[Event "FIDE Candidates 2024"] [Site "Toronto CAN"] [Date "2024.04.06"] [Round "3.1"] [White "Gukesh, D."] [Black "Nepomniachtchi, Ian"] [Result "1/2-1/2"] [ECO "E01"] [WhiteElo "2743"] [BlackElo "2758"] [Annotator "Monokroussos,Dennis"] [PlyCount "80"] [EventDate "2024.04.04"] {[%evp 0,80,26,17,27,-16,0,-20,-23,-16,-11,3,15,43,37,29,26,31,26,31,23,46,37,53,27,31,39,15,58,54,39,15,35,22,44,69,50,66,67,64,48,40,104,93,92,92,93,92,105,99,99,86,88,83,90,86,86,86,81,68,71,72,73,74,87,74,75,86,111,111,97,105,105,90,118,88,83,70,88,64,48,48,48]} 1. d4 Nf6 2. c4 e6 3. g3 d5 4. Bg2 c5 {Almost every line Black chooses against the Catalan aims to play ...c5 sooner or later. This one falls strongly on the side of "sooner".} 5. cxd5 Nxd5 {This avoids being saddled with an isolated d-pawn; the cost is that White can achieve a speedy e4 if he so desires.} (5... exd5 {makes it a Tarrasch.}) 6. Nf3 ({It's too soon for} 6. e4 {, as} Qa5+ $1 7. Bd2 Nb4 $11 {, threatening ...Nd3+, equalizes.}) 6... cxd4 (6... Nc6 {used to be standard, but Nepo's move is the current #1.}) 7. O-O (7. Nxd4 {is probably just as good, but allows the annoying} Bb4+ {.}) 7... Bc5 8. Nxd4 O-O 9. Nb3 (9. Nc3 {is also common, used by Firouzja and Nepo himself the past couple of years. One point is that} Nxc3 10. bxc3 {can be described with equal justification as damaging White's pawn structure and as giving White effective play down the b-file.}) 9... Be7 (9... Bb6 {had been both more common and more successful.}) 10. e4 Nb4 {Threatening to trade queens and win the exchange with ...Nc2.} 11. Nc3 N8c6 12. Be3 Qxd1 13. Raxd1 {The engine says it's equal (though not triple-zeros), but in games where both players were rated over 2000 (or were both rated, full stop) White had gone 5-0 coming into this contest. It's foolish to draw hard conclusions from so little data, but it's worth considering the possibility that for humans White's position is easier to play.} Ne5 $1 14. a3 Nbc6 {Transposing to a game with 2000+ rated players that finished in a draw.} 15. f4 $146 (15. Rfe1 {was played in the earlier games (one featuring an unrated player losing with White to a 1996).}) 15... Nc4 16. Bc1 f6 {Preventing White from playing e5, at the cost of some potential light-squared weaknesses (e.g. d5, after f4-f5) down the road.} (16... a5 {is a good idea, putting into effect what might be dubbed a corollary to Tarrasch's dictum. Tarrasch, in one of his dogmatic moods, wrote that a knight on "N3" (b6/b3/g3/g6, on the knight's own side of the board) "always" stands badly. I'm sure there has been at least one game in chess history where such a knight stood well - giving mate, for example - but there's something to this dictum. The corollary I propose is that however badly the knight stands on N3 (that's the label in descriptive notation, used in the book where I first came across Tarrasch's comment), it's a good idea to kick it with a rook's pawn. Maybe White's knight stands badly on b3, but it might be even worse after ...a5-a4! Here there's a tactical justification: If Black plays ...a4 White can't play Nd4 because Black trades knights and plays ...Bc5, while Nd2 is met by ...Ne3 forking the rooks. The a5 and c5 squares are obviously off limits, so the poor knight would be forced to the ignominious a1 square.