "'xb3 2 1 :b l :a2 ! ,
this game he tries a new idea - the rapid when Black remains a pawn up.
development of his queens ide minor 18 "'b6
pieces. 19 f4 g6
8 �b2 �d7 20 �f3 :f7
9 �e3 �e4 21 :tl :e7
10 exd5 �xc3 22 :g2
11 �xc3 exd5
The dS pawn is unprotected, and White
could have exploited this factor to develop
his queenside initiative by 1 2 b4 ! Instead
he chooses a slow knight manoeuvre, after
which he effectively loses his opening
advantage.
28 :d+
29 �a :e2+
30 �el :xal
31 :xb4 :.7
The defence in inferior positions such
as this one is especially difficult, in that
out of several evils one has to choose the
least. The alternative was 3 1 . . . :xh2 32
l:[xb7 :b2 3 3 b4 h6 34 b5 g5 with a sharp
ending. 37 d4
32 �d2 :.2+ A desperate attempt to bring the bishop
33 �d3 :a into play. If 37 . . .:xh2 3 8 :xb7 :h l then
If 3 3 . . . :xh2 34 :xb7 :h3 , then 35 �e2 39 b5 :b l 40 b6, e.g. 40 . . . h5 4 1 �c2 :b4
:h2+ 36 �e 1 transposes into the variation 42 �c3 :e4 (42 . . . :b l 43 .i.d3 ) 43 :xf7 !
given in the note to Black' s 3 1 st move, �xf7 44 .i.xh5+ �f6 45 b7 and the pawn
since if 3 5 . . .:xg3 White has the un cannot be stopped.
pleasant 36 �f2 :h3 37 �g2, when the 38 :xb7 dxe3
black rook is out of play. It is no better going into a rook ending:
34 :b6! 38 . . . .i.d5 39 :b8+ �g7 40 �xd4 :xe2 4 1
1 34 Averbakh 's Selected Games
1 d4 �f6
2 c4 e6
3 �c3 .tb4
4 a3 .txc3+
S bxc3 cS
known Geller-Euwe game from the 1 953 should have closed the centre by 1 7 d5, so
Candidates Tournament. However, there as then to begin an attack on the kingside
was already no way of deviating. with 1 8 �h5 and g2-g4-g5 . However, it
seems to me that a concretely-thinking
player like the young Lev could hardly
have made such a move, unless it was
prepared beforehand. Probably more to his
taste were the tactical complications that
arose in the game.
17 l:txe4
18 dxeS fxeS
19 "'b3
12 exd4
13 exd4 l:te8
14 f4 �xe4
15 fS f6
16 a4!
And here is the important improvement.
The afore-mentioned game went 1 6 l:tf4
b5 ! 1 7 l:th4 "'b6 with strong counterplay.
By not allowing . . . b6-b5, White seriously
restricts Black' s possibilities. 19 dS
As the 'post-mortem ' showed,
1 9 . . ....c7 ! was also good. Polugayevsky
was intending to reply 20 f6 �xf6 2 1 �fS,
threatening 22 �e3 or 22 �d6. Then if
2 1 . . .d5 there could have followed 22 .i.a3
l:td8 (22 . . . l:ta8 23 .i.d6) 23 .i.e7 l:td7 24
�h6+ �h8 (or 24 . . . gxh6 25 "'g3+ �h8 26
.i.xf6+ l:tg7 27 .i.xe5) 25 l:txf6 (with 25
�f7+ White can force a draw) 25 ... gxf6 26
.i.xf6+ l:tg7 27 "'a3 "'c5+ 28 "'xc5 l:txc5
29 �f5 with a material advantage in the
ending.
But Black has a better defence - instead
16 eS of 2 1 . . . d5 he should play 2 1 . . . .i.xe4 ! 22
In the first instance I decided not to .i.a3 .i.xfS 23 .i.xfS �xfS 24 l:txfS "'c5+
allow the rook to go via f4 to h4. 25 �h l l:tc l + 26 l:tfl l:txa l 27 l:txa l with
17 .i.xe4+ three pawns for the exchange, although the
The commentators criticised Polu insecure position of the opponent's king
gayevsky for this move, reckoning that he leaves White with some saving chances.
136 Averbakh 's Selected Games
h6.
14 .i.g4 lDc7
15 .i.xc8 "xe8
16 lDa
18 lDh4
Thus, White has decided to continue the
attack. Meanwhile, after 1 8 as ! lDe6 1 9
hxg6 hxg6 20 lDa4 he need not have feared
20 . . . lDd4 2 1 lDb6 ..g4 22 lDxd4 f3 23
lDc2 "xe4+ 24 lDe3 , while after 20 . . ...d8
2 1 lDb6 ':'a7 22 .i.c3 lDd4 23 .i.xd4 exd4
16 lDc6 24 "d3 followed by 25 0-0-() the extra
Realising that he has lost the opening piece would be bound to have its say.
battle, Spassky decides on a desperate step 18 "e8
- he sacrifices a piece, in order to gain at 19 hxg6 hxg6
least some chance of counterplay. And 20 "g4 lZ.b8
Averbakh 's Selected Games 1 39
54 l2)f5 ':f8
55 :h5 ':e8
56 �e4 ':gl
57 :h3
The simplifying 57 l2)xg7 �xg7 58 e6+
�g6 59 :xc5 :xe6+ would have favoured
Black.
57 .i.f8
58 �d5 :d 1+
59 �e4 :ct
60 �d5 :d 1+
48 d3+ 61 �e4 :d7
The fifth miracle ! Spassky wins a
whole rook, emerging the exchange and a
pawn ahead, but in so doing he allows
White to activate his pieces and, in the end,
to save the game.
As Tolush showed, 48 ... .i.xe5 ! would
have won. If 49 .i.d2 or 49 :e4 there
follows 49 . . . c4, and it is not apparent how
White can defend against the numerous
threats.
49 �xd3 :xf4
50 .i.c3 :xf3+
51 �e4 :g3
52 �f4 :xg6 62 l2)h6+ .i.xh6
53 l2)e3 :b8 But here the exchange favours White,
Black decides to switch to defence. The since his pieces remain active. However, it
active 53 . . . :b3 could have been met by 54 was dangerous for Black to avoid it
�5, e.g. 54 . . . :e6 55 �f5 �f7 56 l2)c7 because of 63 l2)g4.
and White's initiative is sufficient for a 63 :xh6 :h7
draw. 64 :g6+ �f7
1 42 Averbalch 's Selected Games
65 l:t f6+ � e7 1 c4 e6
66 l:tc6 �d7 3 lLlc3 .i. b4
If 66 . . . %lh4+ 67 �d5 c4, then 68 %lxa6 4 e3 0-0
%ld8+ 69 %ld6 ! l:tc8 (or 69 . . . l:tb8 70 %lc6) 5 .i. d3 dS
70 %lf6 with the unpleasant threat of 7 1 6 lLlo cS
.i.b4+. 7 0-0 lLlbd7
67 l:t xcS l:th6
68 �dS l:tb6
69 .i.aS l:tbS
70 l:txbS axbS
Of course, I did not calculate all these have to play 28 "g5, and after 28 . . . :xg5
spectacular variations at the board, but 29 hxg5 'ii'xfl Black wins easily.
intuition suggested to me that, with
White's king so open and his pieces badly
placed, a tactical solution would always be
found.
