[Event "FIDE World Championship 2023"] [Site "Astana"] [Date "2023.04.27"] [Round "13"] [White "Nepomniachtchi, Ian"] [Black "Ding, Liren"] [Result "1/2-1/2"] [ECO "C84"] [WhiteElo "2795"] [BlackElo "2788"] [Annotator "Europe-Echecs"] [PlyCount "78"] [EventDate "2023.04.09"] [EventType "match"] [EventCountry "KAZ"] [SourceTitle "europe-echecs.com"] [Source "ChessBase"] [SourceQuality "1"] 1. e4 e5 2. Nf3 Nc6 3. Bb5 {La partie Espagnole, avec un instant d'hésitation de Ding Liren avant de continuer.} a6 {Pas de Berlinoise aujourd'hui.} 4. Ba4 Nf6 5. O-O Be7 6. d3 b5 7. Bb3 d6 8. c3 O-O 9. h3 Bb7 10. Be3 $5 {La nouveauté, à haut niveau, de cette partie. Il n'est pas nécessaire d'être 2800 Elo pour se rendre compte que ce coup simple ne va pas révolutionner la variante.} (10. a4 Na5 11. Ba2 c5 12. Bg5 h6 13. Bxf6 Bxf6 14. axb5 axb5 15. Nbd2 Nc6 16. Bd5 {1-0 (48) Nepomniachtchi,I (2795)-Ding Liren (2788) WCh 2023 Astana}) 10... Na5 11. Bc2 c5 12. Nbd2 {Pavel Eljanov pense que Ian Nepomniachtchi a tort de jouer pour le gain : « Étrange décision selon moi de se battre après la catastrophe d'hier. J'ai le sentiment que le match pourrait se terminer aujourd'hui. »} Re8 13. a4 {Viswanathan Anand : « Je ne sais pas si 13...Ff8 14.Fg5 marche ici, mais Ding en a assez des Fg5 dans ces lignes ! Maintenant, si 14.d4 exd4 15.cxd4 cxd4 16.Cd4 Cc4 pourrait être raisonnable. »} h6 {Avec une première réflexion de Ian Nepomniachtchi.} 14. d4 $5 {Après 14 minutes. Logique après avoir posté un Fou en e3 pour supporter cette poussée. Même si jouer d4 aussi tôt dans la partie est inhabituel dans l'Espagnole.} exd4 15. cxd4 {Pour ceux que ça intéresse, Ian Nepomniachtchi a délaissé sa chemise rose pour une blanche aujourd'hui.} cxd4 {Après 8 minutes et 27 secondes.} 16. Nxd4 Nc4 {On voit mieux maintenant le côté négatif de 10.Fe3!?.} 17. Nxc4 $5 {Une idée peu appréciée par les machines.} bxc4 {Le GM américain Robert Hess : « Je commence à m'inquiéter pour Ian Nepomniachtchi. » Viswanathan Anand : «} 18. f3 $6 ({ou} 18. Nf5 $5 Bf8 $1) {est contré par} 18... Bf8 $1 {» Et les Noirs sont mieux ! « Avec h3, le coup f3 ne va pas à cause de l'affaiblissement des cases noires. » GM Fabien Libiszewski} 19. Bf2 $6 {La nouveauté est finalement la cause des problèmes du Russe.} d5 $1 {Le coup que les Blancs cherchent à éviter dans l'Espagnole !} 20. exd5 {Fabiano Caruana : « En fait, je pense que Ian pourrait être déprimé. Et il pourrait s'incliner dans cette position. »} Nxd5 {En revanche, Ding Liren joue libéré et vite : 1h14-1h26} 21. Be4 {Ian Nepomniachtchi semble dépité. L'avantage des Noirs consiste en un meilleur bouclier de pions autour de leur Roi et des cases noires affaiblies chez les Blancs en b2, e3 et g3. Le coup b3 des Blancs est généralement contré par ...c3.} Re5 $5 {Ding Liren a pris son temps - 25 minutes - avant de continuer. Il faut reconnaître que le Chinois ne pouvait rêver d'une meilleure position avec les pièces noires au sortir de l'ouverture.