Annotated Game #130: A tactical finish

This last-round tournament game featured some major swings in the initiative.  Neither myself nor my opponent were on familiar ground in the opening, but we both acheived reasonable positions.  Unfortunately I lost my way first, not having a suitable middlegame plan while allowing Black to take over the initiative and start pressing on the queenside.  The key sequence of the middlegame portion took place in moves 17-19, where Black could have established dominance but failed to see his way through the complications.  After this, Black overpressed and dropped first one pawn, then two, although it appeared like the endgame would be long and hard-fought.  To my pleasant surprise, Black continued to be over-aggressive and lined himself up for a discovered skewer against his king and rook, ending the game immediately.

Although I was a bit fatigued from the previous six rounds' worth of games and bungled the transition into the middlegame, I was able to immediately recover and then successfully calculate several different key sequences, which lead to the eventual victory.  Analyzing this game provided me with a boost to my opening knowledge and a much better idea of what to do in these types of positions, along with how to avoid the same problems in the future.

Having won my last three games, I ended up in the money for the first time in quite a while, which I felt was a vindication of my training efforts and new focus on chess.  We'll see in the next tournament analysis whether that was completely justified (hint: not entirely).

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "ChessAdmin"] [Black "Class B"] [Result "1-0"] [ECO "A37"] [Annotator "ChessAdmin/Houdini"] [PlyCount "111"] {A37: Symmetrical English vs ...g6:4 Bg2 Bg7 Nf3} 1. c4 c5 2. Nf3 g6 3. g3 Bg7 4. Bg2 Nc6 5. O-O e5 {the first time I had encountered this type of play from Black, who was himself in completely new territory.} 6. Nc3 Nge7 7. d3 O-O 8. Bg5 {a wide variety of moves have been played here. More common is the plan to push b2-b4, preparing it with Rb1 or a3. Normally in this variation of the Symmetrical English, White should aim to use his dark-square bishop to exchange (or threaten to exchange) the Ne7 rather than the Bg7.} d6 9. Qc1 { Here Houdini prefers either way of preparing to push b4 (Rb1 or a3).} Be6 $146 (9... f6 {would in fact be the thematic move from Black.} 10. Bd2 Be6 11. Rb1 Qd7 12. Re1 Rab8 13. a3 a6 14. b4 cxb4 15. axb4 b5 16. cxb5 axb5 17. Qa3 d5 18. e3 h6 19. Bf1 Bg4 20. Nh4 g5 21. Ng2 f5 22. Qa2 Kh7 23. d4 exd4 24. Nxb5 { Haffner,A-Schubert,M Kiel 2000 1/2-1/2 (64)}) 10. Bxe7 {I decide to make the bishop for knight exchange in order to decrease Black's control of d5, although this may not be the best idea.} (10. Bd2 Rb8 $11) (10. Bh6 {would be the logical follow-up to the previous move.}) 10... Qxe7 {Black has the pair of bishops, notes Houdini via the Fritz interface. However, White's knights are both good and Black has to watch the d5 square carefully.} 11. Nd2 Qd7 12. e3 {a lot of possibilities here. I chose to play to prevent a Black knight arriving on d4. However, it might be worth simply keeping the move in reserve.} Bh3 (12... a6 {is Houdini's preference, employing prophylaxis against Nb5.}) 13. Nd5 {A classical outpost, notes Houdini. White is quite happy to occupy it. } Bxg2 14. Kxg2 {the exchange has loosened the king position somewhat, the consequences of which will become clearer a few moves later.} Rab8 {Black appears to waste a little time with this and probably should play ...f5 immediately.} 15. Ne4 f5 {Black threatens to win material: f5xe4} 16. Nec3 { I felt that the knight was better placed here to support d5, hence the maneuver.} a6 17. Rb1 {here I fail to accurately assess the position's requirements. Now White's usual plan of b4 is ironically mirrored by Black and more effectively; the text move is simply too slow.} (17. a4 {would restrain Black's pawn advance;}) (17. Nb6 {would emphasize piece play and clear d5 for the other knight.