} 17. a4 {makes sense, and now, amusingly, Black's best may be thumbing his nose at Tarrasch and playing} Nb6 $1 $11) 17. Rfe1 Rb8 (17... a6 {may be best, intending ...b5.}) 18. Bf1 $1 $14 Nd6 19. Nb5 (19. Be3 {is also good.}) 19... Nxb5 20. Bxb5 Kf7 $6 (20... e5 $142 $1 {isn't so easy to play, giving up the a2-g8 diagonal, but it's probably best.} 21. Be3 (21. Bc4+ Kh8 22. f5 a5 $1 23. a4 b6 $11) 21... Be6 22. Nc5 Bxc5 23. Bxc5 Rfd8 $11 {/?}) 21. Be3 a6 22. Be2 $1 $14 {/? The c5 square beckons White's pieces.} b5 23. Nc5 (23. Kf2) (23. Rc1) 23... Rd8 (23... e5) 24. Rxd8 Bxd8 25. Rc1 $16 {Threatening Nxa6. Black's position is unpleasant, and getting worse.} Bb6 (25... e5 $1 26. Nxa6 Bxa6 27. Rxc6 Bb7 28. Rd6 Bxe4 29. fxe5 fxe5 $16) 26. Kf2 Bxc5 27. Bxc5 Bb7 (27... Bd7) 28. Ke3 Rd8 (28... Rc8 $142) 29. Bb6 (29. a4 $18 {was best, loosening Black's queenside. Between White's bishop pair, extra space, and his queenside targets his advantage is probably decisive.}) 29... Rd7 30. Rd1 $6 {Missing the moment. The swap helps Black, and so does the extra time to set up a defense.} (30. a4 $1 bxa4 31. Bc4 $18) 30... Rxd1 31. Bxd1 g6 (31... Ke8 $1 {made sense, looking to defend the pawn from d7 instead of f7, if only to avoid Bb3 + f5 problems.}) 32. Bc5 (32. Bb3) (32. a4) 32... h5 (32... Na5) 33. b3 $6 {The pawn isn't well-placed here, and may just help Black get counterplay at some point with ...Na5, looking to meet a subsequent b4 with ...Nc4+.} (33. Bb6 $16) (33. Bd6 $16) 33... Bc8 $14 (33... e5 $11 {/?}) 34. a4 e5 $1 $11 35. axb5 axb5 36. f5 Kg7 $1 {Among other things, this gives Black the option of playing ...g5 or ...gxf5 without allowing Bxh5 to come with check. We'll see still another point in a moment.} 37. fxg6 {Not with check, thereby allowing} Bg4 $1 ({That said, Black keeps equality after the routine} 37... Kxg6 {as well.}) 38. Bc2 Be6 {Not allowing Bd3, hoping to bring the bishop to c4 after ...b4.} 39. Kd2 b4 $1 40. Bd1 Kxg6 {The first half of the game was very well played by Gukesh, but having outplayed Nepo and achieved a winning position he let his opponent off the hook when he waited too long to play a4.} 1/2-1/2 [Event "FIDE Candidates 2024"] [Site "Toronto CAN"] [Date "2024.04.06"] [Round "3.2"] [White "Abasov, Nijat"] [Black "Nakamura, Hikaru"] [Result "1/2-1/2"] [ECO "D13"] [WhiteElo "2632"] [BlackElo "2789"] [Annotator "Monokroussos,Dennis"] [PlyCount "57"] [EventDate "2024.04.04"] {[%evp 0,57,26,27,27,-14,19,22,21,37,38,15,23,28,42,40,30,25,25,25,22,22,13,11,10,15,27,24,19,5,2,-18,-6,-15,6,0,0,3,4,5,6,5,6,-3,0,0,3,3,1,-5,30,33,16,3,21,14,20,5,14,20]} 1. d4 d5 2. c4 c6 3. cxd5 {Very disappointing. Realistic, perhaps, but still disappointing. Abasov is a heavy underdog in the Candidates, to put it mildly: there's less than a 1% chance, given the ratings, that he will win the event. Still, why not play as if you're a genuine contender, at least