} ({Les alternatives informatiques étaient} 21... Rb8 $1) ({et} 21... Qg5 $5) 22. Rc1 {En seulement 2 minutes et 26 secondes. Incompréhensible après un coup noir inattendu. Daniil Dubov : « C'est juste fou de jouer aussi vite. ». Viswanathan Anand : « Plus je regarde, plus la position des Blancs semble mauvaise. Nepo va devoir faire un gros effort pour défendre ça. »} Rc8 23. Ne2 $5 Qe7 $6 {La bonne pièce mais pas la meilleure case.} (23... Qe8 $1 {conservait l'avantage noir.}) 24. Qd4 $1 {Avec égalité selon les machines.} f5 25. Bg3 $1 {Le seul bon coup, qui force Ding Liren à sacrifier la qualité..} Rxe4 ({Le Chinois pouvait sacrifier temporairement la qualité dans un autre sens.} 25... fxe4 26. Bxe5 exf3 27. Rxf3 Qe6 {avec la menace 27...Fc5.} 28. Rxf8+ $1 {Forcé.} Rxf8 29. Ng3) 26. fxe4 Qxe4 {Viswanathan Anand : « Les Noirs ont peut-être une compensation suffisante, mais est-il mieux ? Je ne le pense pas. Pourquoi Ding a-t-il emprunté cette voie ? »} 27. Qxe4 fxe4 {On voit apparaître toute l'idée des Noirs : implanter le Cavalier en d3.} 28. Rfd1 Nb4 $5 29. Rd7 $1 {Avec maintenant un avantage pour les Blancs.} Bc5+ 30. Kh2 Bc6 31. Rc7 $1 {La bonne idée est en effet d'échanger la Tour restante de l'adversaire.} Rxc7 32. Bxc7 Bd5 {Daniil Dubov : « Je n'aime pas pour les Noirs. Je sais que le moteur dirait que c'est égal, mais il faut être prudent ici. »} 33. Nc3 {Fabiano Caruana : « C'est proche de la nulle. La seule raison pour laquelle ce n'est pas un succès pour Ding, c'est parce qu'il avait une bonne position et qu'il ne devrait pas avoir à se battre pour un match nul. »} Nd3 {Et le Cavalier est finalement arrivé sur la case tant désirée.} 34. Rc2 Bc6 {Avec des pendules à 0h38 et 0h33, les joueurs n'auront aucun problème de temps.} 35. a5 {Viswanathan Anand : « 35.a5 est fort. Le Fou des blancs peut être très utile en b6, et après le contrôle du temps, Ian peut élaborer un plan. Ding a une longue journée devant lui. »} Kf7 {Viswanathan Anand : « L'élément crucial pourrait être le Roi des Noirs. Centralisé, il pourrait facilement maintenir la position des Noirs. »} 36. Re2 Nc1 {Le coup humain.} ({La machine donne le surprenant} 36... Ke6 $1 37. Nxe4 Bxe4 38. Rxe4+ Kd5 39. Re8 Nxb2 {avec 0.00.}) 37. Re1 Nd3 38. Re2 Nc1 39. Re1 Nd3 {Après une ouverture insipide avec les pièces blanches, Ian Nepomniachtchi a profité des imprécisions de Ding Liren pour obtenir un léger avantage, insuffisant toutefois pour le gain. Jour de repos vendredi 28 avril. Rendez-vous samedi pour la dernière partie classique du match.} 1/2-1/2 [Event "Nepo - Ding, WCC 2023: Game 13 - Nepo v"] [Site "lichess.org"] [Date "2023.04.27"] [Round "13"] [White "Nepomniachtchi"] [Black "Ding Liren"] [Result "1/2-1/2"] [ECO "C84"] [Annotator "https://lichess.org/@/rowrulz"] [PlyCount "78"] [EventDate "2023.??.??"] {[%evp 0,78,34,26,25,25,35,13,16,-2,-4,-4,30,4,30,20,23,-1,-15,-7,-1,-1,-5,8,-5,1,13,18,8,3,-13,-5,30,30,8,-23,-20,-29,-7,-46,-30,-43,-6,-9,68,-3,0,6,35,35,22,38,38,34,38,38,38,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]} 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. d3 b5 7. Bb3 d6 8. c3 {Nepo returns to the 8.c3 of game 5 having tried 8.a3 in game 11.