This second-round game taught me another opening repertoire lesson, this time in the Advance Caro-Kann. I had never seen White's move 7 before and thought it must be bad, or at least would provide me with an edge. The most direct response, 7...a5, would indeed have been excellent if it weren't for White's 8. Bb5! with an annoying pin. This is how learning and repertoire-building take place most effectively, however: by filling in repertoire holes whenever they are discovered, and taking the time in home analysis to understand the relevant ideas and comparison positions.
In the late middlegame I finally play actively enough to put my opponent under pressure, but then did not spot variations on a winning endgame idea, which involved sacrificing a pawn to get my knight into the action decisively. An interesting game, in any case, where my opponent gets props for spotting how to take advantage of some incorrect ideas and passive moves - although he then over-pressed with moves like 19. f4, giving me the better chances in the end. This is a common theme seen by Caro-Kann players, and one reason I remain very satisfied with the opening as Black.
[Event "?"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "Class C"]
[Black "ChessAdmin"]
[Result "1/2-1/2"]
[Annotator "ChessAdmin/Dragon 3.2"]
[ECO "B12"]
[WhiteFideId "-1"]
[BlackFideId "-1"]
[PlyCount "68"]
[GameId "2262411026308464"]
1. e4 c6 2. d4 d5 3. e5 c5 4. Nf3 Nc6 5. dxc5 {White accepts the gambited pawn.} Bg4 6. c3 {while this move is multipurpose, it is important to recognize how it opens the d1-a4 diagonal for the White queen.} e6 7. b4 {this was the first time I had seen this move, so I thought it was sub-par and could be refuted by targeting White's pawn structure. However, this is still fine for White.} a5 $6 {has one major flaw, which my opponent finds and for some reason I did not consider. Black has several reasonable options, including:} (7... Bxf3) (7... g6) (7... a6) 8. Bb5 $1 $16 {this pin is now very annoying and prevents Black from consolidating the position. With the previous options, Black would have either prevented Bb5 or kept ...a6 in reserve.} Nge7 {making the best of it, by continuing to develop and reinforce the Nc6.} 9. O-O Bxf3 $6 (9... Ng6 $5 $14 {I considered and is best according to the engine. However, I was worried about White's queen swinging over to a4 to help the Bb5 exert pressure, so I decided to divert the queen to f3.}) 10. Qxf3 {White at this point simply has more space and better piece activity.} Qc7 {targeting e5} 11. Re1 Ng6 12. Qg3 $6 {this actually relieves the pressure considerably, as I now catch up in development and my weak f7 square is no longer targeted.} (12. Nd2 $16 {better to continue with piece development.}) 12... Be7 $11 13. Bg5 O-O {now the Nc6 is unpinned and taking on e5 is threatened.} 14. Bxc6 bxc6 $6 {an example of where the "automatic" recapture should be thought about first.} (14... Bxg5 15. Qxg5 bxc6 $11 {is the engine's preference. Pressure is retained on e5 and the Ng6 helps guard the kingside.}) 15. Bxe7 Nxe7 $6 {played with the idea of placing the knight on f5. However, this takes time and shuts off the queen from a better square.} (15... Qxe7 16. Nd2 f6 $11) 16. Nd2 $14 axb4 $6 {I had a long think here, and mis-evaluated the resulting position.} (16... d4 {is an idea I considered and the engine prefers.} 17. cxd4 axb4 18. Nc4 {the knight having d6 and b6 available just seemed wrong, but it does not have any major threats.} Ra4 19. Nb6 Ra3 $1 $11) 17. cxb4 $16 Ra4 {the best follow-up.} 18. a3 Rfa8 {I realized that I was probably a little worse, which the engine confirms, but thought I was active enough to hold the balance.} 19. f4 $6 {is overly aggressive, as the f-pawn can be effectively shut down by my knight, which now has something to do. White's king position is also now less secure.} (19. Qc3 $16) 19... Nf5 $11 20. Qd3 Qa7 {I spent some time here as well, to good effect, as this tripling of major pieces against the a-pawn forcibly equalizes.} 21. g4 Ne7 22. Qc3 Rxa3 23. Rxa3 Qxa3 {with the win of the a-pawn, material equality is restored and I prefer Black's position, with control of the a-file.} 24. Re3 $2 (24. Qxa3 $11) 24... Qxc3 $6 {I should have preserved the queen here, as White's king is much more vulnerable at this point.} (24... Qa2 $19) 25. Rxc3 $15 Ra1+ {I was getting a bit low on time here, so went with the check on a1.} (25... Ra2 26. Nf3 h6 $15) 26. Kg2 $2 {my opponent was also relatively low on time and under pressure.} Ra2 $1 $19 27. Rd3 {a sufficient defense, since I was not able to find the subtle endgame idea leading to a win.} Rb2 $6 {going for the b-pawn.} (27... g5 $1 {wins, as pointed out by the engine. Getting the Black knight into the fight would be decisive.} 28. fxg5 Ng6 $19 {and e5 will now fall, with a won endgame for Black.}) 28. Rd4 f6 $6 {another miss} 29. Kg3 Kf7 $6 (29... h5 $1 $19 {is the winning idea here, the point being to deflect the defender of f5 and bring the knight in.}) 30. h4 fxe5 $6 {now it's a draw.} (30... h5) 31. fxe5 Ng6 32. Nf3 Rb3 33. Kf2 Rb2+ 34. Kg3 Rb3 1/2-1/2
