Sorry, I was less than precise. Instead of "every GM". I should have said "just about every modern GM".
True or False Chess is a Draw with Best Play from Both Sides
Sort:
Well, that's what Spock thought. There are too many variables so the question is irrelevant, no? Like the irresistable force meeting the immovable object. IMHO
@Smyslovfan
Ok, but it's such weak evidence... so weak I wouldn't want to call it that. Hard work? Professionals? Those words are subjective, they don't mean much. GMs are good compared to other humans, their hard work is to beat their opponents not play perfectly. And if you're playing peers then shouldn't we expect most games to end in a draw, and to contain errors?
If you want evidence lets talk about the drawing margin of most endgames and compare it to the imbalances in the opening (not in the Silman sense). Lets talk about things like forcing moves, dynamism in the position, zugzwang, and how they related to positions we know are drawn and won and then again compare to the starting position.
This is much more compelling than saying there have been an arbitrarily large number of games played to date, and each decisive one contained a mistake.
Feb 5, 2015
0
#1985
I've mentioned the drawing margins before too. That is also evidence.
Your previous statement would negate even the evidence that you now accept.
Feb 5, 2015
0
#1986
There is one minor problem with those other bits of evidence though:
GM Aronian has pointed out that there are some starting positions in Chess960 that are very close to decisive (wins for white). He calls such starting positions "ridiculous", but they suggest that what you claim to be "more compelling" evidence regarding the endgame may in fact not be as compelling to grandmasters.
To those asking the impossible, i.e. a won/lost game that didn't have a mistake, I'd like to ask them the same thing - for a drawn game that didn't have a mistake. Every game has mistakes, and will continue to until computers have mastered chess.
Remember also, that white's only advantage isn't tempo, it's influencing the other side's pawn structure and initial layout, which could be crucial.
SmyslovFan wrote:
There is one minor problem with those other bits of evidence though:
GM Aronian has pointed out that there are some starting positions in Chess960 that are very close to decisive (wins for white). He calls such starting positions "ridiculous", but they suggest that what you claim to be "more compelling" evidence regarding the endgame may in fact not be as compelling to grandmasters.
Well, again, then we compare it to the opening. I assume that in those 960 positions there are a lot of threats / forcing moves available.
In your previous post "negate event he evidence you now accept" I don't understand. Some endgames are solved as a win or draw, this is what I'm talking about.
Feb 5, 2015
0
#1989
eoJ1 wrote:
To those asking the impossible, i.e. a won/lost game that didn't have a mistake, I'd like to ask them the same thing - for a drawn game that didn't have a mistake. Every game has mistakes, and will continue to until computers have mastered chess.
Remember also, that white's only advantage isn't tempo, it's influencing the other side's pawn structure and initial layout, which could be crucial.
Actually, there are many drawn games without any discernible mistakes. It won't take long to dig those up.
Feb 5, 2015
0
#1990
leiph18 wrote:
...
In your previous post "negate event he evidence you now accept" I don't understand. Some endgames are solved as a win or draw, this is what I'm talking about.
Please show me where exactly I wrote that. And remember, you are the one who provided the quotation marks.
eoJ1 wrote: ponz in red
To those asking the impossible, i.e. a won/lost game that didn't have a mistake, You are not being clear here. This is not what we are asking. We are not asking for a drawn game. We are asking if you can find one game out of 100 million games where one side won where there were no mistakes in the game. I'd like to ask them the same thing - for a drawn game that didn't have a mistake. Every game has mistakes, and will continue to until computers have mastered chess. This is not true. There have actually been millions of games played where neither side made a mistake. They were all drawn games. Just look at any very short super grandmaster game which ended up a draw and I challenge you to find a mistake in that game.
Remember also, that white's only advantage isn't tempo, it's influencing the other side's pawn structure and initial layout, which could be crucial. I know what White's advantages are and the total of White's advantages are only worth approximately 1/4 of a pawn and that is not enough to win.