} O-O 9. h3 Bb7 10. Be3 {A new move at this level.} (10. a4 Na5 11. Ba2 c5 12. Bg5 h6 13. Bxf6 Bxf6 14. axb5 axb5 15. Nbd2 Nc6 16. Bd5 Rxa1 17. Qxa1 Qd7 18. Re1 Ra8 19. Qd1 {Was game 5 with some mild pressure for white. Nepo of course won a brilliant game}) (10. Re1 {transposes to well explored systems}) (10. Bc2 {Is another move which Nepo has faced as black in 2022} Re8 11. Re1 Bf8 12. Nbd2 Ne7 13. a4 Ng6 14. Nf1 d5 {Esipenko, Andrey - Nepomniachtchi, I., 0-1, Airthings Masters KO 2022, https://lichess.org/74OIDqqa}) 10... Na5 (10... d5 11. Nbd2 Qd6 12. Nh4 $14) (10... Nb8 {Going for the Breyer manoeuvre} 11. c4 $5) 11. Bc2 c5 {Ding didn't seem phased and played this natural continuation quite quickly} 12. Nbd2 Re8 13. a4 (13. d4 $6 {Black is well setup to face this and the bishop on e3 isn't ideally placed for this} cxd4 14. cxd4 exd4 15. Bxd4 (15. Nxd4 Bf8 {[%csl Re4] White faces some problems defending the centre - not the bishop on e3 gets in the rid of Re1 defending the pawn}) 15... Nc6 16. Be3 Bf8 $15) 13... h6 {Ding cuts out any Bg5 ideas} (13... Bf8 14. Bg5 $5 {Nepo showed a penchant for losing a tempo with the bishop in the previous game - so it is certainly possible this idea was in his prep} h6 15. Bh4 Be7) 14. d4 $6 {Objectively this seems to be already an inaccuracy - going from a line where white might be able to press to a line where it is white who has to be accurate to equalise} (14. axb5 {was more accurate according to the engine} axb5 15. d4 cxd4 (15... exd4 16. cxd4 Nc6 $5) 16. cxd4 exd4 17. Nxd4 {Here the weakness of b5 is an important factor} Nc4 {The first idea mentioned by Caruana} (17... Bf8 18. Nxb5 d5 (18... Nxe4 $6 {Black regains the pawn but at the cost of a passive position} 19. Nxe4 Bxe4 20. Bxe4 Rxe4 {[%csl Ra5,Rf8]} 21. Qd3 $14 {[%csl Ga5,Gf8]}) (18... Ba6) 19. e5 {White gives the pawn back} Rxe5) (17... Qd7 {Also suggested by Caruana but felt it was a little risky for black}) 18. Rxa8 (18. Nxc4 Rxa1 19. Qxa1 bxc4 20. Ba4 Rf8 21. Bc6 Bxc6 22. Nxc6 Qd7 23. Qa6 $14) 18... Bxa8 (18... Qxa8 {is a serious alternative} 19. Nxc4 bxc4 20. f3) 19. Nxc4 bxc4 20. f3 Bf8 {and now a big difference compared to the game is that white has an immediate Ba4} 21. Ba4 Re5 $14 {looks somewhat awkward for black - white has various interesting options and is for choice}) 14... exd4 15. cxd4 cxd4 (15... Nc6 {Is another interesting option} 16. d5 Nb4 17. Bb1 Nd7 {Here the position seriously resembles a Zaitsev except with the white bishop on e3 - this would be quite a risky approach for black}) 16. Nxd4 (16. Bxd4 {was an alternative but black is very comfortable} Nc6) 16... Nc4 17. Nxc4 (17. axb5 Nxe3 18. fxe3 axb5 {Black is clearly more than fine here despite the pawn deficit}) (17. Qe2 Nxe3 18. Qxe3 Bf8 19. axb5 axb5 20. Rxa8 Bxa8 {Also looks easier for black}) 17... bxc4 {Apparently this is a better version of this position for black compared to where the a-pawns