Annotated Game #103: Right ideas, wrong moves

This round two Slow Chess League tournament game saw me go down in flames early in a Caro-Kann Panov variation.  The opening in general can be a little tricky for Black, as if he's not careful White can make some quick threats on the kingside and tactics also loom in the center.  Black should be able to equalize with correct play, naturally, but I was unfamiliar with this particular line and made incorrect strategic choices at key points, helping lead me into a game-ending blunder:

• The first strategic choice occurred on move 10, when I play ...a6.  Not a bad idea in itself, but the logical follow-up of ...b5 and ...Bb7 never happens, stranding the bishop on c8 until its development is problematic (and in fact loses the game for me later!)  The database shows how Korchnoi used the ...Na5 idea, which gains a tempo on the Bc4 immediately, allowing the ...Bb7 development.
• After White goes for the immediate liquidation of his isolated queen pawn (IQP), Black goes for the exchanges on d5 and then goes adventuring with the knight on b4.  The immediate 13...Bf6 or on the subsequent move would have concretely improved Black's pieces and started making threats down the long diagonal.  I eventually play this, but at a point where its impact is lessened and my position in the meantime has been weakened.
• By move 16, Black is faced with another choice under worse circumstances, of how to blunt White's pressure on the h7 pawn.  I chose a superficially active move (...f5) that had a much greater drawback of opening the a2-b8 diagonal, which is what White soon after uses to defeat me.

This game, in addition of being a reminder of the importance of CCT with the losing move, also served to illustrate how one can have some of the right ideas and recognize good candidate moves, but fail to execute them properly (or at all).

