Annotated Game #154: For the want of a pawn...

This second-round tournament game is a good illustration of the importance of the d-pawn (and the d4 square) in French Opening-style pawn structures.  My opponent as White neglected this key aspect and I was able to take advantage of it, winning the d-pawn relatively early on.  Some significant improvements for Black came out in the analysis afterwards, so it was worth looking at in order to tweak my game, for example on move 13.  The strategic endgame errors my opponent made are also typical, primarily involving simplifying down into less complex (but worse) positions.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "Class E"] [Black "ChessAdmin"] [Result "0-1"] [ECO "B12"] [Annotator "ChessAdmin/Komodo 8"] [PlyCount "110"] {B12: Caro-Kann: Advance Variation} 1. e4 c6 2. d4 d5 3. e5 c5 4. c3 Nc6 5. Nf3 Bg4 {Black has achieved a standard setup against the Advance variation played by White, with his bishop outside the pawn chain. This is the basic idea of the 3...c5 variation.} 6. Be2 e6 7. O-O Nge7 {this was my first real think in the game. It's a standard way to develop the knight, but Black has other possibilities.} (7... Bxf3) (7... Qb6) 8. h3 $6 {this essentially forces an inferior continuation for White.} (8. Nbd2 $5 $11 {should not be overlooked}) 8... Bxf3 9. Bxf3 Nf5 {not a bad move, but not the most effective continuation. I should have looked more deeply into the cxd4 option and its various effects.} (9... cxd4 10. cxd4 Qb6 11. Be3 Nf5 12. Nc3 (12. Bg4 Nxe3 13. fxe3 Qxb2 $17) 12... Ncxd4 13. Qa4+ Qc6 $17) 10. Be3 (10. Bg4 $5 $11 {is the reason why Black should have taken earlier on d4. Now White can trade off the knight and inflict some pawn structure damage.}) 10... Qb6 {again, capturing immediately on d4 would be better, although the text move is fine.} 11. Qd2 {now Black is going to be a clear pawn ahead with no compensation.} (11. Nd2 Be7 $15) 11... cxd4 $17 12. cxd4 Ncxd4 {Black does not have to worry about the pin on the Nd4, since it can be dissolved two ways (...Nxf3+ or ...Nxe3).} 13. Bg4 (13. Bxd4 Nxd4 14. Bd1 Rc8 $17) 13... Bc5 (13... Nxe3 $5 {is significantly better, saddling White with additional weaknesses and giving Black additional significant threats, notably to both the weak e5 and e3 pawns. I had focused exclusively on the text move in my thinking.} 14. fxe3 (14. Qxe3 $2 Nc2 15. Qxb6 axb6 $19) 14... Nc6 $19) 14. Bxf5 (14. Bf4 O-O $17) 14... Nxf5 15. Bxc5 Qxc5 {I had focused on this position when originally calculating the sequence. Black has an extra central protected passed pawn and the e5 pawn is weak.} 16. Qd3 {this seems to be too slow. White could have used the tempo to try to develop his pieces quicker.} (16. Rc1 Qd4 $17) 16... O-O 17. Nd2 {evidently the idea behind White's queen move, clearing the d2 square for the knight.} Rac8 18. Rac1 (18. Nf3 Rc7 $17) 18... Qd4 {here I also strongly considered the queen for double rook trade, which I think would be more clearly winning, although the text move also leads to a significant advantage.} (18... Qxc1 $5 19. Rxc1 Rxc1+ 20. Kh2 Rfc8 $19 {looks good for Black, as the rooks control the only open file and should be able to parry any White threats by the queen.} ) 19. Qxd4 Nxd4 $19 20. Rxc8 $6 {I was happy to see this, giving me uncontested control of the c-file.