Luck in Chess

I'm not sure what you mean by that. If you mean the players being uncertain whether they will win or lose, I would agree, but I'm not sure if that is what you are getting at? Basically uncertainty doesn't always tell you the nature of the information that you are uncertain about. You can be uncertain about something due to a lack of skill.

So if you mean that a human playing a perfect information game like chess will cause them to make predictions that they are unsure about, ok, but that sounds more like sociology :)

Agree leiph18.

I still don't see how those who disagree address the argument that mistakes are always made and they are always unpredictable.

Well, yeah, if I was betting on a chess game for the white player to win, for example, there would be chance involved in that bet. But that's not actually playing. What goes into some mistake is a deliberate thought process you have. It may be a right or wrong thought process, but it doesn't operate randomly. A person has full control over their thought process; not full control over whether or not it's good, but that's irrelevant.

Leiph: Chess is a game where players are given complete knowledge of the gameplay. Nothing is certain, unlike poker or backgammon. Now it is up to the players to process the information to come up with a conclusion as to whether this is the best move or not. Can this move give me the highest chance of winning? The players have to come up with a solution.

BUT, The players LACK the ability to derive a conclusion to a position in the middle of a game of chess. This lack of ability causes mistakes in one's game. And the other players has to take advantage of the mistakes. Now the other player has NOT been lucky. He has created circumstances in which the other player will make mistakes. This is not luck or chance.

Yeah I'm not sure why people think mistakes and skill can never go together. It's actually kind of easy to put them together: I made a mistake, because I wasn't thinking very well at the time I made the mistake :)

People who support the theory that luck is involved in chess claim that the chance that the other players will make a mistake is luck. Nothing can be further from the truth.

Lemme elaborate.

Suppose there are two players. Player X and Player Y. Player Y is more skillful than Player X. During the game, Player X makes a mistake. This mistake is taken advantage of by Player Y and thus he goes on to win the game. Was this luck? No. How? Because Player X was responsible for his actions that determined the fate of the game. He had a thought process to use and get the best solution. But he didn't use that well enough and this resulted in a mistake. He had complete control to determine what move to play, it didn't occur randomly. So the circumstances that were created by Player Y helped him win because player X made a mistake giving the game. Player Y may also had made a mistake, but he didn't. His brain functioned as it should have been. His ability was better than Player X.

So, Player X's mistake was not due to luck, but it was legitimate for Y to win. No luck, no chance. Only ability in chess. Explained!

I am not so sure if "a person has full control over their thought process". Large part of chess skill is based on pattern recognition. But if you know a pattern in theory it doesn't mean you will always recognize it during the game - it is probabilistic process. I've lost my recent OTB game against a weaker player because of missed simple tactics (removal of defender). I'd estimate that in 9 cases out of 10 I would recognize this motif. Here something went wrong in my thought process (tiredness, time pressure etc.). Will I see this tactics next time? I'd say with probability of 90% yes but I can't be sure. Of course through proper training I will improve my chances, but the chance of mistake will always be there.

I don't mean the players are uncertain (although they may be), I mean they don't have access to all the information.

How often are your evaluations something like this: "white has some chances, and black has some chances, it's an interesting position where I like my trumps, so I choose this line"

Or annotations like !? which mean interesting. You know the only true evaluation is win, draw, and loss. But this isn't how humans play or think.

I'd call these educated guesses... and to me this is a huge appeal in chess. The mystery.

Like you though I wouldn't call it a lucky game. I don't compare it to poker, because the rules of poker describe a lucky game (unlike chess rules). I'd just say there is a non zero amount of chance when humans play chess against each other.

@Post#127: So you seem to be talking about how performance level can vary. I would just say that skill varies. I don't think that's self-contradictory or anything. It's just saying that there are a number of things that go into the skill of your performance. If I am sick, this will, temporarily, lower my skill because certain mental capacities I have that help me find good moves will decline. So while a person's skill can vary, this doesn't change the fact that it's skill that impacts what happens.

