How much of chess is luck?

Given that in chess the consequences of a move can only be apparent several moves later, and moves that are rigorously calculated to be strong can turn out to have weaknesses that are surely practically impossible for a human to have forseen, it seems to me that Chess involves a degree of luck which is not often discussed. If you have two candidate moves, each of which seems equally strong to you, but one of which is, unbeknown to you, actually weaker, surely then it is just a question of luck which one you happen to randomly choose?

Nope, it is accuracy.

Sensible answers only please.

The candidate move you make usually corresponds to a formulated plan of attack or defense you are envisioning. If you are referring to an extreme case (a Dvoretsky problem- which in my humble opinion are a form of mental masturbation that rarely if ever occur in OTB chess) then yes there is a degree of luck involved. Or let’s expand that out- maybe you will play an opening your opponent doesn’t know and catch him off guard (luck),
Or not get food poisoning or lack of sleep like they did- luck! Luck is overstated. Just worry about playing chess that suits your style and not over thinking moves.

Unless you think that a human can calculate every variation resulting from a move, then there has to be some luck  involved. The only question for me is why people deny this?

0 percent in OTB

Chess is all luck.

Yes there's luck in chess. For example, if I (as a 1550) make a perfect match it's not all skill obviously becouse if it was I would be much higher. Another example, you are completely wining in an easy endgame (10 points ahead or so) and you have time on the clock but you stealmate becouse you wanna finish fast, isn't this good luck for your opponent?

The higher rated you are, the less luck there is

I have a sense that people have such a fixed mindset that they've never even considered this question.

I'll do a worst case first.  As the discrepancy between players get larger the odds that the result is luck decrease.  For two players perfectly matched it is all "luck"in the sense you are using it.

 

 

Green and red represent the results based on normal and logistic distributions respectively depending on where you come down on that issue.   In short I'd say that if X has a probability of p of beating Y then X's result being the result of skill is 2*p-1 for 2*p-1>0 or 0 otherwise. Reversing this the chance it was "luck" is 2-2*p for 2-2*p > 0 otherwise 0.  Using that definition of luck a weaker player beating a stronger or equal player is always the result of luck while with a 1000 point discrepancy luck plays no role at all.  For something more reasonable like 100 point discrepancy, the stronger player winning is 26% skill, 74% luck.  

 

So that's the worst case.  The problem is that there is some additional serial correlation between individual players.  That is if X beat Y even one game X has a moderately higher chance of beating Y in future games than the rating would suggest.  Players "fit" each others styles.  The above assumes that "luck" is part of picking players, like a random blitz game here.   But if you are choosing your opponents based on fit that's a question of skill and not luck.  Which increases the skill percentage.  

Zero luck all skill

Assuming that you play the best moves, you will only win if your opponent doesn't (play the best moves). So how is it anything other than luck that you win rather than draw?

#10

Look OP I'm pretty sure virtually everyone has.

At lower levels, luck is highly involved. You make what you thunk may be a great move- your opponent misses an easy counter-move. But is that luck, or lack of skill, seeing as neither you NOR your opponent have made a move. Probably a bit of both, but mainly luck.

Which colour did you get in your game? That's luck. And, if you only play well when using s certain opening with a certain colour and get that colour and can use that opening, is that luck or skill? Again, it's both. You got lucky you had the OPPORTUNITY to use that opening but you still needed a level of SKILL to carry it out.

If your opponent makes a good, unexpected move, and the only way you can defend against it is by using a piece that just so happens to be close by, is that luck, or skill? Yes, you got lucky that the piece was there, but is it actually your opponents LACK of skill to see that piece in the first place?

If, in an extreme example, you count a 1000 player not seeing that mate in 35 as luck, then yes, virtually everything in chess is luck.

But if you see it as lack of skill, then chess is very skill and accuracy based.

#13

Well your opponents not-playing-perfect-moves may not be unlucky, but simply there LACK of skill.


Again I'm not sure weather the above argument Holds any weight at all , and don't personally have a strong opinion either way

I'm not saying that chess isn't skilled based. Clearly it is. I'm questioning how much of it is luck. Unless you think that the top players can calculate every possible variation leading from each move (which isn't credible), then there must be some luck involved. I'm more interested in why people are in denial about this.

Of course there is luck in chess. No reason to deny that. You can be lucky that your opponent choose to play your favorite opening. You can be lucky that your opponent plays in a way so that there are tactics you can exploit, but you need skill in order to be able to exploit those tactics. Also your examples shows how luck can be part part of a game. So all in all it should be quite clear that you won't be able to win a game without a combination of luck and skill.

Richard_Hunter,

thanks a lot for your post - I agree with you 100%.

Our skill level, reflected in rating is some average. It alows to roughly predict the result of competition on large number of games but for 1 decisive game the luck plays a very big role. Your skill level defines how often you will make mistakes and how serious they will be, but again on average - there is no way to predict mistakes in any given game. And probability of human mistake is never zero.

I once lost an OTB game to a player 150 points below me. Further analysis showed that he's made no mistakes for 30 moves and I was the first one to commit an error. Was this quality of play typical for his level? No. Was it possible? Yes, in one game out of 20 maybe. So pretty sure I'd win a match of 20 against him according to rating prediction, but the reating difference also predicts that sometimes I will lose. And that happened to be that only game. I also have the opposite examples when stronger players committed errors early in the game and I was able to win this game (but probably not the hypotetical match against them).

Unconscious is another big factor. Intuition, patter recognition are all unconscious. Will intuition suggest good ideas? Will pattern matching work in this particular position? That's pretty random, there's no guarantee.

I have already been accused before that by talking about luck I try to avoid responsibility for not being playing good and not training hard enough. On the contrary! It's obvious that by improving our play via hard training, concentration etc. we decrease the probablity of our mistakes and improve our chances to win. But those remain just chances...

For sure there is a little bit of luck in chess. 

Personally, about 20-80between luck and skill

"Luck" being defined as someone screwing up in the worst ways possible.

like the last 4 tournaments I've been to, I was easily winning >5 games in which I had ample time and a wonderful position yet lost inexplicably

 

there is also luck in colors, which i also am never lucky in (70/170 games as white) so yeah...