[Event "WCh 2023"] [Site "Astana KAZ"] [Date "2023.04.26"] [Round "12.1"] [White "Ding, Liren"] [Black "Nepomniachtchi, Ian"] [Result "1-0"] [ECO "D04"] [WhiteElo "2788"] [BlackElo "2795"] [Annotator "Monokroussos,Dennis"] [PlyCount "75"] [EventDate "2023.04.09"] 1. d4 (1. e4 c6 2. d4 d5 3. exd5 cxd5 4. Bd3 Nc6 5. c3 Nf6 6. Bf4 Bg4 7. Qb3 Qd7 8. Nd2 e6 9. Ngf3) 1... Nf6 2. Nf3 d5 3. e3 {For the first time in the match.} c5 4. Nbd2 cxd4 5. exd4 Qc7 6. c3 Bd7 $146 {What is this all about? I'm afraid I haven't a clue. The only thought that came to mind was that it was aimed against lines where White plays Bb5 or Qa4, but those options are just as available a couple of moves later, after Black plays ...Bg4.} (6... Nc6 {is normal, and after} 7. Bd3 (7. Bb5 {is possible but nothing special.} a6 8. Bxc6+ bxc6 9. O-O Bf5 10. Ne5 e6 11. Qa4 a5 $1 12. Nxc6 Bd6 $1 13. Ne5+ Nd7 $44) 7... Bg4 8. O-O e6 $11 {This looks like an improved version (for Black) of an Exchange Caro-Kann - see the note on move 1.}) 7. Bd3 (7. Ne5 $142 {looks good, and unless there's some deep reason why it's not good it looks like just one more argument against repeating 6...Bd7.}) 7... Nc6 8. O-O Bg4 {Hmm. If 6...Bd7 was about neutralizing Bb5 and Qa4 ideas, well, White can play either move now. Yes, White's bishop is going there in two moves rather than one, but the same is true of Black's bishop getting to g4. It made no difference. And as for 9.Qa4, White didn't lose any time at all; in fact, it's probably White's best move.} 9. Re1 (9. Qa4 $142 $14) 9... e6 10. Nf1 Bd6 11. Bg5 $146 (11. a4 O-O-O 12. Ng3 h5 13. h3 Bxg3 14. fxg3 Qxg3 15. Re3 Ne4 16. Qf1 Bf5 17. Ne5 Qh4 18. Nxf7 Rhf8 19. Nxd8 Qxd8 20. Rf3 h4 21. Bf4 g5 22. Be3 Rh8 23. Rxf5 Ng3 24. Qf3 exf5 25. Bxf5+ Kb8 26. Bg4 Rf8 27. Qd1 Qf6 28. Qd3 Ne7 29. Bf3 Ng6 30. Bxd5 Nf4 31. Bxf4+ Qxf4 32. Bf3 Re8 33. Qb5 Re7 34. Qd5 Qe3+ 35. Kh2 a6 36. Qd6+ Ka7 37. b4 Qe1 38. Qc5+ Kb8 39. Rxe1 {1-0 Chau,Q-Cao,S VIE-ch Dalat City 2004 (7)}) 11... O-O $1 {The right move. Nepo is correctly unafraid of the weakening of his kingside, and realizes that if anyone is going to get attacking chances after the swap on f6, it's him.} 12. Bxf6 gxf6 13. Ng3 {[%CAl Bh2h3]} (13. Ne3 {is playable (and arguably better), despite the seemingly loose pawn on h2.} Bh5 {is forced and fine.} (13... Bxh2+ $4 14. Kh1 Bxf3 (14... Bh5 15. g4 $18) 15. Qxf3 Bd6 16. Nxd5 ({The boring} 16. Qxf6 $18 {is winning, but there are better moves, including the flashy 16.Nxd5.}) 16... exd5 17. Qf5 ({Or} 17. Qg4+ Kh8 18. Qf5 {and 19.Qxh7#.}) 17... Rfe8 18. Qxh7+ Kf8 19. Qh8#) 14. g3 Kh8 15. Be2 $11) 13... f5 14. h3 Bxf3 15. Qxf3 {White may be threatening to take on f5, but Black has plenty of good ways to neutralize the threat.