A perfect game of chess is always a draw. Discuss.

Perfect play theoretically can be proven, and perfect play is not a matter of opinion. We just don't have a powerful enough computer at this time to actually prove what happens with perfect play. Maybe 100 years from now a super-duper quantum computer will be able to exhaustively analyze every position and prove whether perfect play is a win for one side or a draw.

@xor_eax_eax05

Nim is a game that can't carry on forever. Try proving that's a draw.

My first sentence in the post is "probably, but it's not known". What is reading attention span? Does anyone know?

I read the first sentence. I was commenting on the other five. Was there meant to be a connection?

A special case of this result was proved in 1913 by E. Zermelo, who showed that the game of chess must always have a winning strategy.

If it were the case that two perfect players would draw, then every game of chess would necessarily end in a draw. However, one player must always have a winning strategy through the proof of backwards induction.

@25
That is not what Zermelo proved.

https://en.wikipedia.org/wiki/Zermelo%27s_theorem_(game_theory)

both players can at least force a draw

Here are some games with perfect play from both sides, all ending in draws.
https://www.iccf.com/event?id=100104

I think this is the perfect answer to this thread. Always. Perfect…

That's because perfect games don't have blunders and mistakes and just may have a few inaccuracies. Forexample today I played this game against stockfish and used the engine helps by hints a lot. I drew [Event "Vs. Computer"] [Site "Chess.com"] [Date "2023-11-19"] [White "LikeChess78"] [Black "Maximum"] [Result "1/2-1/2"] [TimeControl "-"] [Termination "Draw by repetition"] 1. e4 e6 2. d4 d5 3. exd5 exd5 4. c4 Nf6 5. Nc3 Be7 6. Bg5 O-O 7. Bxf6 Bxf6 8. cxd5 Re8+ 9. Be2 c6 10. Nf3 cxd5 11. O-O Nc6 12. Qb3 a6 13. Rac1 Qd6 14. Rfd1 Be6 15. a3 b5 16. h3 g6 17. Qc2 Rac8 18. Qd2 Bf5 19. Re1 Bd7 20. Bf1 Bg7 21. Rxe8+ Rxe8 22. Ne2 Bf5 23. Qc3 Bd7 24. Qd2 Bf5 25. Qc3 Rc8 26. Qe3 Bd7 27. b4 Re8 28. Qb3 Nd8 29. Rc5 Bc6 30. Rc1 Ne6 31. g3 Bb7 32. Bg2 Bh6 33. Rc2 Bg7 34. Qd3 Bc8 35. Qb3 Bb7 36. Qd3 Bc8 37. Qb3 h6 38. Ne5 Bxe5 39. dxe5 Qxe5 40. Qxd5 Qa1+ 41. Kh2 Qxa3 42. Qe4 Ng7 43. Qd4 Ne6 44. Qe4 Rd8 45. Nf4 Nxf4 46. Qxf4 Bf5 47. Rc6 Qa2 48. g4 Be6 49. Qf6 Qd2 50. Rxe6 fxe6 51. Qxe6+ Kg7 52. Qe7+ Kg8 53. Qe6+ Kh8 54. Qe5+ Kh7 55. Qe7+ Kh8 56. Qf6+ Kh7 57. Qf7+ Kh8 58. Qf6+ Kh7 59. Qf7+ Kh8 60. Qf6+ 1/2-1/2

I'm going try to explain why this is fake or misleading IMO or IMAO

I take this position :

If Houdini or another super super calculator calculated everything from this position then we can no fake or misleading talk about a perfect game because artificial intelligence has solved the equation from this position allowing to play the perfect game but in reality Houdini hasn't calculated everything, he knows nothing about this position and like many bots he plays "without really a strategy" and "as in theory he is programmed to be strong against strong bots" he will direct his game towards something where he's more sure he won't lose.

So he'll calculate lots of positions and he will for example get this position :

this position he is not really able to know if it is better or the opposite. It's a dangerous and uncertain position and he still has a timer (he won't think for 10 years) so he's going to go for example in this position :

because in this position he evaluates that there will be a strong chance that he won't lose and he prefers this position because the others are more uncertain giving more chances that he could lose.

