Kramnik,V (2770) - Leko,P (2741) [B12]
Brissago Brissago (14), 2004
[Vladimir Kramnik]

1.e4 c6 In the decisive game of the match Peter chooses an opening in which he has a wealth of experience. In principle, this is a correct decision, but I was happy with his choice - at least it is possible to initiate some play... 2.d4 d5 3.e5! The only move (in the given match situation). 3...Bf5 4.h4 When I made this move during the game, it suddenly occurred to me how absurd it is from the viewpoint of the general logic of opening play. It is strange that during my preparations the move seemed to me to be quite normal... 4...h6 [ 4...h5 is more popular.] 5.g4 Bd7 [ 5...Bh7 is weaker in view of 6.e6! with a strong initiative.] 6.Nd2 [ A novelty of dubious quality, the main basis of which was my desire to deviate from theory as soon as possible. Here play usually develops as follows: 6.h5 e6 7.f4 c5 8.c3 Nc6 9.Nf3 Qb6 10.Kf2 0-0-0 11.Kg3 and so on several games have reached this position, but all this contradicts my chess convictions, and I am happy to leave it to others to play this way.] 6...c5 [ Or 6...e6 7.Nb3 c5 ] 7.dxc5 e6 [ 7...Qc7 was also quite decent, with an unclear game.] 8.Nb3 Bxc5 Otherwise it might not be possible to regain the pawn. 9.Nxc5 Qa5+ 10.c3 Qxc5 11.Nf3 Outwardly White's position looks attractive, but in fact it is not easy for him to complete his development, without losing his e5 pawn in the process. 11...Ne7 12.Bd3 Nbc6 [ 12...Qc7 13.Qe2 Nbc6 is also interesting, and if 14.Bf4 d4! with a reasonable game.] 13.Be3 Black was now threatening d5-d4. 13...Qa5 14.Qd2 [ Played after lengthy and serious thought. I cannot say that the prospect of the coming endgame greatly attracted me, but I could not see anything else. After 14.Qe2 Ng6! White does not make any progress: 15.b4 ( 15.Bd2 Qc7! ) 15...Qa3! an important resource ( if 15...Qc7 there follows 16.Bxg6 Nxe5 ( 16...fxg6 17.Bc5! ) 17.Bxf7+ Nxf7 18.Bd4 ) 16.0-0 ( 16.Bd2 Nf4! ) 16...Ngxe5 17.Nxe5 Nxe5 and there is no way for him to develop his initiative.] 14...Ng6 [ A perfectl y good and natural move, but in a different match situation Black could have considered 14...d4 which gives him fine play: 15.Nxd4 ( 15.Bxd4 Nxd4 16.cxd4 Qd5 17.Rh3 Bb5 ; 15.cxd4 Nb4 16.0-0 Bb5! ) 15...Nxe5 16.Be2 Qd5 ] 15.Bd4 [ Of course, 15.Bxg6 is out of the question.] 15...Nxd4 [ 15...Qc7 16.Qe3 ] 16.cxd4 Qxd2+ [ 16...Qb6 was also possible, when I was intending 17.0-0-0 followed by Kb1 with play on the kingside. The move in the game is logical from every point of view.] 17.Kxd2 And so, as often happens, a sharp and unusual opening has led to a quiet and standard for the French Defence endgame. At this moment I was not very happy with the situation, as I realised how difficult it would be to win this endgame against a high-class player, and especially against Leko, who is famed for his ability to defend faultlessly in such positions. True, in a decisive game it is far more difficult to do this. 17...Nf4 18.Rac1 [ 18.Bf1 (with the idea of Ke3) was weaker in view of 18...h5! and Black adopts a firm stance: 19.g5 ( or 19.Rg1 hxg4 20.Rxg4 Nh5 21.Rc1 Rc8 ) 19...Rc8 20.Ke3 Ng6 21.Bd3 Ne7 followed by g6. For the moment I did not want to fix the kingside pawns.] 18...h5 [ Many commentators judged this move to be a mistake and suggested 18...Nxd3 'with good drawing chances'. It seems to me that after the exchange on d3 there is no guarantee that Black will be able to draw, but there is a guarantee that he will 'suffer' for a very long time. The move in the game is much more critical: Black wants to gain a draw 'without suffering'.] 