until too many losses pile up? His play in rounds 1 and 3 suggest that even with the white pieces the summit of his ambition is to make a draw (at least unless his opponents self-destruct). That's not an amazing achievement for a player of his level. Maybe he's hoping to squeeze a few rating points out of the event. Okay, but to what end? Let's say you're a tournament organizer, and you're running a prestige event and see that Abasov would make a nice fit by rating as one of the lower seeds. Seeing how he played when given a chance to face the world's best, would you think that this is someone who will increase the excitement of your event? Not likely!} cxd5 4. Nc3 Nf6 5. Nf3 {Oof - not even the most interesting way to play the Exchange Slav.} (5. Bf4 Nc6 6. e3 {has more dynamic potential, intending at times to play Nge2 (the knight might go to g3, and the f-pawn may go to f3) after Bd3.}) 5... Nc6 6. Bf4 a6 7. e3 {Another cooperative move. When I was a Slav player I used to love seeing this move, and felt that from here on out I was playing with house money. There would be some small chances to win, essentially no chance to lose, with a draw still the likeliest result. In fact, Black has a plus score from here.} (7. Rc1) 7... Bg4 8. h3 Bxf3 9. Qxf3 e6 {Black will play ...Bd6, and after the bishops are traded there's no danger to him whatsoever. Abasov is a competent player, of course, so there's really no danger to him either. The remainder sees the players swapping pieces and bringing the game to a speedy end by means of a repetition. (They'd otherwise have to pretend to keep trying until move 41.)} 10. Rc1 Bd6 11. Bd3 Bxf4 12. Qxf4 O-O 13. O-O Rc8 14. Qf3 Qd6 15. Qe2 Nd7 16. Qd2 $146 Ne7 17. Ne2 g6 18. Rxc8 Rxc8 19. Rc1 Rxc1+ 20. Nxc1 Nc6 21. Nb3 Qb4 22. Qxb4 Nxb4 23. Bb1 Nc6 24. Kf1 b6 25. Bd3 Nb4 26. Bb1 Nc6 27. Bd3 Nb4 28. Bb1 Nc6 29. Bd3 1/2-1/2 [Event "FIDE Candidates 2024"] [Site "Toronto CAN"] [Date "2024.04.06"] [Round "3.3"] [White "Firouzja, Alireza"] [Black "Caruana, Fabiano"] [Result "1/2-1/2"] [ECO "B30"] [WhiteElo "2760"] [BlackElo "2803"] [Annotator "Monokroussos,Dennis"] [PlyCount "76"] [EventDate "2024.04.04"] {[%evp 0,76,26,37,40,61,61,52,48,49,68,76,69,66,80,59,60,54,96,42,38,39,42,39,50,36,46,52,53,25,27,27,27,39,48,44,44,31,24,33,8,15,57,57,50,59,66,57,51,11,5,-2,12,31,50,58,71,69,56,64,63,53,55,44,48,39,38,30,52,38,34,18,26,31,38,19,26,27,22]} 1. e4 c5 2. Nf3 Nc6 3. Bb5 Nf6 {This used to have a dubious reputation, and was considered inferior to the more usual 3...g6 and 3...e6. If this tournament has shown us anything over the first three rounds, it's that seemingly unplayable or dubious lines can often be rehabilitated.} 4. Nc3 Nd4 5. e5 Nxb5 6. Nxb5 Nd5 7. c4 (7. Ng5 {was considered a huge problem for Black around the turn of the century. It's a tricky line, but solutions have been discovered.} f6 {is the first step in the most reliable recipe, when Black should be prepared for 8.Qf3, 8.exf6, 8.Ne4, and 8.Qh5+ g6 9.Qf3.}) (7. O-O $1 $14 {is one of White's most successful options and is the computer's top choice. Firouzja himself used this against Rapport in 2021.