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

George, I thought heuristics aren't supposed to be any indication of anything with regard to this question. How then is your subjective perception of chess being drawn so much higher than 33.3...%? I would think it should be extremely close to 33.3...% if you don't think heuristic indications have any relation to reality.

At least TetsuoShima was consistent when he did (ultimately) estimate the chances of each result to be one out of three, as that related to his argument that nothing we currently know about chess can be reasonably used to weigh any of the possibilities as more likely than the others.

Sure there are zillions of chess positions and maybe only 100 billion chess games played so far. But to discount the 100 billion chess games played, especially those games played by the top players makes little sense. The top players while not perfect, know a heck of a lot about chess. And when you get a top player with a top machine you have rather closer to perfect chess without errors than most of you assume.

Suspose your had one million balls each the same size in a bin.  You know

some are red and some are blue and some are yellow.  But you do not know how many each of the same color.  Now suspose you have twenty different people to put their hand deep in the bin and pull out a ball.

It just so happens that all twenty pull out a blue ball.  Now is not that some evidence that there are probably more blue balls than any other color?  It does not prove there is a erponderance of blue balls but it is an indication that it is likely there are more blue balls than the other two colors.

Perhaps you mean something different by "subjective probability" than what I am thinking. What it seems to me is that if one is to take your position that heuristics do nothing to clue us in on the answer of this question, then you shouldn't weigh any of the possibilities as more likely than the others. Shouldn't it seem to you that all three of the possibilities are equally likely because you haven't seen a "real" reason to distinguish them from each other?

Perhaps you mean that, for example, the symmetrical nature of chess makes it seem like chess is a draw in an illusory way. But if you know it's an illusion, why then would you be any more surprised with chess being a draw compared to, for example, chess being a win for black?

In other words would you place your bets more confidently on chess being a draw than the other options? Or am I misunderstanding what you mean when you say you would be more surprised if chess is a win for black and less so if it's a draw?

I do not think you understood my pulling blue balls out of a bin 20 times in a row anology.  It has to do with the number of possible chess games compared to the number of chess games actually played. The number of possible chess games exponetialy exceed the number of games actually played.

But we have a sample of 100 billion games played in a very large bin trillions of times larger.  However if every single game out of the 100 billion has the same charistic that gives  us a very good clue.

The characteristic that all these 100 billion games has in common is there is not one game that anyone has ever found where either side won without the other side making an error.  Not one game out of 100 billion where White played and won without his opponent making an error.

Now I cannot look at all 100 billion games, most are by players who make many errors. But I do have the challenge for anybody to find just one game where White won without an error by Black.

On the other hand I have given a game where neither side made an error and it was drawn and I could give many more.

Believe me, if White could force a win from the opening postion, someone would have discovered this by now. 

Is this the place to discuss blue balls?

Where better than on a chess site?

Oh Boy....! (eyes rolling)

Hey, this is fun.  The math jocks keep returning to their ten commandments from GODel, and then they complain when we call them on it.

If the shoe fits, wear it.

Mathematics lost it's certainty about 100 years ago.  Not so with ersatz mathematicians.

But surely you're joking Mr. Feynman?

Can a stronger opponent at least hold a draw (or better) against a weaker opponent? The answer is assuredly yes.

And it doesn't really matter who's playing each side (man or machine, or Centaur).

It's in the nature of the game, and comes to us from 500 years of chess history and experience.

That's all the evidence we have to work with, to date.  Please make a note of it.  

Not sure what history of mathematics you have been reading.  Not sure why food choices matter to this discussion, except perhaps for misdirection.  

Methinks the gentleman doth protest too strongly.

You're an entertaining two-day-wonder of a sock puppet.

The field is yours, @George_Jet.  

Because someone wants to discount the fact that noone has ever shown a game won without a mistake by his opponent.  It does not matter that this is discounted by someone--it is still a very good piece of evidence.

Also it is a fact that there are more and more draws at the highest levels and this is also evidence that points to something. Some are so buried in saying this is not evidence that they deny the truth that there are more and more draws at the highest levels.  Some even fall from grace by giving anedotnal evidence on this.

The evidence is there staring people in the face but some wish to ignore this evidence.

george_jetson5    there is more to my argument than you related as not only do I say that nobody can show a game where somebody won and there were more errors.  I also have shown a game where there were no errors [by error I mean a move which would change the outcome of the game] and the game was drawn and of course I could show more games.

Now you are really burying your head in the sand  [so to speak] when you say I have shown no evidence that over time the games of the best chess players are tending towards more and more draws.

