Monday, September 16, 2013

Chessintuition part 2

In the 3rd round of Open Gent I suffered a strange and unexpected defeat against a 200 points lower rated player, Gilles Regniers. Now I immediately have to add that Gilles is likely underrated as he won very recently the Open Flemish championship despite the fact that several higher rated players were participating, see final positions. So strange and unexpected has to be linked with the course of the game rather than the ratingdifference.

Already in the opening I made a remarkable mistake. During the game I wasn't able to remember the openingstheory. I have that more often when it concerns a variation which I did study but didn't pop up earlier in my practice. During the prizegivings Thibaut Maenhout told me that I am not the only one suffering with this kind of problem. The key to solve this problem is of course a lot of rehearsals till it is branded in the memory but my motivation and priorities ignore this solution. Now forgetting the theory doesn't mean an insurmountable problem if you play the white pieces as long you choose to play pragmatically a solid continuation (14.g3) instead of the most critical one. However I am not a pragmatic player as shown in my blogarticle the scientific approach . So I chose for the risky idea with the exchangesacrifice which I noticed in a similar position, see game below.
[Event "Corus"] [Site "Wijk aan Zee"] [Date "2006.01.24"] [Round "9"] [White "Anand, Viswanathan"] [Black "Van Wely, Loek"] [Result "1-0"] [ECO "B33"] [WhiteElo "2792"] [BlackElo "2647"] [PlyCount "83"] 1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 e5 6. Ndb5 d6 7. Bg5 a6 8. Na3 b5 9. Nd5 Be7 10. Bxf6 Bxf6 11. c3 Bg5 12. Nc2 Rb8 13. a4 bxa4 14. Ncb4 Bd7 15. Bxa6 Nxb4 16. cxb4 O-O 17. O-O Bc6 18. Rxa4 Bxa4 19. Qxa4 Qe8 20. Qxe8 Rfxe8 21. b5 f5 22. b6 fxe4 23. h4 Bd2 24. b7 Kf7 25. Rd1 Bh6 26. Nb4 Ke7 27. Nd5 Kf7 28. g4 Bf4 29. Re1 g5 30. Re2 Red8 31. Nb4 d5 32. Nc6 Rg8 33. Nxb8 Rxb8 34. h5 Ke7 35. Kf1 d4 36. Rc2 e3 37. fxe3 dxe3 38. Rc7 Kf6 39. Rxh7 e4 40. Bc4 Rd8 41. Rf7 Ke5 42. Rd7 1-0" />
Also in my game against Gilles you will notice that I first play a4 to continue afterwards with Ncb4 and cxb4. I was aware of the differences with the Anand-Van Wely game but didn't see a direct refutation of the idea when I executed it. Only a few moves later I already regretted my decision when I discovered that the apparently innocent differences do have a crucial impact on the evaluation of the position.
[Event "Open Gent 3de ronde"] [Date "2013"] [White "Brabo"] [Black "Regniers, G."] [Result "*"] [ECO "B33"] [WhiteElo "2344"] [BlackElo "2106"] [PlyCount "62"] 1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 e5 6. Ndb5 d6 7. Bg5 a6 8. Na3 b5 9. Nd5 Be7 10. Bxf6 Bxf6 11. c3 O-O 12. Nc2 Rb8 {(Bg5, I already once met on the board in my game of 2005 against Arben Dharda. I obviously also studied once Rb8 but during the game it was impossible to remember what exactly is the critical continuation.)} 13. h4 Be7 14. a4 $146 {(I knew g3 was playable but at the same time I also remembered that present theory tells us that white can not hope for an advantage with that move. I also knew that a4 at some moment had to be played but when exactly I could not remember. Eventually I chose to play it already now which is a novelty in this position which weirdly is not that bad. Afterwards I found back in my notes that Nce3 is here critical and in some recent correspondencegames white created chances for some advantage.)