[Event "Slow Swiss #8"] [Site "Chess.com"] [Date "2013.08.27"] [Round "2"] [White "skilledwolf"] [Black "ChessAdmin_01"] [Result "1-0"] [ECO "D40"] [WhiteElo "1477"] [BlackElo "1175"] [Annotator "ChessAdmin/Houdini"] [PlyCount "45"] [EventDate "2013.??.??"] [TimeControl "45"] {D40: Queen's Gambit Declined: Semi-Tarrasch with e3} 1. e4 c6 2. d4 d5 3. exd5 cxd5 4. c4 Nf6 5. Nf3 e6 6. Nc3 Be7 7. Bd3 dxc4 {the typical rejoinder to Bd3, leaving White with an isolated queen's pawn and making him move the bishop twice in succession.} 8. Bxc4 Nc6 9. O-O O-O 10. Bf4 {there are a wide divergence of possibilities here for White. The text move is fairly aggressive and reinforces White's control of e5.} a6 {the first critical strategic choice by Black. The idea of course is to play b5 with tempo on the Bc4 and then develop with ...Bb7, as can occur in similar opening variations. Unfortunately I never get the chance to execute this.} (10... Na5 {occurred to me afterwards as a possibility and it was played by Kortchnoi in the following game, where the Black bishop was in fact successfully developed on the long diagonal.} 11. Bd3 b6 12. Qe2 Bb7 13. Rfd1 Nd5 14. Nxd5 Qxd5 15. Qe5 Rfd8 16. Qxd5 Bxd5 17. Ne5 a6 18. b3 Bd6 19. Bd2 Bxe5 20. dxe5 Bxb3 21. axb3 Rxd3 22. Bxa5 Rxd1+ 23. Rxd1 bxa5 24. Rd6 g5 25. g3 Kg7 26. Kg2 Kg6 27. g4 Rb8 28. Rxa6 Rb5 29. h3 h5 30. gxh5+ Kxh5 31. Ra7 Kg6 32. Ra6 Rxe5 33. Rb6 Re4 34. Rb5 Rb4 35. Rxa5 Rxb3 36. Ra7 f5 37. Ra6 Kf6 38. Ra8 Rd3 39. Ra6 Ke5 40. Ra5+ Ke4 41. Ra6 e5 42. Ra4+ Rd4 43. Ra3 Rd8 44. Ra4+ Kd5 45. Ra6 Ke4 46. Ra4+ Rd4 47. Ra3 f4 48. Ra8 Rc4 49. Ra3 Kd4 50. Ra5 Rb4 51. Kf3 e4+ 52. Kg4 Rb2 53. Ra4+ Kd3 54. Ra3+ Ke2 55. Kxg5 f3 56. Re3+ Kxf2 57. Rxe4 Kg3 58. Re3 Rb5+ 59. Kg6 Kf4 60. Re1 Rb6+ 61. Kg7 Kg3 62. Re3 Rb4 {0-1 (62) Franco Raymundo,E-Kortschnoj,V (2640) Havana 1966 }) 11. d5 {White chooses to go for the immediate liquidation of the IQP. The .. .Na5 idea would also serve Black well here, removing the Bc4's support for d5 by force.} exd5 (11... Na5 $5 {and Black would exchange knight for bishop after either Bb3 or b3.}) 12. Nxd5 Nxd5 13. Bxd5 Nb4 $146 {this is dubious for some classical reasons, including the oldie but goodie about moving the same piece twice in an opening without furthering development. The idea was to finally get the Bc8 developed on the long diagonal, but White simply retreats to e4 and spoils that, so Black is left without any concrete gains.} (13... Qb6 {would have been an improved version of the idea, as the following game illustrates:} 14. Qc2 Nb4 15. Qb3 Nxd5 16. Qxd5 Be6 17. Qd2 Rfd8 18. Qe2 Bf6 19. Ng5 Bf5 20. Rad1 h6 21. Ne4 Qxb2 22. Qc4 b5 23. Qc6 Bxe4 24. Qxe4 Qxa2 25. Qc6 Rac8 26. Rxd8+ Rxd8 27. Bxh6 b4 28. f4 {Gaspersic,M-Orel,S (2040) Kranj 1997 0-1 (38)}) (13... Bf6 {was something I had considered and would have been an easier road to equality. Black could be saddled with the isolated, weak c-pawn but have the two bishops (or BvN with an open board) and plenty of piece activity to compensate.} 14. Bxc6 bxc6 15. Be5 Be6) 14. Be4 Qb6 {in contrast with the previous variation with an earlier Qb6, the move does not come with tempo (without the knight on b4 it would attack the undefended b2 pawn) and White can simply kick the knight in response.} (14... Bf6 $5 15. Bd2 a5 16. a3 Nc6) 15. a3 (15. Re1 {is recommended by Houdini as a more active developing move, although the text move suffices to give White a small advantage.} Bc5 16. Qe2 Be6 17. a3 Nd5 18. Be5 {and the storm clouds are gathering on the kingside over Black's position.}) 15... Nc6 16. Qc2 {here is another point where I had several choices, but all of them defensive now. I try to get creative and avoid the positional compromises that would occur after moving the h- or g-pawns, but it turns out that opening the a2-g8 diagonal is worse.} f5 $6 (16... h6 17. Be3 Qb5 $14) 17. Bd3 {this surprised me during the game, as I thought White would immediately seize the diagonal.} ( 17. Bd5+ $5 Kh8 18. Rfe1 Bf6 19. Rad1 $16 {and Black has major problems, with his problem light-squared bishop and drafty kingside, while White thoroughly dominates the center with his pieces and can use that to launch a kingside attack eventually.}) 17... Bf6 {I start playing more actively, which is good, although it would have been much better several moves earlier.} 18. Rab1 Nd4 { Houdini thinks this is OK, but tactics are looming on the diagonal and it would have been prudent to immediately remove the king from it.} (18... Kh8) 19. Nxd4 Bxd4 20. Rfe1 Be6 $4 {here I played this quickly as being "obvious" to block the diagonal, without doing a CCT/blunder check. After White played his (surprising) next move, only then did I see the fact that White could follow up with the Bc4 skewer.} (20... Kh8 $14 {was possible, notes Houdini.}) 21. Rxe6 $1 $18 {the game is now over, but I play on a couple moves more in the hopes of my opponent blundering back, which doesn't happen.} Bxf2+ 22. Kh1 Qd4 23. Qc4 1-0