(The annotations to this game, by V. L. Korchnoi, are from '64' (№ 43, 1971).

The present game is

also annotated by Yu. L. Averbakh - who was one of Petrosian's seconds in Buenos Aires - in
Shakhmaty v SSSR (№ 1, 1972). His commentary has been incorporated into the notes at key
moments. In both cases, the translation from the original Russian is by Douglas Griffin.)

Petrosian – Fischer
6 match-game, Candidates’ Final, Buenos Aires, 17th October 1971
th

1.Nf3
Already on the basis of the first move made by the ex-World Champion, I drew a conclusion:
Petrosian, as in the 4th game, was not in a fighting mood. This decision, in my opinion, was
absolutely incorrect, taking into consideration the deep depression that had was present in Fischer's
play after the loss in the 2nd game. I am certain that neither Botvinnik nor Spassky would have let
slip the chance to develop their success in the struggle against a demoralised opponent!

1...c5 2.b3 d5 3.Bb2?

Before the start of the match, in an article published in '64', I had written about Petrosian's
shortcomings and, in particular, about his weak, approximate knowledge of opening theory. The
editor of '64' (Translator’s note: This was Petrosian himself!) took a condescending attitude to my
criticism. But, we can agree, it is rarely encountered in grandmaster practice that after the 3rd
move White obtains the worse game!

Grandmaster A. Nimzowitsch, who often employed this opening, continued here 3.e3, after
which Black cannot create a pawn centre ( 3...f6? 4.d4! ).

3...f6!
XIIIIIIIIY
8rsnlwqkvlntr0
7zpp+-zp-zpp0
6-+-+-zp-+0
5+-zpp+-+-0
4-+-+-+-+0
3+P+-+N+-0
2PvLPzPPzPPzP0
1tRN+QmKL+R0
xabcdefghy
Now White cannot prevent the creation by the opponent of a powerful pawn centre - Black has a
clear advantage.
4.c4 d4 5.d3 e5 6.e3 Ne7 7.Be2 Nec6 8.Nbd2 Be7 9.0–0 0–0 10.e4
XIIIIIIIIY
8rsnlwq-trk+0
7zpp+-vl-zpp0
6-+n+-zp-+0
5+-zp-zp-+-0
4-+PzpP+-+0
3+P+P+N+-0
2PvL-sNLzPPzP0
1tR-+Q+RmK-0
xabcdefghy
The manner in which the opening has been played is very reminiscent of the play in the opening
of the 9th game of the Petrosian-Korchnoi match: the same manoeuvre of the king's knight,
...g8–e7–c6, employed by me then, the same painful doubts of Petrosian in the choice of the
defensive stance - first e2–e3, but then e3–e4.

Let us try to figure out Petrosian's idea. The position after 10.exd4 cxd4 he considered to be
unsatisfactory for White, although in this case he has many possibilities of counter-play. The
closed set-up of the Benoni (after 10.e4) holds fewer active possibilities for White, but on the other
hand, in Petrosian's opinion, has a greater margin of safety...

10...a6
Instead of 10...a6 serious attention is merited by the move 10...Na6 with the idea of transferring
it along the route ...c7–e6, and only after this preparing a pawn offensive on the queen's flank.

11.Ne1 b5 12.Bg4 Bxg4 13.Qxg4 Qc8 14.Qe2

Naturally, also after the exchange of queens White's position was without prospects - Black,
possessing an advantage in space, has a firm hold on the initiative. Petrosian retains the queens,
against that chance that he may possibly manage to begin play on the king's flank. Unfortunately,
he later fails to undertake an attempt to become active in the direction of the black king as well.

14...Nd7 15.Nc2?
Why not 15.g3 followed by 16.Ng2? The advance of the pawn to f4 would have diverted
Black's attention from the queen's wing, and would have forced him to conduct a struggle over the
whole board. Instead now, when Petrosian has gone completely over to defence, Black has a
completely free hand.

15...Rb8 16.Rfc1 Qe8 17.Ba3

By attacking the c5–pawn, White holds up the development of Black's initiative - neither
17...Nb6 with the idea of exchanging at c4, nor 17...Bd8 with the idea of posting the bishop at a5
(as occurred in the 9th game of our match) will do, for the time being.

17...Bd6 18.Ne1 g6
White's situation is difficult. He is deprived of any possibility of displaying activity, while Black
unhurriedly strengthens his position, preparing to begin an offensive over the whole board. The
following move by Petrosian was criticised by many commentators; probably they considered that
Petrosian ought to have patiently awaited his fate.

