Luck in Chess

Elubas, what about determinism in chess? Does it exist? Are our mistakes predetermined? Are our good moves predetermined?

I think all this is not predetermined but rather probabilistic.

By the way I often use "random" just to mean "practically unpredictable," or not predictable because of insufficient information. But in chess you have the info, and you are the one who chooses not to derive the answer for one reason or another.

You don't have the info about best move in current position (with the exception of theoretical positions that are in your memory like openings and endgames that you've learned). Can you predict how far your move will be from the best one? Is it deterministic?

EDIT: I don't "choose not to derive the answer". I can't derive an answer because both my mental capacities and my time is limited. I can only derive a guess. How far this guess is from the best move is probabilistic.

But do you agree that there is luck when playing against a computer as shown in the example in comment #257?

The info can be derived. If you are in position x, there is some (or more than one) solution that must be true; it's determined by the resources in the position. It's silly to say that just because a human can't find this necessity that therefore this necessity doesn't exist. Why do people think something can only be skill if the skill of the players is perfect? When I say that I want to play a game of skill, I'm not saying I want to play perfectly; I'm saying I want to beat my opponent by having more skill than them.

It is obvious that best move(s) always exists. But it's silly to say that a human can always find it. We are limited by our mental capacities and by thinking time (same is true for computers). That's why we make mistakes. If you want to beat your opponent you need him to make mistakes that are more numerous and more severe than your mistakes. Having more skill means that probablity of this is higher. But it's still just a probability. That's why 1600 on avarage gets only 76% result against 1400, not 100%.

blunder is luck

No one argues against the fact that skill is still the more important driver of the output of the game. It´s only that it´s not 100% skill, there is a small percentage of luck involved too.

By the way, this is like saying that 2000 years ago sailmen could avoid storms, that it was not a question of luck. It was only their fault that they didn´t have the skill to build a meteorological radar because the information to build it was already there in the nature.

Perhaps a little. But I wouldn't really call it hidden in the case of chess. For example, you don't even need to worry about discovering new physical laws; the "laws" in chess are the literal rules, like a pawn can move here or here, checkmate wins, etc. And there are things that can be derived from those rules. Why don't we? A lack of skill. Not because our rules don't favor some moves over others. It's pretty easy for things not to go according to how they should theoretically: humans mess around and do their own thing -- that's human behavior :)

I just don't think that skill was meant to only apply when the skill was perfect -- otherwise we wouldn't find skill games interesting. Yeah we are not going to be sure about things -- it doesn't mean our skill isn't what's going into everything we do, and our results. It just is doing so imperfectly, and so you will get imperfect results. To a betting man, sure, they might be concerned watching two guys play chess because we don't know whose skill will always triumph. All of these concerns revolve around people applying their skill.

I actually don't disagree. I don't think it's 100% skill, but I think the % of luck is insignificant to the point where it's like saying golf is a little bit of luck because out of the zillions of ways a beginner could hit the ball, they might "just so happen" to hit it the right way. You can't always be rewarded completely proportionally to the amount of skill you put in, but skill is always at the center of things (e.g., it is what makes people play the moves they make, rather than probabilistic forces).

And a lot of things that people think are luck are skill. For example, playing a move that "just so happens to be good" often does have logic behind it, but a more unconscious logic. For example, maybe you can't explain why a move is good, but your unconscious has a hidden pattern recognition/understanding that does allow you to appreciate the move without you being aware of it.

In my opinion, the big difference between chess and other games where randomness is in the rules (like backgammon by rolling dice), is that in chess the possibility to play a game without any luck at all exists (like in the future with super computers that will be able to play perfect games calculating all movements).

I guess I think in a way, whatever "luck" happens is the result of you -- you can sort of pick your odds depending on how much you understand the position. So if you are only 75% sure of something you just have to accept what happens whether it's good or not, because you are the one who put yourself in the shoddy "75%" position when with more understanding you could have been 100% sure. That certainly doesn't happen with coin tossing. You don't "pick your odds" there.

The amount of luck involved in a game of chess is something we can not measure. I agree with you that if I play against a GM then the amount of luck will be insignificant, almost zero.

But when playing aginst someone that is your same level, then it increases. How much? I don´t know, but I wouldn´t say is insignificant.

Whatever "luck" happens is not the result of you - it's the result of your opponent making mistakes. You have some limited control of this (like trying to complicate the game) but very-very limited. The rest of your opponent play is out of your control - just like coin tossing.

It's an illusion to think that we can make a concsious choice between 75% and 100%. Because no human can achieve 100%. You can improve your chances by proper training, good sleep, concentration during the game. So it will go up to 80%. But from time to time you will loose to weaker players and win against stronger ones - because probabilities of those events is far from zero. On large number of games it all evens out and our rating corresponds pretty  well to our chess skill. But when just one game decides who wins the tournament or gets promoted etc. - there is quite a big share of chance/luck involved.

Here is a simple math exercise. According to ELO rating definition 200 points difference means 76% result for stronger player. Let's exclude draws for simplicity. Then 1600 playing a single game vs 1400 has 76% chances of winning. In a match of 3 games his chances of winning are 85%, in a match of 5 games - 91%. Conclusion - luck plays most significant role in an event consisting of one game, as number of games increases the role of luck diminishes, results have better correlation with skill difference.

        The above is rather obvious. The more games played between opponents of different levels of skill, the more accurately that difference will be reflected in the outcome. This will be so in any contest, not just chess.

Of course it's obvious. My point was that outcome doesn't depend on skill level only.

You'd be better making it clear, such as "in the following thought experiment" etc etc. Sorry but I can't be held responsible for replying to what you seem to be saying. Anyway, it isn't derailment but very much to the point, since this discussion is about determinism, which is an incorrect point of view, albeit one that's currently been made fashionable by a handful of misguided philosophers and "logicians".

If "indeterminism" is a negation of "determinism", which is the view that all events are the result of a linear causality which cannot be altered in any way, then indeterminism is "true", especially since asserting determinism to be true is a positive claim needing support and evidence, if it's to be taken seriously. Note that determinism exists inasmuch as cause and effect exist but an assertion that "determinism is true" is more of a universal claim and so is different from the admission that cause and effect exist.

It's common for people to find arguments against their beliefs to be "uninteresting".