New London
1.d4 d5 2.Bf4 e6 3.e3 Bd6 4.Bg3!?
Easily the most popular move in ChessBase's 2019 Mega database. It has been played by many strong grandmasters, including Kramnik, Grischuk and Kamsky, but in a way White is relinquishing the advantage of the first move, ie the pawn-structure is symmetrical, and both players have developed a piece, but it is Black rather than White to move. I guess White's counter-argument to this is that the white queen's bishop has been developed outside the pawn-chain, while Black's is shut in. A sharp alternative is 4.Qg4!?
4...Nf6 5.Bd3 Nc6!?
Threatening the central pawn break ...e5, while trying to get White to spend a tempo protecting the light-square bishop with c3 or a3.
6.f4!?
Normal is 6.Nf3, stopping 6...e5, and if 6...Nb4, then White can play 7.Be2 followed by evicting the b4 knight, probably with a3. The text also stops ...e5, but makes White's dark-square bishop bad.
6...Nb4 7.Be2 Ne4 8.c3 Nc6 9.Nd2 Nxg3?!
I sometimes play as if winning the bishop-pair is an end in itself, rather than the means to an end. The problem with the text is that the knight has spent three moves in order to be exchanged for White's bad bishop while at the same time half-opening the h file for White.
Stockfish10 and Komodo10 like 9...0-0 or 9...Ne7!? They reckon a White capture on e4 is nothing for Black to worry about.
However, Stockfish10 in particular does not like 9...f5!?, when first shown it, although after the engines' main continuation, viz 10.Ngf3 0-0 11.Bh4 Ne7 12.0-0, they reckon the position is equal.
10.hxg3 Bd7 11.Qc2 h6
11...f5?! can be met by 12.Bh5+, when 12...g6?? 13.Bxg6+ is a catastrophe for Black.
12.g4 Qe7 13.b4 a6 14.Nb3 b6 15.Nf3 a5 16.a3
Also interesting is 16.b5 Na7 17.a4 c6 18.bxc6, when the engines prefer White.
16...axb4 17.axb4
Not 17.cxb4? Rxa3 18.Rxa3 Bxb4+ 19.Kf2 Bxa3.
17...Rxa1+ 18.Nxa1 Na7 19.g5 Bb5 20.gxh6 gxh6
White gets a second queen after 20...Bxe2?? 21.hxg7 Rxh1+ 22.Kxe2. However, the engines prefer 20...Rxh6 21.Rxh6 gxh6, as an invasion with 22.Qh7 does not seem to lead to anything.
21.Kf2 Bxe2 22.Qxe2 c6
The engines prefer 22...Qd7, one point being 23.Qa6 can be met by 23...Nb5.
23.Qd3
I intended meeting 23.Qa6 with 23...Qc7, when the engines give 24.c4!, the idea being that 24...Bxb4 25.Rb1 c5 26.dxc5 Bxc5 27.Nb3 leaves Black's position under a lot of pressure. A better reply seems to be 24...dxc4, although the engines still much prefer White after 25.Qxc4.
23...Nb5 24.Nc2
I was more worried about 24.c4 dxc4 25.Qxc4, again with pressure.
24...Bc7 25.Ne5
The engines at first much prefer 25.c4, but my idea was 25...Nd6, when Black has serious counterplay thanks to the threat of ...Ne4+.
25...Bxe5?!
I considered giving up a pawn by 25...Nd6! 26.Nxc6 Qf6, but did not trust it. However, the engines reckon Black has good compensation, although they slightly prefer White.
26.dxe5
White's king is too exposed after 26.fxe5?
26...f5
Stockfish10 suggests 26...Kf8, but much prefers White after 27.g4. Komodo10 gives 26...h5, but again 27.g4 seems a strong reply. I did seriously consider the latter line, but thought Black's h pawn would not last long.
27.exf6
Even stronger, according to the engines, is giving up a pawn by 27.Nd4 Nxd4 28.cxd4!? Qxb4, when White invades with 29.Qa6.
27...Qxf6 28.Nd4 Nxd4 29.cxd4 b5
The engines give 29...Kf7 30.g4 (30.Qa6? allows serious counterplay with 30...Qg6, taking advantage of the fact that Black's king is not the only vulnerable one) Qg6 31.Qxg6+ Kxg6 32.Rc1 Rc8 33.e4 dxe4 34.Ke3 Kf6 35.Kxe4 b5 36.Rh1 Kg7, when Black is passive but it is not clear White is winning.
30.Qc2 Kd7 31.Rc1 Rc8 32.Qh7+ Kd6 33.g3?!
Safety-first, but probably unnecessary. The engines give 33.Rh1, when 33...Rh8 34.Qb7 Qg6 35.Rc1 Qe8 36.Qg7 is their main line. One possible continuation is 36...Qa8 37.e4! dxe4 38.d5! exd5 39.Ra1 Qb8 40.Ra7 with a winning attack - not a line many humans would find over the board.
33...Rc7?
Better was 33...Qf5!, when 34.Qxh6 gives Black a respite from checks, and so he can draw with, for example, 34...Ra8. Also no good for White is 34.Qxf5 exf5 as this is a much better rook-and-pawn ending, from Black's viewpoint, than the note at move 29. White should therefore perhaps try 34.Qb7, although the simple 34...Rc7 seems good enough for Black.
34.Qc2?
34.Qg8 looks strong in all lines, eg 34...Qf7 35.Qxf7 Rxf7 36.Rh1, when the h pawn cannot be defended in the long run, eg 36...Rf6 37.g4 Ke7 38.Kg3 Rg6 (38...Kf7 39.Rc1) 39.Rc1 Kd6 40.Ra1 Rg7 41.Ra8 Rg6 42.Ra7 Rf6 43.Rh7 Rg6 44.Kg3 with Kh4, Kh5 and Rxh6 to follow. This is another long engine line, but Black at no point has any counterplay and must eventually succumb.
34...Kd7?
A mistake, not least because White can transpose to the previous note with 35.Qh7+. Correct was 34...Qf5, which seems to draw.
35.Ra1
This may not be as strong as 35.Qh7+, but is probably good enough.
35...Qg7
Black loses the h pawn after 35...Qf5? 36.Qxf5 exf5 37.Rh1.
36.Ra8
The engines want to take the white king off the second rank with 36.Kf3!?
36...Rb7
36...h5 is suggested by the engines, but they much prefer White.
37.Qc5
The engines' second choice, but 37.g4 is apparently stronger.
37...Rc7 38.Qb6?
Again g4!? is the engines' main line. They continue 38...Qf6 (the checks run out after 38...Qxg4 39.Qf8 Qh4+ 40.Ke2 Qg4+ 41.Kd2 Qg2+ 42.Kc3) 39.Qf8 Qxf8 40.Rxf8 and the h pawn falls.
The text threatens Qxc7+, as well as doubling on the eighth rank, but there is a flaw.
 |
| Black to play and draw |
*****
*****
*****
*****
38...Qg6
Threatens perpetual check starting with 39...Qc2+.
39.Qc5 Qh5
Perpetual check still cannot be stopped. The game finished:
40.Qf8 Qh2+ 41.Kf1 Qh1+ 42.Ke2 Qg2+ 43.Kd3 Qf1+
But not 43...Qe4+?? 44.Kd2 Qg2+ 45.Kc3.
44.Kd2 ½–½
No comments:
Post a Comment