Five draws in Round 6 of the 2021 Sinquefield Cup, despite some pretty hard-fought battles, left the standings intact, but the pace picked up again in Round 7, with two of the favourites winning crucial games and Maxime Vachier-Lagrave emerging as the sole leader before the last sprint.
You can replay all the games from the Sinquefield Cup using the selector below.
And here’s the Round 7 live commentary from Srinath Narayanan, Simon Williams and Andras Toth.
The Round 6 game that attracted the most attention was clearly the one between Maxime and Fabiano Caruana. It could be expected of Fabi to try and strike back with full force after his disappointing loss to Jeffery Xiong in Round 5, and indeed that was the case.
Playing White, Fabi opted for the same line Leinier Dominguez used to beat MVL a few days ago in spectacular style. Maxime seemed unperturbed and confident in his analysis of that game, and it was Fabi who deviated with the quiet but devious move 15.a3.
Both players more or less blitzed out their moves all the way to 26.f4, indicating deep preparation, and even Maxime’s 13-minute think for the ‘unnatural-looking’ 26…Bb5 was probably a case of trying to remember and verify, rather than to invent. It was only after 30.Qd2 that both sides were clearly on their own.
White wants to bring his queen to the commanding central outpost on d5 and create an unpleasant pin on the f7-rook, hoping later to break through with tactical means.
Maxime spent almost half an hour here, needing to decide about the optimal way to transfer his bishop from b5 to e6 via e2 and g4, as well as to create some space for his king with either …h6 or …h5. Very important in all this was, of course, to always have a defence against White’s tactical threats, most notably that of Nc6-e7+, followed by the fall of the d6-pawn. He turned out to be on the right path with his choice, and on move 34 Fabiano was called upon to make some decisions of his own.
It was around here that it dawned on him there is not much prospect for progress, as Black seems to have everything carefully defended. Caruana spent 51 minutes on his next move but could not find anything worthwhile, so he opted for a simpler position with 34.Nc4 Bxc4 35.Rxc4.
After Maxime’s 35…h4 36.Rc2 h3 had quickly created counterplay, the French grandmaster gave his opponent one fleeting chance, when he replied to 37.Ka2 with the outwardly logical move 37…Kh8, unpinning his rook to play …Rh7.
White could have played 38.Rh2 Rh7 here, and then a quiet but venomous move like 39.Qc4!, when it turns out Black is seriously short of constructive moves. Fabiano was concerned about counterattacking tries such as 39…Qg4 40.Bd2 Qg3, but this fails to 41.Qc8!. Instead, he decided to shut the game down with mass exchanges and a draw agreement shortly afterwards.
Leinier Dominguez and Peter Svidler were involved in a very murky and complicated game, starting with Leinier’s early Bg5 in the Italian provoking Peter to push his kingside pawns and temporarily leave his king in the centre. The ins and outs of this opening line are examined in detail in Jan Gustafsson’s Chessable course (“…e5, patzers!”), though not with this exact move order, but navigating such complex positions is never easy, even with prior knowledge and armed with analysis.
It’s not easy to pass judgement on various decisions, such as Svidler’s 12…h5 (12…Nf8 and then …Ng6 seemed more solid, but also slower), or Dominguez’ 19.dxe5 (the engines like 19.bxa6 bxa6 20.Bd5 c6 21.Bxe6 fxe6 22.Ndc4), but the game started to go Black’s way after the unexpected 20…0-0.
Leinier should probably have maintained the tension, as his choice of 21.Bxe6 was met by 21…fxe6!, denying the white e3-knight important squares (d5, f5). Peter quickly freed his position completely and grabbed the upper hand, but missed his chance after what turned out to be a big mistake by Leinier, 26.Qb3?.
Black could have won here with the very hard to spot 26…b5!, the point being that the path of the a7-rook to the kingside is now open, and 27.axb5 fails to 27…Nf3! 28.gxf3 gxf3, and Black is about to give mate with …Rg7+ and …Qh4. White would have to try something like 27.Nb6, when 27…Nf3+ is still pretty strong, but Black can do even better by playing to trap the b6-knight with 27…Na5 28.Qc2 Rb8.
Instead Peter played the natural 26…Raa8, which gave Leinier just enough time to exchange some pieces and cover his weaknesses, and by the time the two players repeated moves there was not much left to play for.
Jeffery Xiong, buoyed by his win against Caruana in the previous round, almost inflicted similar damage on Wesley So. A couple of inaccuracies on Wesley’s part in the quiet London System (12…Qe7?!) already put him close to the edge.