22 .txe5 "xe5
23 �xh3
White does not believe in the
correctness of White's combination. 23
:fe I was relatively best, when I was
intending to reply 23 . . . "h5 with the threat
of 24 . . . :b2 !
23 :b6
24 � gl
There is nothing better. After 24 :g l Therefore White resigned.
there would have followed 24 . . . "h5+ 25
�g2 :g6+ 26 �fl "h3+ 27 �e2 :e8+, Game 44
when major loss of material for White is Averbakh-Fuchs
inevitable, while if 24 :fe I :h6+ 25 International Tournament
'ii'xh6 (25 �g2 :xh2+ 26 �g l "h5 with a Dresden 1 956
quick mate) 25 . . ... f5+ 26 �g2 gxh6. King's Indian Defence
1 c4 g6
2 ttlc3 .tg7
3 d4 ttlf6
4 e4 d6
5 .te2 0-0
6 .tg5 c5
7 d5 a6
8 a4 e6
9 "d2 "a5
24 :h6!
25 h4
If 25 f4 "h5 with decisive threats.
25 :g6+
26 �h3 "f5+
27 �h2 "f6!
The final finesse. It was not yet too late
to go wrong: after 27 . . ... h5 White would
save himself by 28 "f4. But now, to de
fend against the mate, my opponent would
Averbakh 's Selected Games 1 45
10 :a3 !
One of the new ideas that I tried to put
into practice in this variation - firstly, now
it is extremely difficult for Black to play
. . . b7-b5, and secondly, the rook may be
switched, as in the game, to h3 for an
attack on the kingside.
10 exd5
11 exd5 �bd7
12 �f3 �b6
The opponent' s plan is clear - he
intends to attack the c4 pawn with his
knight and queen.
13 0-0 16 �e4!
Unexpectedly the black queen finds
itself in a dangerous position, and in
addition the d6 pawn is hanging. If
16 ... �e5 there would have followed 1 7
"'h3 ! �bxc4 1 8 �f6+ Jlxf6 1 9 Jlxf6
"'d2 ! (the only defence; if 1 9 . . . �d7 White
gives mate in three moves - 20 "xh7+
�xh7 2 1 %lh3+ �g8 22 %lh8 mate) 20
%le3 ! �d7 2 1 Jlc3 �xe3 22 Jlxd2 �xfl
23 �xfl with a considerable advantage for
White.
Having decided that nothing good will
come of active play, my opponent begins
13 Jlg4 retreating over the entire front, but now
Sharp play would have resulted from White markedly strengthens the placing of
1 3 . . . ..,,4 1 4 "'c l ! ( 1 4 as �xc4 1 5 Jlxc4 his pieces.
"xc4 1 6 :a4 "'b3), for example 1 4 . . . �e4 16 �c8
( l 4 . . . �xc4 1 5 �a2 "xb2 1 6 Jlxc4) 1 5
�xe4 Jlxb2 1 6 �f6+ �h8 ( l 6 . . . �g7 1 7
Jlh6+!) 1 7 "'f4 "xa3 1 8 "'xd6 with
complications favourable to White.
However, the move played is no better.
It would appear that Black's entire strategy
is dubious.
14 "'f4! Jlxf3
15 "'xf3 �fd7
Probably B lack should have admitted
the faultiness of his plan and played
1 5 . . . �bd7, but instead he persists,
allowing White to develop a sudden attack
on the king. 17 "'h3 ! "c7
1 46 Averbakh 's Selected Games
20 ltJg3 ! Game 45
White has no reason to hurry, and so he Uhlmann-Averbakh
calmly prepares the sacrifice on h5 . After International Tournament
the immediate 20 .i.xh5 there could have Dresden 1956
followed 20 . . . l::txe4 ! 2 1 "ikxe4 gxh5 22 Nimzo-Indian Defence
l::txh5 ltJf8 and Black holds on.
20 ltJrs 1 d4 ltJf6
Against the attempt to neutralise the 2 c4 e6
opponent's attack by 20 . . . l::txe2 2 1 ltJxe2 3 ltJc3 .i. b4
.i.xb2 I was intending to reply 22 .i.dS ! 4 e3 0-0
.."S 23 l::te 3 . 5 ltJge2 d5
21 1.xh5! 1.xb2 6 a3 .i.e7
The acceptance of the sacrifice 7 cxd5 exd5
2 1 . . .gxh5 could have led to the following 8 ltJg3 c5!
finish: 22 ltJxh5 ltJg6 23 ltJf6+! .i.xf6 24 In my games with Reshevsky and
.i.xf6 ltJxh4 25 l::txh4, when mate cannot Kotov from the 1 953 Candidates Tourna
be avoided. ment I tried to manage without this move,
22 ltJf5! but both times I had an inferior game.
Averbakh 's Selected Games 1 47
10 J.d3
Here Euwe played 1 0 b4, but after
I O . . . d4 ! he ran into difficulties. For
example, if I I lLla4 there could have Black has a lead in development and
followed I I . . . dxe3 ! 1 2 'it'xd8 ( 1 2 lLlxc5? now begins active play.
exf2+ 13 �e2 J.g4+) 12 ... exf2+ 13 �e2 17 b3 g6
-*.g4+ and Black wins. 18 J.h3
10 lLle6 Uhlmann intends to play his bishop to
11 0-0 J.e6 g2, to attack the d5 pawn.
12 lLlee2 J.d6 18 lLle4
19 'it'd3 J.e7
20 g3 J.f6
21 J.e3 :e7
22 :ac1 :ae8
23 :fd l 'ii'd 6
13 lLld4
By giving up attempts to exploit the
weakness of Black' s isolated pawn, White
even ends up with a slightly inferior game.
13 lLlxd4
1 48 Averbakh 's Selected Games
31 he3
Uhlmann risks accepting the sacrifice,
assuming that Black' s attack will lead only
to a draw. He could have defended suc
cessfully only by declining the sacrifice
with 3 1 ':c3 . Then 3 1 . . . b6 ! 32 "xa6 ':xc3
Averbakh 's Selected Games 1 49
1 e4 c5
1 lDt3 lDc6
3 d4 cxd4
4 lDxd4 lDf6
5 lDc3 d6
6 .i.g5 a6
I frequently used to employ this
defence against the Rauzer Attack 6 .i.g5,
trying to manage, initially, without the
move . . . e7-e6, and aiming for the rapid
development of play on the queenside.
Here the game was adjourned, and 7 "'dl .i.d7
White resigned without resuming. After 8 � b5
42 �t2 "'f4+ 43 �e2 "'xc 1 44 "'xd5
"'c2+ 45 �e3 "'c6 46 "'e5+ "'f6 Black
realises his advantage without difficulty.
