The strongest tournament ever. 4/4. What can we say about Fabiano Caruana that has not already been said about Brock Lesnar? They even look quite similar...
by GM Jan Gustafsson
Maurice Ashley's Freudian slip, thanking "World Champion Magnus Caruana" at the end of his interview with Carlsen on the live show sums it up quite well. But let's not get ahead of ourselves.
After the insanity of the first three rounds of the Sinquefield Cup, Round 4 looked more like your regular supertournament. Two Ruy Lopez main lines, a slightly offbeat opening by Magnus Carlsen, two draws and one White win.
What is not regular is for fans to have the privilege of witnessing a run that the chess world only gets to see every few years: Kasparov Linares 2001, Topalov San Luis 2005, Carlsen Nanjing 2009, and now Caruana St.Louis 2014?
Caruana has me lost for words. Fortunately, I know a guy who can tell us how he does it. Calling Peter Svidler!
1. e4 e5 2. ♘f3 ♘c6 3. ♗b5 a6 4. ♗a4 ♘f6 5. 0-0 ♗e7 6. d3 b5 7. ♗b3 0-0 8. ♘c3 d6 9. a3 Another day, another fashionable line of the Ruy Lopez. And this time, one I have some experience with — I might even dedicate a series of mildly entertaining videos to it, at some undefined point in the future. Black has a lot of options here. I (famously?) played 9...Bb7 against Magnus in the last round of the 2013 Candidates, but was then shown the error of my ways by Mickey Adams in the Alekhine Memorial 2014. I also played 9...Be6 a number of times — with decidedly mixed success.
11. ♗g5 c5 12. b4 ♘c6 13. ♘d5 ♗xd5 14. exd5 ♘d4 15. bxc5 ♘xf3+ 16. ♕xf3 dxc5 17. ♖fe1 ♘d7 18. ♗d2 ♗d6 Dominguez Perez-Aronian, Wijk aan Zee 2014. This position is far from clear, but Leinier misplayed it badly and was soon in a lot of trouble: 19. a4 f5 20. ♗b3 e4! 21. dxe4 c4 , and Black is suddenly in control.
13. ♗d2 d5 The in-vogue approach. Rustam Kasimdzhanov's choice gives Black a more fluid structure, and therefore appeals to me more on aesthetic grounds — but it cannot be an accident that players as strong as Aronian and Karjakin are not repeating it.
14. ♖e1 ♕d6 15. ♘a2! The exclamation mark would still apply even if this move wasn't very good — and on first impression, it most likely is. The weirdness factor alone would be more than enough. White replies to Black's build-up in the center by embarking on a three-move knight manouevre which will take it out of action completely until it arrives on b3 — admittedly, by far the best square for it.
15. h3 ♘d7 16. ♘e2 a5 shows Black's plan if White plays normally - after 17. ♖b1 axb4 18. axb4 ♖fb8 19. ♘g3 d4 20. c3 dxc3 21. ♗xc3 ♖a4 22. ♕b3 ♔h8 the position is totally unclear, and Black eventually won in Dominguez-Karjakin, Beijing SportAccord Rapid 2013
15... ♘d7 Levon decides to stick with the original plan. Watching this in real time I felt Black needed to aim for Nd4, with or without dxe4. Making it work is not that easy, though — in fact, in the cold light of morn I failed utterly, and am not even including those attempts in the notes. However, the direct
15... a5⁉ definitely deserved attention. The following is a brief outline of the possibilities of this position — there is a multi-fandom Doctor Who fair in town, and much as I would like to spend a couple of hours in search of absolute truth here, the higher calling is winning out. As things stand, I think White retains some pressure after 16. Qe2:
a) 16. exd5 exd5 17. ♘c3 ♘g4⁉ (17... axb4 18. ♘xb5 ♕d7 19. axb4 e4 20. ♖xa8 ♖xa8 21. ♘bd4 ♘xb4 seems to be OK as well, but Ng4 is much more fun) 18. ♘xb5 (18. h3 ♘xf2! 19. ♔xf2 axb4 20. axb4 ♖xa1 21. ♕xa1 ♘d4!→ ) 18... ♕g6 19. ♕e2 (19. ♘xc7 e4 ) 19... axb4 20. ♗xb4 (20. ♘xc7 ♗c5 ) 20... ♘xb4 21. axb4 ♖xa1 22. ♖xa1 e4 , and Black is doing fine
b) 16. ♘c3 ♖fb8 (16... axb4 17. ♘xb5 ♕d7 18. axb4 ♘xb4 19. ♖xa8 ♖xa8 20. ♘c3 is not a very happy outcome for Black — the pawn on e5 is in trouble) 17. exd5 exd5 18. bxa5 ♖xa5 19. ♘xe5! Otherwise Black is quite comfortable. 19... ♘xe5 20. ♗f4 , and now Black needs to make a choice: 20... ♘fd7 is a more adventurous approach: after (20... ♕d7 21. ♗xe5 ♖ba8 22. ♘e2 ♖xa3 23. ♖xa3 ♖xa3 should be holdable, although White still retains some pressure due to a much safer king) 21. d4 c6 22. dxe5 ♕e6 23. ♘e2 ♗c5! 24. ♗d2 ♖a4 25. ♗b4 ♖f8! 26. ♗xc5 ♘xc5 27. ♘d4 ♕g6=/∞ Black has decent play for the pawn
c) 16. ♕e2⁉
c2) 16... axb4 17. ♘xb4! (17. axb4 d4 18. ♖eb1 ♘d7 , and the weakness on b4 hinders White's plan greatly) 17... ♘d7 runs into 18. exd5 exd5 19. d4! (19. ♗c3 ♘xb4 20. axb4 ♗f6= ) 19... e4 20. ♕xb5 ♘xb4 21. ♗xb4 c5 22. dxc5 ♘xc5 23. ♘d4 , and Black does not have full compensation for the pawn.