}) 7... Nc7 8. Nc3 {Caruana was on his own by this point.} (8. Nxc7+) (8. a4 $5) 8... d5 (8... d6 $1 {was the right way, with equality.}) 9. exd6 (9. d4 $1 {was better, and the reason to prefer 8...d6.}) 9... exd6 $6 {Also chosen in the two other games to reach the position after 9.exd6 (though none involved titled players).} (9... Qxd6 {looks more natural to me, joining the fight for the d4 square. Maybe Caruana didn't like} 10. d4 cxd4 11. Qxd4 Qxd4 12. Nxd4 {, intending to put a knight on b5 (and maybe to first hit the knight with Bf4). It turns out to be harmless though, e.g.} Bd7 13. Bf4 Rc8 14. b3 a6 15. O-O f6 16. Bxc7 Rxc7 17. Nd5 Rc8 18. Rfe1 e5 $1 {Nice, but not necessary - and Black had equally playable alternatives leading up to this point as well.} 19. f4 Bc5 20. Rad1 Kf7 21. fxe5 Rhe8 $44) 10. d4 $14 {A novelty, but not a meaningful one. In the two other games to see 9...exd6 White castled, and after 10...Be7 played 11.d4, transposing.} Be7 11. O-O O-O 12. Bf4 Ne6 13. Be3 Nc7 $1 $146 {A clever move. If White's bishop returns to f4, Black's knight will return to e6. In the meantime, it gets out of the c8-bishop's way and prepares (in some cases) to play ...d5.} (13... cxd4 14. Nxd4 Nxd4 15. Bxd4 Be6 16. Qd3 Bh4 17. Rae1 Re8 18. Re4 f5 19. Re2 Qd7 20. g3 Bd8 21. Rfe1 Ba5 22. a3 Bf7 23. b4 Rxe2 24. Rxe2 Bd8 25. Nd5 b5 26. Bc3 Qb7 27. Qxf5 bxc4 28. Nf4 Be7 29. Ne6 Bg6 30. Qf4 h5 31. Nxg7 {1-0 Hoogkamer,M (1767)-Van Rooijen,M (1785) FIN-NED email ICCF email 2015 [43]}) 14. h3 Bf5 (14... d5 $1 {was playable here.}) 15. Re1 Re8 16. Qd2 (16. Bf4 $142 {doesn't only activate the bishop; it also serves as prophylaxis against ...d5.} Ne6 17. Bg3 $11 {/?}) 16... d5 $1 $11 17. cxd5 c4 18. Bf4 Nxd5 19. Re5 {This enterprising move gives Black a choice of three plausible-looking moves. They're all good, but each leaves Black needing to solve further problems before reaching full safety.} Nxc3 (19... Nxf4 20. Qxf4 {attacks the f5-bishop but also threatens a big pile-up on the e-file. Here the solution is} Bg6 $1 21. Rae1 f6 $11 {, which allows the e7-bishop to move as the e8-rook is supported by the other bishop.}) (19... Be6 20. Bg5 Nxc3 21. bxc3 Bxg5 22. Nxg5 Bd5 $11) 20. Qxc3 Qd7 $5 {Many club players find a repertoire they like and stick to it forever, and don't pay much attention to games in other openings either. I think their digging deep roots in their favorite openings is smart - those who flit from one opening to another like insects pollinating one flower after another generally wind up with only the most superficial understanding of everything they play. Still, those who stick to their repertoire to the exclusion of everything else not only miss out on the joys of variety and the possibility that they will in fact like green eggs & ham; they often miss out on ideas that would help them with their favorite openings. Case in point here. Rather than decide which way of contorting himself to defend both bishops, the rook on e8, and the c4-pawn will be the least unpleasant, he cuts the Gordian knot. Saying goodbye to his c-pawn he puts an immediate end to all his defensive problems, recognizing that the ending he's heading for is akin to one in the Marshall Gambit that decades of practice has shown to be a comfortable draw. While Caruana rarely plays either side of the Marshall (especially in games with a slow time control), he's certainly well aware of it, and puts his understanding of that system to good use here, making an easy draw.