22 fJ
A serious mistake, which leaves White
Thus the game has gone by force into a in a critical position. 22 c4 was still
complicated ending, where White has three correct, and if 22 . . . l:td7 23 cS . Then in
pawns for a bishop. His strategy is simple reply to 23 . . . l:tbd8 White has the brilliant
Averbakh 's Selected Games 151
Game 47
Averbakh-Polugayevsky
25th USSR Championship
Riga 1958
Modem Defence
1 c:4 g6
2 d4 .i.g7
28 l£lb7 ! 3 e4 d6
Here too this manoeuvre leads to the By choosing this move order, Black
win of another pawn, and with it the game. retains the option of developing his king's
29 �b2 l£ld6 knight at f6 or e7, or even at h6.
1 52 Averbakh 's Selected Games
10 �h3!
An important move, the aim of which is
to exchange the light-square bishops.
10 ltJf6
11 "0 a6
Black probably has better chances of
equalising after 1 1 .. .f4 1 2 �d2 h5, as in
the game Portisch-Ree (Amsterdam 1 967),
7 g4 ! but my opponent persists in trying to avoid
This aggressive advance has the aim of the exchange of bishops. With the move
restricting the opponent' s activity. If 7 fl played he plans to switch his queen to the
there could have followed 7 . . . fS, or alter kingside, by freeing it from having to
natively 7 . . . �h6, exchanging the dark defend the c7 pawn.
square bishops. 12 exfS 1i'e8
7 fS 13 ltJge2 "hS
To 7 . . . h5 I was intending to reply 8 g5 14 1i'xhS ltJ xh S
fS 9 fl . Not wishing to reconcile himself to
the cramping of his pieces on the kingside,
Polugayevsky nevertheless makes the
thematic advance of his f-pawn, even
though it involves losing the right to castle.
8 gxfS
Here 8 g5 is no longer effective: after
8 . . f4 9 �d2 h6 ! Black frees himself.
.
8 gxfS
91i'hS+ �f8
9 . ltJg6, trying to avoid the loss of
. .
22 c6
When in a cramped position, one
should not create unnecessary weaknesses.
This attempt to open lines, in order to gain
counterplay, merely hastens Black's
defeat. It was probably better to stick to
waiting tactics with 22 . . . :g8 23 lDef6
:h8, keeping the threat of . . . c7-c6 10
reserve.
23 �d3 cxd5
16 lDg3 ! 23 . . . bS would have merely led to the
This knight move, which had to be activation of the white king: 24 lDhg3 !
foreseen earlier, is the whole point. Now bxc4+ 2S �xc4 cxdS+ 26 �xdS :c2 27
Black faces a difficult choice. If lDfS, and if 27 . . . :xb2 28 :b l .
1 6 . . . lDfxg3 , then 1 7 hxg3 .txh3 1 8 :xh3 24 cxd5 .tr4
lDf6 1 9 �e2 or 1 9 .txf6 .txf6 20 lDe4, The last chance of a defence was
while after 1 6 . . . lDhxg3 there would have 24 . . . :g8, and if 2S lDhg3 :g4. Now,
followed 1 7 fxg3 ! lDd4 1 8 0-0+ �e8 1 9 however, there is no defence against the
.txc8 :xc8 2 0 lDe4, in each case with transference of the knight to fS .
advantage to White. 25 lDg7 :a8
16 lDd4 There is nothing better. If 2S ... :g8 26
17 .txc8 :xc8 lDfS l:txg l 27 :xg l :d8 (defending
Black also has a difficult position after against the threat of 28 :g8+! �xg8 29
1 7 . . . lDxg3 1 8 fxg3 lDf3+ 1 9 �e2 lDxgS 20 lDe7+ �f8 30 lDxc8, winning the d6 pawn)
:afl + �e7 2 1 h4 ! lDfl 22 .te6 lDd8 23 28 lDf6 lDh6 29 lDxh6 .i.xh6 30 :g8+ �e7
.tfS, but this would evidently have been 3 1 :xd8 �xd8 32 lDxh7 and White wins .
the lesser evil. 26 lDfS :d8
18 lDxh5 lDf3+ 27 :g4!
19 �e2 lDxg5 The concluding finesse. By the threat of
20 :agl .th6 28 :xf4 White prevents 27 ... lDh6, and the
21 h4 lDf1 threat of doubling rooks on the g-file
22 lDe4 decides the outcome in a few moves.
1 54 Averbakh 's Selected Games
27 :g8 Game 48
28 lbg8+ �xg8 Averbakh-Fischer
29 :gl+ �h8 Interzonal Tournament
30 :g7 Portoroz 1 958
Black resigns: if 30 . . . �h6 there King's Indian Defence
follows 3 1 �f6 with inevitable mate.
1 d4 �f6
I played against Robert James Fischer on 1 c4 g6
only one, single occasion. This occurred at 3 �c3 .i.g7
the very start of the brilliant career of the 4 e4 d6
future eleventh Champion of the World. 5 .i.e2 0-0
Fischer was just 14 years old when he 6 .i.g5
won the USA adult Championship, a Without particular thought I decided to
whole point ahead of the second-placed employ my variation against the young
Samuel Reshevsky, and gained the right to American. It now seems to me that this
take part in the World Championship decision was not altogether correct. By that
Interzonal Tournament. This tournament time, when the Interzonal Tournament took
took place in the Autumn of the following place, the variation had become firmly
year in one of the Yugoslav seaside resorts. established in tournament play. Fischer
It was there that I saw Fischer for the first undoubtedly knew it, and had prepared
time. He turned out to be a skinny teenager specially for it.
in jeans and sweater, rather shy and even 6 h6
unsociable in his dealings with others. 7 .i.e3 c5
Bobby, as the participants and journalists 8 d5
quickly began calling him, gazed without Only in the mid- 1 970s was it estab
the slightest interest at the colourful lished that after 8 e5 ! White can count on
countryside of the sunny Adriatic. He was an opening advantage. For example:
not attracted either by the golden sand, or 8 . . . �fd7 9 exd6 exd6 1 0 �f3, or 8 . dxe5
. .
the deep blue sea. All his thoughts were 9 dxe5 "'xd 1 + 1 0 :xd I �g4 I I .i.xc5
concentrated on chess. Chess comprised �xe5 1 2 �d5 .
his inner life, and he was completely The move made by me in the game
absorbed in it. And at the chess board this does not promise White much.
Averbakh 's Selected Games 1 55
8 e6
9 h3
Before developing his knight at fl ,
White has to defend against the threat of
9 . . . lLlg4. 14 a6!
9 exdS My opponent accepts the challenge,
10 exdS :e8 although the less sharp 1 4 . . . g5, preparing a
11 lLlo �f5 retreat for the bishop, was also possible.
15 h4
Throwing caution to the winds. Perhaps
1 5 a4 would have been less committing,
and if 1 5 . . . b5 1 6 axb5 axb5 1 7 :xa8
"'xa8 1 8 lLlxb5, also with great
complications.
15 bS
Now commences forcing play, as a
result of which White succeeds in winning
a piece, but Black acquires dangerous
threats.
