[Event "Monaco"]
[Site "Monaco"]
[Date "1995.??.??"]
[EventDate "?"]
[Round "8"]
[Result "1-0"]
[White "Vladimir Kramnik"]
[Black "Jeroen Piket"]
[ECO "B22"]
[WhiteElo "?"]
[BlackElo "?"]
[PlyCount "51"]

1. e4 c5 2. c3 d5 3. exd5 Qxd5 4. d4 Nc6 5. Nf3 cxd4 6. cxd4
e5 7. Nc3 Bb4 8. Bd2 Bxc3 9. Bxc3 e4 10. Ne5 Nxe5 11. dxe5 Ne7
12. Qa4+ Bd7 13. Qa3 Bg4 14. h3 Bh5 15. g4 e3 16. Rh2 Bg6
17. fxe3 Qf3 18. Qa4+ Nc6 19. Qf4 Qd5 20. Rd2 Qc5 21. h4 h5
22. O-O-O Rc8 23. Rd5 Qb6 24. e6 O-O 25. Rb5 Qa6 26. Rxh5 1-0

[Event "Karlsbad"]
[Site "21"]
[Date "1911.??.??"]
[EventDate "?"]
[Round "?"]
[Result "1-0"]
[White "Aron Nimzowitsch"]
[Black "Oscar Chajes"]
[ECO "B22"]
[WhiteElo "?"]
[BlackElo "?"]
[PlyCount "63"]

1.e4 c5 2.c3 Nc6 3.d4 cxd4 4.cxd4 d5 5.exd5 Qxd5 6.Nf3 e5
7.Nc3 Bb4 8.Bd2 Bxc3 9.Bxc3 e4 10.Ne5 Nxe5 11.dxe5 Ne7 12.Qa4+
Bd7 13.Qa3 e3 14.f3 Qe6 15.Rd1 Nf5 16.Bd3 Qg6 17.O-O e2
18.Bxe2 Ne3 19.Kf2 Nxd1+ 20.Rxd1 Qb6+ 21.Bd4 Qe6 22.b3 Bc6
23.Bc5 Qf5 24.Qc1 Rd8 25.Bd6 h5 26.Bd3 Qe6 27.Qg5 Rd7 28.Bf5
Rxd6 29.exd6 f6 30.Qg6+ Qf7 31.d7 Bxd7 32.Qxf7+ 1-0

[Event "Classical World Chess Championship"]
[Site "Brissago SUI"]
[Date "2004.01.06"]
[EventDate "?"]
[Round "5"]
[Result "1-0"]
[White "Peter Leko"]
[Black "Vladimir Kramnik"]
[ECO "D37"]
[WhiteElo "?"]
[BlackElo "?"]
[PlyCount "138"]

