True or False Chess is a Draw with Best Play from Both Sides

I think I see the problem.

Many amateurs do not recognize the difference between good and bad moves and therefore cannot accept that others can. GMs and computer experts know that while they can't always recognize errors, they can be fairly certain after deep analysis whether an error occurred.

Elite players such as Carlsen strive to reach complex positions where errors are more likely to occur. They know that crystal clear play is not enough to win.

As Ponz and others have pointed out, every decisive game has had at least one identifiable error. There have been many draws where neither side made an identifiable error, including many between engines.

@ Kierkegaard14 Before coming in backing someone making dubious claims, maybe try and understand what the point of contention actually is. In no way is he "correct".

No one is disputing that chess is drawn with perfect play - it probably is, though we don't know this with 100% certainty. The claim that is being contested is that many "perfect" games have already been played, and that is something totally different. We don't know yet what constitutes perfect play! GMs aren't saying that there have been games played perfectly - in fact, they're saying the opposite, in that they're speculating on the most likely outcome if such a game were to be played.

Then you don't see the problem. Ponz has maintained that many perfect games have been played. This is complete nonsense. Not because it may nor may not be true, but because until chess is solved we have no way of knowing this with scientific certainty. If chess is solved and it turns out to be a win for white, as unlikely it may be, then that renders every draw played in the history of chess imperfect. And we don't yet know if chess is drawn - not with "perfect" certainty. And as I said earlier, I won't call anything "perfect" until the calculation and analyzation by which we understand is likewise "pefrect". 

That said, I think I see the problem. Illiteracy.

L&L, 

Your argument is that we can't know what perfection is until we've explored every single possibility.

Experts in chess have a different opinion.

@LosingAndLearning81 Sorry, I must have missed where he said that many perfect games have been played already. I would definitely disagree with that sentiment.

 I see now where you clarified before my post. Sorry about that.

In SOME cases. A 1000 level player  can recognize a major blunder that allows mate in one. You don’t need to be a GM, you don’t need an engine, and you don’t need to capable of perfect play.  But what we are talking about is recognizing the perfect moves in all positions, including the starting position. 

 I think we can all agree that the top engines will always play what they calculate are the best moves, and they are orders of magnitude stronger than humans. 

 A couple of years ago, Komodo vs Stockfish. Stockfish played what it thought was the best move, and Komodo announced mate in 59!  Everytime two engines play each other, no matter how strong they are, if one loses that means it demonstrably has played at least one major blunder thinking it was the best move. 

 Current engines can identify blunders that humans can’t identify.  A super-Duper engine of the future (that has solved chess) will be able to identify blunders that the current strongest engines cannot identify. 

Absolutely nobody can know if anybody has ever played a perfect game.  In my opinion, it’s unlikely anybody has.  There are 10 to the 120 power possible game variations 

 That doesn’t at all address what he said. He said GM’s aren’t saying there have been games played perfectly. 

Saying IF a game is played perfectly it ends in a draw, is NOT the same as seeing games have been played perfectly. 

You said it: they have an opinion...on whether chess is a win with perfect play. It's been debated for a long time. By far, the dominant view is that chess is drawn with perfect play. I wouldn't disagree.

But to say that there have already been perfect games played is simply unknowable at this point, precisely because chess hasn't been solved. Games thought perfect even twenty years ago reveal refutations, and at the very least, improvements, using modern computing. Consider the following scenario:

A fifty move game is played; it ends in a draw. The most recent build of Stockfish is in perfect agreement with every move. Chess masters look at the game and deem it perfectly played by both sides.

Okay, but...there were moves that did lose centipawn value. Of course, all that matters is finding the moves that lose the fewest centipawn - it can still be perfect. But what about fifty years from now? A far more advanced chess engine than we are currently capable of building - what if it thinks another move was a little better? Even if it were just .02 centipawn better? The perfect game is gone according to the latest advances in chessic understanding.

We simply cannot look at a game by two chess masters and definitively say that it was "perfect". Applying the attribute of perfection requires a standard far above what any of us are capable of giving, in that, unless we perfectly know, we cannot perfectly bestow the label of "perfect".

It's simple, sound, rational logic.

The claim of a poster was that he does know, for sure, beyond all doubt, that countless perfect games have been played.

Quite frankly, it's preposterous and an insult to reason.

ponz111: “There have been many draws where neither side made an identifiable error”

 And how about errors that are NOT identifiable by current players and current engines?  How many of those have been played and how do you know? 

Ask any GM and he will say that some perfect games [meaning games without error on either side] have been played. Remember there have been billions of games played...

Ask me [someone who has played at the GM level ]

A somber thought: some of these people actually breed. And vote.

Think about that.

 A GM might claim games have been played with no identifiable mistakes, but I doubt any GM would claim to know whether or not a literally perfect game was ever played. 

 Also using GM’s as an appeal to authority doesn’t really work here. All of us know all of the rules of chess and how it is played. A GM having a better understanding of established openings and better pattern recognition does not mean they have a better conceptual understanding of how the game is played and of mathematical probabilities involved.

 I would like an answer to this one please. 

ponz111: “The fact that some games have been misanalyzed does not mean that other games have not been played with no errors”

No, it doesn’t de facto mean that.  But you are the one making the claim that games have been played with no errors. The onus is on you to demonstrate that. 

ponz111: “a perfect game is one where neither side made an error which would change the theoretical result of the game. I could give you examples of this.”

 Nobody on this planet has any possible way of knowing if this has ever happened. 

I have already given two games i have played which were perfect games.

It does not matter if a game is identifiable by current players and current engines as perfect for it to be perfect. A perfect game is a perfect game even if nobody in the whole wide world knows it is a perfect game!

I do not know how many perfect games have been played?  I have played several. I am only one chess player out of millions of chess players. Am sure many others have played perfect games --actually it is easy to play a perfect game as i have shown.

 A very conservative guess is that there have been thousands of perfect games played. [games without error by either side]

 OK, let me break out the crayons.

- A GM will identify mistakes that a 1600 player won’t find.

-  A strong engine will identify mistakes that a GM won’t find. 

- A stronger engine will identify mistakes that a weaker engine (than itself) won’t find. 

- A hypothetical Super-Super engine of the future that has solved chess will identify ALL mistakes than anyone or any engine ever makes.

 Since that last example doesn’t exist, how can anyone possibly know if a game was ever played where this Super-Duper computer would not find any mistakes?