The opinions of grandmasters and super grandmasters that chess is a draw means a whole lot. Virtually all such players believe chess is a draw when neither side makes a mistake.
They believe this for the same reasons I believe chess is a draw.
These players know a whole lot more about chess than most people posting here.
It is pretty well known that the stronger the player, the more likely he will believe chess is a draw when neither side makes a mistake.
Feb 5, 2015
0
#1993
@Grobe: you asked for at least two meanings of proofs. I will give you rour types of proof. There are a lot more types of proofs and every type of proof is a proof with a different meaning. The first one is used very often:
1. you can prove statements using deduction, like
premiss a. all humans are mortal
premiss b. Socrates is a human
conclusion c: Socrates is mortal.
However, it can be proven that the scheme itself has not the capacity to prevent that it can be used to create unprovable conclusions.
You can prove the unprovable using deduction as we see in
premiss a: A barber shaves men who do not shave themselves.
premiss b: The barber is a man.
conclusion c: the barber shaves himself, does he?
2. The second kind of proof is by contradiction:
A simple example is the assertion that there is a largest even number to exist.
You can proof that assertion to be false by showing that that number z is twice another number, lets call that number m. There exists bigger numbers then m (z for instance) and there is no rule that prevent doubling the numbers between m and z. Doubling any number between m and z will create an even number bigger than z. Hence is there no larger even number.
This whole thread is about this type of proof. Ponz asserts that chess is a draw and a lot of other people disagree with that assertion. Both assertions, namely chess is a draw or chess is a forced win for white or black, can not be proven positively. At the current level of technology and knowledge are the only possible types of proof proofs of contradiction. And even those types of proof are not possible yet. But it is kind of weak evidence. If you can prove that chess is not a forced win for black, then does that not imply that chess is a draw or a forced win by white.
3. A third example of proof is proof by implication:
If a causes b and b causes c, then can you conclude that a causes c.
The number of times you multiply a number with another number is equal to the number of times that you can divide the result of that multiplication with that number.
It works, except for zero of course.
4. That brings me to the fourth kind of proof, namely by example:
Altough that can be a fallacy, it can bring forth acceptable proofs.
For instance, the assertion 'all prime numbers are odd' is countered, that is proven wrong, bringing in the example of two, which is said to be the oddest prime.
Smyslovfan was asking for such an example some pages ago.
It's draw.
CookieMonster wrote:
I will say this. You stating there are millions of games where neither player made a mistake proves you do not know enough about chess to make such a statement. There is barely a 5 million game database in existence where none of the games are duplicates, and among those games the majority are between 2000 and 2600. I will GUARANTEE 98% of those games are ridden with mistakes. GUARANTEED!
There are data bases over 6 million. But of course these contain only a tiny fraction of the games played by one player. For example there are only about 100 of my games on the data bases and I have played more than 10,000 games.
Most, the vast majority of players in this world, do not have any of their games on a data bases.
So, it is hard to estimate the number of chess games ever played by earthlings. I would guess a 100 billion--but it is only a guess.
Having said, all this, I will admit that I cannot prove or demonstratee that at least 2 million games have been played without an error. If you count games such as 1. e4 e5 draw agreed--you might come close.
But you are correct 98% of all games played are riddled with errors.
So, I am saying, I made a mistake, More accurate is that thousands of games have been played without error by either side.
[I can admit a mistake]
SmyslovFan wrote:
leiph18 wrote:
...
In your previous post "negate event he evidence you now accept" I don't understand. Some endgames are solved as a win or draw, this is what I'm talking about.
Please show me where exactly I wrote that. And remember, you are the one who provided the quotation marks.
Post #2026
SmyslovFan wrote:
I've mentioned the drawing margins before too. That is also evidence.
Your previous statement would negate even the evidence that you now accept.
Sorry if I'm being slow here, but I didn't understand what you meant by it.