19.Rhg1! Maintaining the tension and retaining the 'dynamics' of the position. 19...Bc6 [ If 19...Nh3?! there is 20.Rg3! Nxf2 21.gxh5 and Black has serious problems. Another approximate variation is also of interest, demonstrating the dynamic resources in White's position:; 19...Rc8 20.Rxc8+ Bxc8 21.gxh5!? Nxh5 22.Ke3 g6 23.Rc1 Kd8 24.Ng5 Rf8 25.Nxf7+ Rxf7 26.Bxg6 ] 20.gxh5 Nxh5 21.b4! a6 22.a4! In accordance with classical principles, White opens a 'second front'. I think that this turn of events came as a surprise to Peter, who was contemplating a quiet and unhurried defence, and suddenly found that he had to calculate variations! In addition, the invasion of the white rook at c7 looks very menacing. 22...Kd8 [ This is, so to speak, the natural reaction of a tired organism, but objectively it may be the decisive mistake. 22...Bxa4 23.Rc7 Bb5 was extremely risky on account of either 24.Rxb7 or ( 24.Bb1!? but after) ; 22...Ke7 23.b5 axb5 24.axb5 Bd7 25.Rc7 b6 Black's position is possibly defensible; at least, there is no direct win, for example 26.Ng5 Rhc8 27.Rxc8 ( 27.Rb7 Rcb8 ( 27...Nf4!? ) 28.Rxb8 Rxb8 with a white edge) 27...Rxc8 28.Nxf7 Kxf7 29.Bg6+ Kf8 30.Bxh5 Rc4! 31.Ke3 Rc3+ 32.Kf4 Rd3 with good chances of a draw.] 23.Ng5 Be8 24.b5! A very important resource, after which Black's position becomes critical. 24...Nf4 [ After 24...axb5 25.Bxb5! it is hard for Black to do anything to oppose the invasion of his position by the white pieces. Here is an approximate variation: 25...Rf8 26.Bxe8 Kxe8 27.Rc7 Rxa4 28.Rb1 Rxd4+ 29.Ke3 Rc4 30.Rcxb7 Rc3+ 31.Kd2 Rc8 32.Ra7 with a mating attack.] 25.b6! Securing the c7 square for the rook. There is no longer any satisfactory defence. 25...Nxd3 [ After 25...f6 26.Nf3 Nxd3 27.Kxd3 Bh5 the simplest is 28.Rxg7! Bxf3 29.exf6 Be4+ 30.Ke2 Bf5 31.Rcc7 with complete domination for White and an easy win.] 26.Kxd3 Rc8 [ 26...Rh7 27.Rc7 ] 27.Rxc8+ Kxc8 28.Rc1+ Bc6 29.Nxf7 Rxh4 30.Nd6+ Kd8 [ 30...Kd7 31.Nxb7! ] 31.Rg1 [ The most direct way to win. 31.Ke3 Rh3+ 32.f3 was also good, but when I noticed the move 34.f4, this dispelled my doubts.] 31...Rh3+ [ It stands to reason that 31...Rh7 32.a5 Ke7 33.Rg6 would not have offered Black any chances.] 32.Ke2 Ra3 33.Rxg7 Rxa4 [ After 33...Ra2+ the following attractive variation was possible: 34.Ke3 Ra3+ 35.Kf4 Rxa4 36.Kg5 Rxd4 37.Kf6 Rf4+ 38.Kxe6 and the white pieces are very active.] 34.f4! The threat of f4-f5 is decisive. 34...Ra2+ [ 34...Rxd4 35.f5 exf5 36.e6 Re4+ 37.Nxe4 dxe4 38.Rc7 and in view of the threat of Rc6 Black loses his b-pawn. ( 38.Ke3 also wins) ] 35.Kf3 Ra3+ 36.Kg4 Rd3 [ 36...Ra1 was somewhat more tenacious, although the endgame after 37.Nxb7+ ( in the computer's opinion, White also wins by 37.Rc7 playing for mate: 37...Rg1+ 38.Kh5 a5 39.Nf7+ Ke8 40.Ng5 but this is neither here nor there...) 37...Bxb7 38.Rxb7 Rg1+ 39.Kh5 is hopeless, for example 39...a5 40.Ra7 Rb1 41.Kg6 Rxb6 42.Kf7 followed by 43.f5.] 37.f5 Rxd4+ [ 37...exf5+ 38.Kxf5 Bd7+ 39.e6 Rf3+ 40.Kg4 ] 38.Kg5 exf5 39.Kf6 I have to admit that I have always had a weakness for raids by my king into the opponent's position (for example, Kramnik-Kasparov, Moscow 1994, or Kramnik-Topalov, Monaco 2003). 39...Rg4 40.Rc7 Rh4 41.Nf7+ [ 41.Nf7+ In view of 41...Ke8 42.Rc8+ Kd7 43.Rd8# mate Black resigned.] 1-0