19.cxb5
White's idea is to turn the queen's flank into the arena of the main fighting action and to thereby
disperse the clouds that are gathering everywhere. Naturally, the engagement with the
excellently-mobilised black pieces on the queen's flank appears far from safe, but Petrosian hopes
that by exploiting Fischer's uncertain play he will manage to escape the worst.

19...axb5 20.Bb2 Nb6 21.Nef3 Ra8 22.a3

This pawn-move could have been readily forced - it is sufficient to transfer the rook along the
route ...f8–f7–a7. Not wishing to direct the opponent towards the correct idea, Petrosian
immediately makes the move of the a-pawn.

22...Na5
Here too deserving attention is 22...Rf7, transferring the rook to the queen's flank - to b7 or c7.

23.Qd1 Qf7 24.a4 bxa4 25.bxa4

XIIIIIIIIY
8r+-+-trk+0
7+-+-+q+p0
6-sn-vl-zpp+0
5sn-zp-zp-+-0
4P+-zpP+-+0
3+-+P+N+-0
2-vL-sN-zPPzP0
1tR-tRQ+-mK-0
xabcdefghy
25...c4?!
At a moment when Black's advantage on the queen's flank is already determined - the move
...c5–c4 cannot be prevented; the white a-pawn turns out to be weak and must soon be lost - Fischer
hurries and lets slip the greater part of the advantage.
 Stronger was 25...Rfc8. However, after 26.Rab1! Rc6 (27.Bxd4 was threatened) 27.Ba3
Petrosian could have defended tenaciously. After 25...Rfc8, worse is 26.Ba3 on account of
26...Qd7 27.Nb3 Nxa4 winning a pawn.

26.dxc4 Nbxc4 27.Nxc4 Nxc4 28.Qe2

XIIIIIIIIY
8r+-+-trk+0
7+-+-+q+p0
6-+-vl-zpp+0
5+-+-zp-+-0
4P+nzpP+-+0
3+-+-+N+-0
2-vL-+QzPPzP0
1tR-tR-+-mK-0
xabcdefghy
28...Nxb2
Another controversial moment. Exchanging the bishop, Fischer goes into an endgame that can
hardly be considered to be winning for Black. In the case of 28...Rfc8 he probably feared the move
29.a5!, and White's pawn becomes dangerous. Instead of the exchange at b2 it was possible to play
28...Na5. After 29.Ba3 White holds on, but his position is markedly worse.

(Translator’s note: - Averbakh calls 28...Nxb2 “A characteristic decision by Fischer”, stating that
“...his thinking is exceptionally concrete. Based on general considerations, it seems that the strong
knight should not be exchanged for White's bad bishop. But Fischer sees that on 28...Rfc8 there
follows 29.a5. Therefore he exchanges the bishop, although this also gives White definite
counter-play.” Averbakh agrees that Korchnoi's suggestion 28...Na5 nevertheless left White with
far fewer chances. He analyses the following variations: 29.Ba3 Bxa3 30.Rxa3 Rfc8 31.Qd1 Qb7!
32.Nd2 Rxc1 33.Qxc1 Rc8 34.Qa1 Rc2; or 31.Raa1 Nb3 32.Rxc8+ Rxc8 33.Rb1 Qc4! 34.Qxc4+
Rxc4, with advantage.
A parallel can be drawn between Fischer's 28th move in the present game, and his 22.Nxd7+ in
the following one.)

29.Qxb2 Rfb8 30.Qa2 Bb4

(Translator’s note: Averbakh calls this decision too an interesting one, awarding the move an
exclamation mark. He points out that after the natural continuation 30...Qxa2 31.Rxa2 Kf7 32.g4
Ke6 33.Nd2 White is able to post his pieces more favourably than in the game.)

31.Qxf7+ Kxf7 32.Rc7+ Ke6 33.g4! Bc3 34.Ra2

(Translator’s note: Here Averbakh suggests instead 34.Rd1!, with the point that Black cannot take
at a4 on account of 35.Rc6+ Ke7 (35...Kd7?? 36.Rxc3) 36.Rc7+ Kf8 37.g5, while if 34...Rc8, then
possible is 35.Rxh7 Rxa4 36.Rg7 g5 37.h4.)

