The last couple of rounds have seen lots of draws in the classical games, with players preferring to try their luck in the rapid and blitz tiebreaks. In today's action, at least, this was not the case: everyone playing White tried to make something of it, though only one player succeeded.
That player was Vassily Ivanchuk, who defeated Anish Giri, though the connection of his win to his possession of the white pieces was tenuous. He did obtain an advantage against Giri's Petroff, but his weird 15th and 17th moves flipped the evaluation, and he was in serious trouble. But then Giri started doing strange things, and frankly both players made lots of errors, possibly due to time trouble. The last serious error was 34...Qf4+, allowing Ivanchuk to trade queens and reach an easily won rook endgame. After Ivanchuk's 41st move, the players had time to take stock, and Giri gave up.
The day's other winner was Vladimir Fedoseev, who defeated Maxim Rodshtein (who may have been a little rusty and emotionally out of sorts after receiving a de facto walkover thanks to "Shortsgate". As with Ivanchuk-Giri, there was no logical line between the opening and the first player to achieve an advantage and the game's result. After Fedoseev's dubious pawn sac on move 22 Rodshtein was better, but White's repeated decision not to initiate the exchange of rooks eventually let his advantage slip away. Even after that the game remained in a precarious balance until Rodshtein's 35.Nc4? missed a nice tactic that had been looming for a while. Fedoseev spotted it, and that clinched it. Again as in Ivanchuk-Giri, once the winner had made his 41st move and time trouble was no longer a factor, it was time to resign.
The other six games were drawn, but all were interesting. Peter Svidler played the Bishop's Opening against Bu Xiangzhi (to avoid the Petroff), but couldn't achieve an advantage and the game was eventually drawn; if anything, Black was a little better through a fair chunk of the middlegame.
Wesley So vs. Baadur Jobava was a Petroff, and Jobava was well-prepared. So's 11.h4 was a rare move, and it was well-met by Jobava's new move, 11...Bc5. It's not clear if there's any advantage to be had for White; if so, it's not with 12.Bd3. Jobava had no problems, and while both players fought well and tried to make something happen, the game never got out of balance.
Richard Rapport and Evgeniy Najer played the longest game of the round. When there were chances, Najer had them, and after Rapport's 44.Nf1 Najer's winning chances were very good. Perhaps 45...Bd3 would have led to a win, and 46...Kf6 would also have given him good chances for the full point. By White's 49th move, however, the draw was an inevitability, provided Rapport stayed alert - and he did.
In all the draws thus far, Black has done very well, and that was also the case in Daniil Dubov - Levon Aronian. Dubov's 20.d5 was too optimistic, and had Aronian played 22....Bd5, or later 30...Qd7, or especially 34...Rxb2, it's quite likely that he would have won. Luckily for Dubov, Aronian played 34...Qf6??, and two moves later the game was drawn.
Wang Hao and Ding Liren played a "correct" draw. Ding was well-prepared on the black side of a Meran, and made a comfortable draw.
Finally, the draw between Maxime Vachier-Lagrave and Alexander Grischuk was anything but comfortable. Vachier-Lagrave went all-out for the attack, sacrificing a rook. The idea was sound, but his 28th move was objectively an error, though not one that was easy to refute. Grischuk very understandably looked for a way to achieve safety, and he found it. He returned the material, with a little interest thrown in, to achieve an easily drawn rook and two pawns vs. rook and three pawns ending, with all the pawns on the same side of the board.
Here are the games, with my comments.