Here is one of the evidences I gave which you apparently just ignore?

or maybe you missed this evidence.  Look at the world championship matches for the last 100 years. You will see that over time there are more and more draws in these championship matches.

Here is another evidence  look at Centaur Chess [which is the very best chess we have] There is also more and more draws in that venue.

It is easy to say I have no evidence if you just ignore the evidence I give.

As I said the evidence is there staring someone in the face but you must have an open mind to look at the evidence raather than saying it is not there.

This topic dosen't matter since everyone will make some sort of mistake in the game

ajian  The very best players have played many games where they have not made a mistake which would change the outcome of the game.

This is many thousands of games.

By the way while there never has been found a perfect game [one without errors] where somebody won --there are thousands of games where nobody made an error which were drawn.

Regarding the fairly rare Qvs Q and P end games  [we are not talking about all such Q vs Q and P endgames.] These have been solved by humans using a 6 or 7 piece tablebase. Humans solved those end games.

But it is quite a stretch to say because humans cannot solve a rare and particular type of endgame this means their evaluations are suspect.

That end game and the samples give are less than one tenth of one percent of all positions.  Not only that but they were and are solved by humans.  

george jetson  I noticed you completely skipped my evidence that chess is more and more tending towards a draw by the top players and you completely skipped the 100 years of world championship matches.

Also, despite what you say, I already knew that chess is immensly rich and offers boundless opportunities for creativity.  Only thing I disagree with is that you think humans cannot see the opportunities for creativity. I have won very strong chess tournaments by using my creativity.  

Re table bases.  Who do you think invented the table bases? It was humans. So humans have invented one more tool to help them evaluate positions.  [these positions are very rare and if you knew more about chess you would know this].

Computer improvements of heuristics.  Who do you think invented these chess engines. Again it was humans.  This is one more tool humans have to evaluate positions.  Houdini is a person with the last name "Houdine".

Tablebases and chess computers are something we can use to evaluate positons and we use them that way.

Chess engines improve our heuristics.  They are the invention of men and also chess players.  A strong human with a strong chess engine can evaluate postions very well.  You act like humans cannot use the chess engines they invented.

Let us say you are correct and 6 and 7 piece tablebases just appeared and they could solves some chess positions which humans cannot solve--so what? what would be your point if that was true?  [it is not true as humans invented the tablebaes]

Your statement that "nobody knows which first moves might constitute errors" You are very wrong about this.  I know which first moves do not constitute erros. Also any grand master can tell you that these first moves are not errors  1. e4  1. c4  1. d4  1. g3  1. Nf3  .  Because you do not know something does not mean nobody knows that something.

I think you are missing the point here. As I said earlier the phenomenon of best play from both sides seldom happens hence it is more of an exemption rather than a rule. You also mentioned something about 10 billion games that are won because one side made a mistake/ blunder, haven't it occurred to you that that was the "best play" from both sides even if it includes errors or mistakes? We are all humans here and errors or mistake happens all the time, it is part and parcel of our being. However it also happens that two players can play a game that are played flawlessly and ends in a draw once in a while. The games between top players specifically those belonging to the 2700 club are more inclined towards this scenario as compared to amateur games. Anyway people will always find ways to improve on the current technology in order to gain an advantage and get the win. Engines evolve and even get better for the same purpose and that is to win. Yeah draws happen with perfect play but you can count on people to find a way to win! That is something that you can expect from every competitive individual. That is something that you can expect from every chess player. 

WalangAlam  I agree that chess theoretically being a draw [or a win for White] has nothing to do with the average player's enjoyment of the game.

I also agree that one should try hard to do the best he can.

I did not say ten billion games are won because someone makes an error as obviosly some of those games were drawn.  

[it was 100 billion].

Most people know that if a game is won, someone had to make an error. 

Sometimes it is the last error which decides the game.

I am not missing your point, I know chess is played for fun by almost all players. 

Well there is something missing from your assumption of "best play from both sides" since chess is played by players of varying degrees of strength and or ratings. Take for example the best play of 2400 player vs the best play of a 2600 player; there is a significant chance of the 2600 player winning than having a draw; the Odds of winning increases with every increase of rating disparity.

          So I would have to assume that what you meant was "best play from both sides of the same playing strength and or rating". Well that would make sense but it cannot be used as an argument to conclude that Chess is a draw. Please don't say anything about centaur chess since the majority of people playing OTB in schools, clubs and tournaments aren't computer assisted. As I said you are stating an exception rather than a rule. Even if all games of the Elite players will end in draws that still cannot be used to conclude that Chess is a draw for the simple fact that Elite players are the exemption and everyone else are not.