16 gS b4
17 gxf6 bxc3
12 g4 18 lLlxe4 ':xe4
I reckoned that after 1 2 0-0 lLle4 1 3 19 fxg7 "'xh4
lLlxe4 �xe4 White has not even a hint of 20 �f1 cxb2
an advantage, and I devised an extremely 21 :bl
double-edged plan involving a pawn attack In this wild position, where we each
on the kings ide, using up a considerable had about ten minutes left for our
amount of time in the process. remaining moves, I suddenly heard almost
12 �e4 in a whisper: 'Draw? ' . I would not like
13 :gl playing such a position in time trouble
Much later in a game with Geller I either with White, or with Black; besides,
played 1 3 0-0 here, but 1 3 . . . �xfl 1 4 my tournament position was quite good,
1 56 Averbakh IS Selected Games
and there was no point in taking a risk. subtleties. My opponent, who was evid
Therefore I accepted the offer. ently aware of this, deviates from the
As is well known, later Fischer himself theoretical 9 . . . d5 in favour of a less well
very rarely used to offer draws. And when studied continuation.
on one occasion he was asked about this
game, he replied:
'Averbakh was afraid of losing to a
child, and I was afraid of losing to a
grandmaster. And so we agreed a draw! '
9 .i.e6
10 �bl "'a5
But here Larsen shows himself to be
not very well versed in the chosen
variation . The move played is a strategic
To be fair, I should mention that after mistake, condemning Black to a difficult
2 1 . . . •h3+ 22 :81 .h l + 23 1%g l .h3+ defence. Geller' s recommendation of
the game could have ended in a draw by 1 0 . . . ttJe5 was better.
perpetual check. 11 ttJxe6 fxe6
12 .i.c4 ttJd8
Game 49 13 ttJd5! .xd2
Averbakh-Larsen 14 ttJxf6+ .i.xf6
Interzonal Tournament 15 1%xd2
Portoroz 1958
Sicilian Defence
1 e4 c5
2 ttJa d6
3 d4 cxd4
4 ttJxd4 ttJf6
5 ttJc3 g6
6 .i.e3 .i.g7
7 a 0-0
8 .d2 ttJc6
9 �
In the 1 950s I myself happily played
the black side of the Dragon Variation, and Thus the game has gone by force into
I think that I had a fair understanding of its an ending in which, because of his weak
Averbakh 's Selected Games 1 57
23 l:df3 !
Threatening 24 e5 . To defend against it,
Black aims for play on the queens ide, but
this only leads to White's initiative on the
kingside developing into a powerful attack. 34 l:h6+! l:xh6
23 �c6 35 jtxh6 l:c4
24 c3 e5 There is no other defence against 36
25 f5 �a5 l:h l .
1 58 Averbakh 's Selected Games
Game 50
Korchnoi-Averbakh
26th USSR Championship
Tbilisi 1 959
Sicilian Defence
1 e4 c5
2 lLlo lLlc6 I decided to switch to normal lines of
3 d4 cxd4 the Scheveningen Variation, although here
4 lLlxd4 .c7 Black had a wide choice of continuations,
such as 7 . . . b5 or 7 . . lLlxd4 8 "'xd4 b5 etc.
.
11 0-0 �f6
12 �d2 tLld4
13 tLlxd4 �xd4+
14 �hl tLlc5
A poor move, which is essentially a
loss of time. I should have castled, and
only after I S .h3 played I S . . . tLlcS .
18 �b6
19 f5
White immediately launches an attack
on the kingside. If 1 9 ... f6 there could have
followed 20 tLlf4 eS 2 1 tLlg6 ! , when the
strength of his attack is demonstrated by
the following variations:
15 :ac t !
This simple move refutes Black's idea,
since, with him being behind in develop
ment, the exchange on d3 favours White.
15 0-0
I 5 . . . �d7 did not appeal to me because
of the same move 1 6 b4, for example
1 6 . . . tLla4 1 7 tLle2 �b2 1 8 :b l , and it is
not possible to capture on a3, as there
follows 1 9 :b3 �b2 20 c3 . However,
1 8 . . . .i.f6 was possible, and if 1 9 c4 tLlb2.
At any event this was the lesser evil, as
after the move in the game Black has no
counterplay at all. (a) 2 1 . . .hxg6 22 fxg6 :e8 23 .hS �e6
16 b4 tLlxd3 24 .h7+ �f8 25 �h6 .e7 26 .h8+ �g8
17 cxd3 .d8 27 �xg7+ .xg7 28 :xf6+;
18 tLle2 (b) 2 1 . . .:e8 22 .h5 �e6 23 :0 �fl
When I chose 1 4 . . . tLlcS I was pro 24 .xh7+! <iifxh7 25 :h3+ <iif g 8 26 :h8
posing in this position to retreat my bishop mate;
to f6, but it was only now that I noticed (c) 2 1 . . .:n 22 11'h5 :c7 23 :0 :xc l +
that after 1 9 tLlg3 .i.d7 ( 1 9 . . . g6 20 fS) 20 24 .i.xc l .e8 (24 . . .•c 7 2 5 .xh7+ ! <iif fl
tLlhS .i.c6 2 1 .g3 ! Black has serious 26 .i.h6) 25 :h3 hxg6 26 .h8+ �n 27
problems. fxg6+ �e7 28 .xg7+ �e6 29 :h8 etc.
However, the move played in the game 19 e5
is no better. 20 .h5 f6
1 60 Averbakh 's Selected Games
32 lLlg3
If 32 h6 there would have followed
32 . . . "'h5+ 33 'ii'xh5 �xh5 34 lLlg3 �g4
3 5 �h2 i.. f2 36 lLlh l i.. d4 with the threat
of 37 . . . �e2.
Strangely enough, this simple reply 32 �f8
came as a surprise to my opponent. 33 h6 gxh6
Averbakh 's Selected Games 161
35 rilh2
Korchnoi does not appreciate the dan
gers of his position and continues to play 39 �h6
for a win. With 3 5 :c l he could probably This loses a piece. However, even after
have achieved a draw, e.g. 35 . . .l:lxc l + 36 the comparatively better 39 �c l �xg3 40
�xc l .h7+ 3 7 .h3 .xh3+ 3 8 gxh3 �b5 .xg3 .h7+ 41 .h4 .xh4+ 42 rilxh4
39 rilg2 �xd3 40 rilfJ etc. :xg2 Black would have had good winning
35 .h7 chances.
36 .h4 39 �xg3
36 .g7+ .xg7 37 �xg7 �e3 ! would 40 �xg7 �xh4
have left the white bishop shut out of the 41 �xh4 :xgl
game, but 36 .h3 was more accurate.
36 .h8!
1 e4 c5
2 lDfJ lDc6
3 d4 cxd4
4 lDxd4 lDf6
5 lDc3 d6
6 .*.g5 e6
7 Wd2 .*.e7
8 0-0-0 �
9 lDb3
11 'irc7
Now, when the d6 pawn is securely
defended, Black sets about implementing
his main strategic plan - a pawn attack on
the queens ide.