Sometimes knowledge has an inverse relationship to performance.

In the face of a surprising move by a foolish opponent, knowing a certain strategic idea may attract you to a usually very correct response. However in this position because one pawn or piece is just a little off, there are deep tactics (which both players will see in 3 moves) which refute the strategic idea. The surprise move was a blunder, but the usually good response is also a blunder.

In fact, a player who had never encountered the usually good strategic idea would be more likely to find the best response.

In this case the game ends with the less skilled and less knowledgeable player winning the game.

This can happen because there is unknown information and moves made by very good players amount to very educated guesses.

And what if Player Y was the first one to make a mistake and Player X takes advantage of this and wins? Probability of this event is not a zero. I don't know what does it mean "it is legitimate for Y to win", I prefer to say "Y has higher probability of winning". Rating is based on probabilities too. For example if Y is 200 points ELO above X he should score 76% on average.

"I mean they don't have access to all the information."

They don't? Pretty much by definition they do, it's just not easy to use.

My distinction between say, chess and poker would be something like this: in poker, some uncertainty is literally inevitable, no matter how good you are; in chess, the uncertainty comes from a lack of skill/understanding. That doesn't change the fact that a person is playing some move because of xyz, not because "that's how the cards turned out." Again a person can be surprised if their move turns out good or bad; that's a consequence of seeing how their subjective interpretation of the position equates to reality. That is not what you do when you call a coin flip, even though that has uncertainty too.

So I guess to summarize my view: I don't see why uncertainty and misjudgments, etc, are mutually exclusive with skill. In fact, it's because of the (imperfect) skill we put into the game that we get the kinds of judgments we do. It's because of different levels of skill that we "probabilisitically" think a GM will beat someone that is not a GM.

Uri65: Yes. But from where does the probability come from. It comes from a lack of skill. A lack of understanding and a lack of ability. Thus, the mistakes are made because of the players lack of all that, and this is not luck. The probability stems from the fact that the higher rated player has a less chance of playing poor and losing. And this is because he has a better understanding of chess.

But yeah, I think there are problems with assuming that any time you merely think something will happen without knowing, some "magical" element of chance is added to that thing. If this were the case, even math itself could have an element of chance, if you're "unsure about which formula represents the situation accurately" or something. Humans can't turn something into chance merely by acting dumb :)

Yes, there would be certain kinds of results we would expect from certain levels of skill. And sure we can quantify it by saying something like "if no other information is available, there is an x percent chance this person will beat this person who is rated 200 points lower." Well yeah, I would certainly think a more skilled player would have more possible scenarios in which he would win than ones in which he would lose :) Seems to be what skill is all about.

I guess we mostly agree. Probability of mistake is a function of skill - on average more skilled player makes mistakes less often and they are less severe.

Sure my first statement seems false, I wanted you to reconcile the seemingly false statement with your own evaluations. I suppose you do a bit when you say "it's just not easy to use." In that case I'll try to use that against you.

A coin flip by your definition has no chance. All the information is in front of you to calculate which side it will land on, it's just you can't do it in that short amount of time.

In cards too, I believe the standard in high stakes in Vegas is to use a new deck which is shuffled a certain number of times (3 I think?). You have all the information there too (if you eye were a slow motion camera).

Chess, by comparison, is many many many many orders of magnitude more difficult to process all the way to the truth.

-----------------------------

I don't think skill and uncertainty are mutually exclusive. There is definitely skill in chess (and not in a coin flip). I'd just say we make educated guesses.

I luckmated you! 1-0.

Yes, that's a key point I think.

Humans have no effect on the thing itself.

Chess itself is a game with zero chance.

But our performance in chess has some small chance.

"In this case the game ends with the less skilled and less knowledgeable player winning the game."

Yes, but that doesn't mean the game itself was probabilistic. The less skilled player made the better decisions, and had clearer thought, for that game. If we wanted to measure overall skill though, we would want the players to play in various different types of positions and so forth. Besides, a player is responsible for using their knowledge correctly; if they don't that's their own problem.