In round 3 of the European Club Cup the top teams, all heavy-laden with 2700s, won their matches - handily. Better still, for them, there were almost no upsets on individual boards either. I only spotted one, but (to employ an archaic colloquialism) it was a real doozy. Another game seemed headed for an upset when the would-be David was killed by his own slingshot.
The successful upset saw Evgeniy Najer defeat Alexander Grischuk. Najer is no slouch, generally rated in the mid-to-upper 2600s, but he is certainly an underdog against Grischuk - especially with black. But in this game, he was the boss. He repelled Grischuk's attack and then took over, steadily gaining ground and dominating his elite opponent in the endgame.
The near-miss was Daniil Dubov vs. Li Chao. The young Russian held a pleasant edge for a long time, and it was starting to grow when, at move 37, he thought it was time for the kill. A brilliant kill, at that. He played 37.Rxe5, threatening 38.Re7 and accurately calculating that 37...Bxe5 38.Qe7+ followed by 39.Bxe5 would lead to a speedy massacre. What he failed to see, however, and with plenty of time on his clock, was the move Li Chao actually played: 37...Qxe5. Oops! Dubov resigned on the move, as the best White can do after 38.Bxe5 Bxe5+ 39.Rc3 Rc8 is a position where Black has a rook and two overpowering bishops against White's queen.
One other game I'll note wasn't an upset. Vladimir Kramnik defeated Ian Nepomniachtchi and moved to 6th on the Live Rating list. If qualifying spots for next year's Candidates' tournament were based solely on the current ratings, Kramnik would be in. It's not, however, and in addition to his having no chance at passing Veselin Topalov for one of the rating qualification spots I don't think he has much of a chance of passing Anish Giri for the second spot, either. But if anyone has more precise information on this, and what would be necessary for Kramnik (or maybe Grischuk, if he can put an end to his year-long slide) to have a chance please let us know in the comments.