12 g4 a6
13 .*.h4
What is this - improvisation or a
thoroughly prepare� variation? At first I
thought that White h ad simply blundered
away a pawn - 1 3 . . . lDxe4 1 4 �xe4 .*.xh4,
Modem theory considers 9 f4 to be the but of course I then saw that after 1 4
strongest, but Bronstein always goes his Ilxe4 ! .*.xh4 1 5 g 5 the black bishop would
own way. However, the move played was be trapped. However, the white bishop is
also not new, and had occurred several not too well placed at h4.
times in events in the 1 950s.
9 'irb6
10 fJ l:d8
11 Ilel
The usual continuation here was 1 1
.*.e3 'irc7 1 2 'irfl, temporarily preventing
1 2 . . . a6 and switching the queen to the
kingside. Thus, for example, Fischer
Benko (USA Championship 1 959/60) con
tinued 1 2 . . . lDd7 1 3 �b5 'irb8 1 4 g4 a6 I S
�5d4 with a slight advantage to White.
However, Black can reply 12 ... d5 13 exd5
�xd5 1 4 �xd5 Ilxd5 1 5 Ilxd5 exd5 with
Averbakh 's Selected Games 1 63
13 b5 22 .c7
14 g5 tDd7 Of course, not 22 . . ....xf2 23 .ixf2 :b8
15 f4 tDb6 24 f6, and White has nothing to fear.
16 ...n 23 .id3 .if8
24 .ig3
22 tDf3
White defends his gS pawn, but he
should probably have sacrificed it by 22
tDbS . After 22 . . . •xf2 23 .ixf2 .ixgS+ 24
�b l :b8 2S tDc7 :d7 (2S . . . :b7 26 .ia6!) 27 .ixc4
26 .ixb6 :xb6 27 tDd s it seems to me that 27 .c3 tDa3+ 28 �c l .xc3 29 bxc3
he has full compensation for the pawn. tDxc2 would have led to a better ending for
1 64 Averbakh 's Selected Games
15 "'e2 ? 22 "'d8!
Now White's position becomes critical . Beginning a tactical sequence that leads
15 .i. g4! by force to a win. The bishop at d3 is
16 B exB threatened, and 23 :xe3 is not possible
17 gxB .i.h3 because of 23 . . . ... g5+. Finally, if 23 .i.e4
"'g5+ 24 "'g3 "'xg3+ 25 hxg3 :xe4 26
fxe4 lilc2 and Black should win this
ending, although this was a more tenacious
defence than that in the game.
23 .i.e1
My opponent hopes for the simplifi-
cation after 23 ......xd3 24 .i.xe3 "'xc3 25
.i.xa7, when 25 . . . :xe l + 26 :xe l "'xa3 is
not possible because of 27 .i.c5, but Black
finds a combinational continuation of the
attack.
18 l:tfel
1 8 l:tfd 1 would have maintained
material equality, e.g. 1 8 . . . lilxd5 1 9 lilxc4
lilxc4 20 :xd5, or 1 8 ... :ac8 1 9 lile4,
although in both cases Black has an
undisputed positional advantage.
18 lilxd 5
19 lilxe4 lilxe4
20 .i.xe4
If 20 "'xc4 "'b6 ! with numerous
threats.
1 66 Averbalch 's Selected Games
23 tL)g4! 6 i.e3 c6
24 i.xh7+ �xh7 7 "'d2 a6
25 fxg4 :xel+ 7 . . . :eS came into consideration, to
26 "'xel :e8 answer S i.h6 with S . . . i.hS.
27 i.e3 :e4 8 i.h6 b5
28 "'bl f5 9 O-O-O!
29 i.f2 llxg4+
30 i.g3 %lg6
White resigns, as he loses a piece.
Game 53
Averbakh-Petrosian
Moscow Team Championship 1961
King's Indian Defence
1 d4 tL)f6
2 c4 g6
3 tL)c3 i.g7
4 e4 d6
5 f.3 A couple of months before this game I
met Petros ian in the 2Sth USSR Cham
pionship, but instead of castling I played
the less accurate 9 h4. After 9 . . . e5 1 0 d5
cxd5 1 1 cxd5 tL)bd7 1 2 g4 i.xh6 1 3 "'xh6
�hS 1 4 h5 tL)gS 1 5 "'e3 g5 ! White did not
achieve anything in particular, and the
game quickly ended in a draw. Therefore
Petros ian was not averse to repeating this
variation, but I had prepared an improve
ment, for the moment preventing 9 . . . e5.
9 i.e6
Here 9 . . . ...a5 1 0 �b l :eS is usually
played.
As the Averbakh Variation gained in 10 h4!
popularity, I myself began playing it In such positions, where White's main
comparatively rarely. Firstly, I did not like task is to get his attack in first, pawns do
repeating one and the same thing, and not count. For this reason Black's previous
secondly, a certain feeling of satiation had move was dubious.
set in. Therefore I began employing the 10 i.xc4
Simisch Variation, especially when I was 11 i.xc4 bxc4
aiming for a win. 12 h5 i.h8
5 0-0 Having a pawn in reserve, Petros ian is
If Black does not castle early, White ready to give up the exchange, but White
carri es out the same plan, aiming for the does not fall for the bait and continues his
exchange of the dark-square bishops. offensive.
Averbakh 's Selected Games 1 67
13 bxg6 fxg6
14 �b3 l:te8
1 4 . . . �bd7 1 5 �g5 is unpleasant for
Black, so he decides to move his rook in
order then to play . . . e7-e6.
IS .e2 e6
16 g4!
Attack, first and foremost! That is
White's motto.
16 �bd7 It appears that White has managed to
17 f4 .as get at the enemy king. After 20 . . . dxe5
there follows 2 1 .c2, e.g. 2 1 . . . rj;f7 22
�g5+ <i>e7 23 .xg6. If now Black replies
23 . . . exd4 24 l:xd4 �5, then 25 %le 1 ! ,
when 25 . . . �xg6 allows 26 l:xe6 mate, and
so Black is obliged to give up the exchange
with 25 . . . %lf8 .
Seeing that the ground is slipping from
under his feet, and striving at any cost to
change the unfavourable course of events,
Petrosian, like a wounded tiger, resorts to
desperate measures.
20 %lxb2
Now it is White who has to defend
18 IS accurately, in order himself not to come
Disregarding possible loss of material, under a strong attack.
White aims to open up the position of the 21 <i>xb2
enemy king. After 1 8 . . . exf5 1 9 gxfS �xe4 Worse is 2 1 .xb2 l:b8, when after 22
I was intending to reply 20 fxg6 ! �xc3 2 1 .c2 .a3+ 23 <i>d2 l:b2 Black wins the
gxh7+ , and if 2 1 . . 5�)xh7 22 .c2+. During queen, while if 22 .a l then simply
the game I thought that after 2 1 . ..<i>f7 I 22 . . . dxe5 .
could sacrifice a piece - 22 .xc4+ �d5 21 l:tb8+
1 68 Averbakh 's Selected Games
22 �c2 lDd5 1 c4 g6
23 'ii'xc4 g5 2 d4 lDf6
If 23 . . . lDxc3 there would have followed 3 lDc3 .t g7
mate in two by 24 'ii'xe6+ and 25 lDg5 . But 4 e4 0-0
now 24 . . . lDe3 is threatened. 5 .t e3 d6
6 f3 lDc6
There are various ways of opposing the
Samisch Variation. The piece set-up em
ployed by Black was then the latest word
in fashion.
