Game index
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Game 8 BGN World Championship
Kramnik - Kasparov [E32]
This was the first encounter that followed the Champions
scenario. Alas, he did not manage to finish off the game...
1.d4 Nf6 2.c4 e6!?
A new opening in the match!
3.Nc3 Bb4
The Nimzo-Indian Defense.
4.Qc2
This is the most popular line at the present time. White avoids
doubled pawns on c3. The drawback of this move is a lag in
development. Interestingly enough, both opponents are recognized
experts in this variation. Half of the theory of this system is
based on their games.
4...0-0 5.a3 Bxc3+ 6.Qxc3 b6 7.Bg5 Bb7 8.f3!
After 8.f3!
This way White restricts the b7-bishop and gets ready to occupy
the center with e2-e4. Thanks to this idea White has scored many
victories in the last 15 years. The f3-square was reserved for
the knight previously, but this leads to an equal game. In this
case Black comfortably develops with h6, d6, Nbd7, g5, Ne4, etc.
8...h6 9.Bh4 d5 10.e3 Nbd7 11.cxd5 Nxd5
Not so good is 11...exd5 12.Bd3 Re8 13.Ne2 and 13...Rxe3? fails
to 14.Bxf6! (but not 14.Bh7+ Nxh7!) 14...Nxf6 15.Bh7+ and White
wins an exchange.
12.Bxd8 Nxc3 13.Bh4 Nd5 14.Bf2
White has the two-bishops advantage. If he manages to complete
his development, White will be better.
14...c5 15.Bb5 Rfd8 16.e4
After 16.e4
Nothing is new so far...
16...Nc7!
This is a good novelty. Black entices White into snatching the c5-pawn.
I recommended this move in "64" magazine a couple of
years ago. However I did not manage to analyze this move as
deeply as Garry did...
The opponent tested this position with reversed colors in the
blitz-match two years ago. That encounter saw 16...Ne7 17.Ne2 Bc6
(17...cxd4 18.Nxd4 a6 19.Be2 Nc5 20.b4 Na4 21.0-0 e5 22.Nb3 Nc3=
Anand-Karpov/Monaco/1999) 18.Ba6 b5 19.a4! bxa4 20.dxc5 Ne5 21.Nd4
and White obtained an advantage. Kasparov-Kramnik/Moscow blitz/1998
17.Bxd7
In case of the white bishops retreat, the black bishop can
move to a6. This is the point of Blacks idea.
17...Rxd7 18.dxc5 f5!
An excellent breakthrough. Black is going to open the center and
hunt for the white king. 18...bxc5 19.Bxc5 f5 has also been
analyzed, but it seems not as dangerous for White.
19.cxb6?!
I think 19.e5! in order to keep the position closed was better.
For example, 19...Rd5 20.cxb6! (not so good is 20.Ne2 bxc5 21.Rc1
Na6 22.Bg3 -- forced -- 22...Rad8 23.Nc3 Rd2 24.Rd1 Rxd1+ 25.Nxd1
Bc6 , etc.) 20...Rxe5+ 21.Ne2 Ba6 (21...axb6 22.Bd4! Rb5 23.0-0
and 23...e5 is met by a highly unpleasant 24.a4! ) 22.0-0-0 Nd5 (the
sequence 22...Bxe2 23.bxc7! Bxd1 24.Rxd1 is fraught with
consequences for Black.) 23.Nd4 -- anyway, in this line it is
Blacks job to prove that the position is about equal, not
Whites, as happened in the game.
19...axb6 20.Ne2
Now, after the exchange on b6 it is too late to play 20.e5 in
view of 20...Ra5! and White is forced to protect the e5-pawn
instead of developing. If 21.Bg3 Rad5, Black is better; if 20.Rc1
fxe4; a cunning 20.exf5 exf5 21.Be3 Nd5 22.Bd2 does not work in
view of the very simple 22...f4! and on any White move, the black
knight jumps to e3. White had another good option, 20.Bxb6!? fxe4
21.fxe4 Bxe4 22.Nf3, which gives White fair chances for equality
due to the fact that there are not many pawns left on the board.
20...fxe4 21.fxe4
If 21.0-0 exf3 22.gxf3 then Black does not capture on f3 but
proceeds with 22...Rd2!
21...Bxe4 22.0-0
Vladimirs desire to evacuate his king from the center is
understandable. However, he allowed the black rook to penetrate
to the second rank. More counterplay was promised by 22.Bxb6 Nd5
(or 22...Bxg2 23.Rg1) 23.Bc5 Bxg2 24.Rg1 with an unclear position.
Now White is in trouble.
22...Rd2
After 22...Rd2
The g2 square is the main aim of the d2 rook!
23.Nc3?!
This is another innaccuracy. After 23.Rfe1! Rxb2 (23...Bb7 24.Rad1!)
24.Bd4! Rc2 (24...Rb3? 25.Nc1) 25.Bxb6 Nd5 26.Bf2 White has more
chances for stubborn defense. Blacks pieces are very
active, but there is nothing real in sight.
