Magnus Carlsen’s unbeaten streak now stretches to 107 games but he admitted to being “completely lost” and only playing for tricks before he ended his year of classical chess with a “harrowing escape” against Levon Aronian. It means he takes a 6-point lead into Sunday’s rapid and blitz, as does Ding Liren, who shrugged off his miss the day before to completely outplay Maxime Vachier-Lagrave again and move within touching distance of the Grand Chess Tour trophy and $150,000.
You can replay all the Grand Chess Tour Finals games using the selector below:
And here’s the day’s live commentary from Peter Svidler, Jennifer Shahade, Alejandro Ramirez and Maurice Ashley:
World Chess Champion Magnus Carlsen remains unbeaten in classical chess since Shakhriyar Mamedyarov defeated him in Biel 16 months and 8 days ago on July 31 2018. Since that time 107 games have now passed, or 105 if you use Magnus math and discount two wins over much weaker opposition in the Norwegian League:
There have been some hairy moments in that time, including during the World Championship match, but seldom has Magnus stared into the abyss as clearly as he did against Levon Aronian on Saturday in London. Magnus was asked if the streak was in the back of his mind:
Not really. I really didn’t play like somebody who was not trying to lose today. I played very, very riskily and quite recklessly and poorly… It’s befitting to end the year in classical chess with a pretty harrowing escape, since I’ve had a few of those, and that’s what you need to keep a streak going.
The chances of an exciting game already looked high Levon went for 3.f3 against the Grünfeld and Magnus responded with 3…Nc6. As Peter explained on air, that wasn’t the move he recommended in his chess24 series, where he felt he should stick to the more Grünfeld-style 3…d5, but it was a line that “leads to very interesting and unbalanced positions which can be a lot of fun to look at”.
For once it was Levon who seemed to win the opening battle, playing the rare 8.Be3 quickly and later not hiding his aggressive intentions with 11.h4. Magnus decided it was time to hit back, but his 11…b5!? pawn sacrifice seems to be both a novelty and a roll of the dice:
Magnus felt he hadn't overstepped the bounds of acceptable risk until 23.Rh3! appeared on the board:
He later commented:
This move I’d completely missed, because I thought I was still sort of doing ok, practically speaking, but after he went for this I realised I’m just lost and I have to play for tricks. I thought I was lost anyway, and so I gave up another pawn, but it didn’t really work out and anybody could see he was winning after that.
Magnus gave up that second pawn with 23…d5!? and was soon objectively losing until he managed to land one of his tricks on the board. Simply 35.Rf5! would have left White’s advantage intact, but 35.Bc5?! looked to be winning on the spot:
It would have been, if not for the only move 35…Rd4!, which suddenly turned the tables. Levon’s edge had all but gone, but after 36.Bxd4 Bxf3 37.gxf3 exd4 38.Rxd4 he got another chance:
The move leading to 0.00 here is eliminating White’s dangerous pawn with the simple 38…Qxf6. There were chess and psychological reasons why Magnus didn’t play that, however. The purely chess one was that he was concerned about 39.Rxd8 Qxd8 40.Kf1 Be7 41.Qe2, but as he noted afterwards 40…Bg7! and e.g. 41.Ne4 Qd3+! solves all Black’s issues. Instead he went for what he called the “completely insane” 38…Rb8!?, though there was some method to the madness:
So I should have calculated that a little longer, but actually at this point I thought it wasn’t a bad practical shot to keep the rooks on, because I felt like he was drifting, and seeing as we don’t have that much to play for I thought why not give it a go and try and see if I can hustle him there a bit!
Maurice Ashley clarified whether Magnus had actually been trying to play for a win:
Not really trying to win, but at least I thought after Qxf6 there are no winning chances and I still have to suffer, and I thought if Rb8 that’s probably a much worse move, but at least to some extent we’re playing for three results.
Instead it once again brought Magnus to the brink of disaster, until Levon was tempted by a trick of his own:
Saving the queen e.g. with 43.Qh2 should still leave White with a won position, but Levon went for 43.f7?! Magnus comented:
f7 is a really nice trick but obviously it’s unnecessary and realistically it just gets me back into the game, which I didn’t really deserve!
After the forced 43…Qxf7 44.Rf4 it seems Black has to accept a much worse ending, but in fact 44…Qg6!! 45.Rf8+ Kg7 46.Qd1 Kxf8 47.Nf4 Qe8 48.Nxe6+ Qxe6 led to a queen ending that was, as the computers tell us, a draw:
Magnus himself wasn’t convinced, but he managed to hang on until the players shook hands on move 81.
He summed up:
First of all it was I think objectively speaking an awful game, quality-wise. I just made so many mistakes and I think he made a number as well. Obviously I was lost, completely lost, but I hung in there and managed to get to this queen ending which was not really what I intended, because I thought there should still be some way for him there, but I was trying not lose by force and I felt that was the only way. It transpired it wasn’t so easy, and once he caught up with me on time as well then that was it.
That meant Magnus is now a huge favourite to convert his 6-point lead into a 3rd place finish in the Grand Chess Tour. The same goes for Ding Liren and winning the event, though Maxime Vachier-Lagrave is a very dangerous customer in speed chess:
Another player might have found it difficult to come back after missing a great winning chance in the first game of the match, but Ding Liren got right back down to business. He was well-prepared in the opening, not just with concrete moves but strategic ideas. He called 17.Ne4! a “very nice idea”:
His plan was to provoke 17…Nxd5 (Nf6+ was of course a serious threat) 18.cxd5 and then bring his knight back to d2, preparing to jump to the c4-square after a4 weakened the black queenside. It was a game of long-term plans, with that one only coming to fruition on move 44.
Ding refused to rush and showed impressive flexibility. On move 27 he played his rook from e1 to a1, but after 27…Bh6…
…he put it back there with 28.Re1, commenting:
This move (27.Rea1) looks so natural, to develop the rook to the open file, but after 27…Bh6 it seems not so easy if I let him take e3, so I just want to keep the tension and go back and find another idea. At least I can just wait to see which idea is the best.
Eventually 33.g4! started operations on the kingside and powerful play culminated in the position after 47…a2:
It wasn’t the only winning move, but here Ding unleashed the best move, 48.Rxg7!
After 48…Kxg7 49.Qg1+ Kf8 he had the clincher, and only winning move, 50.Rf5+! A glorious end to the stop-start journey of the queen's rook:
50…Nxf5 runs into mate with 51.Qg8+ Ke7 52.Qf7#
Ding Liren had been very careful with the finish after missing the win the day before:
I checked it many times! I remembered yesterday I played too quickly, so this time I checked this Rxg7 idea many times.
In this form Ding is a big favourite to win the $150,000 Grand Chess Tour first prize, but a lot can still happen in Sunday’s two rapid and then four blitz games. Garry Kasparov is set to join the show for the first hour, so don’t miss the last day’s action from 15:00 CET here on chess24!
We respect your privacy and data protection guidelines. Some components of our site require cookies or local storage that handles personal information.