} (20... Bd3 21. Rae1 f6 22. R5e3 {forces Black to cope with the threat of d5-d6. The best move here is the odd} Kf7 $1 $11 {, so the queen can take White's pawn if it advances to d5.}) 21. Qxc4 Rac8 22. Qb5 Qxb5 23. Rxb5 Be4 24. d5 a6 $1 25. Rxb7 Bxd5 26. Rd7 Be6 27. Ra7 Ra8 28. Rxa8 Rxa8 $11 {Black's bishops make the draw an easy one. Ideally, Black will advance his pawn to a4, when it will be impossible for White to make progress.} 29. a3 h6 30. Re1 Rd8 31. Be5 Rd5 32. Bc3 Bc5 33. Ne5 Kh7 34. Nf3 Bd7 35. Re5 Rd1+ 36. Re1 Rd5 37. Re5 Rd1+ 38. Re1 Rd5 1/2-1/2 [Event "FIDE Candidates 2024"] [Site "Toronto CAN"] [Date "2024.04.06"] [Round "3.4"] [White "Vidit, Santosh Gujrathi"] [Black "Praggnanandhaa, R."] [Result "0-1"] [ECO "C70"] [WhiteElo "2727"] [BlackElo "2747"] [Annotator "Monokroussos,Dennis"] [PlyCount "90"] [EventDate "2024.04.04"] 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 f5 {What?!! Even the normal Schliemann (3...f5) would be surprising in the Candidates, though it has achieved almost-normal status since Radjabov put it on map for elite players in the '00s. But this?} 5. d4 {The standard move agains the Delayed Schliemann. It's not particularly good against the (regular) Schliemann, while the usual anti-Schliemann weapons aren't so effective against the Delayed Schliemann. Anyway, 5.d4 has been known here for a very long time - there's nothing new yet.} exd4 6. e5 b5 7. Bb3 Na5 (7... Bb7 {is the usual move, though it's not very good. White should play} 8. a4 {to keep his bishop on the a2-g8 diagonal. Black's position is bad, if not yet losing.}) 8. Nxd4 Bb7 {This is only the fourth game to reach this position. Vidit, like everyone else, played 9.Nxf5, but with some outstanding preparation Pragg shows that it's not best.} 9. Nxf5 (9. Nc3 $1 $146 {? may be best. Note that Noah's Arking White's bishop will only make things worse:} c5 $2 10. Nxf5 c4 11. Qg4 g6 12. Bg5 $1 Qc7 13. Nd6+ $1 Bxd6 14. exd6 Qxd6 15. O-O $18 {White will be attacking with five pieces, including all three heavy pieces, and Black's uncastled (and "uncastleable") king will not survive without suffering serious material losses.}) 9... Nxb3 10. axb3 d6 $1 {Played once before. Congrats to the untitled Victor Muntean for coming up with this very nice move, which he followed up correctly.} 11. Qe2 $146 (11. exd6 Qd7 $1 $11 12. Ne3 $2 Bxd6 $17 13. Nd2 $2 O-O-O $19 14. Nf3 {0-1 Ernst,T (2298)-Muntean,V (2067) Karlstad op 2023 (3)} Nf6 $19) 11... Qd7 $1 12. e6 Qc6 13. Bg5 (13. f3 $142 $11 {is a little unpleasant to play, but it made sense to put an immediate stop to Black's evil intentions along the diagonal.