Forums

Is there such thing as "luck" in chess?

Sort:
playerafar
playerafar
Show your flair!
Jan 30, 2025
0
#4381

PD - here's a link to the very first page of this forum 14 years and 5000 posts ago.
Its got some very good posts in it -
https://www.chess.com/forum/view/general/is-there-such-thing-as-quotluckquot-in-chess2?page=1 ---------------------
Suggest you read it. Could take two minutes of your time though. Or more.
Regarding recent posts here about luck in chess there were several good ones including from BC who mentioned about how the less time available on the clocks at the beginning of the game - the more luck will be involved.
And there's other variables affecting how much that luck is a factor.
They could be listed and summarized - just for you - but if you really are interested you could look at page 1 of this forum.
There's also the fact that chess isn't solved - consider that too ....

playerafar
playerafar
Show your flair!
Jan 30, 2025
0
#4382

'Somebody' is trying to push the idea that 'randomness' needs subjectivity.
And he's claiming 'we're talking' 
He didn't 'establish' anything.
He's a lot like the Guy .... but much younger.
---------------------

"Randomness is a fundamental concept in many fields, like mathematics, physics, and computer science, and it doesn't depend on any subjective human experience. For example, in mathematics, random processes are defined rigorously using probability theory. The outcome of rolling a fair die or flipping a fair coin is considered random because each possible result has a predictable probability.

Similarly, in quantum mechanics, the behavior of particles is inherently random. When measuring certain properties of particles, like their position or momentum, the results are governed by probability distributions, and these measurements aren't influenced by any observer's subjectivity."

playerafar
playerafar
Show your flair!
Jan 30, 2025
0
#4383

So now whoever is going to work the word 'randomness'.
There's always more semantics to discuss ...
Dio qualified about players in the game ...
and posted a couple of good links.
And there's this business of 'intrinsic to the game' ...
That phrase hasn't had enough attention yet.

DiogenesDue
DiogenesDue
Jan 30, 2025
0
#4384
ibrust777 wrote:

Here I might be, as Optimissed says, getting too complicated... but I want to speak to a fundamental misunderstanding certain people are having about what a chance event is.
Also... the programmer working on this forum needs to be replaced with someone who knows how to code.

A chance event is incidental, it happens without an apparent cause. Apparent is a very important word here.

Indeterminacy is a state that's not clearly defined, one which could lead to multiple potential outcomes. It's a state that's not the result of a known, measurable cause. It exists with reference to an observer, a context inwhich known and unknown is defined. An indeterminant cause is a cause we infer as acting within a system prior to our observing the state of the system.

Randomness is a statistical manifestation of indeterminacy.

Thus if you're imagining, or trying to argue, that randomness can exist independent of a human observer, that it can somehow be a function of purely physical processes... that just simply isn't true. The very concept of randomness relies on some observable context.

Furthermore, there's an essential incompleteness to the purely physical view, since it presumes that a physical system can fully determine events without reference to its observers, but the observer itself is part of the system, and the event itself is an observed outcome by definition.

Outside of a subjective context indeterminacy can't exist, events can't manifest, and randomness - the statistical manifestation of indeterminacy, doesn't occur. So when we speak of luck, and randomness, we are always by definition speaking with reference to some subjective context.

[AND]

Luck is an evaluation of a chance event. 
Yes, the evaluation can be subjective... but whether a chance event occurred is another question entirely. 
Now, with respect to chess we actually assume some proof of chess exists by which a move can be said to objectively be good or bad. Thus the evaluation of a chess move is not merely subjective. So you're basically wrong there, but that's a digression. 
Because again, we're not arguing over whether luck in chess is good or bad, but whether it exists at all. i.e. whether some chance event occurred at all.

In chess, when you make a move while not knowing what the outcome is, the actual outcome is a function of chance to some degree. 
And as I just established in the previous post, the concept of chance exists in reference to some observable context. 
It is the same when you do anything without knowing the outcome - if you were to whimsically walk into a diner you've never seen, and get food poisoning... you chose the wrong diner, you had bad luck. 
We call this luck because randomness exists in reference to some observable context, otherwise the concept is nonexistent, and the event never manifests. For why, read the previous post.