} Ne7 {Taking Nxf5 ideas off the table.} (15... Kh8 {[%CAl Bf8g8]} 16. Nh5 (16. Nxf5 exf5 17. Qxf5 f6 18. Re6 {Threatening 19.Rxd6, taking advantage of Black's overloaded queen.} Rad8 19. Rxf6 Rxf6 20. Qxf6+ Qg7 $11 {This is probably more dangerous for White.}) 16... Rg8 (16... Ne7 {transposes to the game.}) 17. Kf1 (17. Kh1 Rg6 18. g4 fxg4 19. hxg4 f5 20. gxf5 Qf7 $1 {[%CAl Bg6h6]} 21. Rg1 Rh6 22. Be2 exf5 $15) 17... Rg6 18. g4 Rag8 19. Ke2 (19. gxf5 $2 Rh6 $17 {[%CAl Bg8g5]}) 19... Rh6 20. Rg1 Rg5 $11) 16. Nh5 Kh8 17. g4 Rg8 18. Kh1 Ng6 $1 ({Doubling rooks is playable, but inferior to the text.} 18... Rg5 19. Rg1 Rag8 20. Nf6 R8g6 21. Ne8 {If it weren't for this, Black would be better.} Qd8 22. Nxd6 Qxd6 23. gxf5 Nxf5 (23... exf5 24. Rxg5 Rxg5 25. Rg1 Qh6 26. Qe3 f4 27. Qf3 Ng6 28. Kh2 $14) 24. Rxg5 Rxg5 25. Rg1 Rxg1+ 26. Kxg1 Qd7 27. Bxf5 exf5 28. Qf4 $14 {Maybe this ending is drawn with correct play, but it's not pleasant for Black.}) 19. Bc2 $2 (19. gxf5 $2 Nh4 {Looking to play ...Rg2.} 20. Qe3 exf5 {Now threatening both ...f4 and ...Ng2. The former is the greater threat, so White will have to allow the fork.} (20... Rg2 $2 21. f4 $11) 21. f4 (21. Rg1 $2 f4 {Intending ...Qd7 and ...Qxh3#.} 22. Qe2 f3 23. Rxg8+ (23. Qf1 Rg2 $19) 23... Rxg8 24. Qf1 Qc8 $19 {Black will drop the rook or the knight into g2, and give mate on h3. White can delay mate for a while, but only at an unsurvivable cost in material.}) 21... Ng2 $19) (19. Rg1 $1 {was correct.} Bh2 20. Rge1 (20. Rg2 $4 Nh4 $19) 20... Bd6 $1 (20... Nh4 21. Qe3 {Now it's White who threatens f4, so Black's bishop must retreat.} Bd6 (21... f4 $4 22. Qe5+ $18 {[%csl Oh2]}) 22. f4 $11 {/?}) 21. Rg1 Bh2 $11 {may be best play.}) 19... Nh4 $17 20. Qe3 Rg6 (20... f4 $17 {is also strong.}) 21. Rg1 f4 $1 22. Qd3 Qe7 23. Rae1 Qg5 24. c4 $5 {Not the best move according to the computer, but arguably the best move for a human. If White tries to hold the kingside in a "polite" and well-behaved way, Nepo will sooner or later figure out a way through. So Ding opens the board in search of some counterplay, both on the queenside and along the a1-h8 diagonal, and he eventually gets it.} dxc4 $19 25. Qc3 {[%CAl Bd4d5,Bc2g6]} (25. Qxc4 Nf3 $19) 25... b5 26. a4 (26. Bxg6 hxg6 27. d5+ e5 $19) 26... b4 {Objectively, this is at least as good as 26...a6. But as a practical matter, why change the status quo on the queenside? If White has no ideas, why change the position? He might find some ideas in the new situation.