1. d4 {Grandmaster Peter Leko equalised the score in the World
Championship at Brissago, Switzerland by holding game four and
exploiting Kramnik's endgame errors in game five. Leko won a
pawn in the opening. However, this was known theory and most
experts assumed that Kramnik had worked out in advance how to
draw. Nevertheless, after many hours of play Kramnik committed
an error in a seemingly drawn position that annihilated his
prospects of salvation. (Notes by Raymond Keene.)} Nf6 {1.d4
is a definite surprise. Kramnik chooses a line that he likes
to play against as White, but one that he also knows well. For
example, Kramnik had the position after 11.Bd3 as White
against Nigel Short at the Corus tournament, Wijk aan Zee,
2000. This wasn't Leko's first outing with 1 d4, but 2 c4 was
new for him . "It felt exciting," he said.} 2. c4 e6 3. Nf3 d5
4. Nc3 Be7 5. Bf4 O-O 6. e3 c5 7. dxc5 Bxc5 {White's next move
heads for a small but clear plus. The more complicated
alternative is 8 Qc2.} 8. cxd5 Nxd5 9. Nxd5 exd5 10. a3 {White
has inflicted an isolated dpawn on Black's camp, but black has
free piece activity in compensation.} Nc6 11. Bd3 Bb6 {Kramnik
also chose this move playing Black against Van Wely in the
Melody Amber rapidplay in 2001; Short preferred 11...Be7. I've
played this position myself, and in Keene-Rodriguez, Indonesia
1982, Black came up with the idea 12 h3 d4!?} 12. O-O Bg4
13. h3 Bh5 14. b4 Re8 15. Rc1 a6 {Black does not want to see
his knight driven from its central post and therefore prevents
white from advancing with b5. However, this move permits a
small fork trick combination that enables White to retain an
edge.} 16. Bxa6 {! Outwardly startling, but in fact the point
of White's play.} Rxa6 17. b5 Rxa3 18. bxc6 bxc6 19. Rxc6 Ra7
20. Rd6 Rd7 21. Qxd5 Rxd6 22. Qxd6 Qxd6 23. Bxd6 Bxf3
{Informator 84 gave Kramnik's choice "?!", but it seems likely
Kramnik knew of this Doubling White's pawns would appear to
secure the draw, but when Anand reached this position against
Karpov (Russia vs. Rest of the World 2002) he felt it would be
safer to maintain the bishop pair as compensation. White could
make no progress in that game, which was drawn in 114
moves. White's winning task is complicated by the fact that
all the pawns are restricted to one side of the board, and as
Fine briskly asserted in 1943: "If you are just one pawn
ahead, in 99 cases out of 100 the game is drawn if there are
pawns on only one side of the board." Leko of course knew
this, but as he said afterwards: "Objectively, this is a draw,
but Black needs to prove it there is no forced draw, and also
no draw familiar to everyone as there might be in a Rook
endgame. With the Bishops on the board it is already
unusual."} 24. gxf3 Bd8 25. Rb1 Bf6 {IM Jacob Aagaard,
commenting live on the official website during one of Leko's
long thinks, reminded us that "Kramnik won the World
Championship in 2000 from Kasparov by being better than the
old champion in positions without queens. He frustrated him by
drawing with Black in positions where White had chances, but
could not win. Later, Kramnik talked about positions in which
he knew was worse but also knew how to draw. I think he would
include the [current] position among them."} 26. Kg2 {Only
here is the first new move. 26 Rb5 g6 27 f4 Rd8 28 Bb4 h5 was
Gritsak - Kruppa, Alushta 2002, which was drawn.} g6 27. f4
Kg7 28. Rb7 Re6 29. Rd7 {White manoeuvres carefully, but
ultimately he must gain central terrain with e3-e4 and e4-e5,
committal though this might be.} Re8 30. Ra7 Re6 31. Bc5 Rc6
32. Ra5 Bc3 33. Rb5 Ra6 34. Rb3 Bf6 {That's a hat-trick of
Bf6s in this match for Kramnik...} 35. Rb8 h5 36. Rb5 Bc3
37. Rb3 Bf6 {With the return of the bishop to f6, White makes
his central move.} 38. e4 Ra5 39. Be3 Ra4 40. e5 {The only way
to make progress he has to place his bishop on d4 and his king
on e4, and submit to any tactics Black may be able to conjure
up. Kramnik said afterwards he was right to stay passive, but
lost concentration, and was "already thinking about next
game". After the time control, he found himself with serious
problems.} Be7 41. Rb7 Kf8 42. Rb8+ Kg7 {Kramnik cannot be
tempted to weaken his position, so Leko must try something
else.} 43. Kf3 Rc4 44. Ke2 Ra4 45. Kd3 Bh4 46. Bd4 {This
combined with the threat of e6+ is White's only chance.} Ra3+
47. Kc2 Ra2+ 48. Kd3 Ra3+ 49. Ke4 Ra4 50. Kd5 Ra5+ 51. Kc6 Ra4
52. Kc5 Be7+ 53. Kd5 Ra5+ 54. Ke4 Ra4 {White now gets his Rook
off a dark square, in case of Bd6, forking b8 and f4.} 55. Rc8
Bh4 {Leko has now placed his pieces as well as he can, and
makes his push. Kramnik is obliged to sacrifice the exchange.}
56. e6+ Bf6 57.e7 Rxd4+ 58. Ke3 Bxe7 59. Kxd4 Bh4 {Immediately
after the game, GM Amador Rodriguez commented on Kramnik's own
website www.kramnik.com, "The above position is well-known as
a theoretical draw, and therefore the result of the game was
still unclear." At the top of that web page was the chess
proverb, "A chess master is a good informed amateur": all over
the world there were GMs on the Internet explaining why it
could only be a draw, but Leko was in no hurry to
agree... After the game, he commented "In the Rb1 variation of
the Grnfeld, there is a long variation, which arrives in
exactly this endgame, and this was part of my repertoire as
black. I knew if the bishop stays on long diagonal, it's a
draw, but after the time control, Black is not in time to
place bishop on long diagonal." There are a couple of well
known theoretically drawn positions relevant here known as
'fortresses', because while White has a plus in material and
freedom to manoeuvre, there is no way to make progress. [notes
w/ diagrams omitted] What makes the real difference is that
Black has yet to set up the fortress, and before he can do so
Leko's check on move 61 pushes the Black king to a square that
the bishop would like to use.} 60. f3 f5 61. Rc7+ Kf6 62. Kd5
Bg3 {? Leko in the press conference afterwards suggested
62...Be1 as more awkward: with idea if 63 Rc6+ Kf7 64 Ke5 then
64...Ba5 stopping Rc7+ for the moment, when it's not easy to
make progress. Now, after either 65 Ra6 Bc3+ 66 Kd5 Bf6 or 65
Rc4 Bd8 and 66 Bf6, Black has his fortress. If White
temporises with 64 Rc8, for example, then 64...Bb6. Leko hoped
he might still win this position, but I can't see how.}
63. Rc6+ Kg7 64. Ke5 {Kramnik's next appears to make life
easier for White. After 64.Bh4, it is not at all clear White
can penetrate the black fortress, but GM Shipov gave 64 Bh4 65
Rc7+ Kg8 66 Ke6 Kf8 67 Rc8+ Kg7 68 Rb8! with zugzwang, eg 68
Bg3 69 Rb7+ Kh6 70 Kf7 Bxf4 71 Rb6 Kg5 72 Rxg6+ Kh4 73 Kf6,
etc.} h4 65. Rc7+ Kh6 66. Rc4 Kg7 67. Ke6 Bh2 68. Rc7+ Kh6
69. Kf7 {Black resigns Tom Furstenburg found a neat finish in
analysis, involving an under-promotion: 69 Kf7 Bxf4 70 Rc6 Kh5
71 Rxg6 Be3 72 Kf6 f4 73 Rg8 Bd4+ 74 Kf5 Kh6 75 Kxf4 Bf2 76
Kg4 Kh7 77 Rg5 Kh6 78 f4 Be1 79 f5 Kh7 80 f6 Kh6 81 f7 Bc3 82
f8=N Be5 83 Rg6#. After six-and-a-half hours of play, Peter
Leko has done something that Kasparov could not -- won a WCC
match game from Kramnik.} 1-0