Annotated Game #213: Charting your own path

During this next tournament game, my opponent and I chart our own path early on, outside of what is covered in opening books. But it's not really new ground, once you start looking in the databases. In what is ostensibly a rather offbeat sideline of a Caro-Kann Two Knights, you can find super-GM level games by Carlsen and Topalov on the Black side, which are given below. (And the next Commentary game to be posted will feature a very recent game in the same line, from the 2019 Women's Candidates tournament.)

These days, especially with Carlsen as a model, players seem less obligated to try to duel for a theoretical advantage in main lines, although there's still a lot of opening theory that continues to evolve. I used to have an unhelpful attitude towards opening "deviations", thinking that they should always be punished. Now, I think it's more important to know the key elements of an opening position, both static and dynamic, which will then be your guide - regardless of whether the line you're in is popular or even known. Analyzing your own games when you enter unfamiliar territory is always a good learning experience, since both your knowledge base and insights should grow as a result.

In this game, the main insight for me from Carlsen's different choice on move 5 is how to take advantage of White's Qe2 blocking the standard bishop development. Later on, the queen's early sally on the kingside also offers opportunities for Black on the queenside. I decide to follow a more aggressive plan with opposite-side castling, which offered the clear idea of advancing pawns on the kingside to pressure White. This isn't done in the most effective way, and I also miss some key ideas repeatedly (...Qd5!?), which I'll remember for the future. There's a lot of back-and-forth and White was dangerous in the endgame, but I finally managed to get a draw. My opponent was rated about 100 points above me, so not a bad outcome of an interesting game.

[Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "Class B"] [Black "ChessAdmin"] [Result "1/2-1/2"] [ECO "B11"] [Annotator "ChessAdmin/Komodo 11.2"] [PlyCount "108"] {[%mdl 8192] B11: Caro-Kann: Two Knights Variation} 1. e4 c6 2. Nf3 {this is not in any book variation of the Caro-Kann, but is a legitimate if offbeat choice by White. Here there is a quick transposition into a Two Knights Variation.} d5 3. Nc3 dxe4 4. Nxe4 Nf6 {this offers another transposition into a main line Caro-Kann variation, if White takes on f6.} 5. Qe2 {a surprise here, but it's actually the third most popular choice in the database. Black players always need to be aware of the potential threat of a discovered pin by White's queen on the e-pawn after an early Qe2 is played.} Nxe4 {while exchanging is an obvious move, this is actually not the choice of top-rated Black players. Here's a (very!) high-level illustrative game, with Magnus Carlsen as Black:} (5... Bf5 6. Nxf6+ gxf6 7. d3 Nd7 8. g3 Ne5 9. Nxe5 Qa5+ 10. Bd2 Qxe5 11. Qxe5 fxe5 12. Bg2 h5 13. O-O h4 14. Rae1 f6 15. f4 hxg3 16. hxg3 exf4 17. Rxf4 Bh6 18. Rh4 Bg7 19. Rb4 Bc8 20. d4 Bf8 21. Rc4 Bh3 22. Bf3 Rd8 23. d5 cxd5 24. Rc7 Rd7 25. Rxd7 Bxd7 26. Bxd5 b6 27. c4 e5 28. b4 Ke7 29. c5 Be6 30. Bf3 Bh6 31. Bxh6 Rxh6 32. cxb6 axb6 33. a4 Rh7 34. Rc1 f5 35. Rc6 Bd7 36. Rxb6 Bxa4 37. b5 Rh8 38. Bd5 Rc8 39. Re6+ Kd7 40. Rxe5 Bxb5 41. Rxf5 Rc1+ 42. Kg2 Kd6 43. Be4 Bd7 44. Rf2 Ke5 45. Bf3 Bf5 46. g4 Rc2 47. gxf5 Rxf2+ 48. Kxf2 Kxf5 {1/2-1/2 (48) Vachier Lagrave,M (2783)-Carlsen,M (2851) Leuven 2017}) 6. Qxe4 {now White's queen is no longer blocking the development of the light-squared bishop, which is why immediately exchanging is not the preferred choice.