Annotated Game #297: Opening evolution

The previous tournament I felt I had played rather poorly at the time, but I felt better about my performance at the board following the last series of games analysis; even the opening blunder in game 3 yielded something useful. I also scored reasonably well, and better than statistically expected. This next tournament saw more or less the reverse, although in this first-round game I do well to remember and apply some lessons from the previous Exchange Caro-Kann game, including following through with the ...Bf5 idea. In keeping with the recent post on pawn sacrifices, the 15...b5!? idea is worth highlighting.

[Event "?"] [White "Class B"] [Black "ChessAdmin"] [Site "?"] [Round "?"] [Annotator "ChessAdmin/Dragon 3.2"] [Result "1/2-1/2"] [Date "????.??.??"] [ECO "B13"] [PlyCount "64"] [GameId "2108374627303490"] {[%evp 0,64,25,16,59,71,77,47,47,23,16,13,28,45,44,45,60,29,30,33,30,-18,12,-9,1,-16,-14,-14,10,-8,23,-22,6,-18,-10,-33,-31,-27,-26,-40,-33,-40,-50,-30,-41,-63,-56,-86,-92,-85,-70,-72,-57,-72,-54,-45,-45,-41,2,-5,-9,-6,-1,-8,0,0,0]} 1. e4 c6 2. d4 d5 3. exd5 cxd5 4. Bd3 {the Caro-Kann Exchange Variation.} Nc6 5. c3 Nf6 6. Bf4 g6 7. Nd2 {although this is the third most-played move in the database, it is the first time I have seen it.} Bf5 {I saw no reason to not proceed with the main idea of the ...g6 variation.} (7... Nh5 {is the engine move, and a common idea in this structure, but has relatively few games in the database as tests.}) 8. Qe2 Bxd3 {White has passed up the chance to exchange and inflict some pawn structure damage on Black - which, however, would have some dynamic compensation. I therefore correctly exchange light-squared bishops. White's minor piece is still stronger and therefore a good target, while Black's is definitely "bad", constrained by the pawns on d5 and g6.} 9. Qxd3 Nh5 (9... e6 {it's also fine to develop straightforwardly, with the bishop potentially going to d6.}) 10. Ne2 Bg7 (10... Nxf4 {it's better to simply carry out the threat immediately. Black's unopposed dark-square bishop may be useful in the endgame.}) 11. Nf3 O-O 12. O-O Nxf4 {now I see nothing better and exchange anyway.} 13. Nxf4 e6 {a solid approach, solidifying control of d5 and removing the weakness on e7.} 14. Rfe1 Qc7 (14... Qd6 {is a better square for the queen, as the f8-a3 diagonal is more valuable.}) 15. Ne2 a6 {a waiting move, primarily.} (15... b5 $5 {is a useful idea in this and similar positions. The point being that} 16. Qxb5 Rab8 {and White's b-pawn cannot be defended, while the rook penetrates to the 2nd rank.} 17. Qa4 Rxb2 18. Rab1 Rfb8 19. Rxb2 Rxb2 $15) 16. Ng3 {actually a worse square for the knight.} Rac8 {activating the rook.} 17. a4 {preventing ...b5, but now the knight can leap into action.} Na5 18. Qc2 Nc4 19. Re2 (19. b3 $6 Nd6 $15 {with pressure down the c-file.}) 19... Nd6 {improving the knight and opening up the c-file for the heavy pieces.} (19... b5 {is also now possible.} 20. axb5 axb5 {with now ...Ra8 and ...b4 as follow-up ideas.}) 20. Ra3 b5 {now I actually do find the correct idea.} 21. axb5 axb5 {threatening ...b4} 22. Rb3 $17 {preventing the b-pawn advance but abandoning the a-file.} Ra8 23. h4 {a good idea, to start some counterplay with threats due to the h-pawn "can opener" idea.} Ra1+ {I saw nothing better and it is certainly an effective move.} 24. Re1 Rxe1+ {as usual, exchanging decreases the tension and favors the defending side.} (24... Rfa8 $1 $19 {also finally getting the other rook into play.}) 25. Nxe1 Rc8 {now the threats are lessened.} (25... Ra8) 26. Qd3 {pressuring the b5 pawn.} Qc4 $6 {again with the trading down.} (26... Rb8 {would be more effective, overprotecting the pawn and allowing a follow-up with ...Nc4.}) 27. Qxc4 bxc4 28. Rb6 $15 {now White's problems are largely solved, with a more active rook screening his weak b-pawn.} Nf5 {this simplifies down to a draw quickly.} (28... Bf8 $5) 29. Nxf5 gxf5 {forced, otherwise d5 becomes weak.} 30. f4 Bf6 31. g3 Kg7 32. Nf3 Rc7 {and we both recognize that no progress can be made.} 1/2-1/2