Here 15.Ne4! would have been very unpleasant, especially as Wesley’s intended 15…h6 runs into the powerful blow 16.Ba6! (16…hxg5 17.Nxf6+ Nxf6 18.Bxc8 Rxc8 19.b4 and the Bc5 is pinned). Jeffery missed this but retained pressure, augmented by the strong 27.d5!, which came at a moment when it seemed Wesley was close to finally equalising. But the follow-up, after 27…Bb7, was not easy to assess.
As it turned out, Jeffery was concerned about ideas like …f5-f4 and …e3, but at this right moment he could have increased his advantage with the surprising 28.Qd6!. Indeed, after 28…Qxd6 29.Bxd6 the passed pawn on d5 becomes very powerful, and the white pieces create lots of annoyances for Black, who is bereft of any meaningful counterplay.
Having failed to assess this properly, Jeffery went for the meek 28.Qxe4 Nf6 29.Qd3 Nxd5, and the game slowly but steadily drifted towards equality.
Dariusz Swiercz and Shakhriyar Mamedyarov battled it out in another Berlin, this time the quiet 5.Re1 variation. White always seemed to keep a nagging pull in the resulting symmetrical position, either thanks to his more active pieces or, later, due to Black’s inferior pawn structure. Shakh correctly decided to complicate by grabbing a hot pawn on a2, instead of defending passively, and once he managed to extricate his bishop from there in time the draw was never in doubt.
Nor was the game between Sam Shankland and Richard Rapport too exciting; Richard equalised comfortably with a rare setup in the Scotch, and even chased some slight edge for a while, after Sam made a couple of overambitious moves. This eventually came to nothing in the face of good, active defence, though Richard did miss a couple of opportunities to play for more, most notably one involving a pretty regrouping.
The ingenious 24…Nfe7! would have been strong here, intending to play …Ng8(!)-f6, and suddenly White’s position looks overextended and the h5-pawn in particular is a source of worry. Not easy to spot, but a beautiful positional idea!
With only three rounds left, and a fierce battle for first place developing, the results of Round 7 gave us a sole leader — but also intensified the pursuit of the leader. MVL often gets criticised for his stubborn opening choices, and going into the Berlin Endgame is one of those decisions; but in fact this decision often works out to perfection — more often than he gets credit for. On this occasion he steamrolled Jeffery Xiong in a mere 29 moves, and the whole game was essentially decided in one move.
Maxime tried to grab the bull by the horns with the very direct 18.f5, which should have been met by the precise 18…g6!, slowing White down. Maxime intended to continue in style with 19.e6 fxe6 20.Bd4 Rf8 21.fxg6, ending up with a very strong pawn on g7, but in fact it seems that the creative 20…Rg8! 21.f6 c5!, playing against the powerful bishop that holds White’s position together, would have given Black excellent chances to repel the onslaught.
Instead, Jeffery included the exchange 18…hxg4 19.hxg4 and then 19…g6, which ran into 20.Rh1! and Black has no time to dismantle White’s pawn front. Black’s position is already beyond repair, and Jeffery had to resign soon afterwards.
In what seemed to be a Round 7 thing, Peter Svidler scored his first win of the event (he was the only player remaining winless up to this point) against Dariusz Swiercz following a remarkably similar scenario. The players quickly went down another important theoretical Ruy Lopez main line, this time an Anti-Marshall heavily promoted by Jan Gustafsson in his course, and remarkably Peter (a good friend of Jan’s!) was the one to be caught somewhat unprepared after 17…Qa1 (a typical idea in this line).
Black is essentially trying to keep the white pieces occupied but is also taking some risks by sidelining his queen in this way, and Peter sought to exploit this by playing 18.Ne3, leaving the c3-pawn unprotected. Black could play solidly with 18…Bf6, but Dariusz, after 22 minutes of thought, chose the principled (but very risky) 18…Qxc3, which was met by 19.Nd5. Black is already walking a tightrope here, as his pieces are not well coordinated and somewhat loose, and choosing a square for the queen requires a lot of accurate calculation.
White is already better, it turns out, but 19…Qa1? was a serious error that could have been immediately punished by 20.Ba4!, followed by Re3-a3 trapping the queen. Instead, Peter blitzed out his initial intention of the immediate 20.Re3, which at least gave Swiercz the chance to do some damage limitation with 20…Qa8 21.dxe5 Bd8 — but White remains clearly on top.