Now... I'm going to describe this in set notation, and maybe it'll make this easier.

Player P makes chess move M without knowing whether the move is good, i.e. without having a proof of chess. Thus, in the set of good moves G (moves leading to a win with perfect play), bad moves B, and neutral moves D; given that M belongs to only one of these sets, player P does not know what set move M belongs to - i.e. M ⊆ (G ⊕ D ⊕ B); M ⊆ ?;

Now, the probability P(M) of M ⊆ G (the move being from the set of good moves) is the number of good moves / total moves, i.e. P(G) = n(G) / n(T)
Likewise the probability of a bad move is P(B) = n(B) / n(T), and a neutral move is P(D) = n(D) / n(T).

So, given that move M is partially indeterminate, i.e. its outcome is not fully determined by the player - due to a lack of perfect skill, we can say that M is selected partially at random from the set of moves T.
i.e. M = x ∈ T

We can then then assess, very straightforwardly, whether M ∈ G - i.e. whether M was a good move, i.e. whether player P had good luck in choosing move M.
P(G) tells us exactly how lucky player M was.

Now, it's not a perfect model and I'm not a mathematician, but I'm hoping it leaves little room for ambiguity.

[AND]

If a player doesn't know what the correct move is, and plays any random move... there's a chance they may play a good move, and a chance they may play a bad move. Now, if they're a good player there's a high chance they'll play a good move. They've deeply calculated the correct sequence of moves. Still, they remain uncertain as to what the best move is, because they can't solve chess in its entirety, therefor we still model their move using a probability function, it's just a function weighted more heavily toward good outcomes. Now, over the course of a game - 40ish moves usually - where every move has a high probability of being good, the game will resolve to a good result favoring the good player in almost all cases. Mathematically you'd just describe this as a series of probability functions fed a series of events, where each event had a high chance of a better outcome for the better player, owing to the lesser level of indeterminacy due to his skill. And eventually the game would just resolve to a win for the better player more often than not.

Obviously the result is indeterminate, because the player doesn't have a proof of the game. Just by definition it is. Randomness is just a probabilistic description of that. Luck is an evaluation of that. It's actually not complicated!

Again, as has been pointed out many times - your argument relies on stripping the player out of the equation, setting up an artificial scenario where we think about chess in objective idealized terms. When you remove the player you remove all luck from the game, that is true. But you also just cause your argument to not apply to chess as it's actually played by any person in reality. Again, luck is an inherently human concept, it relies on observers, as does the very notion of randomness. 

Again this traces back to your flawed concept of luck which imagines it exists apart from some observable attribute.

I suggested you try Tarot cards as a way of starting out, just to get you some experience. But based on your argument here it's becoming clear to me that I overestimating you.

You have made a tremendously long argument over 3 posts that I am correct, so thanks for that.

You're apparently a developer, so the idea of chess being a self-contained construct with the player entities passed in as parameters/objects should be quite easy to understand, I mean unless you don't know anything about OOP principles, etc. Humans are not required in a game of chess, physical boards and pieces are not required in the game of chess, and the very laws of physics and this entire universe are not even theoretically required as the venue for a game of chess. So in the end it scarcely matters whether luck exists objectively in our universe, because that is not required for the instantiation of an instance of the logical construct known as chess. You can think of a game of chess as a pocket universe with its own simplified set of laws and properties.

The arguments of many here that you cannot logically separate the human players from the game might have been tenable a century ago, but are completely disproven in today's world. It's a conceit to imagine otherwise.

I'm not sure why you are preaching to the choir about luck being subjective and relying upon observers for context...the post you replied to makes that very clear. I don't agree with your journey to the conclusion, though. It's amusing that you are arguing particular points because when Optimissed figures out you are arguing from a deterministic point of view to make your contorted argument work, the fireworks will be entertaining.

If I am being "overestimated" because I don't accept the power and mystery of a Tarot deck reading, then please, keep right on overestimating me...

DiogenesDue
DiogenesDue
Jan 30, 2025
0
#4385
playerafar wrote:

So now whoever is going to work the word 'randomness'.
There's always more semantics to discuss ...
Dio qualified about players in the game ...
and posted a couple of good links.
And there's this business of 'intrinsic to the game' ...
That phrase hasn't had enough attention yet.