Mastery Concept: Pawn sacrifices for piece activity

This is another in the periodic series of Mastery Concept posts, which highlight chess ideas at the Master level which are often ignored (or even unknown) at the Class player level. Typically these come from seeing a recurring theme in a recent series of game analysis, one which is echoed in other chess study materials that I'm using. Today's theme is the concept of choosing to sacrifice a pawn, in return for compensation in the form of piece activity.

One of my personal chess weaknesses has been an overly materialistic assessment of situations, which means I have inappropriately valued material over positional factors. This is a common phenomenon at the Class level. For example, players may hang onto a pawn for dear life, even if defending it cramps their position and gives their opponent a winning initiative, rather than deliberately jettisoning the pawn in return for freer and more active play for their pieces.

Another situation illustrated below is ignoring candidate moves that "lose" a pawn, without taking into consideration the benefits for the player's overall game. In some cases, these can even be very temporary sacrifices, or the pawn itself is "poisoned" due to a tactical response, which means ignoring the possibility - consciously or unconsciously - is a definite blind spot in the thinking process.

Finally, I'll make the observation that none of these pawn sacrifices lead to forced wins; these are not puzzles to be solved and the game is over. Rather, they are superior to other candidate moves, even if they only end up in a level position. This fact may also contribute to a Class player not considering them, if there is no obvious win or advantage that results.

Firstly from my own play, here are some clear examples. They are followed by a few master-level ones; if you go through any collections of annotated master games, you should be able to easily find more.

Annotated Game #296: The recurring idea of c4-c5, opening diagonals and the c4 square for pieces. I eventually play it, but under less favorable circumstances than I could have earlier in the game - this is also a recurring theme.

Annotated Game #295: The f-pawn sacrifice I should have played comes early enough to be considered a gambit, but the principle is the same - in return for the pawn, the rest of my pieces gain activity that provides full compensation.

Annotated Game #293: The move 15 possibility of playing c4 opens up activity for the dark-square bishop and queen, and works due to tactics (the hanging b-pawn is poisoned) that I should have further considered.

Annotated Game #291: Here's another freeing c4-c5 pawn sac idea on move 15.

Annotated Game #272: This shows the power of the e3-e4 pawn lever in Colle / Stonewall type positions, even when it is a (temporary) sacrifice.

A few Master-level examples, including from analysis:

Tarjan-Kosteniuk (2017) - 15. b4!?

From Bologan's Caro-Kann: A Modern Repertoire for Black (New In Chess, 2018) - analysis, King's Indian Attack; the d-pawn is given up by Black for piece activity, and it is also instructive to look at what happens if White tries to seize the d-pawn earlier.

[Event "?"] [White "Enter New Game"] [Black "?"] [Site "?"] [Round "?"] [Result "*"] [Date "????.??.??"] [PlyCount "32"] [GameId "2128221663227972"] 1. e4 c6 2. d3 d5 3. Nd2 e5 4. Ngf3 Bd6 5. d4 exd4 6. exd5 Nf6 7. dxc6 Nxc6 8. Nb3 O-O 9. Be2 Qb6 10. Nbxd4 Bc5 11. c3 Bg4 12. O-O Rad8 13. Qb3 Bxd4 14. cxd4 Qxb3 15. axb3 Rfe8 16. Bd1 a6 *

From John Nunn's Best Games 1985-1993, Batsford Chess Library, 1995; games 20 (move 19) and 22 (sequence starting with move 21)