7 'ii'd 2 a6
8 0-0-0
Despite the opponent' s obvious inten
tion of attacking on the queenside, White
accepts the challenge and castles long.
8 .td7
Black also plays 8 . . . l:b8 here, or even
8 . . . b5, in any case with a complicated
24 l:d3 lDb4+ game.
25 ci>d l d5
26 'ii' b3 c5
Taking the rook 26 . . . lDxd3 would have
been met by 27 1IIc 2, e.g. 27 . . . lD7xe5 28
dxe5 .txe5 29 'ii'xd3 'ii'xc3 30 1IIg6+ .tg7
(30 . .ci>h8 3 1 .tg7+ .txg7 32 lDxg5+ with
.
12 lLlaS
A poor move. In cut-and-thrust play of
this type, pawns do not count. Black wins a
pawn, but allows his opponent a strong White has a wide choice. One possibil
initiative. After 1 2 . . . hS 1 3 lLldS bxc4 1 4 ity, for example, is 1 7 fxeS dxeS 1 8 hxg6
lLlxf6 1i'xf6 I S d S lLle7 1 6 lLlc3 White fxg6 1 9 dxeS "xeS 20 1i'h6+ �f7 2 1
would have had only slightly the better l:txd7+ lLlxd7 22 1i'xh7+ �e8 (22 . . ...g7 23
chances. l:tf1 + lLlf6 24 l:txf6+ �xf6 2S eS+ �f7 26
13 lLlg3 ! lLlxc4 e6+ �f6 27 lLlge4+) 23 "xg6+ �d8 .
14 .i.xc4 bxc4 However, here I did not see how I could
IS hS "e7 continue the attack, and after greatly
prolonged thought I opted for:
17 dxeS dxeS
18 G :th8
The threat of 1 9 hxg6 fxg6 20 "h6+
looks too dangerous to be ignored.
1 9 h6+ �f8 20 :hfl was tempting, 26 . . . �g8 27 :xg6+) 27 :xf7+, again with
disrupting the coordination of the black mate. My opponent was evidently relying
pieces. However, this would have relieved on the splendid resource 24 . . . .tg4+ 25
the tension on the kingside. I preferred a "'xg4 (25 �e l loses to 25 . . . ...c 1 + 26 �f2
sharper move, after which the opponent "'xd2+ 27 "'xd2 :xb2) 25 . . . :xb2. But
would acquire counter-threats. However, to even so the final word remains with White.
attack, without allowing the opponent any With 26 :xf7+! �xf7 (26 . . . �h8 is no
counter-play, is not something that better: 27 :h7+! �xh7 28 hxg6+ etc.) 27
happens often ! hxg6+ ! (27 "'xg6+ leads only to a draw)
19 hxg6 he mates the black king.
20 "'gS "'b4 Thus 23 :xf6 ! was the quickest way to
The desperate attempt to complicate win, albeit a complicated one, since if
matters with 20 . . . :xb2 is most simply 23 . . . :xf6 White has 24 h6+.
countered by 2 1 :xd7 "'xd7 22 :d 1 . 23 :e6
21 :d2 :b6 24 "'xc7 :b7
22 :f1 :hb8 2S "'r4 .tc6
Taimanov, a resourceful tactician, has If 25 ... :eb6 26 "'e5 :e6 27 "'d4,
the ability to set clever traps, and when threatening both 28 h6+ and 28 �d5 .
playing him you must always be on your 26 hxg6 fxg6
guard. 27 "'gS �h 7
28 �hS+ �h8
29 "'h6 "'e7
30 �r4 "'f7
23 "'xeS
A simple solution to the problem .
Without reducing the tempo of the attack,
White gains a material advantage. At the Black's position is hopeless. White has
same time he avoids a trap - 23 :xf6 both an extra pawn, and a powerful attack.
"'xc3+ and if 24 bxc3 there follows mate My opponent's only chance was that I was
in two moves. This all seems very element in severe time trouble. When there are only
ary, but White has the unexpected reply 24 seconds left on your clock, it is not easy to
�d 1 ! , when it is not clear who has caught calculated even the most elementary
who. Thus if 24 . . . :xb2 25 :xg6+ with a combination. The variation 3 1 :d8+ (the
quick mate. 24 . . . .ta4+ 25 �e2 :xb2 is no immediate 3 1 �xg6+ is also possible)
better because of 26 �fS+ �f8 (or 3 1 . . . .te8 32 �xg6+! flashed through my
Averbakh 's Selected Games 171
Game 5 5
Averbakh-Aronin
Moscow Team Championship 1 956
Sicilian Defence
1 e4 c5
2 �fJ d6 12 :hel !
3 d4 cxd4 White does not waste time on the
4 �xd4 �f6 defence of his central pawn. Its capture
5 M a6 would lead to the opening of the e-file,
6 .i.e2 g6 which would be fraught with danger for
7 .i.e3 .i.g7 Black, as his king is held up in the middle
8 .d2 �bd7 of the board.
1 72 Averbakh 's Selected Games
17 �e6! �xe6
1 7 . . . 0-0-0 leads after 1 8 �xdS exdS to
a bad position without any counter
chances, so Black throws caution to the
winds and accepts the sacrifice.
18 �xe6! fxe6
19 lIxe6+ �d8
16 �xbS! 20 lIdxd6!
This is the point! Black was hoping for
1 6 �xbS axbS 1 7 lIxdS lIxa2, intending if
1 8 �xbS to reply 1 8 . . . lIa l + 1 9 �d2 lIxe l
20 �xe l 'fIxc2, as 2 1 �xd7+ �xd7 22
'fIxfS is not dangerous because of
22 . . . 'fIe4+ and 23 . . . 'fIxdS .
16 e6
If 1 6 . . . axbS White was planning a
complicated combination on the theme of
the double attack: 1 7 �xbS 'fIaS ( 1 7 . . .'fIc6
1 8 lIxdS �b6 leads to a spectacular finish
- 1 9 �xd6+ �d7 20 �4+! ) 1 8 �xd6+!
�d8 1 9 'fIxf8+! �xfS 20 �b7+ �c7 2 1
�xaS lIxas 22 b4 lIbS 23 a4, remaining The climax of White's combination. 2 1
the exchange up. 'fIxfS mate is threatened, and 20 . . . lIe8
Averbakh 's Selected Games 1 73
28 :e2
A clever defence. If 29 1IIxaS there
follows 29 . . . :xd2 30 1IIxd2 �xa4.