} (26... a6 27. b3 Rc8 $19) 27. Qxc4 {Nepo had plenty of time left on the clock at this point - 38:31. Unfortunately, from this point on he starts making bad moves, and quickly. There are two points to 26...b4. One is to chase the queen off the long diagonal, and the other is to chase her away from the defense of the f3 square. So ...Nf3 is the critical move, and indeed, the only move that keeps a winning advantage here. That said, proving the win takes a fairly long calculation. The line is largely forced, so it wasn't out of the question that Nepo could work it out. If he didn't see it, though, then that's again an argument for 26...a6.} Rag8 $2 {Played after 27 seconds. Insane.} (27... Nf3 $1 $19 28. Qc6 (28. Bxg6 Qh4 $1 (28... hxg6 $2 29. Qc6 Qd5 30. Qxd5 exd5 31. Nf6 $11) 29. Kg2 Nxe1+ 30. Rxe1 hxg6 31. Qc6 Rd8 $19 {The knight is lost, and with it the game.}) (28. Be4 Qh4 29. Kg2 Rxg4+ 30. hxg4 Nd2 31. Qe2 f3+ 32. Bxf3 Qh2#) 28... Nxe1 29. Qxa8+ Rg8 30. Qe4 Nxc2 31. Qxc2 Qh4 32. Qd3 (32. Kg2 $4 Qxh5 $19) 32... f5 (32... Qxf2 $2 33. Nf6 Rg6 $2 34. Ne4 $18) 33. Qf3 {It looks like White has everything glued together, but there's one final point.} fxg4 34. Rxg4 Rxg4 35. Qxg4 Qxg4 36. hxg4 {There's no way for Black to trap White's knight. But there's something even better.} b3 $1 $19 {Otherwise White plays b3, and it's equal. Now, though, there's no (good) answer to ...Ba3, and the b-pawn promotes.}) 28. Qc6 $2 {A mistake right back, after less than two minutes' thought. Ding still had a bit over half an hour on the clock, so while there isn't time to take a nap and watch an episode of his favorite TV show, he can spend more than a couple of minutes on this critical position.} (28. Bxg6 $1 hxg6 29. Qc6 $1 gxh5 (29... Bb8 $2 30. d5 $1 {Now White is even winning.} gxh5 (30... Qxd5+ 31. Qxd5 exd5 32. Nf6 Rf8 33. Re7 $18) 31. dxe6 fxe6 32. gxh5 $18) 30. Qxd6 Nf3 31. Qc5 $1 {Maybe both players missed this resource?} hxg4 32. Qxg5 Rxg5 33. hxg4 Nxg1 34. Rxg1 Rd5 35. Rc1 $1 Rxd4 36. Kg2 $11 {/? White's rook will get active, and he will hold a draw despite his pawn deficit - easily.}) 28... Bb8 $4 {After less than 90 seconds, Nepo plays a move that takes him from winning to losing.} (28... Nf5 $1 29. Rd1 {Best. White is *almost* okay here, but not quite.} (29. gxf5 $2 {allows Black to play for mate in two different ways.} Qxh5 (29... Qxg1+ 30. Rxg1 Rxg1+ 31. Kh2 f3+ 32. Ng3 Rg2+ 33. Kh1 Bxg3 34. Qxf3 (34. fxg3 R2xg3 35. Kh2 Rg2+ 36. Kh1 Rg1+ 37. Kh2 R8g2#) 34... Rxf2 35. Qd3 Bf4 $19) 30. Rxg6 Qxh3+ 31. Kg1 fxg6 32. f6 f3 $19) (29. Bxf5 {is a better try, though it's still inadequate.} exf5 30. Re8 Qh4 31. Rxg8+ Rxg8 32. Qf3 {comes close, but it loses - as we've already seen.} (32. Rg2 Qxh3+ 33. Kg1 Rd8 $1 (33... f3 $4 34. Rg3 $1 $18 Bxg3 35. Qf6+ $18) 34. Nxf4 Qb3 {Threatening 35...Qd1+ 36.Kh2 Bxf4+.} 35. Qc1 Qf3 36. Nh5 fxg4 $19 {is winning, and if White tries for a little counterplay with} 37. Qh6 $4 {he loses on the spot.