were exchanged. One important factor is without the a-pawns Ba4 was an important resource for white hitting the rook on e8 and perhaps putting a piece on c6} 18. f3 $6 (18. Qe2 Rc8 19. f3 Bf8 20. Nf5 d5 21. Rad1) 18... Bf8 19. Bf2 {Strategic capitulation - Nepo tries to minimise the damage. To be fair other more direct tries were not working tactically} (19. a5 d5 20. Ba4 dxe4 $1 21. Bxe8 Qxe8) (19. Nf5 d5 20. Bd4 dxe4 21. fxe4 Nxe4 22. Qf3 Qd5 $17) (19. Bf4 d5 20. e5 Nd7 21. Re1 Bc5 22. Kh1 Nf8 $15) 19... d5 20. exd5 Nxd5 21. Be4 Re5 $2 {Very direct! Played after a 25 minute thought. A bit mysterious. In hindsight it is easy to say this was a significant mistake as not only is it objectively inferior to alternatives, it also allows Nepo some active defensive ideas/counterplay which he excels in. I think it would be extremely depressing for Nepo to play the position after 21...Rb8 and we could have seen a collapse there} (21... Rb8 $1 {Simply defending the bishop allowing the knight to move - looks strong. It's hard to say why Ding didn't go for this natural move} 22. Qd2 {Preventing ...Nf4} (22. Bxd5 {Preventing any ...Nf4 ideas} Qxd5 $17 {Black is just much better}) (22. Qc2 Nf4 {is just extremely bad for white} 23. Kh1 Qf6) 22... Nb4 $1 {If the bishops are exchanged on b7 the knight comes to d3} (22... Bb4 23. Qc2 {This mysterious shuffle has a hidden point - the bishop is worse placed on b4 as it blocks the the rooks pressure on b2 in some lines} Nf4 24. Kh1 {and the position is still good for black but white is still in the game})) (21... Nf4 $2 {is flashy but white defends} 22. Bxb7 Qg5 23. g3 Nxh3+ 24. Kg2 Nxf2 25. Rxf2 Rad8 26. Rc1 Rxd4 27. Qxd4 Qxc1 28. Bxa6 {is fine for white}) 22. Rc1 Rc8 23. Ne2 {Played after just 4 minutes. Nepo's time management just seems to fast given the critical nature of the position. This move is objectively dubious but very natural (according to Caruana) and it works out well in the game. The computer gives some alternaives but they are difficult to assess} Qe7 $2 {Tactically inferior to ...Qe8 and also looks more awkward as blocks the ...f8 bishop} (23... Qe8 $1 {This move really makes sense of the ...Re5 move} 24. Bd4 (24. Qd4 f5 {and here white doesn't have Bg3 due to ...Bc5}) 24... Ne3 $3 25. Bxe3 Bxe4 26. fxe4 Rxe4) 24. Qd4 $1 {Nepo finally makes a really good move. This could be met by ...Bc5 if 23...Qe8. the tables have turned somewhat and it is black now who needs to be more careful} (24. Bd4 Re6) 24... f5 (24... Qe8 {may be the best - accepting that ...Qe7 was inaccurate} 25. Rxc4 Rxc4 26. Qxc4 Ne3 $1 27. Bxe3 Bxe4 28. fxe4 Rxe4 29. Bd4 Rxe2 30. Rxf7 Rxg2+ 31. Kxg2 Qxf7 32. Qxf7+ Kxf7 $10) 25. Bg3 $1 {Now black doesn't have...Bc5} Rxe4 {This decision was a bit surprising} (25... fxe4 {would be a much more complex choice - here white has a choice of capture} 26. Bxe5 (26. Qxe5 Qxe5 27. Bxe5 exf3 28. Rxf3 (28. gxf3 Re8 29. f4 Nxf4 30. Rxf4 Rxe5 {with enough compensation for black}) 28... Re8 29. Rxf8+ Kxf8 30. Bd6+ $10) 26... exf3 27. Rxf3 Qe6 {[%CAl Gf8c5] black appears to have enough activity} 28. Rxf8+ {is a way to bail out} Rxf8 $10) 26. fxe4 Qxe4 27. Qxe4 fxe4 {Things have certainly gone wrong for black between moves 21-27, similar to Nepo's collapse yesterday though not quite as dramatic. Perhaps white has some advantage here but black is very active and has a pawn and bishop pair for the exchange} 28. Rfd1 Nb4 {Allows the rook to d7 - quite a responsible decision made fairly quickly} (28... Be7 {[%CAl Ge7g5]}) 29. Rd7 Bc5+ 30. Kh2 Bc6 31. Rc7 Rxc7 32. Bxc7 Bd5 33. Nc3 (33. Rd1 Nd3 34. a5 $1 {Introduces the idea of Bb6} Kf7 35. Bb6 Bd6+ 36. Kg1 Be5 37. Bd4 Ke6 (37... Bc7) 38. Bxe5 Kxe5 39. Rf1 Nxb2 40. Kf2 Nd3+ 41. Ke3 {and at the cost of the b2 pawn white has consolidated and black's position is dangerous}) 33... Nd3 34. Rc2 Bc6 35. a5 Kf7 36. Re2 (36. Ne2 $1 {[%csl Rc4]} Bd5 37. Nf4 Nxf4 38. Bxf4 {A significant achievement to exchange this knight on d3 - still the position might remain a fortress}) 36... Nc1 (36... Ke6 $5 37. Nxe4 {white can avoid this and play something else keeping the status quo} Bxe4 38. Rxe4+ Kd5 {here it seems that white is risking more as the b2 pawn is going to drop after which the c-pawn will be very dangerous - however as Ding pointed out in the press conference he saw there was a perpetual} 39. Re8 Nxb2 40. Rd8+ Ke4 41. Re8+ $10 {black should take the repetition} Kd3 $2 42. Be5 $1 {and white is better}) 37. Re1 (37. Rc2 Nd3 38. Ne2 {white could go back to this variation}) 37... Nd3 38. Re2 Nc1 39. Re1 Nd3 {Draw} 1/2-1/2 [Event "WCh 2023"] [Site "Astana KAZ"] [Date "2023.04.27"] [Round "13.1"] [White "Nepomniachtchi, Ian"] [Black "Ding Liren"] [Result "1/2-1/2"] [ECO "C84"] [Annotator "Crowther,Mark"] [PlyCount "79"] [EventDate "2023.04.09"] {[%evp 0,79,19,31,31,38,21,5,25,1,0,2,16,1,12,10,29,-8,-5,-35,-9,-5,3,17,4,-8,23,3,14,6,14,3,32,42,25,-8,3,-58,-7,-44,-49,-49,-10,-23,34,11,6,0,22,27,26,20,21,21,14,26,23,12,0,0,0,0,0,0,0,0,0,0,0,0,0,0,7,7,0,0,0,0,23,23,23,23]} 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. d3 b5 7. Bb3 d6 8. c3 O-O 9. h3 Bb7 ({Relevant:} 9... Na5 10. Bc2 c5 11. d4 Qc7 12. d5 Nc4 13. a4 Nb6 14. b3 Bd7 15. Be3 bxa4 16. bxa4 Nc4 17. Bc1 Rfb8 18. Qe2 Na5 19. Nbd2 Be8 20. Ra2 Nd7 21. Re1 Nb6 22. Nf1 c4 23. Ne3 Nb3 24. a5 Na4 25. Bd2 Rc8 26. Ra3 Rab8 27. Nxc4 Nxc3 28. Bxc3 Qxc4 29. Qxc4 Rxc4 30. Re3 Nd4 31. Bd3 Nxf3+ 32. Rxf3 Ra4 33. Rxa4 Bxa4 34. Bxa6 Rb1+ 35. Kh2 Bd1 36. Re3 Bg5 37. Re1 Rb3 38. Rxd1 Rxc3 39. Ra1 {Firouzja,A (2785)-Alekseenko,K (2686) chess.com INT 2023 1-0}) 10. Be3 {This hasn't been played at elite level before.} Na5 {Ding didn't take too long over executing standard play.} 11. Bc2 c5 12. Nbd2 Re8 {[#]} 13. a4 $146 ({Predecessor:} 13. Qe2 Rc8 14. b3 Qc7 15. c4 b4 16. Rfd1 Bf8 17. Nf1 g6 18. Qd2 Bg7 19. Bh6 Nh5 20. N1h2 Qe7 21. Bxg7 Kxg7 22. Ng4 f6 23. Qh6+ Kg8 24. Re1 Qg7 25. Qxg7+ Kxg7 26. g3 f5 27. Ngh2 h6 28. Rad1 Rcd8 29. g4 fxe4 30. gxh5 exf3 31. hxg6 Rf8 32. Re3 Rf6 33. Kh1 Nc6 34. Rg1 Nd4 35. Bd1 Rdf8 36. Rg4 Rxg6 37. Rxg6+ Kxg6 38. Nf1 Rf4 39. Kh2 Rh4 40. Kg3 Nf5+ 41. Kh2 Nxe3 42. fxe3 Bc8 {Koay,Y (1154)-Tan,J (1901) Penang 2017 0-1}) 13... h6 14. d4 (14. axb5 axb5 15. d4 {was suggested by Caruana as being more accurate.