Hacker News Reader

Comment by gpm | Hacker News Reader

We expect people to play to win.

In other sports we have rules against playing to draw/stall (e.g. see delay of game in hockey). Likewise in chess (or this and a number of other tournaments anyways), we have rules against agreeing to draw within the first 30 moves to attempt to discourage this sort of behavior.

Not only did Carlsen not play to win the game, he didn't even play to win the championship. Instead he chose to draw, explicitly deciding to split the prize with his opponent.

I, for one, am highly unimpressed.

View on HN

He made a sacrifice and "conceded" one game to a draw in order to increase his chances of winning the championship via his superior rapid play. He lost the battle to win the war. That is complely playing to win.

I dispute this on two points.

He did not fully "win" the championship. Sure, they had to give the title to somebody, so it's him, but splitting half of the prize is not a complete win.

Even if he had fully won, I don't view it as a sportsmanlike strategy to give up a game to put yourself in an advantageous position. That's why we have a rule against agreeing to a draw to early. That's why in hockey stopping trying to score and just passing the puck in circles at center is against the rules (delay of game) - even when it's the best way to win.

> He did not fully "win" the championship

The goal of most title holders is to increase their odds at retaining the title, in whatever way match rules allow. This goal made for some horrible championships generations ago under Soviet rules.

I agree the rules should be "biased" toward rewarding maximum skill, not just maximizing winning odds. But as Caruana said “We work with the match that we have... If the powers that be want to change it, then we’ll work with something else.”

Are you also against sacrificing a piece in a game to gain an advantage? How is this different? He played a sacrifice in the meta game to gain an advantage.

The difference is that:

- This was part of the metagame.

- This was agreeing to a "semi-draw" in the metagame when he did not need to.

I think you misunderstand how the prize money is allocated. They both got 400k for competing, and the winner would take 200k. Since they went to tiebreaks the winner only got 150k while the loser gets 50k. So in this sense Magnus didn’t really split much of the prize with Caruana.

They both get paid 400k for competing. That's payment, and maybe a prize for reaching the championship, but not a prize for the winner of the championship, so I ignore it.

The winner gets 200k in prize money. In the case of a tie the winner gets 150k, the loser gets 50k. That is to say the winner gets 100k, and they split 100k, which is half the prize, which is what I said.

My apologies, I misread your original comment.

In any case, Carlsen said he wasn't aware of the prize money difference after reaching tie breakers, so it probably didn't factor into his decision making.

> In the case of a tie the winner gets 150k, the loser gets 50k.

this literally makes no sense. In a tie, there is no winner and no loser. There was a winner and loser in this case, so there was no tie. I don't know why you keep insisting there was.

Huh, that's why that comment is attracting downvotes...

That has to be the least charitable reading ever. Do you seriously need me to spell out "In the case of a tie after the 12 games under the regular time constraints" while defending my math?

Perhaps Carlson was just being generous to his opponent then. Surely a nefarious deed.

Caring about prize money more than the title is not sportsmanlike, in my view.

Chess is a game that can be played perfectly. There is no perfect Hockey game. If you get two super computers to play chess against one another they will draw a majority of the time, I think it is a testament to the skill of both players that they drew 12 games in a row.

Yep. In chess, you have to make mistakes to lose. Top players don't make very many mistakes, so often nobody loses.

> In chess, you have to make mistakes to lose.

That's a conjecture. Nothing more. Nobody knows whether it is true! It is possible that a perfect player playing White may make you lose every time. It's even possible (though highly doubtful) that this holds for Black instead.

You're right that in theory nobody has proven whether chess is won, lost, or drawn with perfect play. But all available evidence, as well as my personal intuition (and that of almost all top players), screams that it's drawn.

Reply on news.ycombinator.com