Homework
This article is from the archive of The New York Sun before the launch of its new website in 2022. The Sun has neither altered nor updated such articles but will seek to correct any errors, mis-categorizations or other problems introduced during transfer.

The world championship match in Switzerland between Vladimir Kramnik and Peter Leko finished in a 7-7 draw. In the eighth round, a game in which home preparation proved to be a doubleedge sword, Leko took the lead. It seems that Kramnik steered for a position that, remarkably enough, held winning opportunities for black too deep even for a computer to find. Nevertheless, Kramnik won the final game of the match and saved his title.
KRAMNIK VS. LEKO
(white) (black)
Ruy Lopez
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 g5
Theory considers this move, which prevents the rook on e from moving to h4, to be obligatory. In case of 15…Bd7 16.Bc2, white emerges superior. 16.Qf1 Qh5 Leko thought about this retreat for a full 50 minutes. Black obtained sufficient counterplay in several games in which queens were exchanged after 16…Qxf1+. But thinking so long about a theoretical move, Leko, fearing Kramnik’s home preparation, obviously smelled a rat. Kramnik’s homework did not give Leko a rat at all, but a tasty rabbit. 17.Nd2 Bf5 18.f3! Nf6 White’s 18th move is, finally, a novelty. Leko reasonably rejects the exchange sacrifice, because after 18…Bxe4 19.fxe4 Ne3 20.Qf3 Qxf3 21.Nxf3 Nc4 22.Bxc4 bxc4 23.Bxg5 Rab8 24.Nd2! white clearly stands better. 19.Re1 Now 19.a4 Bh3 20.Qf2 Nxe4 21.fxe4 Kg7! preparing the counterattack f7-f5 is not appealing for white. 19… Rae8 20.Rxe8 The immediate 20.a4? Bd3! 21.Qf2 Rxe1+ 22.Qxe1 Re8 is obviously bad for white. 20… Rxe8 21.a4 Qg6! 22.axb5 The attempt to regain the initiative via 22.Ne4 Nxe4 23.fxe4 Bxe4 24.axb5? (The only serious alternative is 24.Bxg5 bxa4 25.Bc4 Bd5 26.Bxd5 cxd5 27.Qf6 with an unclear endgame.) 24…Bd3 25.Bxf7+ Qxf7 26.Qxd3 Re1+ 27.Kg2 Qd5+ would lose. 22… Bd3! 23.Qf2? White should have opted for a draw by repetition here, playing 23.Qd1 Be2 24.Qc2 (not 24.Bc2 Qh5! 25.Qe1 Bxf3 26.Qf1 g4! 27.bxa6 Bxg3! with a mating attack) 24…Bd3 25.Qd1. 23… Re2
(See diagram)
24.Qxe2 The other version of the queen sacrifice: 24.bxa6 Rxf2 25.Kxf2 Qh6 26.Kg1 g4! 27.f4 Bxf4! 28.gxf4 (or 28.a7 Be3+ 29.Kh1 Qh3 30.a8Q+ Kg7 31.Qxc6 Be4+ 32.Nxe4 Qf1#) 28…g3! 29.hxg3 Ng4 30.Nf1 Be4 leads to the inevitable mate of white. 24… Bxe2 25.bxa6 The queen sacrifice was prepared by Kramnik at home. It looks like the a-pawn is unstoppable. Indeed, computers analysis “proves” that white is winning. Alas, this position happens to be too complicated for a computer. Perhaps if Kramnik had not relied on computer analysis, he would have been afraid to enter this position. 25… Qd3! Despite a time shortage, Leko found this move. Black now has a crushing attack against the white king. 26.Kf2 The refutation of the main line, prepared by Kramnik, is 26.a7 Qe3+ 27.Kg2 Bxf3+! 28.Nxf3 Qe2+ 29.Kg1 (29.Kh3 Qf1#) 29…Ng4! 30.a8Q+ Kg7 31.Qxc6 Qf2+ 32.Kh1 Qf1+ 33.Ng1 Nf2# Kramnik’s move also loses beautifully. 26… Bxf3! 27.Nxf3 Ne4+ 28.Ke1 Nxc3! 29.bxc3 Or if 29.Kf2 Ne4+ 30.Ke1 Bb4+ followed by mate in three. 29… Qxc3+ 30.Kf2 Qxa1 31.a7 h6! 32.h4 g4 0-1