[Event "World Cup"] [White "Nunn, John DM"] [Black "Portisch, Lajos"] [Site "Reykjavik"] [Round "8"] [Result "1-0"] [Date "1988.10.??"] [ECO "C73"] [WhiteElo "2625"] [BlackElo "2635"] [PlyCount "65"] [Beauty "8179769202696"] [GameId "1166636377100288"] [EventDate "1988.10.??"] [EventType "tourn"] [EventRounds "17"] [EventCountry "ISL"] [EventCategory "15"] [SourceTitle "CBM 010"] [Source "ChessBase"] [SourceDate "1989.06.01"] [SourceVersion "1"] [SourceVersionDate "1989.06.01"] [SourceQuality "1"] 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 d6 5. Bxc6+ bxc6 6. d4 exd4 7. Qxd4 Nf6 8. O-O Be7 9. Nc3 O-O 10. Re1 Bg4 11. Qd3 Bxf3 12. Qxf3 Nd7 13. b3 Bf6 14. Bb2 Re8 15. Rad1 Re6 16. Qh3 Qe8 17. f4 Rd8 18. Qe3 Nb6 19. e5 dxe5 20. f5 Rxd1 21. Rxd1 Re7 22. Ne4 Rd7 23. Nxf6+ gxf6 24. Re1 Qd8 25. Qe4 Nd5 26. Qg4+ Kh8 27. Bc1 Qf8 28. c4 Nb4 29. Qh4 Qd6 30. Qh6 Rd8 31. h3 c5 32. Re4 Rg8 33. Qxh7+ 1-0

[Event "Olympiad-28"] [White "Pinter, Jozsef"] [Black "Nunn, John DM"] [Site "Thessaloniki"] [Round "12"] [Result "0-1"] [Date "1988.11.26"] [ECO "E99"] [WhiteElo "2535"] [BlackElo "2435"] [PlyCount "92"] [Beauty "7220913783326"] [GameId "1166636396122112"] [EventDate "1988.11.13"] [EventType "team-swiss"] [EventRounds "14"] [EventCountry "GRE"] [SourceTitle "TD"] [Source "ChessBase"] [SourceDate "1994.03.01"] [SourceVersion "1"] [SourceVersionDate "1994.03.01"] [SourceQuality "1"] [WhiteTeam "Hungary"] [BlackTeam "England"] [WhiteTeamCountry "HUN"] [BlackTeamCountry "ENG"] 1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Be2 O-O 6. Nf3 e5 7. O-O Nc6 8. d5 Ne7 9. Ne1 Nd7 10. f3 f5 11. g4 Kh8 12. Ng2 a5 13. h4 Nc5 14. Be3 Ng8 15. Rb1 Bd7 16. b3 b6 17. a3 a4 18. b4 Nb3 19. Nb5 Nf6 20. exf5 gxf5 21. Nc3 e4 22. g5 Nh5 23. fxe4 f4 24. Bd2 Nxd2 25. Qxd2 Qe8 26. Bf3 Ng3 27. Rfe1 Be5 28. Ne2 Nxe4 29. Bxe4 f3 30. Nef4 fxg2 31. Nxg2 Qh5 32. Qd3 Bg4 33. Re3 Qf7 34. Qd2 Qg7 35. Rd3 Rf7 36. Re1 Raf8 37. Ne3 Rf4 38. Ng2 R4f7 39. Ne3 Bh5 40. Rf1 Rxf1+ 41. Nxf1 Rf4 42. Qe1 Bd4+ 43. Kg2 Qe5 44. Ng3 Bg4 45. b5 Bf2 46. Qxf2 Bh3+ 0-1

From Journey to NM - Lightning Strikes Back blog, the round 4 game; technically this isn't master-level play, but it's close and White's sacrifice of the d-pawn for piece activity and an attack is an excellent example.

Don't Just Memorize, Understand! Two Key Moves in the London System - the full video (from Dr. Can's Chess Clinic) is worth watching for its principles, also containing a great pawn sacrifice idea in the central variation presented against the London System, which has a concrete tactical reason behind it.

Annotated Game #296: Activity should win

One of the factors that distinguishes master-level play from typical Class-level is the value placed on piece activity, rather than static factors. This final-round tournament game is an excellent illustration of the superiority of this approach, including sacrificial ideas even when not leading to a forced win; this will also be the subject of an upcoming Mastery Concept post.

Here a common theme that I've identified, unnecessarily holding back the e4 pawn break in a Colle System-type position, also appears. At least I eventually play that, and the interesting recurring idea of a sacrificial c4-c5 pawn break, although later than I should have in both cases. And once again opposite-colored bishops are the deciding factor, in the end.