} Nd7 {a standard developing move, but Black can also immediately challenge the centralized queen:} (6... Qd5 7. Qh4 Qe6+ 8. Be2 Qg4 9. Qg3 Qxg3 10. hxg3 g6 11. d4 Bg7 12. Bh6 Bf6 13. Ne5 Be6 14. O-O-O Nd7 15. f4 Rg8 16. g4 Rd8 17. c4 Bxe5 18. fxe5 Nb6 19. b3 g5 20. Kc2 f6 21. exf6 exf6 22. Kc3 Kf7 23. Rdf1 Rg6 24. Bd3 Bxg4 25. Bxg6+ Kxg6 26. Re1 Nc8 27. c5 Rg8 28. a4 a5 29. b4 axb4+ 30. Kxb4 Rd8 31. Kc3 b6 32. Kc4 Bf5 33. Re3 Bc2 34. Bxg5 Kxg5 35. Rg3+ Kf5 36. Rxh7 Bxa4 37. Rh5+ Ke6 38. Re3+ Kf7 39. Rh7+ Kg6 40. Rc7 Bb5+ 41. Kc3 bxc5 42. dxc5 Kf5 43. g3 Kg6 44. Kb4 Rd4+ 45. Kc3 Rd8 46. Kb4 Rd4+ {1/2-1/2 (46) Bacrot,E (2708)-Topalov,V (2749) Paris 2017}) 7. d4 Nf6 8. Qh4 Bf5 { targeting the weak c-pawn and also preventing the usual development of the White bishop to d3. Without the queen's presence on d1 the doubled d-pawns that would be inflicting on White after a bishop exchange would be a serious weakness.} 9. c3 e6 10. Be2 Be7 11. Bg5 h6 {although this is not actually an immediate threat to the Bg5, because the unprotected Rh8 prevents Black from capturing on g5, it still puts additional latent pressure on White. At least that was my thinking.} (11... Qb6 $5 {is favored by the engines, as Black is in a good position to take advantage of the lack of queenside defenders. For example} 12. b3 Qa5 13. Bd2 Ne4 14. Qf4 Nxc3 $17) 12. Rd1 $146 {it wasn't clear to me what the rook is doing on the d-file. Although the rook is lined up against the queen, the d-pawn in front of it is not going anywhere.} Nd5 { an overly passive approach, aiming for piece exchanges and equality.} (12... Qd5 {is a thematic seizure of the center by Black's queen, and it can't be chased away easily.} 13. a3 (13. c4 $2 Qa5+ $19) 13... O-O $15) (12... O-O { immediately is also good, also essentially forcing the exchange on f6.}) 13. Bxe7 $11 {forced} Qxe7 14. Qg3 {here I thought for a while and decided to take a more aggressive path by castling queenside, since White had declined the queen trade.} (14. Qxe7+ Kxe7 $11) 14... O-O-O {with the dark-squared bishops gone, Black's king is secure on the queenside, allowing kingside expansion.} 15. O-O (15. Qxg7 $2 Rdg8 16. Qe5 f6 {trapping the queen.}) 15... g5 { advancing and protecting the pawn at the same time. Black has only a slightly better position, but it's easier to play and my plan is clear, to do everything I can to try and break through on the kingside.} 16. Rfe1 Nf4 { here I advance my own plan, but allow White to get in a relatively more impactful move.} (16... f6 $5 {takes away the e5 square.}) 17. Ne5 h5 18. Nd3 ( 18. Bd3 Nxd3 19. Nxd3 Rhg8 $11) 18... Nxe2+ {here I should maintain the tension on the kingside, since I have some initiative, rather than help White relieve it.} (18... h4 19. Qf3 Nd5 $15 {here the Be2 is bottled up by White's other pieces and the Re1 is also blocked by it.}) 19. Rxe2 h4 {now this move has less impact.} (19... f6 20. Qe3 $11) 20. Qf3 Qd6 {moving the queen away from the e-file, both to get on the h2-b8 diagonal and to get off the e-file.} 21. Ne5 Rdf8 22. Rde1 f6 {at the time, I judged this to be weakening but not too much so, with that outweighed by the benefit of kicking White's well-placed knight.} (22... Qd5 $5 {is a more solid approach, but I was still thinking more aggressively about a kingside attack.} 23. Qxd5 cxd5 24. h3 $11) 23. Nc4 Qd7 24. Re3 {I thought this was a wasted move.