It has had enough attention, just not in recent pages. Note that what is meant by "luck", what is meant by "in", and what is meant by "chess" have all been pored over many times in this thread. The differences in opinion on these definitions is why this thread is always in a cycle where there's a flurry of posts with people arguing the same things they did last time, followed by periods of almost no posts when people get tired of arguing the definitions.

Kotshmot
Kotshmot
Jan 30, 2025
0
#4386

In our case, @DiogenesDue , I'd like to think we're not in disagreement about the definitions. I think we have are pretty similar view about the terms "chess", "luck", and "in". I actually made a post defining probability, chance and luck some pages back. In our differences I believe it's about logic.

What do you think about the game show "Do you want to be a millionaire"? The one where they pick an answer out of 4 choices and ring a friend who googles the answer for them. Is there any luck involved in the game?

DiogenesDue
DiogenesDue
Jan 30, 2025
0
#4387
Kotshmot wrote:

In our case, @DiogenesDue , I'd like to think we're not in disagreement about the definitions. I think we have are pretty similar view about the terms "chess", "luck", and "in". I actually made a post defining probability, chance and luck some pages back. In our differences I believe it's about logic.

What do you think about the game show "Do you want to be a millionaire"? The one where they pick an answer out of 4 choices and ring a friend who googles the answer for them. Is there any luck involved in the game?

You used to do better on this thread before you fixated on this "chess is like a multiple choice test" idea. The notion has a lot of problems, not the least of which is that the "multiple choices" in chess are not binary, they are graduated, and that there can be multiple right answers in a given chess position.

playerafar
playerafar
Show your flair!
Jan 30, 2025
0
#4388
DiogenesDue wrote:
playerafar wrote:

So now whoever is going to work the word 'randomness'.
There's always more semantics to discuss ...
Dio qualified about players in the game ...
and posted a couple of good links.
And there's this business of 'intrinsic to the game' ...
That phrase hasn't had enough attention yet.

It has had enough attention, just not in recent pages. Note that what is meant by "luck", what is meant by "in", and what is meant by "chess" have all been pored over many times in this thread. The differences in opinion on these definitions is why this thread is always in a cycle where there's a flurry of posts with people arguing the same things they did last time, followed by periods of almost no posts when people get tired of arguing the definitions.

"The arguments of many here that you cannot logically separate the human players from the game might have been tenable a century ago, but are completely disproven in today's world. It's a conceit to imagine otherwise."
'logically' there isn't necessary.
A better issue is whether you can separate players from 'game' regardless of what adverbial 'ally' is assigned or not.
If there's going to be multiple definitions of 'chess' and 'chess game' then yes - sure you can.
-------------------------------
300 years ago - a chess player could have said to another player - 'Say a person just wrote down a series of somewhat random chess moves on a piece of paper and moved the pieces accordingly and continued doing so till the process ended up in a win for one side or a draw.'
the other player: 'say you did - so what?'
first player: 'Could we call that a 'game' of chess? Could we call it 'chess'?'
Reply: 'Sure we could. On both.' - or we can have a game with players'
First: 'The point: Wether with players or without - we don't have to see the players as 'external' to a game of chess when we're using that defintion of two players playing each other at chess.'
Replier: 'Of course you don't. What's your point?'
Initiator: 'That there's Luck in Chess.'
Other: 'Of course there is. The fact that we could define a 'game of chess' as simply a sequence of moves ending in a result and argue that has no 'luck' in it doesn't exclude luck from chess in its much more common definition.'
Orignator: 'You spoiled it by agreeing with me. I wanted an argument.'
Other man: "Sorry. I'll try to disagree next time. What do you want to disagree about?"

Kotshmot
Kotshmot
Jan 30, 2025
0
#4389
DiogenesDue wrote:
Kotshmot wrote:

In our case, @DiogenesDue , I'd like to think we're not in disagreement about the definitions. I think we have are pretty similar view about the terms "chess", "luck", and "in". I actually made a post defining probability, chance and luck some pages back. In our differences I believe it's about logic.

What do you think about the game show "Do you want to be a millionaire"? The one where they pick an answer out of 4 choices and ring a friend who googles the answer for them. Is there any luck involved in the game?