}) 17... b5 {Black now has the initiative and White has no real counterplay, although the position remains balanced.} 18. b4 $2 {played without adequate calculation or feel for the consequences.} (18. f3 b4 19. Ne2 Ne7 20. Nxe7+ Qxe7 21. e4 {is Houdini's recommendation, locking up the position and heading for a draw.}) 18... cxb4 {the wrong pawn capture.} (18... bxc4 19. bxc5 cxd3 $17 {and White will have to struggle to get compensation or regain the material.}) 19. Nxb4 {here I manage to calculate the correct capture.} (19. cxb5 {doesn't work as well.} axb5 20. Nxb4 Nxb4 21. Rxb4 Rfc8 $15) 19... f4 {Black decides to move the kingside attack forward, letting up on the queenside pressure.} (19... bxc4 20. Nxa6 Rxb1 21. Qxb1 $11) 20. h3 $6 { again poor calculation on my part, as I was unnecessarily worried about the mate threat after ...f3.} (20. Nxa6 Rbc8 (20... f3+ $6 21. Kh1 {I simply didn't see his, only picturing the retreat Kg1, which of course is mate after . ..Qh3.} Rbc8 22. Rxb5 $14) 21. Nd5 f3+ 22. Kh1 e4 $14) 20... f3+ (20... bxc4 { again would keep some advantage.} 21. Nbd5 cxd3 22. Rxb8 Nxb8 23. Qb1 $15) 21. Kh2 $11 bxc4 {now Black has no more kingside threats and this is not as effective.} 22. Nxa6 Ra8 (22... Rxb1 23. Qxb1 Rf5 24. g4 $11) 23. Nb4 (23. Nd5 {a tactical possibility that Houdini finds, not exactly obvious for a human.} Rxa6 24. Qxc4 $1 {now Black needs to move his queen away, due to the discovered check threat.} Qf7 25. Qxa6 Qxd5 $18 26. e4) (23. Rb6 $5) 23... Nxb4 24. Rxb4 cxd3 25. Qd2 {White now regains the pawn, one way or another.} Rfc8 26. Qxd3 {I had foreseen the coming pin and calculated that White would do well out of it.} Ra3 27. Qd5+ Qf7 {essentially forced, otherwise white plays Rb7 as a follow-up.} 28. Qxf7+ Kxf7 29. Ne4 {by this point I've solved my problems and have no real weaknesses, with the obvious next step being an exchange of the a-pawn for Black's d-pawn.} Kg8 $6 (29... Rc2 $11 {is what I had expected.}) 30. Nxd6 {this move still gains a tempo by attacking the rook, giving White the initiative while Black will have to focus on trying to regain the material.} Rca8 {the problem for Black is that White can go after the kingside pawns now.} 31. Ne4 Rxa2 (31... R3a4 32. Rxa4 Rxa4 33. Nd2 $14) 32. g4 $16 {White opens up the diagonal for his king to enter the game. Although material is even, Black's f3 pawn is doomed.} R2a4 33. Rfb1 Rxb4 (33... Bf8 34. Nf6+ Kf7 35. Rxa4 Rxa4 36. Nxh7 $16) 34. Rxb4 $16 {I had necessarily calculated this far from move 30. Black cannot avoid material loss.} Bf8 35. Rb7 Bg7 {protecting against the mate threat of Nf6.} 36. Kg3 Rf8 37. Rb5 h6 38. Nd2 g5 39. Nxf3 e4 $6 {an error, severely weakening the e-pawn. Black saw the bishop fork on move 41 but not how White could nullify it.} (39... Re8 40. Kg2 $16) 40. Nd2 $18 Re8 41. Rb4 Bc3 42. Rxe4 $1 Rd8 43. Nf3 {now two pawns up, this should be a win for White, although I expected a long, hard slog for it.} Kg7 44. Rc4 Ba5 $2 {this takes the bishop effectively out of the action.} ( 44... Bf6 $18) 45. Nd4 Rd7 46. Kg2 Bd8 47. Ne6+ {I immediately took advantage of the chance for further simplification, figuring that the bishop (if active) would be more difficult to overcome with my knight, rather than the pure rook endgame.} ({Houdini prefers to win a pawn with} 47. Nf5+ $5 Kh7 48. Rc6 h5 49. gxh5 {which is something I confess I did not look at.}) 47... Kf6 48. Nxd8 (48. Rc6 {is another way to win the h-pawn that I missed.} Kf7 $18 49. Nd4 Rb7 50. Rxh6) 48... Rxd8 49. Rc6+ Kg7 50. Rc2 Re8 51. Re2 {playing very defensively, which is not the way to go in a rook ending. Rook activity is most important.} (51. Ra2 Rc8 $18) 51... Kf6 (51... h5 52. gxh5 Kh6 53. Kg3 $16) 52. Kf1 (52. Ra2 Rc8 $18) 52... Ke5 $2 {Black gets too aggressive and lines himself up for a tactical response.} (52... h5 $142 $5 $16) 53. f4+ $1 $18 {surprise! Says Houdini.} Ke4 (53... gxf4 54. exf4+ {wins the Re8 for White with a skewer tactic.}) 54. Kf2 gxf4 $4 {and my opponent misses the tactic while overpressing.} 55. exf4+ Kxf4 56. Rxe8 1-0