Do you think that to beat someone at chess, you have to replay, hypothetically, all the possible combinations of the board, or to cold read the person in front of you?

Certainly not. The first choice quickly becomes impossible, and depending on tricks such as reading your opponent is stupid. It’s the quality of your opponent’s play that you must read.

To win you just have to play better than your opponent and to see one step deeper into the game. AI has been beating human beings at chess for many years but that doesn’t mean it plays a perfect game. It just has to play better than humans. It doesn’t have to read humans or understand how humans play.

You have to play better than their last play! You do it enough you’ll win.

Of course not, for the reasons stated by others above.

Moreover, neither are what chess players generally do. They do try to think a few moves ahead, but without considering all moves. And the reading of opponents isn’t “cold reading” but thinking of what it seems like would be good and likely countermoves by them.

But, do you think that, by calculating all moves, an AI could, theoretically, play a perfect game?

I always tried to gauge the other player and would setup a long-term strategy based on how good they likely were.

Ignore your apponent and cocentrate on the board.

The strategy that works best for me is to be aggressive and try to keep the other player on defense.

“But, do you think that, by calculating all moves, an AI could, theoretically, play a perfect game?”
At that point, you’re talking about something humans can’t do, so it really is a problem for computers and theory.

I think that at the point where you have that powerful a computer opponent at work, you’re reaching odd limits of the game that I am not familiar with, and if you want the actual answer, you might do well to read the people who programmed or studied that level of AI, e.g. see the link in @ragingloli’s last answer to this related question .

My expectation however is that:
* considering every possible move and counter-move to any length is actually still too large a problem space, and also not necessary, partly because many paths can be rejected and rules out.
* there are ways of rating a board position that also mean you don’t need to analyze many paths beyond a certain point
* to try to play “a perfect game”, you need to consider playing against a similarly-capable programmed opponent
* to maximize the chances against a similarly-capable programmed opponent, you may need to learn and adapt to that opponent
* the attempt to make the best AI possible might devolve into some situation where different AI styles perform better against other AI styles, and/or devolve into a kind of stalemate. I don’t know – someone would need to try, and it might take many generations to prove someone was objectively better than all others, if that’s even possible.

It’s not really humanly possible to consider all possible board combinations. In chess pattern recognition and understanding your position is more helpful than trying to calculate every possible position. Indeed, if you understand your current position well then a great many moves can be ruled out right away.

As far as reading your opponent: chess isn’t poker. There are no tells and nothing is hidden. All information is in plain sight to both players. You play the board, not the person. Indeed, with the prevalence of online chess now you can play and never even see your opponent.

@Darth_Algar: Something is hidden in chess: What’s your opponent’s next move?

@luigirovatti

You don’t understand chess, do you?

@Darth_Algar: I don’t get what you mean. I certainly understand the rules. All the possible moves, man, I don’t really know.

Rules are easy. The rules can be learned by a small child in a matter of minutes. But what I’m saying is that the notion that your opponent’s next move is hidden, while true in an absurdly literal context, fails to get at the heart of the matter. If you understand the current position well you can anticipate your opponent’s next move. Calculate your best move, then calculate your opponent’s best move in response and assume that he’s going to play that move. This is why so many games between top level players end in draws.

I think Luigi has bought in to the show business aspect of grandmaster level tournament chess. To drive up the “box office” and merch receipts great effort is expended in reducing the game to a conflict between 2 nerdy egos—a concept pro wrestling fans readily appreciate. And if you pay attention, you will notice that in the predictable cycles of public interest in the game, the uptick is always distinguished by the sudden arrival of volumes on the so called “psychological” aspects of chess. It’s the same as any other sport, game or competition. Here’s some sound chess advice Luigi: leave that bullshit for the rubes.

Lord knows. Look at the way the media try to play up Magnus Carlsen as some hyper-genius savant. The man himself has stated that he’s fairly average, intelligence-wise. And his interests outside of chess lean more towards the “common” (basketball, football/soccer) than they do “nerdy” (like mathematics, etc).