A very important game in a central match of the Cup. With Garry Kasparov in brilliant form, we knew that the head-to-head is our best chance of stopping Ladya Kazan, and a lot depended on the white boards.
1.e4
c5
2.Nf3
d6
3.d4
cxd4
4.Nxd4
Nf6
5.Nc3
a6
No surprises so far.
6.f3
e6
And this should not have been one as well, had I called Alex Grischuk a bit earlier that morning. As it happened, we met in the elevator on our way to the playing hall, and I asked him what he thinks Viorel will play. 'Some Najdorf' came the reply, ' or maybe this e6 and Nc6 thing they all seem to play these days'.
7.Be3
Be7
8.g4
Nc6
9.Qd2
0-0
10.0-0-0
Nxd4
11.Qxd4
Another problem you're faced with when playing the Kasparov's team is the question whether he has been involved in the preparation. Since Viorel normally plays 6...e5 - he did that in our meeting with these colours in 2002 - I felt that the answer to the question above might be 'yes', and that prompted me to refrain from 11.Bxd4, which I have played in this position before.
11...Nd7
Played immediately again. The point of this early withdrawal is the line 12.h4 Ne5!, and White is forced to play 13.Be2, taking an important square away from his knight. I thought for about 5 minutes here and decided I have an interesting, if somewhat risky, alternative.
12.g5!?
Now White is ready to meet 12...Ne5 with 13.f4, and in case of 12...b5 13.h4, normal play will commence.
12...Bxg5
Of course, the most principled reply.
13.Rg1
[ During the game I felt this is is the only possible follow-up, but 13.f4
is actually not as stupid as it may look. Still, after 13...Be7!?
( 13...e5
14.Qxd6
is not very good; 13...Bf6
14.Qxd6
Bxc3
15.bxc3
Qa5
16.Kb2
is also very risky) 14.Rg1
g6
it is not clear how to develop the attack - Black looks very safe.]
13...e5?!
But this is a bit of a cop-out. As it has been pointed out to me later, Black should have played [ 13...Bxe3+
14.Qxe3
, and now after the strongest 14...Ne5
( worse is 14...Qe7
15.f4
( there's also an interesting possibility of 15.Qh6
g6
( 15...f6
16.Rxd6
Ne5
17.Be2
) 16.Qd2
depriving the knight of the g6 square 16...Ne5
17.Rg3
Rd8
18.f4
Nc6
19.Rd3
, winning the pawn back) 15...b5
16.Bd3
( 16.Qd4
e5
17.Nd5
Qh4
) 16...Nc5!
( after the natural 16...Bb7
White breaks thru: 17.e5!
dxe5
18.Bxh7+
Kxh7
19.Rxg7+
and mate) 17.e5
Nxd3+
18.Rxd3
dxe5
( 18...d5
19.f5
f6
20.fxe6
Bxe6
21.Nxd5
Bxd5
22.Rxd5
, and the game is equal) 19.fxe5
f6
( 19...f5
20.Ne2
, and White has some compensation for the pawn) 20.exf6
Qxf6
21.Ne4
, and once again White probably has enough compensation for equality, but certainly not more.) 15.f4
Ng6
16.e5
d5
17.Bd3
( 17.f5
is insufficient: 17...exf5
18.Nxd5
Qh4!
( 18...Qa5
loses to the spectacular 19.Rxg6!!
fxg6
20.Bc4
Be6
21.Ne7+
Kf7
( 21...Kh8
22.Nxg6+
hxg6
23.Bxe6
) 22.Rd7+-
) 19.Rd4
Qxh2
, and Black is on top) 17...Qh4!?
( 17...Bd7
is too quiet: 18.h4!
, and White finally has something going; 17...d4
18.Qh3
f5
( 18...Qa5
19.Bxg6
fxg6
20.Rxd4
) 19.exf6
( 19.Ne2
b5
is probably good for Black) 19...Qxf6
20.Ne2
e5
21.fxe5
Qxe5
22.Qg3
, and once again the game is roughly equal) 18.Ne2
Bd7
, and Black seems to be doing quite well.]
14.Bxg5
[ 14.Qd2
was also possible: 14...Bxe3
15.Qxe3
Nf6
16.Qg5!
( 16.Qh6
g6
) 16...g6
17.f4
Qe7
( 17...exf4
18.Qxf4
( 18.e5
Nd7
) 18...Be6
19.Rxd6
Qc7
deserves attention too) 18.f5
b5
19.Bd3
Bb7
, with some play for the pawn - but since the endgame in the game was quite inviting, I did not hesitate for long.]
14...exd4
15.Bxd8
dxc3
16.Be7
During the game I was very upset not to have played [ 16.Ba5!?
cxb2+
17.Kxb2
,but in fact after 17...Nc5!
18.Rxd6
Be6
the bishop will never reach the long diagonal]
16...cxb2+
17.Kxb2
Re8
18.Bxd6
With two bishops and nice open files for his rooks, White is of course better here - but Black has no real weaknesses, and achieving progress is not by any means easy
18...g6
Black could also play [ 18...Nf8
immediately here. A possible reply to that is 19.h4!?