[Event "?"] [White "ChessAdmin"] [Black "?"] [Site "?"] [Round "?"] [Annotator "ChessAdmin/Dragon 3.2"] [Result "1/2-1/2"] [Date "????.??.??"] [ECO "A47"] [PlyCount "129"] [GameId "2099915250376809"] 1. d4 Nf6 2. e3 e6 3. Bd3 c5 4. c3 (4. Nf3 {This is the way to go to transpose to the Colle-Zukertort.} b6 5. O-O Bb7 6. Nbd2 Be7 7. b3) 4... b6 5. Nf3 {This is now a Colle System.} (5. f4 {would keep the Stonewall Attack going, although Black now has the option of trading bishops with ...Ba6!?}) 5... Bb7 6. Nbd2 Be7 7. O-O O-O 8. b3 {hesitating before committing to the center.} (8. e4 {is the critical idea in the Colle System.}) (8. Re1 $5 {would prepare the e-pawn advance further.}) 8... d6 {Given my opponent's opening setup, I had expected he would refrain from ...d5, but I had wanted to confirm his intentions before moving a central pawn.} 9. Bb2 $11 Nbd7 10. c4 {this is the second most played move in the position, according to the database, and is essentially a tempo loss by the c-pawn. This is not critical; however, it does mean that I can't recapture with a pawn on d4 after the later e4 push, which is important.} (10. e4 $14 {this remains the key idea in the Colle setup, followed by Qe2.}) 10... Re8 11. Qc2 {developing the queen to a useful square to clear the first rank and form a battery on the b1-h7 diagonal.} Qc7 12. Rfe1 h6 {removing the threat from the battery on the diagonal.} 13. e4 {a little late in coming, but still good to play.} cxd4 {best, as other moves give up more space to White.} 14. Nxd4 Nc5 15. Nb5 Qd7 {now I had a long think, since there was not an obvious plan to pursue.} 16. f3 $6 {solid but inflexible, and opening the a7-g1 diagonal is an additional weakness.} (16. Bf1 {this removes the bishop from the knight's attack and keeps it out of the way of the other pieces.}) (16. Rad1 {activating the rook would also be logical.}) 16... a6 $15 17. Nc3 (17. Nd4 $5 {I did not even consider, but it looks marginally more active.}) 17... Rac8 {Black has an easier time finding a middlegame plan, simply getting heavy pieces more centralized then pushing pawns.} 18. Nf1 {looking to improve my worst piece, although it's still better to retreat the bishop.} d5 19. exd5 exd5 $6 {not a terrible move in objective terms,, but now my light-square bishop has more life in it, and I am able to take the initiative.} (19... Nxd3 20. Qxd3 exd5 21. cxd5 $15) 20. Bf5 Ne6 $11 {I confess I missed the backwards knight move here, being too excited by the skewer on the diagonal.} 21. Rad1 {the logical follow-up, adding another pin against the queen.} Bc5+ 22. Kh1 d4 {I underestimated the strength of this move. The position is still equal, but the rest of the game revolves around the struggle over the d-pawn.} 23. Ng3 (23. Na4 $5 {I also considered, but eventually decided it was better to reactivate the other knight.}) (23. a3 $5 {is also a consideration, taking away the b4 square from the Bc5.}) 23... g6 24. Nce4 {while this offers up some chances for Black to go astray, it essentially simplifies the game. Some other possibilities involved exchanging on e6.} (24. Bxe6 Rxe6 25. Rxe6 Qxe6 26. Nce2 $14 {I had not calculated this far and did not see this would be an effective move.}) 24... Nxe4 25. Nxe4 Bxe4 26. Bxe4 Bb4 {illustrating how an earlier a2-a3 would have been useful; the bishop moves with tempo.} 27. Rf1 f5 28. Bd3 {not a bad square for the bishop, with some ideas of sacrificing on the kingside in the future.} Rcd8 29. a3 {played after a good deal of thought. The engine approves.} Bf8 30. Rfe1 {also played after some thought, but this is the "safe" follow-up.} (30. c5 {is the active idea I play later on. The pawn is temporarily sacrificed for piece activity, including access to the c4 square and the now-open a6-f1 diagonal.} bxc5 31. Bc1 {another key idea in the position, to contest f4, with the ideas of g2-g4, Qc4 or Bxa6 to follow up.} (31. Bxa6 {immediately is also possible.})) 30... Bg7 31. b4 {taking the c5 square away from the knight.} Nf4 32. g3 {played after a long think. In calculating the variations I was concerned about back-rank threats. It is also forcing in nature.} (32. Bc1 {looks like a more solid choice, leading to further exchanges.