[Event "FIDE World Championship"] [Site "https://lichess.org/study/1NIB7sSp/ZviuiZ6f"] [Date "2021.12.01"] [Round "5"] [White "Nepomniachtchi, Ian"] [Black "Carlsen, Magnus"] [Result "1/2-1/2"] [WhiteElo "2782"] [BlackElo "2855"] [Annotator "John Watson"] [Variant "Standard"] [ECO "C88"] [Opening "Ruy Lopez: Closed, Anti-Marshall"] [Source "https://lichess.org/study/1NIB7sSp/ZviuiZ6f"] [Orientation "white"] 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 { Carlsen mixes it up. This move is relatively rare, played only about a tenth as much as } (8... Bb7 { , which was seen in game 3, and }) (8... b4 { , which we may yet see in a future match game. }) 9. axb5 axb5 10. h3 { Unusual. In the game, it transposes to } (10. c3 d6 11. h3 (11. d4 { is the common alternative })) 10... d6 11. c3 b4 { With a dynamic defensive idea in mind. } (11... Bb7) (11... Bd7 { and }) (11... Re8 { have all been played here. }) 12. d3 { The most challenging move. } (12. d4 bxc3 13. bxc3 exd4 14. cxd4 d5 15. e5 Ne4 { is double-edged, but Black has sufficient play. }) 12... bxc3 { Bringing the b8 rook into play. } (12... d5 { is similar but gives White more options: } 13. exd5 Nxd5 14. Nxe5 (14. c4 Nb6 15. Nxe5 $14) 14... Nxe5 15. Rxe5 Be6 16. Bxd5?! (16. c4 Nb6 17. Rea5 $16) 16... Bxd5 17. cxb4 Bf6 18. Re1 Rxb4 { Jakovenko,D (2644)-Khalifman,A (2653) Moscow 2005 }) 13. bxc3 d5 { A new move, at least in my database. It makes sense to play this in combination with ...bxc3. There are options, including these: } (13... Be6 14. Ba4 (14. Bxe6 fxe6 15. Nbd2 $14) 14... Bd7 15. Nbd2) (13... Nd7 14. Be3 (14. d4) 14... Nb6 15. Nbd2 Kh8 16. Bd5! Qd7 17. Bxb6 Rxb6 18. Nc4 Ra6 19. Qb3 { with the initiative and a large advantage, Balashov,Y (2565) -Janocha,W (2330) Wisla 1992 }) 14. Nbd2 (14. exd5 Nxd5 { is critical. Carlsen doubless has reams of analysis going all the way into the ending, but let me just give what I see as the critical continuations: } 15. Ba4 (15. Nxe5 Nxe5 16. Rxe5 Be6 { with ideas of ....Nxc3 and ...Bf6 offers plenty of compensation }) (15. Qc2!? { looks slow but threatens e5 and deserves consideration. }) 15... Ra8 (15... Rb6) 16. Ba3 Ra6! 17. Bxc6 Rxc6 18. Nxe5 Re6 { gives Black compensation for a pawn and is close to equality. }) 14... dxe4 15. dxe4 Bd6 { Now one idea is ...Ne7-g6, but it's not clear if that equalizes. } (15... Bc5 { is more active and was suggested by Svidler and Kramnik. Then } 16. Ba4 Qd6 17. Nb3!? (17. Nc4 Qxd1 18. Bxd1 Bb7) 17... Qxd1 18. Rxd1 Bd6 19. Be3 { would keep an edge, albeit quite a small one. }) 16. Qc2 (16. Ba4 { is interesting: } 16... Ne7!? (16... Ra8 17. Nc4 Be6 18. Nxd6 Qxd6 19. Qe2 Ne7 { leaves White perhaps slightly better }) 17. Nc4 Ng6 18. Ba3! Bxa3 19. Qxd8 Rxd8 20. Rxa3 { hits e5, which is hard to defend. There might follow } 20... Bb7 21. Bc2 Ra8 (21... Re8 22. Ra5! $16) 22. Rxa8 Rxa8 