, hoping to open the h-file at some point.]
19.Rg5
I decided this rook's future was no longer connected with the g-file, and transferred it to the other side of the board. [ 19.h4
could have been met by 19...Nf6
, and Black is hanging in there.]
19...Nf8
20.Rc5
Ne6
21.Rc3
Bd7
22.Be5
Bc6
23.Bf6
Rec8
Phase one is completed. White controls the d-file, and his pieces are well-coordinated. What to do next, though, is a very difficult question.
24.Bc4?
In the end, I decided that the only way forward is to trade off the the light-squared bishops by transferring my bishop to d5. If Black takes, then I will have a passer, and if he allows me to take on c6, I will have a target to work on. Or so went the reasoning... [ 24.Rd6
Ng7
or; 24.Bh3
Ba4
did not seem to do too much, but still it was better than the text.]
24...Nc5!
Somehow this square has completely evaded my sight. Now Black is almost out of the woods.
25.Re3
A sad necessity. I would much prefer starting with [ 25.Kc1
, but here Black has an immediate equalizer: 25...Bxe4!
26.Bd4
Na4
27.Bxf7+
Kxf7
28.Rxc8
Rxc8
29.fxe4
, and draw soon]
25...Na4+?!
An unfortunate decision. I was very pleased to find that the attempt to force matters by playing [ 25...Nd7
26.Bd4
Bxe4
is refuted by 27.Bb3!
Bc6
28.Re7
Rf8
29.Be3!
, and the threat of 30.Bh6 decides, since 29...Rae8
loses to 30.Bxf7+
; However, Black could start by playing 25...b5
, and after 26.Bd5
( it may actually be better to retreat 26.Bf1
here, but admission of error to such extent would not come easily) 26...Na4+
27.Kc1
Nb6
, and Black has won an important tempo.]
26.Kc1
b5
27.Bb3
Not 27.Bd5 now, of course.
27...Nb6
[ 27...Nc5
28.Bd5
is still some way from the draw]
28.Rd4
[ White would in general prefer to stop 28...Nc4 by 28.Rc3
but here Black has 28...Na4!
]
28...Be8
[ An important alternative was, of course, 28...Nc4
29.Bxc4
( 29.Rc3
Be8
) 29...bxc4
30.Rxc4
Bb5
with great chances for a draw, but Black was understandably reluctant to suffer this ending. He was already somewhat short of time here.]
29.f4
But now White has a chance to launch some sort of an offensive. I still thought it should be a draw here, until after
29...Rc6
[ 29...Nd7
30.Bh4
Nc5
31.Bd5
Bc6
32.f5
is not much fun as well, although it was probably preferrable to the text]
30.e5
Rac8
31.f5
Nc4
(all almost forced) it has occured to me that White should not play something idiotic like 32.Rh3 - he only has to protect himself from Nc4-a5.
32.Re2!
After playing this move I was able to access the position calmly, and realized I am almost winning. The plan is to trade on g6 and play e5-e6 at the best moment, while Black has no counterplay at all - a5 is always met by a4, undermining the knight.
32...R6c7
Time-trouble, but unless Black forces White's hand, he can improve his position forever: h4, a3 and so on. [ 32...Na5
is of course met by 33.Bd5
]
33.fxg6
fxg6
34.e6
The winning plan is something like h4, Rf4, Bd4, Ref2, and penetration along the f-file.
34...Rc6
[ 34...h6
35.h4
did not improve the position too much, but at least Black no longer runs the risk of getting mated.; while 34...Rc5
runs into 35.Rd7
R5c7
36.Bxc4
I must admit I was so reluctant to part with the bishop during the game that I missed that resource 36...bxc4
37.Red2
, and Black is unable to parry the threat of 38.Be5]
35.Bg5?!
Once again White could play 35.Rd7 here, but since I did not see 36.Bxc4 I could not resist the temptation to improve on this shot a little bit.
35...Kg7?
And this is the final mistake. [ 35...Rc5
36.h4
Rf5
would allow Black to continue the fight.]
36.Rd7+
Kg8
37.Bh6
Now White will keep both bishops.
37...Nd6
[ 37...Rc5
38.Rg7+
Kh8
39.Rf2
Rf5
40.Rxf5
gxf5
41.Rf7
is no improvement.; and neither is 37...g5
38.Rg7+
Kh8
39.Rf7
Bxf7
40.exf7
Rxh6
41.Re8+
]
38.Rxd6
Rxd6
39.Rf2
With mate to follow. Black resigned.
As it happened, it was only enough to save the match for us - Kasparov continued his excellent run by beating Grischuk in a hard-fought game, and Paco Vallejo missed great winning chances for us on board 5 - so we needed some help from the Beer-Sheva team, who held Ladya in the next round, to see us through. You always need a break somewhere, though - and I think we deserved the Cup in the end. 1-0