}) 32... Nh3 $6 {this was surprising and I thought must be bad, but unfortunately I did not find the best solution.} (32... Nxd3 33. Qxd3 Re6 34. Rxe6 Qxe6 $11) 33. Qg2 {good enough for equality, but there were better options.} (33. Kg2 Ng5 (33... f4 $2 {is what I was concerned about, as the Nh3 is protected and the king's pawn shield attacked, but} 34. Bxg6 fxg3 35. hxg3 Ng5 36. Qf5 $1 $18 {forces exchanges into what should be a winning endgame.}) 34. h4 Nf7 35. c5 $16) (33. c5 $5 {this idea works again.}) 33... Ng5 34. h4 Nf7 35. Qf2 {this just loses a tempo, since the queen has to go to f1 shortly.} (35. Qf1) (35. f4 $5 {I considered but didn't like giving up the e4 square.}) 35... Ne5 36. Kg2 $6 (36. Bf1 {and} d3 $2 {is a mistake due to} 37. Bxe5 Bxe5 38. Rxd3 $18) 36... Qc6 (36... Nxd3 {would seem to be a more consistent follow-up, taking away the best blockader of the d-pawn, but my opponent apparently is more interested in using the knight for a kingside attack.} 37. Rxd3 Rxe1 38. Qxe1 g5 $17) 37. Qf1 {defending c4 again} Ng4 {unexpected and safe due to the pin on the f3 pawn, but not actually threatening.} 38. Bc1 {played after some thought, the correct idea and essentially forced. The obvious threat is the fork on e3.} Ne3+ 39. Bxe3 dxe3 {now under pressure, I became too concerned about physically blockading the advanced pawn.} 40. Qe2 $6 {this unnecessarily cramps White.} (40. c5 $11 {here's this idea again. With more active pieces come compensation in the form of tactics. For example} bxc5 41. Rxe3 Rxe3 42. Bc4+ Kh7 43. Rxd8 $11 {and if} cxb4 $2 44. Bg8+ $1 $18) 40... Bc3 $1 $17 41. Rg1 Qf6 $6 (41... Bd2 $17) {My opponent clearly felt that he was squeezing me to death, but here I finally buckle down and play the freeing move} 42. c5 $1 $11 bxc5 43. bxc5 Bd2 44. Rb1 {now the active rook on the b-file makes a big difference, so I prioritize that over capturing the a6 pawn, although that also works.} Qc3 45. Bc4+ Kh8 {I was unsure of what was best in this open position and was in the sudden death time control, so just chose to capture material and try to hold onto my a-pawn afterwards. Continuing with active play is better.} 46. Bxa6 (46. c6 $5 {and if} Rc8 47. Bxa6 $14 {and the c-pawn is tactically defended, due to the skewer on the a4-e8 diagonal.}) (46. h5 $5 gxh5 $6 47. Rh1 $14) (46. Bf7) 46... Qxc5 47. a4 (47. Rb7 $5 {I felt was too risky, with the e-pawn so advanced, but the engine demonstrates that it's a draw, given mutually exposed kings. For example} Qxa3 48. Qc4 Re7 49. Rxe7 Qxe7 50. Rb1 $11 {the e-pawn will be captured if it advances, due to the reverse battery on the a6-f1 diagonal.}) 47... Bb4 48. Rgd1 {finally getting the rook back into play.} Rd2 49. Rxd2 Bxd2 {now time pressure really starts to be felt} 50. Rb7 f4 $2 {my opponent however by this point has even less time on the clock.} 51. g4 {I saw this would be safe.} (51. Qd3 $1 {this correct but risky move was beyond my ability to calculate at the time.} e2 52. Qxg6 $18 {wins, although it's scary.} Qf2+ 53. Kxf2 e1=Q+ 54. Kg2 Qxg3+ 55. Qxg3 fxg3 56. Kxg3 $18 {with a won endgame.}) 51... Bc1 {best guess here, due to time scramble handwriting} 52. Bb5 (52. Qa2 {the engine says is winning, but again risky and sacrificial.} e2 53. Rh7+ Kxh7 54. Qf7+ Kh8 55. Qxe8+ $18) 52... Rd8 53. Rd7 Rxd7 54. Bxd7 Qd4 {with the exchanges and opposite-colored bishops, now a draw seems inevitable.} 55. Bb5 Bd2 56. Bd3 g5 57. h5 Qxa4 {while I had missed in the time scramble that my opponent could pick up the pawn, as we'll soon see it does not matter, due to the blockade of e2 on the light squares.} 58. Bc4 Qc2 59. Bd3 Qc3 60. Bb5 Kg7 61. Qc4 {forcing the issue, as Black's king has no screening from checks.} Qxc4 62. Bxc4 Kf6 63. Kf1 Ke5 64. Ke2 Kd4 65. Ba6 {I offered a draw here and my opponent declined, although he had just seconds on his clock to my minute or so; there was a 5-second delay. After my bishop bounced back and forth on the diagonal several more times, I offered another draw and he accepted. This was all a bit strange for me, since the position was obviously blocked and he was the only one who could lose (on time).} 1/2-1/2