} bxa4 15. Ncb4 $2 {(I play the same idea as in the topgame Anand - Van Wely played in Wijk aan Zee 2006 but this is here nonsense as black already castled. Playable are Nce3 or even the cool Ra2.)} (15. Nce3 $5 Rxb2 16. Qxa4 Nb8 17. g3 Be6 $13) 15... Nxb4 16. cxb4 f5 { (Here I already realized that something was incorrect in my opening as in the variation which I remembered, black plays Bd7 to prevent white taking easily on a4. )} 17. Rxa4 fxe4 18. Bxa6 {(Neither g3 solves the problems so I prefer to play on with full stomach. )} (18. g3 Kh8 $1 19. Bg2 e3 $1 20. Nxe3 Bd7 $1 21. Rxa6 Bb5 22. Ra3 d5 23. Nxd5 Bc4 $17) 18... e3 $2 {(Black wants to remove the pawns on the queen-side but here the cool Bxh4 was a lot stronger. )} 19. Nxe3 Bxa6 20. Rxa6 {(Of course I wanted to play Qb3 in the game but I indicated afterwards correctly that it is not good. )} (20. Qb3 $6 d5 21. Qxd5 Qxd5 22. Nxd5 Bb5 23. Nxe7 Kf7 24. Ra7 Ra8 {(The computer shows a funny alternative here, Ke6 as the knight can not escape and will be eaten by Rf7. White however keeps some survivingchances. It reminds me on my blogarticle: de paardenlokker.)} 25. Nc8 Rxa7 26. Nxa7 Ra8 {(This trick I saw in my calculations. My opponent told me afterwards that he missed it but I am pretty sure that he would have seen it if the line popped up in the game. )} 27. Nxb5 Ra1 $15) 20... Rxb4 21. g3 Rxb2 22. Qd5 Kh8 23. O-O Qd7 24. Ra8 Rb5 25. Rxf8 Bxf8 26. Qa8 Kg8 27. Rc1 Qb7 28. Qe8 Rb1 29. Qe6 Kh8 30. Rxb1 Qxb1 31. Kh2 Qg6 *" /
Black has lost the big advantage and in the final position I can now make an easy draw with exchanging the queens. Engines have still problems today to notice that white has a fortress in this type of positions. However I wasn't satisfied with the draw and I assumed playing risk-less for a win, was still possible. Afterwards my opponent was surprised to hear that from me but I had some good arguments. First I possessed the advantage of Capablanca. He claimed in 1932 that the tandem queen+knight was stronger than queen+bishop, see the historical article from Edward Winter. Hereby I immediately have to add that the correctness of this claim is being disputed today, see e.g. this article. More important is that white controls the key-square d5 to place a dominant knight which can't be exchanged. The strength of such trump can be seen well in the correspondence game below which I won.
[Event "EU/M/1280"] [Date "2000"] [White "Brabo"] [Black "Kevicky, M."] [Result "1-0"] [PlyCount "89"] 1. e4 Nc6 2. Nf3 e5 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 d6 8. c3 O-O 9. h3 Be6 10. d4 Bxb3 11. axb3 Re8 12. d5 Nb8 13. c4 c6 14. dxc6 Nxc6 15. Nc3 Qb6 16. Bg5 Nb4 17. cxb5 axb5 18. Qe2 Rxa1 19. Rxa1 Rb8 20. Bxf6 Bxf6 21. Qxb5 Qxb5 22. Nxb5 Be7 23. Nc3 Nd3 24. Nd2 Nxb2 25. b4 g6 26. Nd5 Bf8 27. Ra3 Bh6 28. Rb3 Bxd2 29. Rxb2 {(White has a dreamposition which he has no problems to convert in a full point.)} Bg5 30. Kf1 f5 31. f3 Bd8 32. Ke2 Kf7 33. Ra2 Rb7 34. Kd3 Ke6 35. Ra8 Kd7 36. Kc4 fxe4 37. fxe4 h5 38. b5 g5 39. g4 hxg4 40. hxg4 Bc7 41. Rg8 Bd8 42. Rg7 Kc8 43. Rxb7 Kxb7 44. Ne3 Bb6 45. Nf5 1-0" />
Finally I knew from a previously on this blog discussed Svechnikovvariant that white in similar pawnstructures can keep on playing for a win with (temporarily) a pawn down. A recent example from correspondence in which white had success, can be replayed below.