If you notice most of thier game they each more aless sacrifice the same pieceis and left with same more aless at the end but thats my opinion but when ive been watching some games of magsun carlsun hes just blowes me away with pure skill and talent i still would'nt mind to play him even though I would lose but just to say ive played him!
CookieMonster wrote: ponz in blue
ponz111 wrote:
CookieMonster wrote:
I will say this. You stating there are millions of games where neither player made a mistake proves you do not know enough about chess to make such a statement. There is barely a 5 million game database in existence where none of the games are duplicates, and among those games the majority are between 2000 and 2600. I will GUARANTEE 98% of those games are ridden with mistakes. GUARANTEED!
There are data bases over 6 million. But of course these contain only a tiny fraction of the games played by one player. For example there are only about 100 of my games on the data bases and I have played more than 10,000 games.
Most, the vast majority of players in this world, do not have any of their games on a data bases.
So, it is hard to estimate the number of chess games ever played by earthlings. I would guess a 100 billion--but it is only a guess.
Having said, all this, I will admit that I cannot prove or demonstratee that at least 2 million games have been played without an error. If you count games such as 1. e4 e5 draw agreed--you might come close.
But you are correct 98% of all games played are riddled with errors.
So, I am saying, I made a mistake, More accurate is that thousands of games have been played without error by either side.
[I can admit a mistake]
I am glad we are getting somewhere. However, I am not even giving you thousands out of a 5 million game database. who said out of a 5 million game database? I am saying out of ALL games ever played which is more like 100 billion, not 5 million. We find improvements all the time in past games. And even the improvements are said to have possible improvements. sure we find "improvements" of games played. But these are usually improvements in practical play. To change the result of a game played would have to be a very big improvement. The fact that we can find improvements [which is a vague term] has little to do with my statement that there have been thousands of games played where neither side made an error. We are defining "error" as a move which would change the result of the game played.
Here I will give you an example of a perfect drawn game...Carlsen had a 6 game winning streak in the Tata Steel Chess Tournament. Then he was scheduled to play Ivanchuck where Ivanchuch had White. Ivanchuch decided to not risk playing for a win and delibertly played for a draw. The game was drawn and I am making a very strong guess that it was a perfect game where neither side made a error which could change the result of the game.
The closest we have to games with no errors are games where the game was drawn in the opening where we know the game is as perfect as possible. For example: There very well may be 10 games in this database where the games ended after 1. e4 e5 2. Nf3 Nc6 3. Bb5 Nf6 <--- and the games go 10-15 moves into theory of the Berlin. The problem is those games are not real. Sure they are real just because you call them GM draws does not make them unreal. They are GM Draws that are effectively attempts at avoiding tiring games that their energy could be best saved for round 5 and 6 in a 6 or 7 round tournament. After move 15-20 the risk of anyone making a mistake in the middle game increases exponentially depending on how familiar they are with said opening they are studying. Also you are just posting about 5 million games when the number of games played on this earth is many times 5 million--it is in the billions.
Want proof? Magnus Carlsen has proven time and again he is the best player in the world. But, how many mistakes can computers find in his games just by him that we mortals don't understand? It depends on what you mean by "mistakes" which you are not defining here. I know countless attempts by cyborgs. Time and again you go into a dedicated chess site, lets say, FICS or ICC where kibitzing is on super levels. At least 80% of all kibitzers in there are Cyborgs pretending to be average players. All of those 80% believe they can make people believe they know when Carlsen makes an easy mistake. This is probably true. They probably can determine when he makes a mistake, but it's when their engine tells them.it does not matter who or what points out a mistake-a mistake is a mistake. Why is this proof? proof of what? What point are you trying to make? sure we can understand why an engine can see a mistake of a gm quickly . also it does not take 3 pt advantage to win. I can't make sense of what you are trying to say in this paragraph? Because engines have seemingly demistified GM's. Obviously we in mortal levels can't understand why an engine can see a mistake of a GM quickly, but the illusion that computers make it easy makes them believe chess is easier than it seems. But the proof is that those cyborgs are generally correct, up to and around 80-85% of the time. When it's an evaluation above 3 pawns. And You can do the same thing with pretty much all over the board master level games.
