Isn’t chess a “finite” game where no luck is involved? Are computers able to calculate perfect strategies for both players?
And if yes, then which side is at the advantage?
In: 11
I am no chess master. But after listening to a few chess documentaries. Chess masters tend to like to go first. That should tell you some things.
Chess is a finite game, meaning there’s a finite number of possible positions and a finite number of unique games. However, even though it’s finite, the number of unique games is unimaginably huge, and it’s far larger than our current computer technology could process in a reasonable amount of time. Like, it might take billions of years for a modern computer to “solve” chess and be able to play it perfectly. It might be solvable, but it’s unlikely that any of us will see it solved in our lifetimes. Which is good, because once a game is solved, it becomes boring and feels pointless to play.
If chess was solved, I’d expect white to win every game. It’s common knowledge that white has a small advantage because they get to move first. If you start with a small advantage and you know how to play perfectly, you’re pretty likely to win. However, winning and losing are not the only possible results of a chess game, you can draw too. It’s possible that black could force a draw with perfect play, even if white plays perfectly too. We won’t know until chess is solved. Don’t hold your breath.
mathmatically, i believe it was deduced that white has the highest possiblity if both sides equally played at the same level. White having the OPENING move, means it gets to direct the game and theory that will be played and therefore get to CHOSE the next move that will be countered by the perfect response, therefore knowing what to play next. the black would be on the ultimate never ending defense, every single time. So white would have this advantage, ever single time. based on the fact they were both playing perfectly and without human errors.
For a long time there was a big argument over whether chess is a win for white or a draw. Nowadays we think it’s a draw because better and better computers are drawing each other more and more often.
This doesn’t mean that it’s definitely a draw though. Lots of things in maths have a surprising outcomes when they reach their perfect or optimal state. For example, the sum of 1+1+1+1… Is infinite, but if you ever stop it, it’s finite. Something similar could be true for chess; near perfect play is a draw, but perfect play is a win.
I think it’s a draw. But I can’t prove it, and it’s unlikely if not impossible we ever will.
Part of the reason that it is difficult to use computers to model the game is that you can make moves that don’t actually progress the game. They can get include moves like, move piece a forward two spaces, the next turn move piece a back two spaces, then next turn move piece a forward two spaces, ect… When you actually use an intelligence to limit the number of moves to ones that progress the game in some way it drastically cuts down on the number of moves. So at the moment humans are the only intelligence that will make intelligent moves naturally, most of the computers that learned to play chess were programed with moves or learned from watching human players.
There’s a concept in computer science called “intractable”. That means a problem that theoretically has a solution, but you can’t solve it in practice with the computing power available to you.
The solution to chess is intractable. Since chess is a finite game, there is definitely a perfect way to play any given position. (That doesn’t mean you can always win – sometimes perfect play would still lead to a draw, and some positions are lost no matter what you do, unless your opponent blunders.). But we can’t actually calculate that perfect solution, except for endgame positions with a maximum of 7 or fewer total pieces between both sides (white has a king and three other pieces, black has a king and two other pieces, etc.)
It is very strongly believed that if both sides played perfectly the outcome would be a draw. There are a number of reasons for this: (1) just having the first move probably isn’t enough of an advantage to be able to force a win against a perfect opponent, (2) games between human grandmasters are mostly draws, (3) have the strongest computer program play both sides of a game and it will mostly draw, and (4) in hundreds of years of chess no one has found any kind of forcing win. But none of this is a perfect proof, and we can’t rule out that a perfect game would be a white win or even a black win.
Well white has a starting advantage so theoretically if they continued to make perfect moves from then on white would simply maintain that small advantage until the end of the game resulting in more wins than black.
The simple answer is “No”. If both sides were to each play an absolutely flawless game, white would not be the deciding factor.
That would be like saying that the winner in a 100 meter race would always be determined with the very first step.
In addition to what has been said; I found the bar chart of AlphaZero vs Stockfish on this page very interesting. 29% win rate as white, mostly draws as black:
[https://www.deepmind.com/blog/alphazero-shedding-new-light-on-chess-shogi-and-go](https://www.deepmind.com/blog/alphazero-shedding-new-light-on-chess-shogi-and-go)
Maybe someday AI will see chess the same way we see Tic Tac Toe.
Theres too many positions and lines to follow. With computers playing the “best lines” it usually leads to a draw.
However, there is some Mathmatician that estimated there are about 10^120 possible chess positions available. Thats 120 zeros. You cant fathom how large that is. Like when a computer calculates millions of moves a second, it comes nowhere close to that number. A million is less than an atom compared to 10^120.
I will add there are some lines the computer will miss, but these are usually very rare circumstances.
Latest Answers