In theory it is possible to consider all possible moves in chess then all possible replies by the opponent right down to the end of the game. There are so many possibilities however that it would take white around 10 to the power 70 years to make the first move.

Faster computers might reduce this timescale but physical constraints mean the first move could never be made before the universe came to an end.

No. I assume that by all the moves you mean all possible board positions. That does not tell how one board position relates to one several moves away, so this knowledge alone would be useless.

If you’re talking about all moves as @flutherother described it then yes you would always win or draw. It would take a crazy amount of time for the fastest supercomputer or a brain the size of Jupiter.

Nope, everyone makes mistakes.

It sounds like someone needs to watch The Queens Gambit on Netflix.

That’s called a computer, right? They know all the possible moves, and if you were to play against one, you would lose or draw.

It depends on what you mean by “knowing” a move, and how well the person would have access to thinking about those moves. Anyone who knows the rules and has an analytical enough brain can know all of the moves in any position. What enables effective play is more about being able to think about the many future moves, counter-moves, and their resulting positions. So if you knew all possible series of moves in such a way that you could look at a position and evaluate them all and distill that into an answer of what move would be best to make next, then I’d think yes. I would not call that merely “knowing” the moves.

@snowberry It would be a computer. But all chess computers were beatable by the best human players until recently, when the best chess computers beat the grand master human.

Even if you knew all the possible moves, wouldn’t the problem be knowing which one to choose at a particular point in the game? Isn’t it really about knowing the best possible move rather than all possibles?

@kneesox ” Isn’t it really about knowing the best possible move rather than all possibles?”

If you know all possible moves (and we’re talking about all possible future moves of you and the opponent too) you can derive the best possible move. Basically it’s a tree of all possible futures and your algorithm “prunes” away any branch that leads to a loss.

@gorillapaws at any given point, is there more than one “best” possible move?

@kneesox I think the answer is yes, but since no computer has solved for all possible moves (because it would take eons to work out) I don’t know if that question can be truly answered (though maybe it can). It may turn out that there is one (or more) sequence of moves, if made, that will always win, or draw (just like there is in tic-tac-toe). Edit: I missed the “any given point part.”

@kneesox Continued from above. We know mathematically that in certain scenarios late into the game there are multiple moves that are equally “best.” So if the “given point” is one of those moments we know the answer is true. But if the given point is the very first move, my inclination is to say yes, but I don’t believe it’s been proven (or can be proven until a computer is powerful enough to work out all possible moves in some reasonable amount of time). Again, I might be wrong about this and there may be other means for proving this mathematically.

@gorillapaws thanks, very interesting, especially the “tree of all possible futures”. If two computers (using different algorithms) play against each other, does the universe melt down?

Watch The Queens Gambit and then get back to us.

If you knew all counter measures, you could.
You don’t have to ponder every move. You only need to consider every move made so far in a game.
I used to watch guys playing sidewalk chess in SF. They told me that if you know the counter to your opponent’s moves, you will win. A kid wanted to play one day. Nobody would take money from a kid. His mom was with him, and I got the impression she wanted to see how he could do with an opponent. I asked the guy nearest if I could play the kid on his open table. I explained that I was an ametuer, so no master. I told mom I would play for real, so if I win or lose, it would be an honest outcome. We played.
I remembered what I was taught as just an observer. I didn’t worry about what I should do to win. I looked at what was the counter to his moves.
Yeah, I beat the kid.
Do I feel bad? No. We both learned something that day.

It’s been 16 years since a human beat the best computer at chess. As everyone has pointed out, despite this, computers are a good ways away from being able to work out every possible move. However, it is pretty clear that the human mind just doesn’t have the bandwidth to perform calculations of the magnitude you suggest.

Not if your opponent new all the moves as well.
Or two computers compeating…endlessly.

@Inspired_2write – it is possible that chess is a game where white, played perfectly, will always win.

No. Because if you know all possible moves, so does someone else. The best you might hope for is a draw.