Now.. If Magnus Carlsen can make more than a few, lets say 8-15 mistakes in a single game being the world champion x 3 AND being the highest rated player currently and the current record holder.. How can he make so many mistakes, yet there is thousands of games with no mistakes? The answer? You are "Assuming" there are that many games with no mistakes.
Wanna know something that made me happy? I had an IM trainer once. Recently. I asked him flat out, "How should I go about looking at the world championship games in a mind set to learn from them?"
He told me equally flat,
"Grab a cup of coffee and watch them like you were watching a sports game. I don't fully understand the games myself."
I knew this answer when I asked, but I guess I wanted him to tell me. If an IM can admit this, why can't you?;-) Obviously he knows there will be mistakes. But since we are not on world class levels, (Of course he's a lot closer), why not admit our mortality in chess?
The likes of Carlsen do admit their mortality, but they also acknowledge their "Demi-god" status at the same time I am sure.
Did you know the top 50 are in the group of the "Less than 1%" group??
How about this one. I am on the top 10% in both my state and national. How can you believe that people between 1% and 5% play even one game where no errors exist when the "less than 1%" can't produce games with no errors? It's illogical.
BTW: You can't consider all games humans play when they aren't documented. That's like taking a pole of how many people know how to play chess and 10% of the world poles in, but you consider that the ones who didn't pole must have knowledge of chess. That is also poor form.
CookieMonster
YES we can consider all games played -not the less than one tenth of 1 percent you want to consider.
Your trainer gave you bad advice to not look and study the games of the very best players unless your skill levels were to short to do this.
Yes, I can identify my own mistakes in my games. If you cannot identify your mistakes you will not improve much.
But regardless of that, a super gm with the aid of the best chess engine can identify mistakes. In fact, a super gm can identify his mistakes without using an engine.
If you that Magnus makes 8 or more mistakes in a game, you are simply wrong.
There have been drawn games of 2 or 3 moves which are undocumented but contained no mistakes.
Ponz asked for those arguing against the premise that chess is a draw to provide even one example.
So far, over 2000 posts have been made, and not a single decisive game without errors has been produced.
Which sets up a completely circular line of reasoning when your definition of a mistake is a move that results in one side or the other losing....
You don't really understand what "circular line of reasoning" is apparently.
He is asking for EVIDENCE that chess is not a draw. After hundreds of millions of games, SOME evidence should have turned up by now.
A lack of evidence after 100s of millions of games isn't final proof, but it's a strong suggestion that the truth lies in the other direction.
I don't understand why 100 million games is proof of anything. It's like saying mathematicians have tried to solve a problem for 100 years, so I can declare it's unsolvable. I mean, it's just meaningless. Humans aren't perfect, and chess is so complex, that no number of games means anything.
He may be asking for evidence, but how can you give evidence before chess is solved? You can give reasonable assumptions (I made a few clever ones myself a few pages back, why chess is a draw) but that's it.
But ok, if we choose to accept we don't like each other's phrasing, then one side can stop the topic. Grobe has said though that he doesn't want new members to read the phrasing ponz uses and get the wrong idea. Maybe that's why he was unwilling to stop for 100 pages.
Don't expect to see Grobe respond though, ponz apparently blocked him.
You're making the same mistake, conflating evidence with proof.
The 100s of millions of games isn't a random selection, but the very best efforts of the very best chess players (and many others) to play the perfect game. This is indeed evidence. It's not final proof, but it is evidence that chess is most likely a draw. All the thousands of hours of analysis by top GMs and engines have yet to produce a single line that gives White a decisive edge.
That's not a final proof, but it is evidence. So yes, evidence is available. It's not sufficient for some, but for every GM, it's sufficient for them to accept that chess is a draw with best play.