(1) Kramnik,V (2743) - Topalov,V (2813) [D12]
WCh Elista RUS (12), 12.10.2006



1.d4 d5
With the exception of the first game of the match, Topalov faced serious opening problems in the Catalan. There is little wonder that for the decisive game he chose the rock-solid Slav.

2.c4 c6 3.Nf3 Nf6 4.e3 Bf5 5.Nc3 e6 6.Nh4 Bg6 7.Nxg6 hxg6
Today I connected with some delay to the playchess.com server. When I saw the current position my first thought was that I had clicked on an earlier game in which Topalov had White. Indeed, this is the first time in the match when the players repeat a previously tested variation with reversed colours.

8.g3
Kramnik had occasionally played this variation with White, but only in rapid events. Unlike Topalov, he does not seem inclined to make experiments, preferring to follow the well known paths.

8...Nbd7 9.Bd2
[Kramnik had also played 9.Bg2 which leads to a position typical for the Catalan gambits after 9...dxc4 when Black can defend his extra-pawn with a further ...Nb6.]

9...Bb4N
Topalov is consequent with his handling of this position with White. We can understand now that one of the purposes of his 8.a3 had been to prevent ...Bb4, which he apparently considers the most efficient way to counter White's strategy, based on placing several pawns on dark squares. Previously, Black developed the bishop to e7 or d6.

10.Qb3 Bxc3 11.Bxc3 Ne4 12.Bg2 Nxc3 13.Qxc3 f5
By switching to the Stonewall structure, Black not only restricts the enemy bishop even more, but also creates himself chances for a king side attack.

14.0-0 Qe7
By delaying the castle, Black keeps his opponent under some psychological pressure regarding the possibility of a brutal attack along the h-file.

15.cxd5
We can feel that Kramnik's treatment of the position differs from that of his opponent in radical way. In the previous game Topalov blocked the queen side by means of c5, while two games earlier his whole opening play prepared this move, eventually causing Black to give up the centre with ...dxc4. It is remarkable that Kramnik refrained from blocking the queen side in the last 10th game, even though that is the almost automatic reaction to Black's ...b5 in the Catalan. There, too, he exchanged on d5, in order to retain some initiative in an half-open position.

15...exd5 16.b4
The minority's attack remains a simple and dangerous plan in spite of simplifications. Black cannot avoid the weakening of his queen side.

16...Nf6 17.Rfc1 Ne4 18.Qb2 0-0 19.b5 Rac8 20.bxc6 bxc6
White can count on a small but stable advantage because of his more compact structure.

21.Qe2
White's general plan consists of pressure along the c- and b-files, but he should not neglect prophylactic measures on the other wing. By placing the queen on e2 he inhibit to a certain extent the attack based on ...g5 and ...f4 (when the queen would place the knight under unpleasant pressure and keep the weakened g4-square under control). Besides, the last move is useful for active purposes, too, by enabling the transfer of the queen to a6 in the most rapid way.

21...g5 22.Rab1
[The continuation of the same policy. Before choosing the concrete plan, Kramnik improves the position of his pieces. After taking the b-file under control, the threat Qa6 becomes unpleasant. The more straightforward 22.Rc2 would have allowed Black continue his king side play with 22...Rf6 ]

22...Qd7
A multi-purpose move. Black avoids the pin along the e-file (enabling ...f4 in certain cases) and over-defends the c8-rook and the c6-pawn in order to enable ...Rf6 without fearing Qa6. However, the removal of the black queen from the e7-square is a small success for White, who will not have to care about an eventual sacrifice on g3 in case of a further f3.

23.Rc2
[White overdefends the second rank (the f2-square in first line) and prepares doubling rooks along the c- or b-file. Abstractly speaking, 23.Rb2 looks less adequate because it would leave the rook over-charged by the control of the second rank and the b-file, although from practical point of view it would have probably made little difference.]

23...Rf6 24.Rbc1
[In case of 24.Qa6 f4 25.Rb7 Qf5 Black attack could become dangerous.]

24...g4
[Black decides to eliminate the possibility of White's f3. 24...Rh6 was possible, too, when after 25.Qa6 (threatening to win a pawn with f3 already) he would have had a choice between transposing to the game with 25...g4 or play 25...c5 ]

25.Rb2?!
[For the same aforementioned reasons, 25.Rb1 , might have been more accurate. The difference will become clear at a later stage of the game.]

25...Rh6
A logical but very committal plan. If the attack along the h-file will fail, the rook will be left out of play, leaving White free hands on the opposite wing.

26.Qa6 Rc7
[Now, 26...c5?! would be strongly met by 27.Rb7! when the weakness of Black's central pawns would make itself felt.]

27.Rb8+ Kh7 28.Qa3
We can notice now another effect of the move Qe7-d7. This important diagonal became available to the white queen. Suddenly, the threat Qf8 puts the black king in some danger.

28...Rb7 29.Qf8?!
[After this move the game will soon enter drawish paths. White's last chance for an advantage consisted of 29.Ra8 , when exchanging queens on e7 or d6 would not absolve Black from all his strategic problems.]

29...Rxb8 30.Qxb8 Qf7 31.Qc8 Qh5 32.Kf1 Nd2+
[With the rook on c2, this check would not have been possible. Besides, against 32...Qg6 (which would have been a worthy alternative in the game), White could have continued his invasion along the b-file with 33.Rb2. In the game, 33.Rb1?? is impossible because of the fork on d2.]

33.Ke1 Nc4
Covering the weakness of the c6-pawn.

34.Bf1
Well, at least temporarily...

34...Rf6 35.Bxc4 dxc4 36.Rxc4
White has managed to fragment Black's structure even more, but now the light squares around his king are weak, allowing Black display strong counterplay.

36...Qxh2 37.Ke2 Qh1 38.Rc5 Qb1 39.Qa6 Qb2+ 40.Kf1 Qb1+ 41.Ke2 Qb2+ 42.Kf1 Rh6 43.Qd3
After the return of the queen to the centre, the white king is out of danger.

43...g6
[43...Qa1+ 44.Ke2 Qxa2+ 45.Kf1 g6 would win a pawn, but after 46.Qc4 he would have to give perpetual anyway. The rook ending would be microscopically worse in view of White's better structure and the inevitable loss of the a7-pawn.]

44.Qb3 Rh1+ 45.Kg2 Rh2+ 46.Kxh2 Qxf2+ 47.Kh1 Qf1+ 1/2-1/2