}) 13... g6 14. Ne3 h6 15. Bf4 (15. Bh4 $142) 15... Ne7 $15 16. c4 b4 {Not only keeping lines closed, but restricting the knight on b1. Vidit is getting a taste of his own medicine - recall his game with Nakamura in the preceding round, when it was Nakamura who, with White, found it very difficult to develop his queenside.} 17. Qg4 Qc5 18. O-O Bg7 19. Nd2 O-O-O $2 (19... O-O $1 $15 {looks a bit nuts at first sight, but when you ask how exactly White is going to attack Black's king it's hard to find anything resembling a persuasive answer.}) 20. h4 $2 (20. Nd5 $1 {was a strong move. Black can neither abide the knight nor safely take it with the bishop, so his next move is forced.} Nxd5 $8 21. e7+ Rd7 22. cxd5 h5 $1 23. Qg3 ({or} 23. Qxg6 Rxe7 24. Rfe1 $14) 23... Qxd5 24. Nc4 Rxe7 25. Bxd6 h4 (25... cxd6 $4 26. Nb6+ $18) 26. Qh3+ $14) 20... Rde8 $15 (20... Bxb2 $5) 21. Rad1 Bxb2 22. Nd5 {This was better two moves ago, but it's still a good idea.} Nxd5 23. cxd5 Bxd5 24. e7+ $6 (24. Rfe1 $1 Bc3 25. Ne4 $1 Bxe4 26. Rxe4 Qh5 $1 27. Qxh5 gxh5 28. Bd2 $15 {White should probably make a draw here.}) 24... Kb8 $17 25. Be3 Qb5 (25... Qc6 $142 $17 {/-+}) 26. Nc4 Bc3 27. Bd4 Rhg8 $2 (27... Rh7 $1 28. Nxd6 $1 cxd6 29. Bxc3 bxc3 30. Qxg6 Bxg2 $3 31. Kxg2 Rexe7 32. Qxd6+ Rc7 $1 $17 {is not a simple line to work out, but it does leave Black with a substantial plus.}) 28. Bxc3 bxc3 {Here White has one move that equalizes; alas for Vidit, it's highly counterintuitive.} 29. Qd4 $2 {A very natural move, but a missed opportunity.} (29. Na3 $3 Qb7 30. Rd4 $1 c5 {Preventing 31.Rb4 and gaining a tempo, but weakening the d-pawn.} 31. Rd3 Rxe7 32. Qg3 Rd8 33. Nc4 Re6 34. Rxc3 $11 (34. Na5 $11)) 29... Bb7 $6 (29... Ba8 $1 {was better, the main point being to meet} 30. Rfe1 {with} (30. Qxc3 Rxe7 $17 {/-+}) 30... Qb7 31. f3 Qa7 $17 {/-+ , ensuring that White will only win the c-pawn at the cost of a queen trade and his e-pawn.}) 30. Qxc3 $15 Rxe7 31. Na5 $1 {Vidit's position is objectively better than it was, but despite the engine evaluation White's position remains extremely difficult. He's down a pawn, his king is weak, and Black's rooks are more active.} Re5 32. b4 $2 {This is a useful move if White can have it for free, but he can't. The cost is an important tempo that gives Black's attack wings to fly.} (32. Nxb7 $142 Qxb7 33. Rd4 $1 $11 {/?}) 32... g5 $17 (32... Ba8 $1 $17 {/-+}) 33. h5 $6 (33. Nxb7) 33... g4 (33... Ba8 $142 $1 $19) 34. Rfe1 $2 $19 (34. Rc1 $142 Qb6 35. Nxb7 {was the last chance to (kinda, sorta) keep the game going.}) 34... g3 $1 35. Nxb7 {Too late.} (35. fxg3 Qb6+ 36. Rd4 (36. Kh2 $2 Rxh5#) 36... Rxg3 $19) 35... gxf2+ $1 36. Kxf2 Qxb7 (36... Reg5 $1) 37. Qf3 Reg5 38. Qxb7+ Kxb7 {The rest is a mopping-up operation that Praggnanandhaa handles masterfully. (Or rather, supergrandmasterfully.)} 39. g4 Rxg4 40. Re6 Rh4 41. Rxh6 Rg5 42. Rh7 {This allows Black to force a trade of rooks, though it's not as if there was some good alternative waiting in the wings.} Rhxh5 43. Rd7 Kc8 44. Re7 Re5 45. Rg7 Rhf5+ $1 {Black plays 46...Rg5+ next, swaps the queens, and wins the resulting endgame with ease.} 0-1