Annotated Game #295: No need to panic

Panic in the opening is never good, as this next tournament game miniature shows. Because there is more to it than just a one-move blunder (although not much more), I think it is worth publishing, as it highlights a useful gambit for Stonewall players to know, along with the unnecessarily deleterious psychological effects of panic. The unexpectedness of receiving two Black pairings in a row I believe contributed to my discombobulation, although that is not an excuse.

[Event "?"] [White "Expert"] [Black "ChessAdmin"] [Site "?"] [Round "?"] [Annotator "ChessAdmin/Dragon 3.2"] [Result "1-0"] [Date "????.??.??"] [ECO "D30"] [PlyCount "17"] [GameId "2099915250376808"] 1. d4 d5 2. c4 c6 3. Nf3 e6 {The move-order for a Slav Stonewall, although Black is not committed necessarily to that.} 4. e3 Bd6 (4... f5 {allows the provocative} 5. g4 $5) 5. Qc2 f5 {now we have the Stonewall formation.} 6. b3 Qe7 {this move-order is not necessary to prevent Ba3, a common idea for White, due to tactics. However, the queen will probably go to e7 anyway.} (6... Nf6 7. Ba3 $2 (7. Bd3 O-O 8. cxd5 {and either recapture is fine} exd5 9. Bxf5 Bxf5 10. Qxf5 Ne4 11. Qe6+ Kh8 12. Nbd2 Rf6 $11 {and White's queen cannot stop getting kicked around by Black's pieces.}) 7... Bxa3 8. Nxa3 Qa5+ $19) 7. Bd3 Nf6 8. cxd5 {here I panicked over the loss of a pawn, not recognizing that this is actually a standard Dutch gambit; see also the above variation with Bd3.} cxd5 $4 {saves the pawn, but loses the bishop. An example of failing to consider tactics in the opening phase.} (8... exd5 $1 9. Bxf5 Bxf5 10. Qxf5 O-O $11 {follwed by ...Ne4; this is a completely legitimate gambit.}) 9. Qxc8+ 1-0

Annotated Game #294: What the Caro-Kann is good for

This second-round tournament game shows off what the Caro-Kann is traditionally good for - a solid, semi-open game for Black that allows you to punish an overly-aggressive White player. This is exactly what happens during the game, featuring the Exchange Variation, although I over-complicate the final phase and miss forcing the win.

