|
[20 Nov 2009|04:59am] |
When I was a young chessplayer, I did not want to study Capablanca. I wanted to study David Bronstein, the most artful but starcrossed grandmaster.
In 1951, Bronstein was within an inch of winning the world chess championship, but the Soviet chess leaders would have preferred that the allegiant champion Botvinnik keep the title. Under "great psychological pressure", as Bronstein put it, he pulled the biggest choke job in world chess championship history, enabling Botvinnik to tie the score when the match rules left the title with the champion in case of a tie.
Even on his deathbed in 2006, 15 years after the dissolution of the Soviet Union, plus the collapse of communism, and the disappearance of the secret police, Bronstein would not talk about threats — implied or explicit — to himself or his family in 1951.
Perhaps they held the title beyond David Bronstein's reach, but they couldn't take away his art. Bronstein had some of the fanciest opening ideas ever, and when his notions appealed to me, I played them.
For example, I have won many nice games with 1. e4 e5 2. Nf3 Nc6 3. Bc4 Nf6 4. Ng5 d5 5. exd5 Na5 6. d3 h6 7. Nf3 e4 8. dxe4 Nxc4 9. Qd4 Nb6 10. c4, when White has a monster pawn formation for his sacrificed piece, and it conveniently controls the white squares, so who needs the bishop. However, I have never had occasion to play that bishop sacrifice in a tournament game.
Once I got into trouble because I follow Bronstein's recommendations without examination. In his beautiful book 200 Open Games, Devik made this suggestion against the Bishop's Opening: 1. e4 e5 2. Bc4 Bc5 3. Qg4 d5 4. Qxg7 Qh4 5. Qxh8 Qxf2+ 6. Kd1 Qxg2 "with hectic play", he said. But one opponent simply played 5. Qxe5+, and I was lucky to survive as Black.
A month ago, one of the stronger players at our chess club played me to a draw in a Sicilian Wing Gambit after 1. e4 c5 2. b4 cxb4 3. d4 d5 4. exd5 Qxd5, when White cannot play Nc3 to budge the black queen. If this position arose again, I thought I would attempt the more accepted move order 1. e4 c5 2. b4 cxb4 3. a3 to get that b4-pawn out of the way. Then 3…d5 4. exd5 Qxd5, and White has to avoid 5…Qe5+ snaring a rook, so 5. Nf3, and the most testing move for Black is 5…e5, taking the bigger share of the center.
I never liked this position with White after 6. axb4 Bxb4, but it was necessary to try it, so I searched a database for White's options at move 7. I saw that four times in the database, some nutbar played 7. Ra3 to give away a rook.

I glanced at that, and noted that after 7…Bxa3 8. Bxa3, White's bishop temporarily traps the black king in the middle of the board, and White is going to jump on the black squares with Nc3-b5, then Nc7 or Nd6. Hm, I thought, and then I went to see which four wackjobs in chess history played that sacrificial 7. Ra3.
In all four games, white was David Bronstein playing against the legendary supercomputer Deep Blue (the best 'puter of its time, the one that made world news by beating Kasparov) and its ancestor Deep Thought.
Bronstein was a great supporter of computer chess. In the 1960s, he helped Russian programmers develop the M20 chess computer, and in the '80s and '90s, he participated often in the AEGON man vs. machine tournaments. Bronstein probably jumped at the chance to test Deep Thought and Deep Blue, and of course, it was just like Devik to experiment with that 7. Ra3 sacrifice against 'puters that don't miss a trick.
I said, "If I ever reach that position, I am going to make Bronstein's rook sacrifice."
Yesterday — November 19 is Capablanca's birthday, incidentally — I had to play Steve Svoboda again, the fellow who drew in that earlier Wing Gambit. We played:
1. e4 c5 2. b4 cxb4 3. a3
Black paused for a moment because White has immediately varied, which hints at home cooking. Even so, he played the right way.
3…d5 4. exd5 Qxd5 5. Nf3
He paused again, and I thought, "Please don't play the lesser alternative 5…Bg4". Steve played the most challenging move.
5…e5 6. axb4
A third think, and he did the right thing again.
6…Bxb4 7. Ra3
I could not conceal a grin.
7…Bxa3 8. Bxa3 Bg4
The Deeps tried 8…e4, 8…Bd7, and 8…Qa5. 8…Bg4 looks sensible because Black, who has rook plus pawn against a bishop, cannot be stopped from exchanging a piece while he is ahead in material.