23...Bb7
Obviously not 23...Bc6? 24.Rac1
24.b4
After 24.Rad1 Rxb2 25.Rd7 Rc8 White cant regain the pawn in
view of 26.Na4 Rc2 27.Nxb6 Rf8! and there is no reasonable
defense against Bb7-a6.
24...Rf8!
This looks like the best way to keep the initiative. In case of
24...b5 White would have equalized with 25.a4! bxa4 26.Rxa4 Rc2
27.Rxa8+ Bxa8 28.Nd1; not very convincing is 24...Nd5 25.Ne4! Rc2
26.Nd6 Bc6 27.b5 Bd7 28.a4 and Blacks bishop is out of play.
If 24...Rc2 then not 25.Rfc1? Rxa3! but 25.Rac1! with an equal
game. I suggested 24...e5 with the idea of transferring the
knight to e6 and f4 in my on-line comments. Later I found a good
reply: 25.a4! Ne6 26.Rfd1! and White starts chasing the black
rook.
25.Ra2
Probably White could have started his own counterplay with 25.a4
Rc2 26.Nd1 (26.Rac1? Rfxf2) 26...Nd5 27.a5, but in order to do so
one would need nerves of steel.
25...Rxa2 26.Nxa2 Nd5
Black is threatening a knight-jump to f4.
27.Bd4 Ra8!
After 27... Ra8!
An excellent move! The rook completed its work on the f-file and
forced the white knight to move to a2. It is time to go back to a8!
Not so promising was 27...Nf4 28.Rf2 b5 29.Nc3 Bc6 30.Bc5 Rf7 31.Bd6
g5 32.Ne2, etc.
28.Nc3?!
With time trouble looming, Vladimir decided to jettison a pawn in
the hope of drawing the opposite-color bishops endgame. It is not
the best decision from a chessic standpoint, but it turned out
that Vladimir was right! His dream came true with a little help
from Kasparov. The continuation 28.Bb2 Ne3 29.Rc1 Bd5! 30.Nc3 Bxg2
does not solve all Whites problems. The best move for White
is 28.Rf3, though it is met with the unpleasant 28...g5! and the
black knight is heading for f4. However, White might hold in the
line 29.Nc3! Nf4 30.Re3 Rxa3 (30...Bxg2 31.a4!) 31.g3 Ra1+ 32.Kf2
Nh3+ 33.Ke2 Kf7 34.Bxb6 Rh1 35.Rd3! with some counterplay for a
pawn. Would Kramnik have found this in the time trouble?
28...Nxc3 29.Bxc3 Rxa3
Black has emerged a pawn up. With the rooks on the board, White
has problems defending this position.
30.Bd4 b5 31.Rf4!
An excellent maneuver! The rook is defending the g2 square and
eyeing the g7 pawn. Passive defense is doomed in such positions.
After 31.Rf4!
This is the key position of the game. Garry has a position with
good winning chances and a one-hour advantage! He could have
drank some coffee, looked through the window at London, thought
everything over...
31...Rd3?
was quickly played instead. After 31...g5 32.Rf6 Ra2 33.Rg6+! Kf8
34.Rxe6 Rxg2+ 35.Kf1 Black cant preserve his advantage
because White is threatening Bc5+, Re8+ and Re7+, killing the b7-bishop.
It is not so clear how good 31...Rb3 (which I thought was the
best during the game) is. The thing is that after 32.Rg4 (32.Be5?
h5!) 32...g5 33.h4 Rxb4 34.hxg5 h5 35.Rf4! White has the quite
unpleasant threat of g5-g6-g7 followed by Rf8+ at his disposal.
For example, 35...Rb1+ 36.Kf2 Rc1 37.g6 Rc2+ 38.Ke3 Rxg2 39.Rf7
Bd5 40.Rg7+ Kf8 41.Bf6! e5 42.Bxe5 All these variations are not
necessary. In any case it is clear that White has good drawing
chances. For this reason, I guess Black should have played 31...h5!,
not allowing the white rook to take up the g4 square. In this
case White is doomed to a tough defense. For example, 32.g4 h4 (or
32...Bf3 33.gxh5 Bxh5) 33.g5 Ra2 34.Rxh4 Rg2+ 35.Kf1 Rxg5 and
Black gradually builds up the pressure.
32.Rg4 g5
After 32...Rd1+ 33.Kf2 Rd2+ 34.Ke3 Rxg2 35.Rxg2 Bxg2 36.Be5!
White builds a fortress.
33.h4 Kf7 34.hxg5 hxg5 35.Kf2!
This move saves the game! White is going to place his bishop on e3.
35...Rd2+
Garry quickly made this move. Too quickly... After 35...Kg6 36.Ke2
Rb3 37.Be3 Rb2+ 38.Kf1 Black loses his g5 pawn.
36.Ke3 Rxg2 37.Rxg2 Bxg2 38.Be5!
Final position
Only at this point the champion sank into reflection. Too late!
The position is absolutely drawn. White maintains a blockade on
the e5 and g3 squares. Draw!