34...Rc8
XIIIIIIIIY
8r+r+-+-+0
7+-tR-+-+p0
6-+-+kzpp+0
5+-+-zp-+-0
4P+-zpP+P+0
3+-vl-+N+-0
2R+-+-zP-zP0
1+-+-+-mK-0
xabcdefghy
35.Rxc8
35.Rxh7 fails to give a clear draw. Black has two paths to the win. For example, 35...d3
36.Rb7 Ba5! (weaker is 36...Rc4 37.Rb3, and White holds on) 37.Rb5! (defending against the
threats of 37...Rc2 and 37...Rc4) 37...Rc1+ 38.Kg2 Bc3! 39.Rb3 Rc8! and no satisfactory defence
against the threat of 40...Rc2 is apparent. Also possible is the more prosaic path 35...Rc4 36.Rb7
Rcxa4 37.Rxa4 Rxa4 38.Rb6+ Kd7! and White is forced to return the rook to b1 and to seek
salvation in a difficult endgame, since no good is 39.g5 d3 40.gxf6 Ra1+ 41.Kg2 d2 42.f7 Ke7
43.Rb7+ Kf8 44.Ng5 Rg1+, and Black wins.

(Translator’s note: Averbakh examines the variations given by Korchnoi, and clarifies them as
follows: “In the first variation, instead of 39.Rb3 much stronger is (35...d3 36.Rb7 Ba5 37.Rb5
Rc1+ 38.Kg2 Bc3) 39.Rd5 Rc2 40.Ra3 d2 41.Rb3, and it is not apparent how White can lose.
And also in the second variation White can play more strongly. Thus, instead of 39.g5 possible is
(35.Rxh7 Rc4 36.Rb7 Rcxa4 37.Rxa4 Rxa4 38.Rb6+ Kd7 ) 39.Rb5! and, posting the rook at d5,
White saves the game.”
He concludes his comment as follows: “Thus, the impression is created that with the move
35.Rxh7 White could save the game.”)

35...Rxc8 36.a5 Ra8 37.a6 Ra7 38.Kf1 g5!

An important move. Otherwise, if the king leaves the king's flank, there follows g4–g5.

39.Ke2 Kd6 40.Kd3 Kc5 41.Ng1

White does not have time to play 41.h4. Black could reply 41...gxh4 42.Nxh4 Kb4 43.Nf3
Kb3, winning the a-pawn.
41...Kb5

XIIIIIIIIY
8-+-+-+-+0
7tr-+-+-+p0
6P+-+-zp-+0
5+k+-zp-zp-0
4-+-zpP+P+0
3+-vlK+-+-0
2R+-+-zP-zP0
1+-+-+-sN-0
xabcdefghy
In this position the game was adjourned.

42.Ne2
This sealed move is fairly obvious.The resultant endgame is very difficult to assess, especially with
such limited time. It may only be said with certainty that White's task is very difficult.

(Translator’s note: Averbakh mentions an interesting possibility that was suggested by Moscow
chessplayers - 42.f4 gxf4 43.Nf3, with the idea of g4–g5. However, since a different move had
been sealed, not attention was paid to this at the time.
He states that Petrosian was not in a happy mood at the end of the first session - the a6–pawn is
lost, while a forced path to a draw was not apparent. Their analysis began at 11pm.)

42...Ba5
Naturally, after the exchange of rooks Black has no winning chances at all.

43.Rb2+
At White's disposal there was another defensive plan, beginning with the move 43.Ng3, and
followed by Ng3–f5 and the transfer of the knight either to d5 or to c4. Possible are the variations:
43.Ng3 Rxa6 44.Nf5 Kb4 45.Ne7; or 44...Rc6 45.Rc2 Rc3+ 46.Rxc3 dxc3 47.Ne7, and White
achieves a draw. His task is more complicated after 43...Kxa6. For example, 44.Nf5 Rc7 (but not
44...Rb7 45.Nd6 Rb3+ 46.Kc4) 45.Rc2 Rb7 46.Rc6+ Ka7 47.Rc5 Be1 48.Ke2 Bc3, and Black can
count on success.

Another plan of defence, 43.Nc1, is suspect. After 43...Rxa6 44.Nb3 Kb4 45.Rb2 possible is
45...Bb6 with the threat of 46...Ra3, as well as 45...Ka3 46.Nc5 Rc6 47.Rb3+ Ka2 48.Rb5 Bb4
49.Rxb4 Rxc5 50.Ke2 Rc2+ with a large advantage.
 (Translator’s note: Nonetheless, Averbakh indicates that it was on this latter variation that the
adjournment analysis initially centred: “Fairly quickly, we found the continuation 42...Ba5 43.Nc1
Rxa6 44.Nb3 Kb4 45.Na1! Rc6 46.Nc2+ Kb5 (46...Kb3 47.Nxd4+!) 47.Na3+ Kc5 48.Nc4 Bb4
49.Ra8, and the active piece position gives White every justification to count on a draw. We tested
this continuation, as they say, inside and out, and each time found salvation for White. Since the
hour of the clock was all the time getting further past midnight, the decision was taken to interrupt
the analysis, in order to proceed to it the following morning with fresh force. But Petrosian could
not rest, and later the same night into his head there came the following variation: 44...Ra8! (instead
of 44...Kb4) 45.Ra1 Ra7 (45...Kb4 46.Rb1) 46.Ra2 Kb4! 47.Na1 Rc7! 48.Nc2+ (48.Rc2 Rxc2!
49.Kxc2 Kc4 etc.) 48...Kb3! 49.Ra3+ Kb2, and White has difficulties. The morning analysis began
with the testing of this variation that had been found by Petrosian. It became clear than instead of
47.Na1, it is necessary to play 47.Nc1, and if 47...Rc7, then 48.Rc2. However, Black can continue
47...h5, beginning active play on the other flank. In short, there once again proved to be no clear
draw.
Then there arose the idea of playing simply 43.Rb2+ Kxa6 43.Rb1, trying to create a a sort of
fortress. Admittedly, here too a clear draw was not found, although we spent in total not less than
ten hours on analysis...”.)

