[Event "FIDE Candidates"] [White "Vachier Lagrave, Maxime"] [Black "Grischuk, Alexander"] [Site "Yekaterinburg"] [Round "4"] [Annotator "Daniel Fernandez"] [Result "1/2-1/2"] [Date "2020.03.21"] [PlyCount "106"] 1. e4 {} e5 2. Nf3 Nc6 3. Bb5 Nf6 4. 0-0 Nxe4 5. d4 Nd6 6. Bxc6 dxc6 7. dxe5 Nf5 8. Qxd8+ Kxd8 9. h3 Ke8 10. Nc3 h5 11. Bf4 Be7 12. Rad1 Be6 13. Ng5 Rh6 14. Rfe1 {This is the critical move, setting up an existential problem for Black: if the bishop on e6 moves, White can play e6; if he goes ...Rg6 then White gains time to push kingside pawns, and if he does not cut the kingside Gordian knot then he has more pieces tied up there than White does. The main plan in response to all this is ...h4, but not immediately.} (14. g3 Bxg5 15. Bxg5 Rg6 16. Bf4 Nh4 $11 {Fedorchuk,S-Malakhov,V Brest 2018} )Bb4 {Before granting White a mobile kingside majority Black makes sure that the balance of queenside structures favours him.} (14... h4 15. Re4 Rd8 16. Rxd8+ Bxd8 17. Ra4 a6 18. Nce4 Bc8 19. Nf3 Rh8 20. Bg5 Bxg5 21. Nexg5 Ke7 22. Rf4 c5 23. Kh2 b6 24. g4 hxg3+ 25. fxg3 Nh6 26. Re4 Bb7 27. Re2 a5 28. Nh4 Rd8 29. g4 Ba6 30. Rf2 $14 {Motylev,A-Bacrot,E Shamkir 2014} )15. a3 Bxc3 16. bxc3 h4 {Up to this moment both players have been following Grischuk's round 2 game, of which the opening phase was rather in his favour.} 17. g4 $5 (17. Kh2 Kf8 $1 {Black initiates a standard plan. The king must be evacuated so that the rook can get to e8, which in turn frees up the e6-bishop to move without the pawn thrust e6 being necessarily an issue.} 18. Rb1 b6 19. Rb4 Re8 20. Ra4 a5 21. Bd2 c5 22. Rf4 Rh5 23. Rf3 Kg8 24. Bc1 Ne7 25. g4 hxg3+ 26. fxg3 Nc6 $15 {White has more or less played without an idea here and the central pressure he usually gets in return for the fractured queenside structure is not materializing. Nepomniachtchi,I -Grischuk,A Yekaterinburg 2020} )hxg3 18. fxg3 Ne7 ({ Also perfectly possible and a bit more principled is:} 18... Rd8 19. Rxd8+ Kxd8 20. g4 Ne7 $11 )19. h4 Nd5 20. Bc1 {Probably this is still preparation (!) but how worried need Black be?} Nxc3 $5 {Black takes the bait!} (20... Rd8 $11 )21. Rd3 Na4 22. Rf3 {Clearly this was White's intention. Black needs to respond incisively to the doubling on the f-file.} Bd5 23. Rf4 (23. Rf5 Rh8 24. Ref1 Rf8 $11 )Nb6 24. Ref1 Rg6 25. Rf5 Bc4 26. Re1 {White intends h5 and there are various ideas of a4, Re4 and so on that he can use too. His compensation is not in doubt.} ({I wondered for a while whether some kind of 'perpetual rook' draw was on the cards, but after} 26. R1f3 $2 {Black probably responds simply with} Rd8 $1 $17 {activating the second rook, and White's compensation begins to look really doubtful.} )Ke7 27. h5 (27. e6 $5 { At first the engines like this move, but it's nothing special compared to the more solid text.} f6 28. h5 Rh6 29. Ne4 Rh7 (29... Rhh8 30. h6 Rag8 31. hxg7 Rxg7 32. Rxf6 Rgh7 $13 ))Rh6 28. g4 {Now the position has reached peak complexity and both sides understandably begin making mistakes.} Rhh8 $6 { Black opts for a move which is 'safe', but still wrong.} (28... Be6 $1 29. Nxe6 Rxe6 30. Bg5+ Kf8 31. Ref1 Kg8 32. Rxf7 Nd5 $11 {Black has fair winning chances in this position too, because White often finds himself a pawn down.} )29. a4 $5 (29. e6 $1 f6 30. a4 $1 {It is very far from clear that this order is any better for White. Why would you voluntarily deprive yourself of Rxf7+ in all the lines? But the engine is impossible to argue with.} fxg5 31. Ba3+ Ke8 32. Rxg5 Rh7 33. Rf5 Rh8 34. Rf7 $14 {White has a massive attack, and Black's time-pressure (in no small part caused by spending 43 minutes on move 18) compounds the problems.} )Nxa4 $4 (29... Be6 30. Ba3+ Ke8 31. Rf4 Nd5 32. Nxe6 fxe6 33. Rf3 $44 )30. Ba3+ $6 {White had a win which will doubtless go down in the calculation books for decades to come.} (30. Re4 $1 Be6 (30... Nb6 31. Ba3+ (31. Rxc4 Nxc4 32. Rxf7+ Ke8 33. Rxc7 $14 {is also possible, but a long way from being decisive.} )Ke8 32. Nxf7 $1 {This is the least intuitive part of the whole sequence and so I assume it is this MVL missed.} Bxf7 $6 33. e6 $18 )31. Nxe6 fxe6 (31... Nc3 32. Ng5 Nxe4 33. Rxf7+ Ke8 34. Rxg7 $18 {Black has managed to exchange one of the attackers but the remaining two, plus the passed pawns, are too much.} )32. Bg5+ $1 (32. Ba3+ c5 )Ke8 33. Rf3 $1 {Specifically this move, so that Black's knight can't get to c3 and then d5.} Nb6 34. Ref4 Nd5 35. Rf7 b5 36. Rxg7 $18 )c5 31. e6 f6 32. Bxc5+ Nxc5 33. Rxc5 fxg5 $6 {White gets a little nibble after this, though the game never leaves the drawing margin.} (33... Rhc8 $1 $11 )34. Rxc7+ Kd6 (34... Kf6 $4 35. Rf7# $1 {is mate.} )(34... Kf8 $5 35. Rxc4 a5 {seems a touch more solid; White has no way to press meaningfully despite a number of lines where he retains an extra pawn.} )35. Rxc4 a5 36. Rd1+ (36. Rd4+ $5 Kc6 37. Red1 Rhe8 (37... a4 38. Rc4+ Kb5 39. Rc7 $16 )38. Rc4+ Kb5 39. Rc7 Rxe6 $14 {Black should probably draw this, but there are various ways to pose problems.} )Ke7 37. Re4 Rhd8 38. Rb1 Rdb8 39. Rb5 a4 40. Rxg5 Rg8 (40... a3 $2 41. Rxg7+ Kf6 42. Rf7+ Kg5 43. Rf5+ Kh6 44. Rf1 $16 {sees White make it back to a1 in time.} )41. h6 gxh6 42. Rxg8 Rxg8 43. Rxa4 h5 $1 {Making sure that there are no more tricks based on White's rook being marginally more active.} 44. Kf2 Rxg4 45. Rxg4 hxg4 46. Kg3 Kxe6 47. Kxg4 Ke5 48. Kf3 Kd4 49. Ke2 Kc3 50. Kd1 b5 51. Kc1 b4 52. Kb1 b3 53. cxb3 Kxb3 1/2-1/2