16. ♕e2 d4 I always tried to steer clear of this structure when playing these types of RLs with Black, and making this move in particular would fill me with dread. Black is totally committed to creating something on the queenside, and soon — because if he doesn't, he will be stuck with a totally immobile pawn structure and passive pieces tied to the protection of it (as the game amply demonstrates). However, it is very easy to underestimate White's next move (that weirdness factor again).
17. ♖eb1! With the center now completely static, White switches his full attention to the prevention of pawn breaks on the queenside. The casual
17... ♘b6 This is too compliant, I think. Once the knight gets to b3, Black is essentially an observer, waiting for White to formulate a plan, and hoping to be able to parry it somehow. After
17... ♖fb8 18. ♘c1 (18. c4 dxc3 19. ♗xc3 ♘d4 is far from clear) 18... a5 White is probably better, but at least Black has his own trumps: 19. bxa5 ♘xa5 20. a4 b4 21. ♘b3 ♕c6 (21... h6 22. ♗e1 c5 23. ♘fd2 ♘b6 is another possibility — the weakness on a4 gives Black reasonable counterplay.) 22. ♗g5 (22. ♗e1 ♘b7 23. ♘fd2 ♖xa4 24. ♖xa4 ♕xa4 25. ♖a1 ♕c6 26. ♘c4 gives White ample compensation for the pawn - but Black should be OK: 26... ♘d6 27. ♘ba5 ♕a6 ) 22... ♗xg5 23. ♘xg5 h6 24. ♘f3 ♘b7 25. a5 ♖a6 26. ♘fd2 ♘d6 27. ♕g4 ♖e8 28. ♖c1 ♖e7 , and compared to the game, White's control is far from absolute.
18. ♘c1 ♘a4 19. ♘b3 ♖f7 From this moment on, it is hard to comment on the game for a stretch, due to Black's total lack of constructive options. We have courtside seats for a Fabiano clinic on how to exploit this advantage, though. If Black wanted to pass, you could argue he should have started with 19...h6, covering the g5-square — but since one of White's plans is h4-h5, h6 is hardly an attractive option either.
19... ♘c3 was the last active try available to Black, and I think he should have gone for it: 20. ♗xc3 dxc3 21. ♖d1 (21. ♕e1 ♘d4 leads to a similar position) 21... ♘d4 22. ♘fxd4 exd4 23. e5 ♕d5 24. ♕e4 ♖ad8 25. ♕xd5 ♖xd5 26. ♖e1 c5 27. ♖e4 , and although the computer prefers White, I think Black should be able to hold this — and in any case, I would prefer any semblance of double-edgedness to what followed in the game.
20. ♖c1 The rook won't stay long on c1, but the threat of c3 forces Black to spend a tempo on Rd8, allowing an important regrouping.
20... ♘c3 21. ♕e1 ♖xf3 22. gxf3 ♖f8 injects some excitement — but it's hard to believe Black should be desperate enough for that at this point... However, 21. ..Na4, asking White to at least find something more useful than 22. Qe2, was worth considering. I guess Levon just did not like the change of structure arising after Bxc3 dxc3, followed by a knight trade on d4...
21. ♘g5 ♖f6 I may be guilty of results-oriented thinking here, but I would probably prefer to trade some pieces — it's getting rather crowded in Black's camp. However, the horrible e7-bishop could be very useful later on when the game opens, and you need to protect all those weak pawns parked on dark squares...