23. Nfxe5 Nxe5 24. Nxe5 $14) 16... h6 (16... Qe7 17. Nf1!? (17. Ba4 Na5! { is fine for Black }) (17. Bd5! Nd8 18. Nc4 $14) 17... Be6 18. Ba4 (18. Bxe6 Qxe6 19. Bg5 Nd7 20. Ng3) 18... Na5 19. Ne3 Bb3) (16... Ne7 17. Nc4 (17. Nf1 Ng6 18. Ng3) 17... Ng6 18. Nxd6!? (18. Be3) 18... cxd6 19. Ba3 Qc7 20. Red1 (20. Qd1 Rb6 21. Bb4 Qc6) 20... Rb6 $13) 17. Nf1 { A big decision. } (17. Nc4 { is met by } 17... Be6 { , pinning the knight, when a crazy variation I've analyzed goes } 18. Ra6 (18. Be3 Ne7 19. Reb1 Ng6) 18... Ne7! 19. Nfxe5! Qc8! (19... Bxe5 20. Rxe6! fxe6 21. Nxe5 $16) 20. Ra5 Bxe5 21. Rxe5 Ng6 (21... Nd7 22. Rxe6 fxe6 $13) 22. Ra5 Bxh3!? (22... Nh4!?) 23. gxh3 Qxh3 24. Re3! Qh4 { and although still a piece down, Black's ideas of ...Ng4 and...Nf4 secure an equal game, because White's queen has to keep an eye on the bishop on b3 and can't help in the defense: } 25. Rg3! Ng4 26. Be3 Qh2+ 27. Kf1 Qh1+ 28. Ke2 Nxe3 29. Nxe3 Nf4+ 30. Kd2 Rfd8+ 31. Rd5 Nxd5 (31... h5!?) 32. exd5 Rb6 { , which the engines assess as equal. }) 17... Ne7 (17... Be6 18. Bxe6 fxe6 19. g3 Qe8 20. Kg2 Qg6 21. N1h2 $14 (21. Ra6!?)) 18. Ng3 Ng6 19. Be3! { Suddenly the threat c4-c5 with a powerful bind is hard to meet. } (19. Nf5 { might run into } 19... Nxe4!? (19... Qe8 { is safer }) 20. Nxh6+! gxh6 21. Bd5 Kg7!? (21... Kh7) 22. Qxe4 { , when } 22... f5? (22... Qf6 $14 { limits White to a slight advantage }) 23. Qe3 f4 24. Qa7 Rb6 25. c4 { is virtually winning. }) 19... Qe8!? { Magnus took a long think here, falling 40 minutes behind Nepo. But it proved an excellent practical decision, as White immediately went wrong. The options were not pleasant: } (19... Bb7 20. Ra7! Qe7 21. Nd2 Bc8 22. Nf5 $14) (19... Re8 { might be objectively as good as 19...Qe8: } 20. c4! Be6! (20... c5 21. Ba4 $16 { ; Black bishop on d6 is a terrible piece }) (20... Bb4? 21. Red1 Qe7 22. c5! { with the idea Rdb1 and the bishop on b4 is in trouble }) 21. Ba4 Bd7 22. c5 Bf8 23. Red1 Bxa4 24. Rxa4 { with enough pressure to give Black ongoing problems. }) 20. Red1?! { Played fairly quickly, this inaccuracy provd to be the turning point of the game. } (20. c4! { is both obvious and effective. When asked at the press conference, Carlsen later confirmed that this was 'definitely' what worried him. Nepo said that he thought he could play this later with just as much effect, but the rook move gives Black time to unwind. White has a nice plus after 20 c4, for example, } 20... Be6 (20... c5 21. Ba4 $16) 21. Ba4 Bd7 22. c5 Bxa4 23. Rxa4 Qc6 24. Rea1 Be7 25. Qa2 $16) (20. Nf5 { allows } 20... Nxe4! 21. Nxh6+! (21. Ba4 Qe6 22. Nxh6+ gxh6 23. Qxe4 f5) 21... gxh6 22. Bd5 (22. Bxh6!? Rxb3! 23. Qxb3 Nc5 $15) 22... Kh7! 23. Qxe4 f5 { with counterattack and equal chances. }) 20... Be6 21. Ba4 (21. Bxe6 Qxe6 { is quite comfortable for Black. }) 21... Bd7 22. Nd2 { White thought a long time here. The advantage has pretty much disappeared, as can be seen by these variations: } (22. Bb3 Be6 23. Ba4) (22. Nf5 Bxa4 23. Qxa4 (23. Rxa4 Rb2! 