Magnus Carlsen and Ian Nepomniachtchi are locked at 1.5:1.5 going into the first rest day after Game 3 of their match ended in a quiet 41-move draw. Ian varied on move 8 from what he'd played in Game 1, but Magnus once again had a surprise in store, 10...Re8, an old and, in his words, "really, really dumb move". He was hoping to get a game, but found Ian well-prepared there as well, and in the end there was nothing better than to liquidate and end the game in under three hours.
Replay all the games with computer analysis.
Replay the day's action with Judit Polgar and Anish Giri...
...or with David Howell, Kaja Snare and Jovanka Houska.
And here's Danny King's recap of Game 3.
“It’s just a fight, but we’ll see eventually if the match will settle down a bit”, said Magnus Carlsen after two bruising battles in the first two games of his 5th World Championship match. The general impression was that the players were likely to take their feet off the throttle in Game 3, and that’s just what happened.
The opening guesswork was easier this time round, since we’d seen Magnus showcase a new idea in the first game of the match against the 8.h3 Anti-Marshall. Unless Team Nepo had already found a convincing antidote it was likely they’d alter their approach, and Erwin l’Ami once again called it.
Nepo went for the main line with 8.a4.
That of course came as no surprise to Magnus, and he continued his ploy of deviating from the trodden path at the earliest possible moment — before his opponent could spring a surprise of his own.
10…Re8 had been played at the top level, but not in a very long time.
Magnus told Tania Sachdev immediately after the game that he played a sideline to try and get a game — i.e. a position in which his opponent wouldn’t know exactly what to do because he’d studied it with his team. It wasn’t about the objective strength of the move:
Re8 is a really, really dumb move, because usually you would try to go Re8 without d6, but it turns out even here he was well-prepared, and he didn’t give me any even slight chances to play.
As Magnus notes, however, Nepo was ready, with his team no doubt very clear by now that they have to look at almost every possible non-losing move for Magnus in a position. Ian commented afterwards:
In some way I expected everything, so I guess after such long preparation you more or less try to be familiar with most of the moves, but I’m sure there are also some worthy tries which haven’t been tried yet. Re8 is one of the pretty logical sidelines. This whole plan is very healthy. Re8, Bf8 h6, so a little bit like the Zaitsev variation, but still very nice for Black.
Move 14 was a moment of what might have been, since Anish Giri pointed out that the neural engine he was looking at gave 14.g4!? as the top move. He was certain Team Magnus had gone very deeply into the possible wild complications of a move that, at first glance, seemed right up Ian’s alley.
Instead, however, Ian quickly played 14.c4, a safe, positional move that made it clear he was happy to try and squeeze out a small edge and leave Magnus with the option of passive defence or trying to generate counterplay, at the risk of over-pressing.
Ian pointed to 16…a5!? as “surely positionally quite questionable”, which may be code for the first move that Ian’s team hadn’t looked at in their analysis. Magnus was then laughing and seemingly talking to his pieces before he decided to redirect his bishop with 17…Bc8.
Magnus would later say:
I was obviously making some fairly ugly moves, but it seems all to work out reasonably well. At least I couldn’t see any concrete way to pose serious problems, so I think it was a reasonably logical game.
Given the way the game developed from this point onwards, Ian had some regrets about playing 18.d4 instead of a slower alternative, but as he said himself, it was “very natural”, since “Black has all his pieces on the last rank”. Nepo’s first and only long think of the game then came after 20…Be6.
Anish Giri explained that Magnus could be playing himself or Wesley So instead, given how the games had gone, but he also said he hadn’t expected to see that quick, impulsive, attacking Ian in the match. On the other hand, it’s early days!
In the position it seems Nepo wasn’t considering the computer’s clear first choice, 21.Qd3, but mainly 21.Nd2!? and the mysterious move he eventually played, 21.h3. He noted that he could no longer prevent the d5-break, and that the move could be helpful in some lines, not least for preventing checkmate on the back-rank.
