[Event "FIDE Candidates 2024"] [Site "Toronto, Canada"] [Date "2024.04.04"] [Round "1.3"] [White "Firouzja, Alireza"] [Black "Praggnanandhaa R"] [Result "1/2-1/2"] [WhiteElo "2760"] [BlackElo "2747"] [Annotator "WGM Tatev Abrahamyan"] [Variant "Standard"] [ECO "C83"] [Opening "Ruy Lopez: Open, Classical Defense"] [Source "https://lichess.org/study/Mez4h4WA/KVHwtV6A"] [Orientation "white"] { Annotations by WGM Tatev Abrahamyan } 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Nxe4 6. d4 b5 7. Bb3 d5 8. dxe5 Be6 9. c3 Be7 { So far, the players are following main line theory which has been played thousands of times. } 10. Bc2 Bg4 11. Qe1 { An unusual way of getting out of this standard pin! Interestingly enough, this move was played by none other than Pragg's sister Vaishali in last year's World Cup. } 11... Nc5 (11... Bxf3 { looks like a logical way to try to punish White's previous move. } 12. gxf3 Nc5 13. f4 Qd7 14. Qe2 Qe6 15. Be3 f6 16. b4 (16. f5! Qxe5 17. f4 Qd6 18. Rd1 { would have given White a big advantage. } 18... O-O 19. b4 Nd7 20. Rxd5! Qxd5 21. Bb3 $18) 16... Na4 17. f5 Qxe5 18. f4 Qd6 19. Qh5+ Kf8 20. Rd1 Nb2 { was eventually a win for Black, in 55 moves, in Vaishali – Muzychuk, Baku, 2023. }) 12. Nbd2 Ne6 13. Kh1 { Truly a strange-looking move! The idea becomes clear in the next few moves. } (13. Qe2 { Pragg had faced this move against MVL but went on to lose despite getting a decent position: } 13... O-O 14. Re1 Qd7 15. Qd3 g6 16. Qf1 f6 17. h3 Bxh3 18. gxh3 fxe5 19. Nxe5 Nxe5 20. Rxe5 Nf4 (20... Bc5! $19) 21. Re3 Bc5 22. Rf3 { and White eventually came back to win in 44 moves in Vachier-Lagrave – Praggnanandhaa, Chess.com, 2023. }) 13... Bh5 (13... O-O 14. Ng1 f6 15. exf6 Rxf6 16. f4 g6 17. f5 Bxf5 18. Bxf5 gxf5 19. Ndf3 f4 20. Nh3 Qd7 21. b3 Raf8 22. Bb2 Bc5 23. c4 d4 24. Nf2 bxc4 25. bxc4 Rf5 26. Rd1 Ng5 27. Nxg5 Rxg5 28. Ne4 Re5 29. Nxc5 Rxc5 30. Ba3 Re5 { and a draw was agreed in Gukesh – Yakubboev, Ankara, 2023. }) 14. Nb3 Bxf3 15. gxf3 { And now we finally see the idea behind 13. Kh1: it is to allow this pawn structure and, in turn, use the g-file for the rook. How does one evaluate a position like this? White has compromised his structure and has long-term weaknesses. At the same time, simply based on the speed Firouzja was playing, we know that that the computer evaluates this as at least playable for White. The position is truly a reflection of modern engines' influence on the game. } 15... Bg5 (15... O-O 16. f4 { Immediately, Black comes under a huge attack: } 16... f5 17. exf6 Rxf6 18. f5 Nf8 19. Rg1) 16. Rg1 Bxc1 17. Nxc1 { The correct capture, as the knight on b3 was useless and needs to rejoin the game via the d3-square. } 17... Qh4 18. Rg4 Qh5 19. Nd3 O-O 20. Qe2 f5 21. exf6 Rxf6 { The position is double-edged, as it is unclear whose king is going to be weaker. White has the semi-open g-file, but the g7-pawn is well targeted. On the other hand, his own king looks vulnerable and the pawns on the f-file are visibly weaker now. } 22. f4 Rh6 23. f3 Rf8 24. Qf2 Rff6 25. Re1 Qf7 26. Qg3 Rh5 27. a4 Rfh6 28. Re2 Rh3 29. Qe1 Rxf3 30. Nc5! { White is posing some difficult questions for Black. } 30... Qh5? (30... Nf8 31. axb5 axb5 32. f5 { The position remains complex: Black is up a pawn, but White has control of the e- and g-files. }) 31. Nxe6 (31. Reg2! { would have forced Black to find series of computer-esque only moves: } 31... Ne5 (31... Nxc5 32. Rxg7+ Kf8 33. Rg8+ Kf7 34. Qe8+ Kf6 35. Rf8+ Qf7 36. Qxf7#) (31... Qf7 32. f5! Nxc5 33. Rxg7+ Qxg7 34. Qe8#) 32. fxe5 Qf7 { with the threat of ... Rf3-f1. } (32... Nxc5 33. Rxg7+ Kf8 34. Bd1! $18) 33. Bd3 Nxc5 34. Rxg7+ Qxg7 35. Qg1 Ne6 36. Rxg7+ Nxg7 37. Qg5 $16) 31... Qxg4 32. Ng5 Ne5! { This is the correct path to a draw. } (32... Rxf4? 33. Re8+ Rf8 34. Qf2!! { What a shot! This would end the game. } 34... Rxh2+ (34... Rf6 35. Bxh7+ Kh8 36. Rxf8+ Rxf8 37. Qxf8#) (34... Rxe8 35. Qf7+ Kh8 36. Qxe8#) 35. Qxh2 Qxg5 36. Qxh7+ Kf7 37. Bg6+ $18) 33. Rxe5 Rxh2+ 34. Kxh2 Qh5+ 35. Kg2 Qg4+ 36. Kh1 Qh5+ 37. Kg2 Qg4+ 38. Kh1 Qh5+ 39. Kg2 { 1/2-1/2 The game is a draw. } 1/2-1/2