[Event "Larry Evans Memorial"] [Site "https://lichess.org/study/F73s9jnE/JsKEB2jH"] [Date "2023.04.09"] [Round "6"] [White "Tarjan, Jim"] [Black "Atwell, Rose"] [Result "1-0"] [WhiteElo "2448"] [BlackElo "2203"] [Annotator "GM Jim Tarjan"] [Variant "Standard"] [ECO "D31"] [Opening "Queen's Gambit Declined: Charousek Variation"] [Source "https://lichess.org/study/F73s9jnE/JsKEB2jH"] [Orientation "white"] { Annotations by GM Jim Tarjan } 1. c4 e6 2. d4 d5 3. Nc3 Be7 4. cxd5 exd5 5. Bf4 c6 6. e3 Bd6 { As she has played in this position before. } 7. Bxd6 Qxd6 8. Bd3 Nf6 { Straightforward, and what I have played in this position. But the top players have used a different plan: } (8... Ne7 { intending ... Bc8-f5, and if } 9. Qc2 { then } 9... b6 { , successfully trading off the bishop via the a6-square. One can argue that the slight weakening of Black's queenside pawns means White at least has something to play for. }) 9. Nge2 O-O { [#] } 10. Qd2 { Believe it or not, this seems to be new, as everyone plays 10. 0-0, including Sorokin against me at the recent North American Open. With 10. Qd2, I am telegraphing my intention to repeat Sorokin's plan of a quick f2-f3 and e3-e4. I am also feinting the possibility of queenside castling (would that be so bad?). I suspect that Rose's next move is an anticipatory response to that possible 0-0-0. } 10... a5 { Perhaps Black could try } (10... b6 { to trade off the bishops with ... Bc8-a6, but with Black's knight on f6, White's plan of f2-f3 and e3-e4 still may have some bite. }) (10... Re8 11. O-O b6 { was Sorokin – Tarjan (by transposition). } (11... Nbd7 { Later I decided Black should have played this move. Then, if } 12. f3 { (White could also wait a bit on this), Black can play } 12... c5 { , hoping the weakened e-pawn gives Black some counterplay. }) 12. f3 { and, faced with an impending e3-e4, I switched plans with } 12... c5 { but this is definitely not right. }) 11. f3 c5 { Atwell correctly realizes that Black needs to get this in, but taken together, the last two moves don't make a good impression } 12. Nb5? { A small but important oversight. Very obviously I should have traded pawns right away: } (12. dxc5 Qxc5 { and Black will lose time with her misplaced queen. But I overthought, figuring I could do even better by gaining Nc3-b5 first. }) 12... Qe7 13. dxc5 Nbd7 { So the proper piece gets to c5. And now even ... a7-a5 makes sense in this position, cementing the c5-knight. Even so, I would rather play the white side, but it is only a normal, small edge. } 14. Ned4 Nxc5 15. O-O Re8 { This creates a problem with Black's normal development because of the possible Nb5-c7. I would snap the bishop off and develop: } (15... Nxd3 16. Qxd3 Bd7 { or instead just }) (15... Bd7 { , but my sense is it's right to take off White's bishop. There's something here about the two remaining white knights' duplicating effort. }) 16. Rfe1 b6 { So Black has to develop the bishop somewhere other than d7, though this ... b7-b6 will probably be played in any case. } 17. Bf1 { Given the chance to save this bishop, I will take it! } 17... Bb7 { I was expecting } (17... Ba6 { . }) 18. Nf5 { These are very subtle points, but just as Black should have taken my bishop when she had the chance, I make sure to take hers — even if the bishop cosmetically looks "bad." } 18... Qd7 19. Nfd6 Re7 20. Nxb7 Qxb7 21. Rad1 Rd8 { The rook is prone to pins on the d-file and perhaps should have stayed elsewhere, if possible. } 22. Qf2 { An ineffective attempt to improve the position of my queen. Another idea would be } (22. e4!? { already, while another plan in the position is }) (22. a3 { to drive away the c5-knight with b2-b4. }) 22... Ne6 23. Qg3 Red7 { May I confess I was hoping she would trade queens with } (23... Qb8 { , thinking the old man would have better chances of beating this talented junior without queens on? Didn't turn out that way, actually. }) 24. Qf2 Ne8 { Black has improved the position of her pieces, and now she wants to trade off the b5-knight for the less useful f6-knight, while leaving her remaining knight well-placed on e6 or c5. This is a decision that shows Atwell's talent. } 25. Rd2 Nd6 26. Red1 Nc4 { Perhaps it's better to just take the opportunity to trade: } (26... Nxb5 27. Bxb5 Rd6 { but then there is } 28. e4 { . }) 27. Rc2 { Naturally Black saw } (27. Bxc4 dxc4 28. Nd6 Qc7 29. Nxc4? Rxd2 30. Nxd2 Qc2 { and wins. }) 27... Ne5 { The knight looks good in the center, but I didn't see its prospects from this square. I think she should wait, or go back and trade on b5. For example, } (27... g6 { seems useful to me, making luft in a way that prevents a white piece from using the f5-square. }) 28. b3 { Not a move I necessarily wanted to make, but I did want to keep that knight out of c4. } 28... f5?! { She is trying to stop e3-e4, and also do things herself, such as a ... f5-f4 push. But I was happy to see this move as the opening of the kingside pawns may backfire, especially with all the major pieces still on the board. } 29. Nd4 Nxd4 30. exd4! { Time to switch plans, and in particular, to drive Black's knight out of the e5-square. } 30... Nf7 31. Rdc1 Nd6 32. Qg3 { The computer wants to play } (32. Rc6 { here, and I was looking at it move by move. You have to make sure Black never has an effective } 32... Nc4 { , and, at this particular moment, White has } 33. Re6! { with a great advantage. }) 32... Re7 33. Qf4 g6 34. h4 Qd7 35. h5 Rc8 { The computer criticizes this, but Black is clearly not happy and at some point has to try to trade off one of those doubled rooks. } 36. Rxc8+ Nxc8 37. h6 { And in any case her moves worked as here I choose the wrong plan. It turns out that the correct path was } (37. hxg6 hxg6 38. Bb5! Qe6 { and here I didn't see how to go further, as the black queen gets active. The computer points out that White has } (38... Qxb5 { is the other move White had to have a satisfactory reply to. White must first find } 39. Rxc8+ Re8 40. Qd6! Rxc8 41. Qxg6+ Kh8 42. Qh6+ Kg8 43. Qe6+ Kg7 44. Qxc8 { and then figure to win this endgame. }) 39. Rc6! Qe1+ 40. Bf1 { and Black can resign. }) 37... Nd6 { The computer gives White only a slight edge now, but most evidently that is not a realistic assessment in a game between humans. Black remains under pressure. } 38. a4 { I had this idea in mind all along, to finally get the bishop going with Bf1-b5. } 38... Kf7 39. Kh2 Kf6 { A sign of a rather good player, it seems to me, realizing that the king belongs on f6. But here perhaps } (39... Qd8 { is more comfortable, anticipating White's next move. }) 40. Bb5 Nxb5 { The infamous 40th move. After } (40... Qd8 { the computer blithely calls it equal, but, again, easy for a computer to say. With accurate moves, Black holds on; for example } 41. Rc6 Re6 42. g4 (42. Qc1 g5) 42... g5) 41. axb5 Re6 42. Rc7 Qd6? { Despite plenty of time on the clock now, this is the losing move. To be fair, Black can only hold because of a nice resource, and that "all rook endgames are drawn:" } (42... g5! 43. Rxd7 (43. Qg3 { and only then } 43... Qd6 (43... Qe8)) 43... gxf4 44. Rxh7 Kg6 { and Black should hold this. A likely continuation is } 45. Rd7 (45. Rh8 Re3 46. Rb8 Re6!) 45... Kxh6 46. Rxd5 Kg5 { and with bad pawns, bad rook, and a bad king, this rook endgame is certainly drawn. }) 43. Qxd6 Rxd6 44. f4! { This ends it. The weaker } (44. Rxh7 Kg5 { is more of a fight. }) 44... g5 { Hopeless, but so is everything else. } (44... Rd8 45. Rb7! { most accurate } 45... Rd6 46. Rxh7 g5 47. Rc7 { and Rc6 }) 45. Rc6 { 1-0 White wins. } 1-0