Magnus could have played the immediate 21…d5, but instead went for 21…c6, which Anish considered an example of the World Champion’s willingness to keep positions alive. He felt it's +ev for Magnus, at least in the long run.
The central break wasn’t delayed for long, however, with 22.Bc2 forcing 22…d5!, when the nuance 23.e5! kept some hope for White. Our commentators wondered if they needed to revise their assessment of c6 from a clever try to a serious inaccuracy.
Queens disappeared from the board with 23…dxc4! 24.Qxd8 Rexd8, however, and it very quickly became clear that Magnus had everything worked out. Despite his broken pawn structure he was fast enough to advance with his king and push his f-pawn so that he could set up an impenetrable barrier.
All that was left was for the players to make it to move 40, since the regulations don’t allow them to offer a draw before that point. In less than three hours it was all over.
So after the huge tension of the previous two days there was no great drama on Day 3, except perhaps when the players learned that they had to attend a doping test.
That would be one ever present topic of the post-game press conference. The players were asked if traditional doping could give chess players an advantage, with Nepo pointing out the obvious disadvantage.
Well, but if you use the doping you’re getting banned from the event, so that’s how it works!
He was sceptical that substances could help, while Magnus wasn’t so sure.
I think there are probably experiments to be made. People use drugs to both prepare for and to do exams, for instance. I assume they could possibly help in chess as well. If at some point my level of chess drops drastically I might start experimenting, but for now I don’t see the need!
When it was put to the players that their game had been one of the most “accurate” chess games ever played in a World Championship match, they both had good responses:
Magnus: Thank you very much. It feels great, it fills me with pride… but unfortunately only half a point on the score table!
Ian: I must say it’s quite a murky question before the doping test!
That test was also interfering with Magnus’s plans for the evening before the rest day.
I can instead tell you about my plans, or the immediate plans after the game today, if it were not for the doping control, which sort of messes it up. The first step of the plan was Chelsea - Manchester United, then the next step was Real Madrid - Sevilla, and then Golden State - Clippers, and then I thought we’ll kind of take it from there on the free day.
The other topic was having had three draws so far in Carlsen-Nepomniachtchi, 12 in Carlsen-Caruana, and two at the end of Carlsen-Karjakin, meaning 17 classical World Championship draws in a row since Magnus beat Sergey in New York on November 24th 2016 to hit back in that match.
We’d had two fantastic fights in the first two games, and Ian didn’t help matters by generalising after the much quieter third game:
This is the current status of chess theory — it’s hard to find some advantage.
“Any game could explode, just not today”, said Magnus, though he did repeat, when asked, his wish that the World Championship could become more fast-paced:
I think there are many ways to play, but the most obvious one is just shorten the time control. At least make it a mix of classical and rapid.
Another question about draws became the straw that broke the camel’s back. Ian gave a standard response about chess being drawish and requiring mistakes for decisive results, while Magnus commented:
I would say that so much has been written and said about this. I really don’t have a lot to add — we try!
The thought of draws was also on his mind when he answered the final question of the press conference. What will people remember about Magnus Carlsen 50 years from now?
I don’t know. I think talking about legacy during a match is a rabbit hole I don’t want to go down… but hopefully he will be somebody who won a classical game in a World Championship match after the year 2016!
There are still 11 classical games to go this match, with the score tied at 1.5:1.5.
The next chance will come on Tuesday, after the rest day, when Magnus has the white pieces. It also happens to be his 31st birthday! Things didn’t go so well a year ago when he lost to Wesley So in the final of the Skilling Open, but he has better memories when it comes to the World Championship, since in 2016 he beat Sergey Karjakin in playoffs on his birthday to retain his title.
All the action will be live right here on chess24 from 16:30 local time (7:30 ET, 13:30 CET).
We respect your privacy and data protection guidelines. Some components of our site require cookies or local storage that handles personal information.