28... ♗g7 Black is almost there — if he gets in Qd6, his position will be very hard to break. However, he is not quite in time.
29. ♘a5! Fabiano is quick to spot a way to exploit the temporary increase in Black's discoordination, and the whole edifice is about to come crashing down. After
29. ♘h2 Black would have a sad choice between 29... g5 and (29... ♔h7⁉ 30. ♘g4 ♖f4! 31. ♗xf4 exf4 32. ♕f3 e5 , closing the position down entirely and hoping it will hold, even a full exchange down (I suspect Levon was rather hoping for this — it would appeal to him greatly, with its Petrosianian spirit). 30. h4⁉ gxh4 31. ♕g4 ♔h7 32. ♕h5 ♖g6 33. ♘g4 , and despite an extra pawn, Black has very little in the way of productive plans. However, the text is much stronger.
39. ♕g6 The sheer mass of White's pawns makes defence impossible here.
50... ♖xf7 51. ♖h7+! both of his knights are lost. A very impressive victory for Fabiano — a stunning new geometrical idea in a position where players tend to go for age-old templates without much pause for thought, and flawless conversion once Levon did not react to it in the best way. On a more personal note — back in 1995 (I am tempted to say 'before Fabiano was born', but poetic license does not stretch quite THAT far) I, too, started a tournament with 4/4, and a 3K+ performance. I have a nagging suspicion, however, that my achievement then is somewhat less impressive than what FC is showing so far in St. Louis. In fact, I will probably never be able to look back at it with the same fondness now. DAMN YOU!
The World Champion was not all that pleased with his effort.
Today was not real chess. It was just moving the pieces around.
Just my kind of game!
3... ♗b4 4. ♘f3 A line we have seen Carlsen use quite a bit when in "d4-mainline mode". He lost to Naiditsch at the Olympiad with it recently, but only after getting a great position and showing a rare interesting idea in one main line. How about the other one?
4... b6 The move that made me stop using 4. Nf3. Black has been in excellent shape here theoretically.
5. ♗d2⁉ Whoa! This move is about as popular as Michael Bay at a movie critics convention. What is the point? For starters, it avoids doubled pawns and breaks the pin. This allows White to develop with g3 and Bg2, establishing a standoff on the long diagonal — a relationship that tends to favour the g2-bishop. On the flipside, Bd2 is slow. It doesn't prepare e4, it doesn't pin the knight, and it might have to move again later. We haven't seen this move much since the days of Smyslov and Petrosian for a reason. I am still intrigued, and I do believe that this is not just a "let's play something rare" approach, but the result of serious work. It could well mark the beginning of a new trend.
5. ♗g5 used to be the move here, but it is deeply analysed nowadays with Black doing fine after e.g. 5... h6 6. ♗h4 g5 7. ♗g3 ♘e4 8. ♕c2 ♗b7 9. e3 d6 10. ♗d3 ♗xc3+ 11. bxc3 f5 12. d5 ♘a6 1/2-1/2 Nakamura,H (2751)-Anand,V (2810), Wijk aan Zee 2011
5. ♕c2 used to be Magnus' weapon of choice. The following game sums up the state of theory quite well: 5... ♗b7 6. a3 ♗xc3+ 7. ♕xc3 0-0 8. ♗g5 d6 9. ♘d2 ♘bd7 10. f3 ♖c8 11. e4 h6 12. ♗h4 c5 13. ♗d3 cxd4 14. ♕xd4 ♘c5 15. ♗e2 e5 16. ♕f2 ♘e6 with double-edged play in Carlsen,M (2826)-Karjakin,S (2763), Moscow 2011
6... c5 An important decision, forcing White to clarify the situation in the center before he is developed. Very logical. However, I would not be surprised if this is already the spot where people will choose to deviate in future, as Carlsen got a very reasonable position soon.
7. dxc5 What else?
7... bxc5 leads to a structure where Black would much rather have his bishop back at home on e7, guarding the d6-square.
10... ♘a6 We have arrived at the well-known theoretical position mentioned on the third move, with a minor difference: the white queen is on d1 instead of a4! This has its pros and cons. Pro: Nc5 will not be coming with tempo, forcing the queen to move. Con: The useful Rfd1 is not legal in this position.
11. ♘b5 Very direct play. With the knight threatening to jump into d6, Topalov is forced to react immediately. There were some quiet alternatives which we might see in future:
11... d5 Forced.
12. ♘e5 Magnus can be happy with the outcome of the opening. Black has to be precise to neutralise the active white pieces.
12... ♘c5 Covering b7 and abandoning the rim is a good start.