[Event "EU/WS/M/026"] [Site "ICCF"] [Date "2011.01.25"] [White "Haas, Bernhard"] [Black "Hansen, Lars-Henrik Bech"] [Result "1-0"] [WhiteElo "2190"] [BlackElo "2133"] [PlyCount "147"] 1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 e5 6. Ndb5 d6 7. Bg5 a6 8. Na3 b5 9. Nd5 Be7 10. Bxf6 Bxf6 11. c3 Bg5 12. Nc2 Ne7 13. h4 Bh6 14. a4 bxa4 15. Ncb4 O-O 16. Qxa4 Nxd5 17. Nxd5 a5 18. Bc4 Bd7 19. Bb5 Be6 20. Bc6 Rb8 21. b4 axb4 22. cxb4 Kh8 23. b5 Bxd5 24. Bxd5 Qb6 25. O-O Qxb5 26. Qxb5 Rxb5 27. Ra6 f6 (27... f5 28. Rxd6 fxe4 29. Bxe4 Rb4 30. Re1 Rb2 31. f3 Bf4 $11) 28. Rxd6 Rc5 29. Rd7 g6 30. Kh2 Rc1 31. Rxc1 Bxc1 32. Kh3 f5 33. f3 Be3 34. Re7 fxe4 35. Bxe4 Bd4 36. h5 Rf6 37. g4 Rd6 38. g5 Kg8 39. Kg4 Bc3 40. Re8 Kf7 41. Rh8 Kg7 42. Rc8 gxh5 43. Kxh5 Ba5 44. Ra8 Bb6 45. Rb8 Bc7 46. Rb7 Rd7 47. Ra7 Rf7 48. Ra8 Rf8 49. Ra6 Rf7 50. Rh6 Kf8 51. Rc6 Bd8 52. Re6 Re7 53. Rd6 Bc7 54. Ra6 Kg7 55. Ra8 Rd7 56. Re8 Bd6 57. Bc6 Rf7 58. Bd5 Rf8 59. Re6 Rd8 60. Be4 Bf8 61. Rh6 Kf7 62. Rxh7 Bg7 63. Kg4 Rh8 64. Rxh8 Bxh8 65. Kh5 Kg7 66. Bc2 Kf7 67. Bd3 Kg7 68. Bc4 Kh7 69. Be6 Kg7 70. Bf5 Kf7 71. Kh6 Bg7 72. Kh7 Kf8 73. Be6 e4 74. f4 1-0" />
Adding up everything, made me very optimistic about my chances which explains why I avoided several easy drawingvariations later during the course of the game. Initially it looked all very nice till I pushed too far and lost my way in the complications. Even then a draw was still possible but switching to defense with little time remaining didn't work anymore.
[Event "Open Gent 3de ronde"] [Date "2013"] [White "Brabo"] [Black "Regniers, G."] [Result "0-1"] [ECO "B33"] [WhiteElo "2344"] [BlackElo "2106"] [SetUp "1"] [FEN "5b1k/6pp/3pQ1q1/4p3/7P/4N1P1/5P1K/8 w - - 0 32"] [PlyCount "63"] 32. Qc4 $5 {(Qxg6 is an easy draw but I wanted more what a risky strategy is with little time on the clock and a pawn less.)} (32. Qxg6 $5 hxg6 33. Kg2 Kg8 34. Kf3 Kf7 35. Ke4 Ke6 36. g4 Be7 37. f3 Bxh4 38. Nd5 {(I showed this position to my opponent after the game. Engines still give a clear advantage to black because of 2 pawns extra but it is clear that white has a fortress on the white squares. )}) 32... Qf6 33. Kg2 g6 34. Qc6 Kg7 35. Nd5 Qf5 36. Qc4 e4 {(Black neither is playing for a draw. Despite the textmove is giving a pawn away, nothing changes to the evaluation of the position.)} 37. Qd4 $5 {(I noticed that Nc3 wins a pawn but also queens are exchanged after it so removing any winning chances. Again I chose for the risky continuation.)} (37. Nc3 $5 Qf3 38. Kg1 e3 39. Qd4 Kf7 40. Qxe3 Qxe3 41. fxe3 $11) 37... Kf7 38. Nf4 Be7 39. Qc4 Kg7 40. Ne6 Kh6 41. Qc1 g5 42. Nd4 $6 {(Now I am pushing too far. Here I had to exchange with hxg5 to an easy defendable queen-endgame. )} Qd5 43. Qe3 Bf6 44. Ne2 Qf5 $6 {(The engines recommend Qd3 to keep some advantage but for practical play I prefer Qf5.)