Instead, Dariusz spent a whole hour on this and his next move, but 20…Qa7 21.Ba4 Bd8 22.Ra3 was already curtains, with the black pieces hopelessly tangled and material loss unavoidable. Peter concluded the game with a steady hand, just like MVL did; two games essentially decided in one move. Peter was visibly relieved afterwards.
Wesley So is leading the Grand Chess Tour overall standings and is guaranteed to win it if he finishes at least in third place. This probably affects his play somewhat, especially in the latter rounds of the event; a certain lack of sharpness and ambition often becomes evident in his decisions. His game against Sam Shankland was a case in point: having obtained a pleasant edge from the opening, with more active pieces and an extra central pawn, Wesley seemed a bit more reserved than the position demanded. One moment in particular was somewhat puzzling.
Sam has just played the very weakening 21…a5, and one would expect Wesley to immediately pounce on the newly created weaknesses with 22.Na4! — but instead he remained focussed on his long-term plan of advancing his kingside majority. Even then he seemed hesitant, and in the end he decided to not even try to make progress, in a position where under any other circumstances he would probably have kept on playing for quite a while.
Leinier Dominguez got the extremely rare opportunity, at least in top-level chess, of uncorking a big surprise as early as the 5th move, in a rare line of his pet Queen’s Gambit Accepted. The move 5…e5!? did not seem to catch Richard Rapport entirely unawares, though, and the players quickly moved along an original path where they both castled long. An important moment arose on the 18th move, after a repetition.
Richard decided not to repeat again with 18.Qf5+ Be6 19.Qc2 Bb3, and played on with 18.Qc3; but after 18…Qxc3+ 19.bxc3 Rxd2 20.Kxd2 Rd8+ 21.Ke2 Rd5 Leinier was poised to regain the pawn on e5 and stood no worse. Richard decided to pull the emergency break and exchanged down to an opposite-coloured bishops ending (even giving up another pawn in the process), one that was completely drawn.
The other player, apart from MVL, who benefitted from this round’s battles was Fabiano Caruana, who put paid to Mamedyarov’s GCT hopes with a nice, fighting win with Black. It all started with Shakh reverting to the 4.f3 Nimzo-Indian, a system very familiar to his opponent from the white side of the board as well. Fabi in fact followed a game he played in 2020 from the other side against Levon Aronian, deviating on move 10 by including the exchange on d5 before playing …Nh5 and …f5.
White’s recapture 11.exd5 came as a surprise (11.cxd5 looked more natural), and Caruana’s active pieces assured him of some initiative in the resulting position, with White having problems coordinating his pieces.
In such strange, complex positions it is very hard to understand exactly how to proceed, and passing judgement on the decisions of such strong players by simply comparing them with the choices of the engines is a notoriously treacherous path. However, one particular engine suggestion should be pointed out, if only because of its admirable consistency with the by now well-known trend of engines to push rook pawns all the way! On move 17, instead of allowing the game to open up, Fabi could have tried an interesting positional idea.
Here 17…g4!? would have kept both white bishops restricted, and after 18.b3 Ne4 19.Bb2, as Fabi assumed, the engine wants to play 19…a5, intending to push further with …a4-a3 or open the a-file.
The game went on, with both players slowly going down on the clock and, with lots of difficult decisions to make, the tension culminated on move 26.
Mamedyarov played the very unexpected 26.Qe2, walking into a discovered attack, when 26.Bxe4 Bxe4 27.Qe2 could have posed some questions without much risk. Fabi replied with the obvious and forcing 26…Nf6, and here Shakh made a crucial mistake in choosing a square for his queen with 27.Qd1?, which ran into 27…Bxd3 28.gxf6 Bxf1 29.Qg4+ Kh8 30.Rxf1 Rxf6, with a clear advantage for Black, which was eventually converted.
The correct move would have been 27.Qf2! instead; following the line above, the queen would have checked with 29.Qg3+ instead, leading to this position.
The big point would have been that White would be almost winning with 31.Be5!; thus 30…Qxf6 would have been preferable, when 31.Be5! again would be good, but only equal. Fabiano explained all the intricacies of this hard-fought game in his interview.
The penultimate round is the stage for the clash between Vachier-Lagrave and So, the two leaders of the overall GCT standings; this game could determine the eventual winner, both of the tournament and the Grand Chess Tour.
Don’t miss the live commentary from 16:00 ET/22:00 CEST each day!