Chess is less complicated than a Rubiks Cube which means that every game is likely a draw because its too simple. People think that chess is complicated but the reality is that it isn't complicated at all. It is all simple patterns. For people like Magnus Carlsen do you really think they see chess as complicated? Obviously not. When your blitz rating tops 3,000 it basically means the game is a piece of cake for you. Is his brain overclocked or something? No. The truth is he was born, for whatever reason, with the ability to see the pattern, just like some people are born naturally gifted to solve Rubiks cubes really fast. It's the exact same thing, some people will sit there with a Rubiks cube trying a million bad moves, and others will just know the pattern intuitively and solve it in a few seconds. Now, its kind of a silly game, but, imagine pitting two Rubiks cube masters against each other. Who will win? Well, one will be a second or two different, or even less than a second, but in the end they'll be basically identical. Of course the actual percentage difference between them is going to be less than a thousandth or ten thousandth of a percent in accuracy and speed but that tiny bit adds up to victory. The nice thing about chess when compared to Rubiks cubes is that it can end in a draw very easily. Whereas with rubiks cubes there are rarely draws because the time to solve would have to be exactly equal which never actually happens. That is partially why when it comes to blitz chess you see some players rise higher in the rankings, its the time to solve quotient. They're just faster to move, faster to see the pattern for whatever reason. Basically what I'm saying is that Magnus sees the pattern so there's very little chance of him losing, ever, its not analysis for him so its not fatiguing. It requires no thought at all on his part. He simply sees the correct moves. In some cases he follows rules he has learned or been taught after many years of coaching and playing. So, clearly, chess is a draw because its a pattern game. The solution is the same pattern over and over again, so there's no chance of a win except to play wrong moves on purpose. So that's the inherent nature of chess, playing wrong moves. Once you realize that chess isn't about good moves, then you know what chess is, it is a drawn game. The only way to win is to lose without losing.

None of which makes any sense at all.

Perfect games can have more than just a few inaccuracies.

This game follows Syzygy tablebase optimal moves from the mate in 2 set up position. Syzygy plays perfectly (that's what it's for).

And it can be a lot more inaccurate than that!

By the inherent nature of the game, playing accurate moves should never lead to a loss. That's a fact. I've seen programs beat other programs then the games were analyzed by Stockfish and the end result is that when the computers lose it's because they played inaccurate moves. This is why many engines now have engine detection built in because if they're playing an engine they need to play much more accurate overall. And that's basically why some engines were able to beat stockfish for awhile because the engines basically pretended to be worse than they were, but, now stockfish looks for engine moves and plays accordingly. Now stockfish is basically unbeatable by engine or human.

The key thing to realize is that the subset of "accurate" moves may be much, much smaller than you think. In fact, I'd be willing to bet that there's really only a handful of openings that result in a draw if every possible move on the board is considered. It may even be the case that there is only one accurate opening.

That's Stockfish 15 trying to beat me from a theoretically winning position with only 5 of the original 32 men on the board. Doubt if it's improved as much as you say since then.

Almost guaranteed the next version of Stockfish will beat the current version; what Stockfish does is pretty much irrelevant to the topic.

Two knights can't mate so this obviously a draw or stalemate. There's nothing for either side to do and blacks only hope is that white misses the chance to take the pawn.

Two knights can mate.

Check here.

Three games.

(a) 1.White resigns.

(b) 1.Draw agreed.

(c) 1.Black resigns.

One of those games is perfect.

@40

neither (a) or (c) can be perfect games.
(b) can be a perfect game, but also (d) a 7 men endgame table base draw is reached or (e) a position is repeated 3-fold. Look at the perfect games I linked to.

@23

"This game follows Syzygy tablebase optimal moves"
++ This is not a game. A game starts with the initial position.

@28

Those players are ex aequo World Champions and nobody could take a point from them so far.
It is correspondence chess at 5 days/move average, engines allowed.