} (24. h3 {would have prevented my next idea.}) 24... g4 25. Qf4 Rd8 {addressing in a simple manner the new threat of Nd6+ by adding to the protection of the d6 square.} (25... Qc7 {also is fine, but is much more complicated, because if} 26. Nd6+ $2 (26. Qxc7+ $11) 26... Kd7 27. Qxf5 exf5 28. Re7+ Kxd6 29. R1e6+ Kd5 30. c4+ Kxc4 31. Rxc7 Re8 32. Rce7 Rxe7 33. Rxe7 {and Black is a pawn up in the rook ending.}) 26. a4 {now White shows interest in getting his own pawns going against my king.} h3 {I thought for a long time here, since it wasn't clear to me how best to continue on the kingside. I don't in fact have any breakthrough possibilities, though.} (26... Rh5 $11 {is suggested by the engine, with the point that the rook can now move along the 5th rank to good effect, prior to committing with ...h3.}) (26... Qc7 {admitting that the position is even would also be a solid approach.}) 27. Qg3 $6 {other reasonable moves by White lead to equality. The text now allows me to open the h-file and get attacking chances.} (27. Rg3 $5 $14 {must definitely be considered}) 27... hxg2 $15 28. Qxg2 Rh3 {another significant think here, as there were several good-looking options. Naturally I'd like to double on the h-file.} 29. Rxh3 $2 (29. a5 $5) 29... gxh3 $19 {now I have a real advantage, but can't figure out the best way to proceed.} 30. Qg3 {forced} Qd5 $2 {this one-move threat against the Nc4 does nothing for me.} (30... e5 {this pawn lever is the key, although it is not easy to see the consequences.} 31. dxe5 (31. Ne3 Be6 32. Qh4 f5 $19) 31... Qe6 $1 {a subtle move that pins White's e-pawn, attacks the Nc4 and simultaneously threatens Rg8.} 32. Qf3 Qg8+ 33. Kh1 Bg4 34. Qf4 Be6 35. Ne3 fxe5 $19) (30... Qh7 {would also be good, again now enabling Rg8.}) 31. Ne3 $11 {now I'm forced into an awkward sequence to maintain equality.} Qe4 32. f3 Qd3 33. Nxf5 Qxf5 34. Re4 Rh8 {here I was thinking in too much of a static defensive fashion. Activity and counter-threats are better to pursue here.} ( 34... c5 $5) (34... Qh7 {also is good, because if} 35. Rh4 $2 Qb1+ 36. Kf2 Qxb2+ 37. Kf1 Qxc3 $19) 35. Qg4 (35. Kf2 Qh7 $11) 35... Qxg4+ {at this point I judged that I would be better off in a rook endgame.} (35... Rh5 $5) 36. Rxg4 { the rook endgame looks generally balanced, although my advanced h-pawn is a weakness.} Rh7 (36... Kc7 37. Rg7+ Kd6 38. Rxb7 Rg8+ 39. Kh1 Rg2 $11) 37. Kf2 Kd7 38. Kg3 Kd6 39. Rh4 Rg7+ 40. Kxh3 e5 {not a good choice.} (40... f5 41. a5 $14) 41. dxe5+ Kxe5 42. Rg4 $16 {now exchanging would give White a won pawn ending.} Rh7+ 43. Kg3 Kf5 (43... f5 44. Rg5 $14) 44. b4 Re7 45. h4 {passed pawns must be pushed!} Rh7 {I'm trying to protect the 7th rank and stop the h-pawn at the same time.} 46. c4 (46. a5 {would put more pressure on the queenside.}) 46... Ke5 {at least my king is centralized and fighting.} 47. Kf2 $2 {this allows me to force the rook away from protecting the h-pawn.} (47. Re4+ Kf5 48. a5 $16) 47... f5 $11 {now my opponent realized what he had done.} 48. Rg8 Rxh4 $15 49. Rg7 Rxc4 {this rushed move is not optimal, as it would have been better to preserve the two queenside pawns together.} (49... b6 $5 50. Rxa7 Rxc4 51. b5 cxb5 52. axb5 Rb4 53. Rb7 Rxb5 $15 {technically speaking this should still be a draw with best play, but I'd rather be Black.}) 50. Rxb7 $11 {now we're back to equality.} a5 51. b5 (51. bxa5 Rxa4 52. Ra7 Kd4 $11) 51... Kf4 {here I play it safe.} (51... c5 $5 52. Rc7 Kd6 53. Rc6+ Kd5 54. Rf6 f4 55. b6 Rxa4 56. b7 Rb4 57. Rxf4 Rxb7 $17) 52. bxc6 Rxa4 53. Rb5 Rc4 54. Rxa5 Rxc6 1/2-1/2