Wow.

**P.S.** In the comment thread, Peter Dorman has an interesting discussion of Carlsen’s errors so far during the tournament.

Posted by Andrew on 21 November 2016, 7:12 pm

Wow.

**P.S.** In the comment thread, Peter Dorman has an interesting discussion of Carlsen’s errors so far during the tournament.

## Recent Comments

- Mark Samuel Tuttle on “Have Smartphones Destroyed a Generation?” and “The Narcissism Epidemic”: How can we think about the evidence?
- Daniel Lakeland on The curse of dimensionality and finite vs. asymptotic convergence results
- Corey Yanofsky on Zoologist slams statistical significance
- Adam Schwartz on 3 quick tricks to get into the data science/analytics field
- RJB on “Deeper into democracy: the legitimacy of challenging Brexitâ€™s majoritarian mandate”
- Chris Wilson on The curse of dimensionality and finite vs. asymptotic convergence results
- Keith O'Rourke on Zoologist slams statistical significance
- Keith O'Rourke on Zoologist slams statistical significance
- Austin Fournier on The Lab Where It Happens
- Valentin Amrhein on Zoologist slams statistical significance
- ojm on The curse of dimensionality and finite vs. asymptotic convergence results
- Bob Carpenter on The curse of dimensionality and finite vs. asymptotic convergence results
- Terry on The graphs tell the story. Now I want to fit this story into a bigger framework so it all makes sense again.
- Terry on The graphs tell the story. Now I want to fit this story into a bigger framework so it all makes sense again.
- Peter Dorman on “Deeper into democracy: the legitimacy of challenging Brexitâ€™s majoritarian mandate”
- Sander Greenland on Zoologist slams statistical significance
- Mark Schaffer on Zoologist slams statistical significance
- Carlos Ungil on The curse of dimensionality and finite vs. asymptotic convergence results
- Carlos Ungil on The curse of dimensionality and finite vs. asymptotic convergence results
- Carlos Ungil on The curse of dimensionality and finite vs. asymptotic convergence results

## Categories

Pretty new to chess..why does this spell victory for Karjakin?

1. Qxa2 Ng4+ and White is dead:

(2. Kh1 Qc1+ 3. Bf1 Qxf1#)

(2. Kh3 Qg1 threatening Nf2# and Qh2# and the White can’t defend both without losing his queen)

I wouldn’t’ve seen it myself! It was only after Carlsen resigned that I could try to put together the steps, and even then I was getting stuck.

Oh, I get it now. Thank you! I was thinking Qxa2 Nf3+ and getting stumped. Of course Ng4+ is what does it.

I had no idea chess moves looked so much like chemistry. Nerds.

Ok fine, I’m a nerd too. Here’s an awesome move I only sorta get: https://en.wikipedia.org/wiki/Ear-reddening_game

Jrc:

OK, you out-nerded me on that one! I can play go, but really really badly. I don’t know any joseki at all.

Actually the “ear-reddening move” is the one in the figure below – the stone marked (1). The joseki part is just set-up (it probably took a long time to determine that that was a big turning point – I don’t think it was obvious at first).

But just look at that stone marked (1) in that second figure. It is just, like, sitting there, floating out in space, far from all the apparent action, but indirectly affecting and influencing every other position on the board. Beautiful and odd and out of place. I feel like even if you didn’t know go, and you were watching that game, that move would look surprising. Honestly, my first reaction would be to look at the guy playing white to know whether it was a great move or a terrible move, so I’m not that surprised someone noticed his ears getting red.

That was pretty nutty. Carlsen had an easy perpetual earlier but went for the win.

Very cool! After KH3, White cannot perpetual check either, because QG1 blocks QA7+ as well as QB6 (after QB2+, KG6).

Carlson has made a number of fairly obvious errors in this match, obvious to me at least, and back when I used to play I was only a national master. Perhaps the most shocking was his …f4 in game 4, a move I would never have played; it was so antipositional it could be justified only by concrete analysis, but Carlson by his own admission didn’t even bother to try.

I’m sure people will want to psychologize this. One hypothesis is that Carlson was way too complacent going into the match (shades of Hillary?) and thought he would coast through his title defense without having to do any work. This is supported by his apparent lack of opening prep as White.

I sense a variation on this. When I think about his errors, they all have the property of a sort of hazy optics. By this I mean, if you imagined the positions many moves ahead as part of your calculation, they have vague features that look favorable or at least not toxic. For instance, after …h5 in the latest game, Carlson is dead in the water, but that wouldn’t be so obvious from a distance. If that position appeared as a possibility 5 or 10 moves into the future (in chess one move for each side together is called a “move”) you might not notice that it’s lost. Similarly with that notorious …f4. You might think, “Oh, I have a protected passed pawn on the kingside and bind on the queenside, so there has to be a win there somehow.” The nearly locked up nature of the position, the lack of access squares for the minor pieces in particular, might not be so visible at a distance. What I’m getting at is a type of laziness or distraction: it looks to me like Carlson has not been dutifully recalculating variations as he proceeds down them. If you’ve done the work to foresee a position several moves ahead, as you get closer to it you have to do more work, recalculating to see if your earlier judgment was correct. I get the impression that Carlson hasn’t been doing this, at least not consistently. I can’t think of any other reason he would make moves that a much weaker player like me can see as faulty.

Jerry from ChessNetwork on Youtube gives a great analysis of this game: https://www.youtube.com/watch?v=ECYkkpANUe0

SS: We spoke about Anand, Kasparov, Karpov and other world class players (in part I). What are your impressions about Magnus Carlsen?

MarkDvoretsky: Recently I gave an interview to the Russian chess news. When they asked me about Magnus Carlsen I told them about my visit to Norway. I was there to train the best players of Norway (apart from Magnus), and I met Simen Agdestein. He is a very strong player, former coach of Magnus Carlsen and a very nice guy. When we spoke, he told me some very interesting points about Magnus. His first feature is the fantastic intuitive instinct about the game. He feels the position perfectly. Simen said that Carlsen’s calculation is not the best. Even he (Simen) could sometimes calculate better than Magnus [laughs], but his feeling and vision of the position is perfect. The second special thing about Carlsen is that he absolutely doesn’t have any fear about losing a game or a bad result. This helps him to conserve his nervous energy and fight in any situation. Intuition and fearlessness are two of Carlsen’s biggest assets.

SS: Magnus is also very strong in the endgames.

MD: Yes, this is a logical consequence of his superb intuition. But in the past there have been instances where he has made some very basic mistakes in rook endgames. If you look at those games closely you will realize that these mistakes were not a result of intuition but lack of calculation [Smiles] So most of Carlsen’s errors can be seen when concrete calculation was involved.

Even more interesting is Karjakin’s miss of the draw in the game he lost – mental blocks, trusting your opponents’ calculations, etc. Chess decision-making is fascinating. Here is the chess equivalent of Buridan’s ass – “”World Champion Magnus Carlsen is top favorite in Paris but the tournament did not start well for him. In the first round Carlsen played with White against Wesley So and with bold and creative play the World Champion reached a completely winning position in which he had two queens against queen and knight. But then he could not make up his mind which winning move he should play and lost on time – a bitter loss particularly so because the rapid games are played with a time-limit of 25 minutes for the whole game and a 10-second increment per move.”