}) 14... exd4 15. cxd4 cxd4 16. Nxd4 Nc4 $1 {Black seems to have excellent play now.} 17. Nxc4 bxc4 18. f3 $6 {This seems to be a fairly serious inaccuracy.} (18. Qe2) (18. Nf5 Bf8 19. Re1) 18... Bf8 19. Bf2 {Nepomniachtchi admitted he really didn't like his position here. It was actually pretty obvious just by looking at him that he thought that.} d5 20. exd5 Nxd5 21. Be4 Re5 $6 {Played after a 25 minute thought. Ding admitted after the game he missed 24.Qd4. Here black has a lot of options two of them are:} (21... Rb8 {is the computer's choice setting up a lot of threats. It's probably not the first move you look at.}) (21... Qg5 {is also very interesting.}) 22. Rc1 {Nepomniachtchi started playing a bit more confidently.} Rc8 23. Ne2 Qe7 $6 (23... Qe8 {was far more accurate not limiting the scope of the bishop but still having the advantages of Qe7.}) 24. Qd4 $1 {Now it's black that's under pressure to find good moves.} f5 25. Bg3 Rxe4 {an exchange sacrifice, probably necessary by now.} (25... fxe4 {was another way and it may have been slightly more accurate.}) 26. fxe4 Qxe4 27. Qxe4 fxe4 28. Rfd1 Nb4 29. Rd7 Bc5+ 30. Kh2 Bc6 31. Rc7 Rxc7 32. Bxc7 Bd5 {This last sequence of moves are all computer approved and Bd5 was necessary.} 33. Nc3 (33. a5) (33. Rd1) 33... Nd3 34. Rc2 Bc6 35. a5 Kf7 36. Re2 $11 (36. Ne2 Bd5 37. Nf4 Nxf4 38. Bxf4 {and white may still be pressing.}) 36... Nc1 37. Re1 Nd3 38. Re2 Nc1 39. Re1 Nd3 40. Re2 {Neither side has any reason to play on here.} 1/2-1/2 [Event "FIDE World Chess Championship 2023"] [Site "Chess.com"] [Date "2023.04.27"] [Round "13"] [White "Nepomniachtchi, Ian"] [Black "Ding, Liren"] [Result "1/2-1/2"] [ECO "C84"] [WhiteElo "2795"] [BlackElo "2788"] [Annotator "Rafael"] [PlyCount "75"] [EventDate "2023.??.??"] [TimeControl "7200+30"] 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. d3 b5 7. Bb3 d6 8. c3 {Nepomniachtchi returns to the variation he employed in the fifth game of the match.} O-O 9. h3 Bb7 10. Be3 {This natural development move is rare and has never been played in grandmaster games. It deviates from conventional opening theory at the same moment as in their previous duel in this line.} (10. Re1 {is the main move.}) (10. a4 {was played in game five of the match.}) 10... Na5 11. Bc2 c5 {This is the standard Spanish plan.} 12. Nbd2 Re8 13. a4 {This move is officially a novelty. Once again we see one of the hallmarks of this match: a battle outside the most well-known opening lines. It's quite a feat to find a new position in the Ruy Lopez so early without playing any eccentric move.} h6 {A usual move, avoiding a possible pin with the bishop on g5 after the planned Be7-f8.} 14. d4 {It's debatable whether it's better to capture on b5 before playing this.} (14. axb5 axb5 15. d4 {Let's suppose Black plays in the same way as in the game, although this is not forced at all.