} (20. Kh2 $5) 20... Rxc8 21. Nf3 {a common endgame error by Class players - simplifying down material only magnifies your opponent's advantages, in this case the protected passed d-pawn. The doubled f-pawns also do nothing for White.} Nxf3+ 22. gxf3 {at this point the endgame is completely winning for Black, so I concentrated on making sure that I would be safe and maintain the advantage, rather than on winning in the most rapid way.} g6 {cautiously giving the king an escape square off the back rank, while also dominating the f5 square. While not the absolute best move - probably that would be an immediate ...Rc2 - the text move keeps the win firmly in hand while lessening the number of things that I have to worry about. This is what Dan Heisman refers to as the "go to sleep" principle in the endgame.} 23. b3 f6 {the idea here is to eliminate the e-pawn and open the f-file, although White could have defended against the ideas.} (23... Rc2 $5) 24. exf6 {allowing me to execute my idea.} (24. f4 $5) 24... Kf7 25. Re1 Kxf6 {now the Black king is more centralized and White's weak f-pawns are isolated.} 26. Re2 e5 27. Rd2 Rd8 28. Kg2 {White now commits his king to the kingside, rather than to stopping Black's central passed pawn. He had no good options here, but this makes my task easier in the center.} d4 {passed pawns must be pushed!} 29. Rd3 Kf5 30. Kg3 Rc8 (30... e4 {is also possible.} 31. fxe4+ (31. Rd2 d3) 31... Kxe4 32. Rd2 d3 $19 {and now Black's king can't be stopped from going to c3 and clearing the way for the pawn to queen.}) 31. a4 $6 {this just weakens White's queenside pawns.} Rc3 32. Rd2 Rxb3 33. Rc2 e4 $6 {this is unnecessarily complicated and opens up the king to checks from the side.} (33... Rb4 { is better in consolidating the position.} 34. a5 d3 35. Ra2 Rd4 $19) 34. Kg2 { White misses his chance to harass my king and put up some more active resistance.} (34. Rc5+ Ke6 35. Kf4 e3 $19) 34... Rxf3 35. a5 (35. Rc5+ { is now not as effective.} Ke6 36. Rc7 d3 $19) 35... Rc3 36. Rb2 Rc7 37. Rb5+ Ke6 38. Rb4 Rd7 39. Kf1 Kf5 40. Rb5+ (40. Ke1 {desperation} d3 41. Rc4 $19) 40... Kf4 41. Ke2 e3 (41... d3+ {is more to the point.} 42. Kd1 Rc7 {and now the f-pawn will fall if White keeps the king on the d-file.}) 42. fxe3+ dxe3 43. Rb3 Rd2+ 44. Ke1 Ra2 (44... b6 {makes it even easier for Black} 45. axb6 axb6 46. Rxb6 $19) 45. Rxb7 Rxa5 46. Rxh7 Rh5 {this seemed obvious at the time but it allows White to take the outside passed a-pawn, which is less advantageous trade.} (46... Ra1+ 47. Ke2 Ra2+ 48. Ke1 a5 $19) 47. Rxa7 Rxh3 48. Rf7+ Ke5 49. Ke2 g5 $2 {with Black's two passed pawns only two files apart, this gives White good defensive chances.} (49... Ke4 {is the only move that preserves the advantage, in fact.} 50. Re7+ Kf4 51. Rf7+ Kg4 52. Re7 $19 g5 { now the pawn advance comes with a guaranteed win.}) 50. Re7+ $2 (50. Kd3 $11 { and White is still in the game, even equal according to the engine. The White king and rook combine well to shut off Black's king from making meaningful progress. For example} Ke6 51. Rf8 {and Black has no way to advance the e-pawn without losing it.}) 50... Kf5 51. Rxe3 $2 {immediately loses the king and pawn endgame.} (51. Rf7+ Kg4 $19) 51... Rxe3+ 52. Kxe3 Kg4 53. Kf2 Kh3 54. Kg1 Kg3 55. Kh1 Kf2 (55... Kf2 56. Kh2 g4 57. Kh1 Kg3 58. Kg1 Kh3 59. Kf2 g3+ 60. Kf1 Kh2 61. Ke2 g2 62. Kd3 g1=Q 63. Kc4 Qg5 64. Kb4 Kg2 65. Kc4 Kf3 66. Kb4 Ke3 67. Kc4 Qa5 68. Kb3 Kd3 69. Kb2 Qb4+ 70. Ka2 Kc3 71. Ka1 Qb2#) 0-1