43...Kxa6 44.Rb1 Rc7 45.Rb2

Possibly, deserving attention is 45.f3, hindering the regrouping of the black pieces.

45...Be1
XIIIIIIIIY
8-+-+-+-+0
7+-tr-+-+p0
6k+-+-zp-+0
5+-+-zp-zp-0
4-+-zpP+P+0
3+-+K+-+-0
2-tR-+NzP-zP0
1+-+-vl-+-0
xabcdefghy
46.f3
A very important moment. Much stronger looks 46.Rb1. If now 46...Bxf2, then 47.Rf1 Be3
48.Rxf6+ Kb5 49.Re6 Ra7 50.Rxe5+ Kb4 51.Nxd4 Ra3+ 52.Ke2 Bxd4 53.Rxg5. This position is
clearly drawn. To mate the white king is hardly possible, while the h-pawn cannot help Black to
win. Thus, after 46.Rb1 Black should return the bishop to a5 and plan a further piece
re-arrangement. It seems that such a restructuring was possible. Avoiding the exchange of rooks,
he can transfer the king to c6, forcing White to occupy the a-file, and then invade with the king
along the b-file. Naturally, this plan is complicated, and it would be difficult for Black to
demonstrate whether or not his position is winning. If his position is won, then the exchange at
the 28th move (28...Nxb2) should be recognised all the more.

(Translator’s note: Averbakh states that in analysis they had reached the conclusion that White
should play 46.Rb1, and that f2–f3 should be played only in the analogous situation with the black
rook at c6. “Evidently, the analysis of this exceptionally complicated position had exhausted
Petrosian, and as a result he unexpectedly played 46.f3 with the rook at c7.”)

46...Ka5 47.Rc2 Rb7 48.Ra2+ Kb5 49.Rb2+ Bb4 50.Ra2 Rc7 51.Ra1 Rc8
XIIIIIIIIY
8-+r+-+-+0
7+-+-+-+p0
6-+-+-zp-+0
5+k+-zp-zp-0
4-vl-zpP+P+0
3+-+K+P+-0
2-+-+N+-zP0
1tR-+-+-+-0
xabcdefghy
52.Ra7?
Now the white rook is cut off from its camp. More tenacious was 52.Ra2. In this case Black
could continue the assault either with 52...Be7 (or 52...Bf8; but not 52...Bc3 on account of 53.Rc2
), or by means of 52...h5.

(Translator’s note: - Averbakh considers that it is not apparent how Black can win after 52...h5
53.gxh5 Rh8 54.Ng3 Be1 55.Rb2+ Kc6 56.Rc2+ Kd7 57.Ra2 Bxg3 58.Ra7+. He goes on to state
that “Clearly stronger is 52...Be1, but on this possible is 53.Rb2+ Ka4 54.Rc2!, and Black is still
faced with a lot of work in order to demonstrate the possibility of winning.")

52...Ba5 53.Rd7
On 53.Rxh7 there would have followed the same reply.

53...Bb6 54.Rd5+ Bc5 55.Nc1 Ka4 56.Rd7 Bb4 57.Ne2 Kb3 58.Rb7 Ra8 59.Rxh7 Ra1
60.Nxd4+ exd4 61.Kxd4 Rd1+ 62.Ke3 Bc5+ 63.Ke2 Rh1 64.h4 Kc4 65.h5 Rh2+ 66.Ke1 Kd3
White resigned. (Translator’s note: Averbakh concludes: "The 6th game was destined to be a
turning point in the match. The impression is created that this long and difficult game deprived
Petrosian of too much strength. Later, he was short of energy, and he played the following
encounters significantly beneath his possibilities.")