} exd4 16. cxd4 cxd4 17. Nxd4 Nc4 18. Nxc4 bxc4 {Now White has an interesting option available:} 19. Ba4 {Although it seems not even this is enough for an advantage after ?Rf8}) 14... exd4 $1 {[%c_effect d4;square;d4;type;GreatFind;persistent;true]} (14... Bf8 $2 {[%c_effect f8;square;f8;type;Mistake;persistent;true] is a clear positional mistake, as after} 15. d5 $1 {[%c_effect d5;square;d5;type;GreatFind;persistent;true] White gets a dream position, shutting the b7-bishop from the game.}) 15. cxd4 cxd4 16. Nxd4 Nc4 $1 {[%c_effect c4;square;c4;type;GreatFind;persistent;true] This accurate move gives Black a good position.} 17. Nxc4 {The computer also considers other moves, but for a human, it's almost automatic to capture such a knight.} bxc4 18. f3 $6 {[%c_effect f3;square;f3;type;Inaccuracy;persistent;true] This leads to an unpleasant position.} (18. Qe2 $1 {[%c_effect e2;square;e2;type;GreatFind;persistent;true]} Rc8 19. f3 {is similar to the game but an improved version for White.}) 18... Bf8 {Now Black is ready to play ?d6-d5.} 19. Bf2 (19. Bf4 {Nepomniachtchi mentioned he wanted to make this work but didn't like the position after} d5 20. e5 Nd7 21. Re1 Qc7) (19. Nf5 {this move was also discussed in the press conference.} d5 20. Bd4 dxe4 21. fxe4 Nxe4 {The compensation for the pawn is not enough.}) 19... d5 $1 {[%c_effect d5;square;d5;type;GreatFind;persistent;true]} 20. exd5 Nxd5 {White's pawn structure is not good, especially on the kingside, as the knight can jump to f4 and attack the king. Careful play is necessary.} 21. Be4 {A natural move, pinning the unpleasant knight.} Re5 $6 {[%c_effect e5;square;e5;type;Inaccuracy;persistent;true] A strange move.} (21... Nf4 {is interesting, but the sacrifice is not enough for an advantage.} 22. Bxb7 Qg5 23. g3 Nxh3+ 24. Kg2 Nxf2 25. Rxf2 Rad8 26. Qc1 {The simplest.} Rxd4 27. Qxg5 hxg5 28. Bxa6 {Obviously Black didn't sacrifice a piece to go into this endgame.}) (21... Rb8 {is a good move. The knight gets ready to go to f4, so it must be captured.} 22. Bxd5 Bxd5 23. Qd2 {Black has some advantage due to the pair of bishops.}) (21... Qg5 22. Bxd5 Bxd5 {is also slightly better for Black.}) 22. Rc1 Rc8 23. Ne2 $6 {[%c_effect e2;square;e2;type;Inaccuracy;persistent;true] The knight offers the d4-square to the king or bishop while getting ready to go to c3. But the move is a bit passive.} (23. Qd2 $5 {[%c_effect d2;square;d2;type;Interesting;persistent;true] is a better option.}) 23... Qe7 $6 {[%c_effect e7;square;e7;type;Inaccuracy;persistent;true] This is an important moment, as the move played throws away the advantage. It's likely that Ding missed White's next move.} (23... f5 {is not dangerous since White can remove the bishop.} 24. Bb1) (23... Qe8 $1 {[%c_effect e8;square;e8;type;GreatFind;persistent;true] is the best way to keep the initiative. The queen is better placed here as it doesn't interfere with the bishop on f8. The position is dangerous for White, as he can't play in the same ways as the game since the queen is lost after} 24. Qd4 f5 25. Bg3 $4 {[%c_effect g3;square;g3;type;Blunder;persistent;true]} Bc5) 24. Qd4 $1 {[%c_effect d4;square;d4;type;GreatFind;persistent;true]} f5 {The natural followup, but now White has a forced saving move.