Carlsen-Pelletier. Position after 40… Kxf6. Puzzle. Can White, to move, draw?
No. But he almost did.
White tried 41. Rf5+, and black cannot ever take the rook because of stalemate on the back rank. So, black declines, playing Kd5. Luckily for him, the knight guards c5, so the rook cannot keep simply checking the king on the fifth rank. If the relative positions had been one rank up the board, with a Kf7 and Rb6, then it would have been drawn, simply. If the King flees to h7 in that scenario, then white goes Rg7+, Kh6 (Kh8 is no better, producing Rg8+ and he gets chased back again) and then Rh7+ wins the rook on h2, or produces stalemate. Instead, the game continued…
42. Re5+ Kd6
43. Rd5+ Kc6
44. Rd6+ (c5 is covered by the knight!) Kc5
45. Rc6+ Kd5 (surely Kb4 or Kd4 is quicker? NO! Kb4/d4 is met by Rxc4+! Now, if Rxc4, Nd2+ picks up the rook via a handy fork)
46. Rd6+ Ke4
47. Re6+ Kd3
48. Re3+ Kd2
49. Resigns. Nothing works now. Re2+ loses to R or Kxe2 and the white king suddenly has a2 to move to. Rd3+ also loses to cxd3, but not Kxd3??.
Carlsen played horribly in this game. He deliberately (I think) sacrificed a piece for two pawns on move 16, but never seemed to have enough for a win. He struggled to find any compensation at all. Was it overconfidence? Did he miscalculate? Was the original sacrifice unsound? Doubtless we will find out shortly.
[Opening “Nimzo-Indian: classical variation”]
1. d4 Nf6 2. c4 e6 3. Nc3 Bb4 4. Qc2 O-O 5. a3 Bxc3+ 6. Qxc3 b6 7. Bg5 Bb7 8. e3 c5 9. dxc5 bxc5 10. Ne2 a5 11. Bxf6 Qxf6 12. Qxf6 gxf6 13. O-O-O Rd8 14. Nc3 d6 15. Bd3 Kf8 16. Bxh7 f5 17. e4 Kg7 18. exf5 Kxh7 19. fxe6 fxe6 20. Rhe1 Bxg2 21. Ne4 Bxe4 22. Rxe4 Ra7 23. Rxe6 Rad7 24. Rd3 Nc6 25. Rh3+ Kg8 26. Rhh6 Ne5 27. f4 Nxc4 28. b3 Nxa3 29. Rhg6+ Kh8 30. Rh6+ Kg8 31. Rhg6+ Kh8 32. Rh6+ Rh7 33. Rxd6 Rxd6 34. Rxd6 Rxh2 35. Ra6 Nb5 36. Rxa5 Nd4 37. Kb1 Nxb3 38. Rb5 c4 39. f5 Kg7 40. f6+ Kxf6 41. Rf5+ Ke6 42. Re5+ Kd6 43. Rd5+ Kc6 44. Rd6+ Kc5 45. Rc6+ Kd5 46. Rd6+ Ke4 47. Re6+ Kd3 48. Re3+ Kd2 Black wins 0-1