You used to do better on this thread before you fixated on this "chess is like a multiple choice test" idea. The notion has a lot of problems, not the least of which is that the "multiple choices" in chess are not binary, they are graduated, and that there can be multiple right answers in a given chess position.

There are many ways to go about this discussion but it is my belief that this is the simplest to any observer.

Let's put aside that problem you pointed out, regardless if I think if it's relevant, we can return to it. Just kindly provide an answer to the question and then we can proceed. The question is not to immediately make a link to chess but for now to also test/verify if our definitions match.

DiogenesDue
DiogenesDue
Jan 30, 2025
0
#4390
ibrust777 wrote:

a) in quantum physics, which you clearly know nothing about, quantum indeterminacy states that without an observable - i.e. some subjective context inwhich events are observed - a systems physical state cannot determine events, i.e. it remains indeterminate. Which is exactly what I described above, and exactly opposite of your point.

b) probability theory rigorously defines the known vs. unknown in the form of a sample and population. The sample consists of the observables, the population consists of the unobserved. i.e. it relies explicitly on the subjective context of the mathematician, that data which is known to him vs. unknown. Without this distinction there can be no mathematical probability. So no, wrong. Try again.

c) Flipping a coin or rolling a dice, similarly, is an act of indeterminacy insofar as human beings cannot predict the outcome, but when it comes purely to the physical forces acting on the dice or coin, the outcome is not indeterminate.

You fail, and you demonstrate that you have zero understanding of the argument, but all along have just been hopping on board the backs of others points. Infact, you've now done a 180 twice in this thread, and I'm still not sure you realize it. But never has it mattered since I've yet to see you make an original argument.

Keep trying

a) Go ahead and tell me your Schrodinger's cat theories about chess positions, I guess wink.png.

b) Go ahead and attempt to connect this to something I have argued in a way that is not nonsense. That is, don't spout a bunch of stuff, then state at the end "so no, you're wrong" when I have not discussed what you are pontificating on.

c) So...you are a hardcore determinist then when it comes to dice? How does that jive with your quantum indeterminacy?

You keep wandering farther and farther off into the weeds. Demonstrate the 180, if you think you can.

Now, I have already told Koshmot that if he wants to argue that a random move generator can beat Carlsen, and therefore there is luck in chess, then I'll agree that there's about a one in a trillion trillion chance of getting "lucky" enough to do that much the same way as infinite monkeys can type all of Shakespeare's work, etc. So if you want to go after that level of a technicality, I'm not really interested that type of meaningless discussion.

Kotshmot
Kotshmot
Jan 30, 2025
0
#4391

Now, I have already told Koshmot that if he wants to argue that a random move generator can beat Carlsen, and therefore there is luck in chess, then I'll agree that there's about a one in a trillion trillion chance of getting "lucky" -Dio

Just to note, I haven't argued it is realistic that a random move generator could beat Carlsen. I have argued that it demonstrates that luck cannot be excluded in chess and that it manifests in various practical situations. For example, a single "lucky" move is enough, a whole game is not needed for the element to exist.

playerafar
playerafar
Show your flair!
Jan 30, 2025
0
#4392
Kotshmot wrote:
DiogenesDue wrote:
Kotshmot wrote:

In our case, @DiogenesDue , I'd like to think we're not in disagreement about the definitions. I think we have are pretty similar view about the terms "chess", "luck", and "in". I actually made a post defining probability, chance and luck some pages back. In our differences I believe it's about logic.

What do you think about the game show "Do you want to be a millionaire"? The one where they pick an answer out of 4 choices and ring a friend who googles the answer for them. Is there any luck involved in the game?

You used to do better on this thread before you fixated on this "chess is like a multiple choice test" idea. The notion has a lot of problems, not the least of which is that the "multiple choices" in chess are not binary, they are graduated, and that there can be multiple right answers in a given chess position.

There are many ways to go about this discussion but it is my belief that this is the simplest to any observer.

Let's put aside that problem you pointed out, regardless if I think if it's relevant, we can return to it. Just kindly provide an answer to the question and then we can proceed. The question is not to immediately make a link to chess but for now to also test/verify if our definitions match.