29 l:xe2 :xe2
30 'ifxaS �e4
This seemingly active move simplifies
White's problems. However, even after
30 . . . �xg2 3 1 'ifa7+ �b7 32 as �c8 33
White's attack has led to an original 1IIc 5+, followed by the advance of his
position with an unusual balance of forces: queenside pawns, he would have won.
queen and three pawns against rook, 31 1IIe5+ � b7
bishop and knight. One might argue about
which side has a material advantage, but
the fact that the black king has no pawn
protection and has to be defended by
pieces forces the position to be evaluated
in favour of White. However, in order to
realise his advantage he must act
energetically, otherwise the black pieces,
after uniting, may be able to assail the
opponent's king with great force.
22 l:d2 l:ae8
23 b3 l:e1+
24 �b2 l:fe8!
After doubling on the e-file, Black 32 �a3 !
intends to exchange White's only rook. Due to the pin, the c2 pawn is immune.
25 1IId 4 �b6 32 �dS
The threat of 25 1IIa7+ has to be 33 as �c7
parried. 34 a6+!
26 a4 as The concluding finesse. The pawn
27 111c3 l:leS cannot be taken: if 34 . . .�xa6 35 1IIxc7, or
28 f4 34 . . . �xa6 3 5 111b 5+.
1 74 Averbalch 's Selected Games
Before this the main continuation was brings fresh reserves into the attack. Less
considered to be 9 .a4+ lbc6 1 0 lbxc6 clear is I S lbc7+ �d8 1 6 lbxa8 WcS ! with
.i.xc3+ I I bxc3 .i.d7 with an acceptable the two threats of 1 7 . . . Wxc4 and
game for Black. 1 7 . . . lbf2+.
9 .i.xc3+
10 .xc3 lbxe4
The little show I put on had achieved its
aim - my opponent did not even suspect
that he had fallen victim to a prepared
variation. Now Black comes under a very
strong attack.
15 lbef6
If I S ... lbecS, defending the e6 pawn,
then 1 6 .i.e3 1i'h4 1 7 lbd6+ and 1 8 .i.gS+
leads to the win of the queen.
16 .i.xe6! .xb2
The bishop could not be taken because
of mate in two moves .
11 lbb5! •c5 17 %let
I think it was at this move that Estrin
terminated his calculations. By attacking
the f2 pawn he was hoping to gain time for
the defence. However, the check at f2 is
not yet mate, and there are only two pieces
participating in Black' s attack. Therefore,
without concerning himself over defence,
White continues his offensive.
12 .xg7! %lf8
13 .i.h6
Continuing to ignore the opponent' s
threat.
13 .xfl+
Probably the toughest defence was A tragi-comic position ! 1 7 . . . •xbS
13 lbd7 1 4 0-0 .xc4 IS a4 ! a6 1 6 %lac l
... loses to 1 8 .i.c4+, and meanwhile White is
axbS, giving up the queen. Now Black threatening 1 8 .i.xd7+ �xd7 19 %lc7+ �d8
loses by force. 20 .xf8+, a line which also wins after
14 �d l lbd7 1 7 . . .•b4 - 1 8 .i.xd7+ �xd7 1 9 %lc7+ �d8
15 l:tel ! 20 .xf6+ and mates. Therefore Black
At d I the king feels fine, and White resigned .
1 76 Averbakh 's Selected Games
Game 57
Khasin-Averbakh
Moscow Team Championship 1964
Sicilian Defence
1 e4 cS
2 lLlf3 lLlc6
3 d4 cxd4
4 lLlxd4 g6
Inviting White to play 5 c4, to which I
usually used to reply 5 . . . .1g7 6 .1e3 lLlf6
7 lLlc3 lLlg4 S 1Wxg4 lLlxd4 9 1Wd l lLle6
with a complicated game, but my opponent
prefers a more proven alternative. 10 lLlxd4
S lLlc3 .1g7 11 .1xd4 bS
6 .1e3 lLlf6 12 gS lLlhS
7 .1c4 0-0 13 .1xg7 �xg7
8 .1b3 14 1i'd4+
After exchanging the dark-square
bishops, my opponent thought that with
this check he would gain the advantage.
However, Black finds an excellent
resource.
15 .xd6 'ifxg5
16 'ifxd7
1 6 'ifd2 was more cautious, of course,
but my opponent evidently still had
pleasant memories of our meeting in the
29th USSR Championship (Baku 1 96 1 ).
There, in a game from the last round that
was exceptionally important for me, he had
boldly accepted my sacrifices and won.
However, history does not always repeat
itself!
5 d4 d5 13 . c2 !
6 a3 After the immediate 1 3 c4 Black could
have exchanged queens, when White's
chances of an attack are reduced prac-
tically to nil.
13 h6
1 3 . . . g6 is also unattractive.
14 c4 dxc4
1 4 . . . d4 loses a pawn after I S .te4,
while 1 4 . . . .te6 I S lUd I is unpleasant for
Black. It would appear that he has to take
on c4.
15 .txc4
6 lDe4
In reply to White's unpretentious
opening, Black has several good contin
uations: 6 . . . .te7, 6 . . . a6 and finally
6 . . . cxd4. As for the move chosen by Black,
I do not particularly like it, although it is
sometimes played.
7 .td3 lDxc3
8 bxc3 .te7
9 0-0 0-0
10 .tb2 b6
This leads to difficulties. The natural 15 �h8
I O . . . dxc4 I I .txc4 b6 would have led to a How else can Black defend against the
complicated struggle, in which White has threat of 1 6 .g6?
only slightly the better chances. He loses a pawn after I S . . . .te6 1 6 Wc3
11 cxd5 exd5 .f6 1 7 .xf6 gxf6 1 8 .ta6, while if
12 dxc5 .txc5 I S . . . lDe7, then 1 6 1Ud l .c7 1 7 .c3 lDfS
1 8 e4.
16 .e4
The attack develops of its own accord.
First of all White switches his mighty
queen to the kingside, and he does it with
gain of tempo.
16 .tb7
Already this is probably the decisive
mistake. By playing 16 . . . .td7 Black could
have hoped for a successful defence,
whereas now he comes under a powerful
attack. Events develop by force.
17 :tfd l .c8
1 80 Averbakh 's Selected Games
18 i.d3 !
Forcing the opponent to open up still
further the position of his king.
18 IS
19 "'h4 "e6 23 "'gS
Mate follows after 1 9 . . . �h7 20 lLlg5+ The main variation of the combination
�g6 2 1 i.e2. was much more interesting: 23 . . ...h7 24
20 i.c4 "'g6 i.d5 l:ac8 25 i.e6 ! l:ce8 26 i.xfS ! l:xfS
21 l:d7 i.e7 27 l:xg7 l:b5 28 lLlg6+ "xg6 29 l:xg6+
There is no other way of defending the l:xb2 30 l:xh6+ �g7 3 1 l:xc6, and White
g7 pawn. After 2 1 . . . lLld4 22 exd4 i.xf3 23 wins.
"g3 or even 23 l:xg7 White wins a piece. 24 f4 "'g4
If 23 . . ....g4 immediately, then 24 f3
"'g5 25 f4 wins.
22 l:xb7 ! ?
