Magnus Carlsen won his fifth world championship in resounding fashion today, beating Ian Nepomniachtchi in the eleventh game to close him out by a score of 7 1/2 – 3 1/2. Congratulations to him on a really well-deserved victory! He is now turning in to one of our most durable world champions.
As pretty much everybody already knows, this was two matches in one. The first five games were all drawn, but they were anything but “GM draws.” Carlsen demonstrated a willingness to sacrifice material for long-term pressure, even when the compensation was unclear. For his part, Nepomniachtchi showed extremely accurate technique in navigating through the mine fields that Carlsen laid for him. Some people lamented that five consecutive draws showed that “classical chess is broken,” but I don’t think that any true chess fan would agree. If the players continued with the tense and hard-fought chess of the first five games, something would be sure to happen.
And in game six, something did. It was surely one of the most epic struggles ever in a world championship game, and just on the basis of move count it was the longest (136 moves). In the middlegame Nepo offered Carlsen the opportunity to trade his queen for two rooks. This is one of the toughest trades in chess to evaluate, and first of all I have to give credit to Nepo for playing this interesting and unforced idea, which took the game out of routine pathways and made it historic.
I’ve never really understood before when two rooks can beat a queen. I understand when a queen can beat two rooks. That happens when the side with the two rooks has too many loose pawns or an exposed king. I understand when two rooks can draw a queen. That happens when the two rooks can defend all the weak pawns and can double up on a rank or a file. But how do you play for a win with two rooks?
Well, the answer is that your opponent’s pawn structure has to be bad enough that you can double your rooks up against multiple targets. In the position after move 40, Nepo had a passed pawn on a3 that could not go anywhere. It could be attacked by all of White’s pieces: rook on a1, rook on a2, knight on c2. Black had only two pieces to defend it with, the queen and the bishop. So goodbye, passed pawn. In addition, Nepo had weak pawns on f7 and f6. Once again, Carlsen could threaten to double up on f7, and Black wouldn’t be able to defend. What actually happened was kind of cute — Carlsen managed to sac the exchange to win both f-pawns. This left him with a knight, rook, and two passed pawns against the lone queen.
The last 80 moves or so of this amazing game just involved Carlsen slowly, with infinite patience, moving the passed pawns forward while avoiding perpetual check threats. The computer says that Nepo could have drawn almost to the end. But humans are far worse than computers at telling the difference between a near-perpetual check and a real perpetual check. Especially when you’re playing on an increment of 30 seconds a move. It’s absolutely no surprise that Nepo eventually let Carlsen slip out. This was not Nepo playing badly, but Carlsen playing in finest Carlsen style.
After such an incredible battle and ordeal, you had to wonder how Nepo could come back. The answer, sadly, was that he couldn’t. Over the last five games of the match he simply went to pieces, losing games 8, 9, and 11 with blunders. Actually I think that the games might have been more interesting for amateurs and beginners to watch, because he made the kind of tactical oversights that trip up sub-2200 players over and over. Game 8, he overlooked a tricky combination involving a back-rank mate. Game 9, Carlsen sacked a pawn and Nepo tried too hard to hang onto it, resulting in a piece getting trapped. Game 11, Nepo played a line that would give him the advantage unless Carlsen saw that he could sac an exchange for a mating attack. Of course, Carlsen doesn’t miss that stuff, nor would almost any grandmaster. You have to wonder if at this point Nepo was just trying to go down in a blaze of glory, rather than actually trying to win.
More qualified people than me will surely dissect what went wrong with Nepo. Sam Shankland, on chess.com, wrote that he thinks there is a “Nepo A” and a “Nepo B.” Nepo A was the one we saw for the first six games, and Sam considers him the second-strongest player in the world. Nepo B was the impulsive, mistake-prone Nepo we saw in the last five games. After game six, Nepo was maybe trying too hard to make something good happen and losing touch with the facts on the board. And in the last game, as I just said, I think that Nepo was just looking for an honorable way to end it.
What did the match mean for Carlsen? I compare this match to the two Alekhine-Bogolyubov matches (1929 and 1934), where Alekhine thrashed his favorite whipping boy soundly. These matches didn’t really do anything to enhance Alekhine’s legacy. Perhaps they shouldn’t have been played: Alekhine was ducking a rematch against Capablanca. Likewise, I think that history will say that Nepo was out of his league in this match; he was not at Carlsen’s level. The one good thing about the match is that it does show us that Carlsen is back on Carlsen’s level. There were doubts about this after the colorless matches against Karjakin and Caruana, where Carlsen had to go to tiebreaks. I think, possibly, that there might have been an upgrade in Carlsen’s play after he saw how the computer AlphaZero played. There were definitely AlphaZero-esque moments: the way he massaged a win out of game 6, the way that he repeatedly posed difficult defensive challenges to Nepo with long-term strategic sacrifices. He won by creating so many threats that Nepo lost track of them all, and that’s kind of similar to how AlphaZero won.
What did the match mean for Nepomniachtchi? Unless he can somehow channel “Nepo A” more often and banish “Nepo B,” I think he is done as a serious world championship contender. Just getting to the championship match was his real triumph, as it was for Nigel Short when he qualified to play against Garry Kasparov in 1993. Kasparov won by a wide margin, 12 1/2 – 7 1/2, but I don’t think that anyone considers it a great failure for Short. Like Nepo, he was out of his league.
Finally, one of my students said something that I thought was interesting and tend to agree with. He said that this match might make us appreciate what Fabiano Caruana achieved against Carlsen a little bit more. For twelve games Caruana stood up to the Carlsen pressure and didn’t crack a single time. At the time we tended to focus on the fact that there were no wins, but we lost sight of the fact that there were also no mistakes.
And I’ll leave you with a question: When do you think we’ll see a match between Carlsen and Alireza Firouzja, the 18-year-old phenom who is now the world’s #2 rated player? And who do you think will win?
{ 3 comments… read them below or add one }
Enjoyed your analysis of the match. Carlsen is amazing!
I think Firouzja will be pushed aside in the next cycle, but in 2024 he’ll be there.
As for the recent match it smells of Lasker – Janowski, Zeitgeist adjusted.
I have been following the Starcraft 2 games of AlphaZero’s sibling AlphaStar, and your comment “He won by creating so many threats that Nepo lost track of them all” is exactly how AlphaStar won its match against the top human player Serral 4-1.
I found the games interesting when I first saw them–commentary makes it possible for a non-player to follow the action fairly well–but then I learned to play the game, and now I just wince in sympathy. Being attacked all over the board like that is so, so hard. AlphaStar didn’t otherwise seem to play that well (it has a bizarre tic of starting to build things, canceling them, and then restarting them which seems purely harmful) but it does not mind micromanaging attacks in five sectors at the same time. They choked it to make no more actions per minute than the human (around 400-500, which is astounding) but it still had a big edge in its ability to split its focus and attention. And toward the end Serral made some unforced mistakes, like Nepo.