} Qd1#) 32... fxg4 33. Rxg4 Rxg4 34. Qxg4 Qxg4 35. hxg4 b3 $1 $19) 29... Qh4 $1 30. Qf3 Ng3+ $1 31. Rxg3 (31. Kg2 Nxh5 32. Bxg6 Rxg6 33. Kf1 Nf6 $19) 31... fxg3 32. Bxg6 (32. Kg2 f5 $1 33. fxg3 fxg4 $1 $19) 32... Qxh3+ 33. Kg1 gxf2+ 34. Kxf2 Qxf3+ 35. Kxf3 fxg6 $1 36. Ng3 (36. Nf6 Rf8 {is why Black recaptures with the f-pawn on move 35.} 37. g5 Be7 $19) 36... Rf8+ 37. Kg2 Rf4 38. d5 exd5 39. Rxd5 Bc7 40. Rd7 Be5 41. Ne2 Rxg4+ 42. Kf3 Rg5 43. Rxa7 Bxb2 $19) 29. Qb7 $4 $17 {/-+ After five minutes, Ding returns the favor, though he's not as badly off as he was two half-moves ago.} (29. Bxg6 $1 hxg6 (29... fxg6 30. Rxe6 gxh5 31. Re8 $18) 30. d5 $1 $18 {wins for White, as seen above.}) 29... Rh6 $2 $11 {Played after a minute.} (29... Nf5 $1 30. Rd1 (30. gxf5 $4 Qxh5 31. Rxg6 (31. fxg6 $4 Qxh3#) 31... Qxh3+ 32. Kg1 hxg6 33. f6 f3 $19) 30... a5 $17 {/-+}) 30. Be4 Rf8 $2 {Nepo spent 10 minutes on this move. It's a mistake, though less damaging than his last three moves.} (30... f3 $1 {was the best move, hinting at possibilities like ...Rxh5 followed by ...Qf4.} 31. Qxf7 Ng2 $1 {[%CAl Bg5h4] Not primarily about the rook on e1, though Black is threatening to take it, because the bigger threat is ...Qh4 followed by mate on h3.} 32. Qxf3 Nxe1 33. Rxe1 Qh4 $11 {It seems as if both sides are at a standstill.}) 31. Qxb4 $16 Qd8 32. Qc3 (32. d5 $142 e5 33. Rc1 $16) 32... Ng6 $14 33. Bg2 (33. Re2 $142) 33... Qh4 $6 (33... Bd6 $142 $1 $11 {Next up, ...Qe7, to prevent White from mobilizing his queenside majority.}) 34. Re2 $14 {/? Black has largely minimized the damage over the past few moves (though the last move hurt a little), and while White is a little better it's a three-results position. Anything can happen. Unfortunately for Nepo, "anything" does happen, and in the worst way.} f5 $4 {Apparently Nepo completely overlooked White's next move, or just somehow assumed it was impossible.} (34... a5) (34... Rg8) 35. Rxe6 $18 {Everything else loses, but this is *completely* winning for Ding.} Rxh5 (35... fxg4 {goes down without a fight.} 36. d5+ Be5 (36... Kg8 37. Qg7#) 37. Rxe5 Rxh5 38. Rxh5+ Qf6 39. Rf5 Qxc3 40. Rxf8+ Nxf8 41. bxc3 $18 {might not be winning if Black gets away with surreptitiously removing White's rook from the board. Sticking to the realm of the legal and ethical, however, it's over.}) (35... f3 36. d5+ Be5 37. Rxe5 $1 fxg2+ 38. Kxg2 Nxe5 39. Qxe5+ Rhf6 40. Nxf6 Rxf6 41. g5 Qxg5+ 42. Kh2 Qh6 (42... Qh4 43. Qe8+) 43. d6 Qf8 44. Rd1 Qd8 45. d7 Kg7 46. Rd6 $18) 36. gxh5 Qxh5 {White is up the exchange and a pawn, and has an attack (and Black doesn't). It's over.} 37. d5+ Kg8 38. d6 (38. d6 f3 {Now basically any sensible-looking move except for 39.Bxf3?? wins:} 39. Bf1 $18 (39. Bxf3 $4 Qxh3#) (39. Qxf3 $18) (39. Qc4 $18)) 1-0