Spectacular, but by no means obliga
tory. White could have gained a material 25 i. e2 !
advantage in a less complicated, prosaic, Black resigns
Averbalch 's Selected Games 181
If 2 5 . . . ...xe2 there follows mate in two does so only at the cost of allowing White
moves, while after 25 . . ....xh4 26 J..xg7+ decisive scope elsewhere. The various
�g8 27 J.. c4+ %lf7 28 %lxf7 Black has no catastrophes which could have befallen
defence against the numerous threats, e.g. Black, had he not kept a very careful watch
28 . . . lLla5 29 J.. e6 %le8 30 %le7+ �h7 3 1 on these pawns, are made clear in the
J.. f6+ %lxe7 32 J.. xf5+ �g8 33 J.. xh4) 26 notes.
lLlg6+ etc.
Game 59
Averbakh-Hug
International Tournament
Palma de Mallorca 1972
Dutch Defence
57 ':'g7+ and mates) 57 e6 winning easily. Naturally White does not oblige by
A clear warning to Black that the pawns exchanging the passive black rook, unless
will rush forward at the slightest excuse. he can gain some substantial compensating
54 :f8 � g7 advantage such as the freeing of his pawns;
55 :c8 ':'f7 besides, he has plans for manoeuvring
56 �c1 behind Black's lines with both rooks.
For the moment White's threats on the 59 :f7
kingside are contained, so the king starts 60 ':'g5 :15
its long journey to c5, after which the 61 ':'c7+
combined pressure on both wings is almost
certain to break Black's resistance.
56 �h6
57 :g5!
61 � g8
If 6 1 ...�h6 White calmly proceeds with
his king's walk - 62 �a3 ! , since Black
cannot capture on g5 . If 6 1 . . . .:.f7 62 ':'xf7+
White is in no hurry, so he takes the i.xf7 63 f5 ':'e8 64 e6 i.g8 65 �a3 and
time to set a trap without endangering Black must remain totally passive while
anything. The following variations illus the king strolls in: he cannot even chall
trate most pointedly the perils to which enge rooks at g7 because of the mate at h5.
Black is exposed: And to sacrifice the bishop back at any
(a) 57 . . . :fS 5 8 ':'h8+ �g7 59 ':'hxh5 stage would lead to a hopeless rook
and White wins immediately, as Black ending.
cannot defend himself against the threat of [The rook ending after 64 e6 i.xe6 65
f4-fS, whether he moves his rook away or ':'xh5+ �g7 66 ':'g5+ �h7 67 fxe6 ':'xe6
exchanges; 68 ':'g3 �h6 does not look altogether clear.
(b) 57 . . .i.f5 58 ':'h8+ i.h7 (or 58 . . . :h7 It seems more consistent for White to
59 ':'f8 i.e4 60 fS and wins) 59 fS ':'ee7 60 maintain the bind and continue infiltrating
:g6 mate! with his king: 64 �a3 , and if 64 . . . :g8 65
57 �h7 ! llxg8 �xg8 66 �b4 i.e8 67 �c5 �f7 68
Black is alive to the danger, so White �d6 a3 69 f6 with a zugzwang position
must revert to his main plan, the invasion similar to that in the diagram below.]
by the king. The critical variations occur if Black
58 �bl ':'15 plays here 63 ....:.e7 64 �a3 i.e8 65 �b4
59 :gl ':'g7, whereupon White's king gets a real
Averbakh 's Selected Games 1 83
If 69 . . . �g6 70 l:th6 �f7 7 1 l:txe6 �xe6 finish off a fine game like this one with the
72 �d6 �f7 73 l:tg7, with e5-e6 to follow. greatest possible precision, so White looks
for and finds an alternative mating net.
74 :b6!
Threatening 75 l:ta8+ �d7 76 l:tb7
mate, or 75 . . . �c7 76 l:tab8 with inevitable
mate, so Black must move the rook on e6
to make room .
74 :g6
75 fS!
The pawn has the last word after all; it
covers e6, thereby threatening the same
mate as given above, while capturing it
allows a different finish.
[75 e6, forcing the rook to return to e6
The move played releases the passed and allowing the above mate, also wins.]
pawn, which is certain to have equally 75 :xfS
serious consequences before long. 76 :a8+ �c7
70 l:tgg7 �b8 Or 76 . . . �d7 77 l:tb7+ �e6 78 :Xe8
71 :b7+ �a8 mate.
72 :a7+ �b8 77 l:tab8
73 :hb7+ �c8
4 g3 g6 10 h4!
Despite the fact that Black has
preserved his bishop from exchange, White
begins an attack on the opponent' s king.
10 lDc6
11 lDa
Here the capture on c6 would not be
logical : White has already revealed his
plans. Now he forestalls the black knight
manoeuvre c6-e5-g4.
11 e5
11 �xc8 1Wxc8
2 1 ...:xc8 22 as �S 23 J.h3 is
unpleasant for Black.
22 as �dS
23 a6 bS
24 1Wcs 1Wd7
2S :c1 l:ae8
26 �e4
14 1We7
15 .i.gS 1WfT
16 .i.e3 :e7
17 �fgS 1We8
18 .i.xd4 exd4
19 a4!
White's queen needs to reach the a2-g8
diagonal, but the direct 1 9 1Wc2 is parried
by 1 9 . . . 1WbS . Therefore he begins an 26 :b8
attack on the queenside. 20 1Wc2 is now a A trap. If White plays 27 1Wxc6 1Wxc6
threat. 28 :xc6, then by 28 . . �b4 29 l:d6 J.eS
.
20 �d6!
White is not afraid of ghosts! In reply
to 20 . . . :xe2+ 2 1 �fl 1We7 there would
have followed 22 :xh7 J.g7 (22 . . . 1Wxd6
23 :xh8+ ! ) 23 �de4, winning the
exchange.
20 1Wd8
Averba/ch 's Selected Games I S7
Year Event + - =
Place
Year Event + - =
Place
Year Event + - =
Place
Fischer 1 54 Ragozin 1 1 1
Fuchs 1 44 Ravinsky 5 1
Roizman 1 64
Gligoric 86
Goldberg 49 Simagin 1 3
Golovko 54 Smyslov 72
Golz 1 42 Spas sky 1 3 7
Stahlberg 8 1
Hug 1 8 1 Suetin 57, 1 1 3
Szabo 45, 99
I1ivitsky 64
Taimanov 25, 1 05, 1 3 6, 1 68
Keres 97 Tolush 95
Khasin 1 76
Kholmov 75 Uhlmann 1 46
Korchnoi 1 1 9, 1 5 8 Unzicker 79
Kotov 3 6
Vasyukov 1 49
Larsen 1 5 6 Veresov 3 1
Lilienthal 34
Zak 22
Mikenas 20
Moiseev 52
Index of Openings
Benoni Defence 1 3
English Opening 1 1 6, 1 84
French Defence 1 7, 1 27
King ' s Indian Defence 34, 69, 1 22, 1 25 , 1 3 7, 1 44, 1 54, 1 66, 1 68
Modem Benoni 95
Modem Defence 1 5 1
Ragozin Defence 64