} 25. Bg3 $1 {[%c_effect g3;square;g3;type;GreatFind;persistent;true]} Rxe4 (25... fxe4 {This way of capturing is also possible.} 26. Qxe5 (26. Bxe5 {is also possible, although it's scary to leave the queen on d4 with the danger of ?Bf8-c5 coming.} exf3 27. Rxf3 Qe6 28. Rxf8+ $1 {[%c_effect f8;square;f8;type;GreatFind;persistent;true] One of the advantages of being up an exchange is that you can give it back at the right moment.} Rxf8 29. Ng3 {with an equal position.}) 26... Qxe5 27. Bxe5 Bc5+ 28. Kh2 (28. Bd4 $2 {[%c_effect d4;square;d4;type;Mistake;persistent;true] loses to} Nf4 $1 {[%c_effect f4;square;f4;type;GreatFind;persistent;true]}) 28... Ne3 29. Rfe1 exf3 30. gxf3 Rd8 {Black has good compensation, enough for equality.}) 26. fxe4 Qxe4 27. Qxe4 fxe4 {Black has good compensation for the exchange, but it's clear that he didn't get the most out of the promising position achieved before.} 28. Rfd1 Nb4 $6 {[%c_effect b4;square;b4;type;Inaccuracy;persistent;true] This makes things a bit harder for Black.} (28... Bc5+) (28... Be7 $5 {[%c_effect e7;square;e7;type;Interesting;persistent;true]}) 29. Rd7 $1 {[%c_effect d7;square;d7;type;GreatFind;persistent;true]} Bc5+ 30. Kh2 Bc6 31. Rc7 {Exchanging a pair of rooks is usually a good strategy when you are an exchange up.} Rxc7 32. Bxc7 Bd5 {The computer is initially quite happy with White's position, but for human eyes, the compensation is decent with a knight coming to d3, the king ready to be centralized, and two strong bishops.} 33. Nc3 (33. a5 $1 {[%c_effect a5;square;a5;type;GreatFind;persistent;true] is a good possibility, giving the b6-square to the bishop. The game can continue:} Nd3 34. Rb1 Kf7 35. Bb6 Bd6+ 36. Kg1 {and my machine continues to claim an advantage for White.}) 33... Nd3 34. Rc2 Bc6 35. a5 Kf7 36. Re2 (36. Ne2 {is a way to keep fighting for an advantage.} Bd5 37. Nf4 Nxf4 38. Bxf4 Ke6 {Black should be able to make a comfortable draw.}) 36... Nc1 {The players decide to repeat moves and this is an understandable decision.} (36... Ke6 {is also possible, playing for a win. If White captures on e4 the game may end in perpetual check after} 37. Nxe4 Bxe4 38. Rxe4+ Kd5 39. Re8 Nxb2 40. Rd8+ Ke4 (40... Kc6 $2 {[%c_effect c6;square;c6;type;Mistake;persistent;true]} 41. Be5) 41. Re8+ Kd5 (41... Kd3 $2 {[%c_effect d3;square;d3;type;Mistake;persistent;true]} 42. Be5 c3 43. Rd8+ Kc4 44. Bxg7) 42. Rd8+) 37. Re1 Nd3 38. Re2 {A nervous game in which both players had some chances, but Ding probably could have pushed harder from the position he got in the opening. The tension grows and now we have a rest day before the final classical game. Will Ding play aggressively to fight for a win with White or is it better not to risk it and continue the battle in the tiebreaks $2 We'll know soon enough.} 1/2-1/2