24. Nxd6 cxd6 25. Qxb2 Qxa4 26. Rxd6 Nxe4) 23... Qxa4 (23... Ra8 24. Qc4!?) 24. Rxa4 Ra8 25. Rda1 Rxa4 26. Rxa4 Rb8 27. Nd2 Nf4 28. Kf1 Ne6) (22. Bxd7 Qxd7 23. c4 Qc6! 24. c5 Rfe8 { and White isn't able to increase the pressure. }) 22... Bxa4 23. Qxa4 (23. Rxa4 Rb2! { (It's possible that White missed this when thinking about 20 Red1) } 24. Qxb2 Qxa4 25. Ra1 Qc6) 23... Qxa4 24. Rxa4 Ra8 25. Rda1 Rxa4 26. Rxa4 Rb8 27. Ra6 (27. Kf1 { leaves more play on the board, but Black should be fine after } 27... Nd7 28. Ra7 (28. Ke2?! Nh4! 29. Kf1 Bc5) 28... Nf4) 27... Ne8!? { Just when things seemed to be winding down to a draw, a surprise. This move and the idea behind it astonished several of the commentators. Carlsen allows what seems to be a powerful bind for White. } (27... Nd7 { with the idea ...Bc5 was a logical solution, according to analysis during the game and confirmed by the computer. Crucially, it would be a lot easier for the ordinary player to handle. For Carlsen, as it turns out, the choice of 27...Ne8 was equally practical - see my introduction to the game. }) 28. Kf1 Nf8 (28... f6) 29. Nf5 Ne6 (29... h5 { with the idea ..g6 (and discouraging g4) might lead to } 30. Nc4 Ne6 31. Ncxd6 Nxd6 32. Nxd6 cxd6 33. Rxd6 Rc8 34. Rd3 Rc4 35. f3 g5 { . This may be objectively drawn, but would require hours of defense at the board. }) (29... Nd7 30. Nc4 Bc5 31. f3 { is almost equal, but the game continues. }) 30. Nc4 Rd8 (30... Bc5!? { is complicated if White plays in forcing fashion, but simply } 31. f3 { gives Black some problems. }) (30... h5!? 31. f3 (31. Ncxd6 Nxd6 32. Nxd6 cxd6 33. Rxd6 Rc8 34. Rd3 Rc4 35. f3 g5) 31... g6 32. Ncxd6 cxd6 33. Nxd6 (33. Ne7+ Kg7 34. Nd5 Rb1+) 33... Rb1+ 34. Ke2 Rb2+ 35. Bd2 Nxd6 36. Rxd6 Kf8 { should be okay for Black. }) 31. f3 { Some options: } (31. Na5 Rb8) (31. Ke2 h5 32. Na5 Rb8 33. Nc6 Rb2+ 34. Kd3 f6 (34... g6 35. Ra8) 35. Ra8 Kf7 { seems to equalize. }) (31. g4 f6 32. Ke2 { Svidler thought that this would be tough for Black to play. The engines are happy enough, however, because White lacks a specific plan. Compare the game. }) (31. h4 h5 32. f3 f6) 31... f6 (31... h5 32. Ncxd6 cxd6 33. Ra7 $14) 32. g4 (32. h4 h5! { looks fine, because } 33. g4 g6 34. Nh6+ Kg7 35. gxh5 Nf4 36. Bxf4 exf4 37. Ng4 gxh5 38. Nf2 Bc5 { is level. }) 32... Kf7 33. h4 Bf8 34. Ke2 (34. h5?! Rd3 $15) 34... Nd6 35. Ncxd6+ (35. Nfxd6+ Bxd6 36. h5 Be7 37. Ra7 Rb8) 35... Bxd6 36. h5 Bf8! { Remarkably, Carlsen had arrived at this position in his analysis on move 27 and realized that it was an unbreakable blockade, with or without the rooks on the board. } 37. Ra5 Ke8 38. Rd5 Ra8 39. Rd1 Ra2+ 40. Rd2 Ra1 41. Rd1 Ra2+ 42. Rd2 Ra1 43. Rd1 { 1/2-1/2 The game is a draw. } 1/2-1/2