} (44... Qd3 $1 45. hxg5 Bxg5 46. Qxd3 exd3 47. Nc3 $15 {(White should have little problems to draw. The fact that black has the wrong bishop for the h-pawn, significantly increases the drawing chances. )}) 45. hxg5 Bxg5 46. Nf4 $2 {(The precise Qb3 is recommendeded by the engines and leads to full equality but such moves are very hard to find in timetrouble. )} Qg4 $6 {(Not a bad choice from practical point of view but technically Bxf4 is more critical. )} (46... Bxf4 $1 47. gxf4 Qg4 48. Kh2 d5 49. Qc1 Qh5 50. Kg1 d4 51. Qc8 Qd5 52. Qf8 Kh5 53. Qg7 d3 54. Qxh7 Kg4 55. Qg7 Kxf4 { (A nice position for Finalgen which tells me that black should win in maximum 85 moves !! The program Finalgen does not permit today to tell us if this is in order with the 50 moves rule. In any case if I follow the first choice of Finalgen each time then I can find a path in which white can claim a draw with the 50 moves rule. However it is an insurmountable task to check all the alternatives manually. Because of the limited practical value of this endgame, I am satisfied with the verdict that black has good winning chances. )}) 47. Qxe4 Bxf4 48. Qd4 $2 {(White misses a fantastic draw in this strange position. )} (48. Qe7 $1 Be5 49. f4 Bb2 {(After Bxf4 follows Qf6 and black can not retract himself in a proper way from the checks.)} 50. Qxd6 Qg6 51. Qxg6 hxg6 $11 {(White can still barely draw with g4. )} (51... Kxg6 $11)) 48... Kh5 $2 {(During the game I noticed Qg5 as black can save the bishop without giving up a pawn. )} 49. Qh8 $2 {(White misses a final opportunity to stay in the game. The remaining part of the game, I had to play in 1 minute while my opponent still had a half hour so obviously a desperate task.) } (49. Qe4 $5 Kh6 50. Qe7 $1 {(I discussed the correct method to draw in the analysis of the previous white move. )}) (49. Qd5 $5 Qg5 (49... Kh6 50. Qe4 $1 { (Of course no repetition as otherwise black can correct his mistake. )} Qg5 51. Qe6 Kg7 52. Qd7 Kh8 53. Qf7 {(The rest we can see in the mainvariation from move 63 onwards. )}) 50. Qf7 Kh6 51. Qf8 Kg6 52. Qg8 Kf6 53. Qf8 Ke5 54. Qe8 $1 Kd4 55. Qa4 Kc5 56. Qa7 Kc6 57. Qa6 Kd7 58. Qb7 Ke8 59. Qc8 Kf7 60. Qd7 Kf8 61. Qc8 Kg7 62. Qd7 Kh8 63. Qf7 Be5 64. f4 Qg8 65. Qe7 {(Black can not save the bishop and prevent perpetual check and keep all the pawns on the board which means white can make a draw. )}) 49... Qg6 50. Qd8 Bg5 51. Qd7 Qe4 52. f3 Qe7 53. Qg4 Kg6 54. f4 h5 55. Qf3 Bf6 56. Qd3 Kg7 57. Qf5 Qf7 58. Kh3 Kf8 59. Qc8 Qe8 60. Qf5 Ke7 61. Qh7 Qf7 62. Qd3 Qg8 63. Qe4 {(With a few seconds remaining on the clock, I did not wait for Qe6 and resigned. )} 0-1" />
This game clearly shows that trusting different patterns, themes,... is no guarantee for success. In an earlier blogarticle about chessintuition somebody made the remark that intuition is just applying learned knowledge. Often it isn't that simple. Except in particular openings, no position is exactly the same as another one. So each position has its own characteristics which means you can't blindly apply some knowledge. Applying the right knowledge at the right moment is also intuition. This you can't just learn from a book. Obviously experience helps to better evaluate. A similar sound can be heard by the present worldchampion in this youtubemovie.


No comments:

Post a Comment