[Event "?"] [White "Class A"] [Black "ChessAdmin"] [Site "?"] [Round "?"] [Annotator "ChessAdmin/Dragon 3.2"] [Result "1/2-1/2"] [Date "????.??.??"] [ECO "B13"] [PlyCount "82"] [GameId "2085142363553942"] 1. e4 c6 2. d4 d5 3. exd5 cxd5 4. Bd3 Nc6 5. c3 Nf6 6. Bf4 g6 7. Nf3 Bg7 (7... Bf5 {is the main idea behind the ...g6 variation, although it is not obligatory.}) 8. Qc2 {this is not a logical place for the queen in the long term, with the diagonal blocked, although it does prevent ...Bf5.} Nh5 $11 {another key idea in this variation, chasing the Bf4.} 9. Be3 Qc7 10. Qb3 Nf4 {although the knight has moved multiple times, each time White is also forced to lose a tempo retreating, and the knight is at a better square. At this point, it is now best for White to exchange off the advanced knight.} 11. Bxf4 Qxf4 12. Nbd2 {the knight would be best on c3, but of course that is occupied.} Qd6 {with d5 pressured, this simple retreat is best.} 13. O-O O-O {the game is now very even.} 14. Qc2 {not very productive, going back to the blocked diagonal.} Bd7 {solid development.} (14... Rb8 {followed by a minority attack would be a good plan.} 15. a3 b5 $15) 15. Rfe1 Rfc8 {activating the rook and choosing the queenside for making future efforts.} 16. Qb3 e6 {again, solid play.} (16... Rab8 $5) 17. a4 {clearly interested in preventing the advance of my queenside pawns.} a5 {this is too optimistic. Now that the White a-pawn is advanced, I will not be expanding successfully on the queenside. There are several other calm possibilities.} (17... Rc7) (17... a6) (17... Rab8) 18. Qd1 Qc7 (18... e5 $5 {is now a possibility, but I had no particular desire to play an IQP position.} 19. dxe5 Nxe5 20. Nxe5 Bxe5 21. g3 h5 22. Bf1 $11) 19. Nf1 Re8 {more of a waiting move. Black has no real breakthroughs available, so playing neutrally is actually a good strategy.} 20. Ne3 Rab8 21. Ng4 Re7 22. Nge5 {White attempts to get something going on the kingside.} Nxe5 23. Nxe5 Bc6 24. Qf3 {White is slightly overpressing here, but I continued with the more solid option, after some thought.} Bxe5 (24... f6 $5 25. Ng4 Rf8 {shuts the center and kingside threats down thoroughly.}) 25. Rxe5 {the position is still completely even, but White can now think about attempting something on the h-file, combined with a bishop sac.} Kg7 26. Qg4 b5 {I thought for a while here and found the correct, active reaction.} 27. axb5 {White was very low on time here.} Bxb5 28. Bxb5 Rxb5 29. Qe2 $2 (29. Re2 $15) 29... Qb6 $19 {activity is again key, pressing against White's queenside weakness, while there is also a back rank problem for him.} (29... Qb8 {would be even better, allowing Black to triple up on the b-file with Alekhine's gun.}) 30. Ra2 Rb7 {Black is now winning.} 31. c4 {an active defense that allows me to make a bad choice.} dxc4 {played after some thought, with some time pressure on my side. This should still be enough to win.} (31... Rxb2 $1 {simplest is best.} 32. Rxb2 Qxb2 $19 {with a continuation likely such as} 33. h3 {needed luft for the king} (33. Qxb2 Rxb2 34. Kf1 a4 {and the outside passed pawn decides it.}) 33... Qxe2 34. Rxe2 dxc4) 32. Qxc4 Rb4 {still maintaining threats to the b- and d-pawns.} (32... Rc7 {I hallucinated that this would not work, although the line was admittedly difficult to visualize.} 33. Rc5 Rcxc5 34. dxc5 Rxc5 35. Qd4+ Kg8 36. Qd2 Qb4 $19) 33. Qc3 Rc7 $2 {played one move too late; I missed White's next, as well.} (33... Rxb2 $1 {simplest is again best.} 34. Rxb2 Qxb2 35. Qxb2 Rxb2 {and the back-rank mate threat allows Black to win.} 36. g3 a4 37. Ra5 Rb4 $19) 34. Rc5 $11 Rd7 35. h3 Rbxd4 36. Ra4 {now White eschews simplicity, but I miss the chance to complicate.} (36. Raxa5) 36... Qd6 (36... Qxc5 $5 37. Qxc5 Rxa4 38. Qe5+ Kf8 $17 {and Black has whatever winning chances are in the position.}) 37. Rxd4 Qxd4 38. Rxa5 Qxc3 39. bxc3 Rd1+ 40. Kh2 Rd2 41. Kg1 Rd1+ 1/2-1/2