By: Lubomir KavalekPeter Leko, the Hungarian challenger, leads the titleholder Vladimir Kramnik 5 1/2 to 4 1/2 at the 14-game Classical World Championship in Brissago, Switzerland, after yesterday's game, the 10th. Leko won the eighth game on Thursday; two weekend games were drawn.
In case you wondered, the word "classical" in the title of the match refers to a time control. "It includes games played with an average of three minutes or more per move," was the definition when the term "classical chess" was coined during a 1988 Grandmasters Association meeting. Garry Kasparov and Valery Salov even felt that those who played a tournament with a faster time limit should be morally condemned. Those were the times!
Had Kramnik remembered what kind of match he was playing, he might have averted his loss in the eighth game by spending more time on his opening moves in the complicated Marshall Attack in the Spanish. Instead, the world champion played too fast, finding himself in a lost position after 23 moves. Was Leko performing brilliantly or was Kramnik guided by a hidden hand of former FIDE world champion Ruslan Ponomariov, who analyzed the position incorrectly to white's advantage two years ago? Or was the culprit a computer analytical engine unplugged too soon? After the game, Kramnik was not sure whether he confused something or should have put the blame on analytical mistakes.
Kramnik-Leko, Game 8
1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.0-0 Be7 6.Re1 b5 7.Bb3 0-0 8.c3 d5 9.exd5 Nxd5 10.Nxe5 Nxe5 11.Rxe5 c6 12.d4 Bd6 13.Re1 Qh4 14.g3 Qh3 15.Re4 (This variation, where the rook is often sacrificed for positional gains, troubled black attackers for some time.) 15...g5 (Preventing 16.Rh4, but creating weak squares on the kingside. Now 16.Bxg5? is met by 16...Qf5!) 16.Qf1 (An old continuation from 1949. The new one 16.Qe2 was convincingly steered to a draw in the game Ponomariov-Anand, Linares 2002: 16...f5 17.Bxd5+ cxd5 18.Re6 f4!! 19.Rxd6 Bg4 20.Qf1 Qxf1+ 21.Kxf1 Rae8 22.Bd2 Bh3+ 23.Kg1 fxg3 24.hxg3 Re2 25.Be3 Rxe3 26.fxe3 Rf1+ 27.Kh2 g4 28.Rxd5 and black has a perpetual check with the rook.) 16...Qh5 (Sharper than 16...Qxf1+.) 17.Nd2 Bf5 18.f3 (Kramnik is not backing off.) 18...Nf6 (After 18...Bxe4 19.fxe4 Ne3 20.Qf3, white soon gets a second pawn for the exchange, since 20...Ng4 is met by 21.Nf1, threatening 22.Bd1.) 19.Re1 Rae8 20.Rxe8 Rxe8 21.a4 (Shortly after he became FIDE world champion, Ponomariov defeated Michael Adams's Marshall Attack in Linares in 2002. In the analysis to that game, Ponomariov reached the position that Kramnik and Leko had now on the board..) 21...Qg6! (After 21...g4 Ponomariov gives 22.axb5 gxf3 23.Nxf3 Bh3 24.Bxf7+! Kxf7 25.Ng5+ Kg7 26.Nxh3 and white wins.)
22.axb5?! (Perhaps the problem for Kramnik began here. Ponomariov was mainly interested in clearing the position with 22.Ne4!?, claiming that white is better either after 22...Bxe4?! 23.fxe4 Nxe4 24.axb5 axb5 25.Ra7; or after 22...Nxe4 23.fxe4 Bxe4 24.Bxg5 bxa4!? [On 24...Qxg5? 25.Qxf7+; or 24...Bd3? 25.Bxf7+! Qxf7 26.Qxd3 Qh5 27.Bh4 wins.] 25.Bc4, but Kramnik thought after the game that with 25...Bd5 black is fine.) 22...Bd3 23.Qf2? (Losing. After 23.Qd1!? Ponomariov gives 23...Be2 without saying anything about the position. Playing for a win with 24.Bc2? lands white in trouble after 24...Qh5! 25.Qe1 Bxf3 26.Qf1 g4!, threatening 27...Bxg3!. But after 24.Qc2, black can draw, repeating the moves: 24...Bd3!? [24...Qh5 is ill-advised because of 25.Qf5!] 25.Qd1 Be2 etc.) 23...Re2! (On 23...axb5? comes 24.Bd1.) 24.Qxe2 (Giving up the queen with 24.bxa6 also loses. The computer program Fritz8 gives 24...Rxf2 25.Kxf2 Qh6 26.Kg1 Qh3!, tying up the knight on d2 and setting up a mating attack, for example 27.a7 Bxg3! 28.a8Q+ Kg7 29.hxg3 Qxg3+ 30.Kh1 g4! 31.Qxc6 Qh3+ 32.Kg1 g3, threatening 33...Qh2 mate. Or 26.Kg2 g4 27.fxg4 Qe3! and black wins.) 24...Bxe2 25.bxa6 Qd3! (A decisive invasion.)
26.Kf2 (Kramnik first thought that at most black may have a perpetual check until he suddenly realized that after 26.a7 Qe3+ 27.Kg2 Bxf3+ 28.Nxf3 Qe2+ 29.Kg1 Ng4!, black has a mating attack, for example 30.a8Q+ Kg7 31.Qxc6 Qf2+ 32.Kh1 Qf1+ 33.Ng1 Nf2 mate. The computer program Fritz8 initially shows that white is winning big after 26.a7, but within 1 1/2 minutes finds the win for black. Did someone in Kramnik's camp unplug it too soon?) 26...Bxf3! (A clincher, paving a way for the knight.) 27.Nxf3 Ne4+ 28.Ke1 Nxc3! (The simplest.) 29.bxc3 (On 29.a7 Qe2 mates; and after 29.Ng1 Ne4, white doesn't have a good defense against 30...Bb4+.) 29...Qxc3+ 30.Kf2 Qxa1 31.a7 h6 32.h4 g4 White resigned.
European Club Cup
NAO Chess Club of Paris won the 20th edition of the European Club Cup on Saturday in Izmir, Turkey. Led by grandmasters Adams, Alexander Grischuk and Etienne Bacrot, the team scored 12 match points. They finished one point ahead of Bosna Sarajevo and two Russian teams, Ladya Kazan and Max Ven Ekaterinburg. Kasparov at the helm of Max Ven scored only three points in six games, losing to grandmaster Sergei Rublevsky.
Solution to today's study by A. Herburg (White: Kf2,P:g2,h4; Black: Kh5,P:f7): 1.Kg3 f5 (On 1...f6 2.Kh3 f5 3.Kg3 wins.) 2.Kf3!! Kxh4 3.Kf4 wins.