[Event "World Championship"] [Site "Dubai"] [Date "2021.12.04"] [Round "7"] [White "Nepomniachtchi, Ian"] [Black "Carlsen, Magnus"] [Result "1/2-1/2"] [ECO "C88"] [WhiteElo "2782"] [BlackElo "2856"] [Annotator "Monokroussos,Dennis"] [PlyCount "81"] [EventDate "2021.11.27"] [EventCountry "UAE"] 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 O-O 8. a4 Rb8 9. axb5 axb5 10. h3 d6 11. d3 $146 {A new move in the match (11.c3 was played in game 5), and technically a novelty, though 11.d3 transposes to games that reached this position through a different move-order.} h6 12. Nc3 Re8 13. Nd5 Bf8 $146 {Another "novelty", again transposing to a predecessor - just one, in this case.} 14. Nxf6+ $146 {The last novelty of the game. Based on the players' time usage, it seemed they were both out of prep at this point, at least out of preparation they could easily recall.} (14. c3 {1-0 (62) Vachier Lagrave,M (2718)-Telecom Sud Paris Paris 2013}) 14... Qxf6 15. c3 Ne7 {After 20 minutes. The point for Black is twofold, to swing the knight to g6 and to clear the way for ...c5. A possible drawback is that it invites White's rook to a7, but it doesn't seem to lead anywhere concrete.} 16. Be3 (16. Ra7) ({and } 16. d4 {are both possible, but neither move promises all that much.}) 16... Be6 (16... Ng6 $2 {ought to be avoided, as it costs the exchange after} 17. Ba7 Rb7 18. Bd5 {Black should avoid this, even if it's not quite the disaster it would otherwise be, as White's bishop on a7 is in a pickle after} c5 19. Bxb7 Bxb7 $16) (16... c5 {is a reasonable alternative to the text.} 17. Nd2 ({or} 17. Qe2 {.})) 17. d4 {I suspect that it would have been more practical - if Nepo had the energy - to delay pushing the pawn for the time being.} (17. Bc2) ({and} 17. Ra6 {may not be better, but they keep the game going. As we saw in game six, sometimes that's all one needs.}) 17... exd4 $1 {A little unusual unless Black intends to follow with a counter-thrust in the center (...c5 or .. .d5). Here it's a combination of ...c5 and counterplay against White's e-pawn that makes it work.} 18. cxd4 Bxb3 19. Qxb3 Ng6 20. Rec1 {After almost 18 minutes.} (20. Rac1 {was highlighted by everyone as a little better, as it prevented the equalizer we'll see in the game. Also highlighted by everyone: the fact that it doesn't make much of a difference.} Rxe4 (20... c5 $2 21. dxc5 dxc5 22. Bxc5 $16 {is a free pawn. This is the argument in favor of 20.Rac1 over 20.Rec1.}) 21. Rxc7 Re7 {and after White's rook retreats somewhere -} 22. Rcc1 {, for example - he continues to enjoy a very slight edge. Nepo acknowledged this in the presser without being particularly impressed by it, and Carlsen's reaction was similar.}) 20... c5 $1 {With full equality.} 21. e5 {This is in effect a draw offer, and most of the rest of the game happened a tempo, with pieces flying off the board as if they're the featured merchandise in a Black Friday sale.} Qf5 22. dxc5 (22. exd6 $2 c4 $17) 22... dxc5 23. Bxc5 Bxc5 24. Rxc5 Nxe5 25. Nxe5 Rxe5 26. Rxe5 Qxe5 27. Qc3 Qxc3 28. bxc3 Rc8 29. Ra5 Rxc3 30. Rxb5 {In an ideal world the game would have abandoned here, if not around move 23. In this world, players need a little motivation to avoid premature draws, and so the players are not permitted to offer a draw before move 40. (Draws can still take place in fewer moves if, for example, there's a repetition.) The game continues:} Rc1+ 31. Kh2 Rc3 32. h4 g6 33. g3 h5 34. Kg2 Kg7 35. Ra5 Kf6 36. Rb5 Kg7 37. Ra5 Kf6 38. Rb5 Kg7 39. Ra5 Kf6 40. Ra6+ Kg7 41. Ra7 1/2-1/2