14... dxc4 Topalov correctly decides to release the tension and go for the ending. Due to the slight weakening of the black queenside, only White can be better, but creativity will be required to prevent the game from petering out.
17. ♘c6⁉ Here comes the spice! Carlsen decides not to recapture the pawn, trying to paralyse the black pieces instead.
17... ♖de8 After 6 minutes of thought, Topalov goes for this clumsy-looking move. It does keep the extra pawn at the risk of giving the f8-colleague a mild claustrophobia attack. The alternative was
18. ♖fd1 Carlsen's idea is becoming clear. He is a pawn down, but try to suggest a sensible move for Black!
18... h6! The most sensible there is.
19. e4⁈ The obvious follow-up. By threatening e5, Carlsen initiates a brief infight. I don't like it, though! He could have asked Topalov to come up with another useful move by going
19. ♔g2! Not an easy task at all - the best Compy comes up with is 19... ♔h7 when the "I can pass more easily than you" game could have been taken further with (19... ♗c5 does not work as long as there is no target on e4 20. ♘a4 ) 20. h3!
19... ♗c5! Veselin happily grabs the chance to untangle, even if this means giving back the extra pawn.
23. ♗e3⁉ This might look ugly, but it is a touch of class. The bishop wasn't doing much on f4, so it is exchanged for its colleague, complicating Topalov's task.
26. b3 As Carlsen has compromised his pawn structure, we have the familiar situation that he has to play dynamically — opening the position — before his static weaknesses become a factor.
27... ♖c7 should also keep the balance.
37... ♖b5 38. ♖d8 ♘c5 39. h4 He keeps continuing in French Montana "I ain't worried about nothing" style. The balance is not disturbed yet, but the risks of Carlsen's lust for fighting outweigh the rewards this time.
40. ♔g2 still kept things at 0.00.
40... ♔d6! Problems! a7 or b8, the caves of Moria or the pass of Caradhras?
41. ♘b8 The caves it is. Let's hope the black king does not turn into a Balrog...Ok, I'm done with the LOTR stuff.
43... ♖b5! Eyeing Alfred pawn before going after its big brother.
44. ♔e2 was more precise. The difference is shown in the line 44... ♘e6 45. ♘xa6 ♖xa5 46. ♘b4 ♖b5 47. ♘c2 ♖b2 where having the king on e2 allows 48. ♔d2 with a helping hand. No such thing in the game.
45... ♖c5! Carlsen's life would be hanging by a thread, but even here he is not lost yet! 46. ♖xc5! (46. ♖e8 ♖xa5 is hopeless.) 46... ♘xc5 47. h5! Strictly the only move! The black pawns have to be fixed to provide future inroads for Harald V. (47. ♔g3 g5! ; 47. g5 h5 48. ♔g3 ♘d3! 49. ♘xa6 g6! and the Dark Knight does the amazing job of restricting both king and knight.) 47... ♔c7 48. ♘xa6+ ♘xa6 49. ♔g3 ♔d6 50. ♔f4 ♘c5 51. ♔f5 and the active white king combined with Alfred gives him just enough play to survive. 51... ♔e7 52. ♔e5
48... ♔c6 49. ♘b4+ ♔c5 50. ♘c2 ♔c4 51. ♔e2 g5 52. hxg5 hxg5 53. ♔d2 ♘c5 54. ♘a3+ ♔b4 55. ♘c2+ ♔c4 56. ♘a3+ ♔d5 57. ♘b5 ♘d7 58. ♘c3+ ♔e5 59. ♘d1 ♘f6 60. ♘f2 ♔d5 61. ♘h3 ♘xg4 62. ♘xg5 ♘xe3 63. ♘xe4 and the resources have been fully exhausted.
I feel bad for not annotating the Frenchman's game for the second day in a row. He is in clear second place!MVL could easily have been the MVP so far if it weren't for that American-Italian from Madrid. His game against Nakamura was what we call a
In a well-known line of the Ruy Lopez, Vachier came up with the novelty 17.Nd2.
Nakamura reacted precisely, starting with 17...Rb8! and after some exchanges, a balanced position occurred. The players shook hands on move 30.
You can check out the whole game here.
After the game, Nakamura said he remained unfazed by Caruana's massive lead after four rounds.
"I just have to beat him."
Yes, Caruana is two points ahead of the field.
Winning all his games keeps helping his rating, too. The gap to the top is still significant, but it sure did shrink!
Man, am I hooked on this tournament. More tonight!
Don't miss the live commentary with GMs Maurice Ashley, Yasser Seirawan and WGM Jennifer Shahade, here on chess24!
To pass the time, why don't you check out the full show of Round 4?
We respect your privacy and data protection guidelines. Some components of our site require cookies or local storage that handles personal information.