9. Nc3 Bxf3 10. gxf3
White's king will also have difficulty finding a safe home, therefore.
10…Qa5
The queen is closer to her subjects with 10. ... Qd7 11. Nb5 Ne7 12. Nd6+ Kf8 13. Qb1 Nbc6 14. Qb3 Nd8, and Black is more solid.
11. Nb5 Ne7
A natural move, blocking the a3-f8 diagonal, and then the plan is …Nbc6 plus …O-O, where White has nothing to show for the sacrifice.
12. c3
White's positional logic is that he will be forced to play Nd6+, and then his bishop will be unguarded. 12. c3 prepares Qb3 to protect the bishop, and to coordinate with the knight against f7. 12. c3 also readies Bb4 if necessary.
12…Nbc6
According to plan, but 12…Nf5 means that Black won't have to move his king after the check on d6.
13. Nd6+ Kf8 14. Qb3 Qd5
Uh oh. I discounted 14…Qd5 because of 15. Bc4, but I didn't notice that …Qd5 vacates a5 for a knight fork.
I felt like I might've been losing now, and went into the tank for 30 or 40 minutes. First White has to think about giving Black what he wants by 15. Qxd5 Nxd5 16. Nxb7+ or 16. Bc4, but that's like letting the other basketball team get an emotional lift from making a defensive stop.
The alternative is 15. Bc4 Qxf3, when White has to consider giving the other rook away, but that's not good, so 16. Rg1, and then Black has his stupid fork 16…Na5. In my long think, I envisioned 17. Qb4 (the queen sacrifices 17. Nxf7 and 17. Bxf7 go nowhere) Nec6 b6 18. Qc5 b6 (18…Nxc4 is better, but I was fixed on the notion that Black would force a trade of queens if he could) 19. Qd5 Qxd5 20. Bxd5, and that looks OK for White because he keeps his pressure on f7 while pinning c6-knight and a5-knight is out of play.
15. Bc4 Qxf3 16. Rg1 Rd8
So the 30+ minutes I spent thinking about 16…Na5 was wasted. 16…Rd8 gets another piece out, and leaves …Na5 as a threat. Because Black did not make the threatening move right away, White has to break in on f7 while he can. Not with 17. Bxf7, because 17…Rxd6 18. Bxd6 Qxf7 and White is sunk.
17. Nxf7 Rxd2
Completely unexpected. When a player is totally surprised by a move, he's lucky if it doesn't kick his ass. In this case, White takes two rooks while Black takes one. 18. Kxd2 Qxf2+ 19. Be2 Qxg1 20. Nxh8 Qf2 Better to get to the f-file (White threatens Qf7#) by checking: 20. ... Qg5+ 21. Kc2 Qf5+ 22. Kd1, but White is still preferable.
21. Qxb7
Threatens mate in two after 22. Qa8+, and to capture on c6.
21…Qf4+ 22. Kd1 22. Kc2 is much better, heading for shelter on the queenside: 22…Qa4+ 23. Kb2 or 22…Qe4+ 23. Kc1 Qf4+ 24. Kb1 Qe4+ 25. Ka1. 22. ... Qa4+ 23. Ke1 Qh4+ 24. Kf1
I thought I'd blown it, because if he played 24…Qh3+, I didn't see anything better than conceding the draw to 25. Kg1 Qe3+ 26. Kf1 Qh3+. White can keep playing for a win by 24…Qh3+ 25. Kf2 Qxh2+ 26. Ke1 Qh4+ 27. Kd2 Qf4+ 28. Kc2, but White already had blindness toward Kc2 at move 22.
24…Qf4+
Unlucky to throw the wrong check, and now Black loses. 25. Kg2
Black will run out of checks. 25…Qg5+ 26. Kh1, or 25…Qe4+ 26. Bf3 Qc2+ 27. Kh1, and the three checking squares on the back rank are covered by three white pieces.
25…g6 26. Qxc6 Qg5+ 27. Kh1 Kg7
White wouldn't have been startled by 27…Qc1+.
28. Qe8 Ng8 29. Bc4
Admittedly, White relaxed. 29. Qf8+ Kxh8 30. Bc4 is precise.
29…Qe3 30. Qf8+ 1-0
|
|