Our definitions of what chess is don't have to match.
The point is that there are multiple definitions of what it is.
And another point is that nobody is confined to one definition either.
Unless he or she so self-confines.
An absence of luck in one definition doesn't have to cause such absence in another.
Is there luck in chess?
Depends on how you want to define the terms - 
but in the most common usage of the word 'chess' the answer is Yes and luck occurs in multiple ways in that case.

Kotshmot
Kotshmot
Jan 30, 2025
0
#4393

Of course the definitions don't have to match. But if they do match, and you still find yourselves in disagreement, then the problem lies somewhere else and can be identified. This way progress is possible if both parties are willing to contribute.

playerafar
playerafar
Show your flair!
Jan 30, 2025
0
#4394
SymphonicKnight wrote:

First of all you have to define what you mean by "luck" before your question can be engaged on that level, but if you mean that one is not knowledgeable about the outcomes to deliberately choose them, and all outcomes no known, this happens 100% of the time. However, skill creates luckier players than others, and it is going to happen that if one can calculate 7 moves deep, and the result foreseen, this will usually be better than the best move at 6 moves deep, although such values do flip in engine evaluations. Furthermore, positional evaluations do matter, but, again, they are going to be subject to pragmatic re-evaluations at certain levels. Just look at how an engine changes evaluations at different depths in many positions, and see that one's search depth will look like look all the time, unless one actually sees the checkmate.

'on that level'. What level?
'knowledgeable about outcomes' ...
'luck in chess' ... and the possible range of meanings of those three words.
But other words (and phrases) are being brought in ....
'skill'. 'randomness'.
Bringing the word 'skill' into the conversation doesn't somehow zap 'luck'.
Does 'luck' need 'subjective'?
I would say no.
Computers can't care about anything - but two strong computers playing each other - well one of them could still have luck by getting a won result by choosing a move that its opponent failed to play a sufficient reply to.
The computer that won could have chosen that move even though it had a lower evaluation than another move - which was played in a previous game but led to a draw so the computer varied this time. Got lucky.
Because 'chess isn't solved'.

DiogenesDue
DiogenesDue
Jan 30, 2025
0
#4395
ibrust777 wrote:

Trying to be brief since I have to sleep... there are multiple problems with your bad argument here.

1) your argument is not confined to chess, it actually applies to all of reality. i.e. it argues that luck does not exist. For example, your argument claims that luck doesn't exist in poker - because with humans removed the card that's drawn is just the result of physical forces determining the outcome. When someone flips a coin... the side the coin lands on is just a result of physical forces, and is fully determined by them. Thus, your argument reduces to an absurdity, it claims luck doesn't exist in anything. But really all it does is deny the definition of luck any validity, essentially you are defining yourself as correct as Optimissed pointed out earlier, but this is meaningless.

2) From the beginning of this conversation we have acknowledged that the concept of chess with the players removed is possible to imagine... however it's a meaningless scenario, it has no relation what anyone here actually does when they play the game.

3) there have been arguments made regarding how chess can even be interpreted without some subjective context, i.e. the very concept of a "good move" or "winning the game" relies on a player with a specific priority, that priority exists apart from the games rules. For example, can you explain how the inclination to win - or the concept of a good vs. bad move - can exist in this purely physical model of chess that you're imagining?

4) the argument I made actually does not support your conclusion, what it suggests is that consciousness is an inherent part of reality, and that any reality inwhich we remove that is incomplete. Hence the argument I made regarding incompleteness, i.e. you presume a physical system can fully determine events without reference to its observers, but the observer itself is part of the system, and the event itself is an observed outcome.

1) It's beyond reaching to try to claim that separating human players from the game of chess itself equates to determinism. Your poker argument is a straw man, especially since I have used Poker more than once as an example of a game which requires skill, but without perfect information, ergo introducing random outcomes. Similarly, it absurd to try to argue that I am speaking about all of reality when I took great pains to define the logical construct of chess and why it is *not* tied down to a physical reality.

2) It's meaningless? The best chess players on the planet are all non-human and have been since at least 2006 if not earlier. What's meaningless is thinking that playing chess matters for human beings as anything more than a fun diversion at this point, and that's even assuming you thought that a board game had any significant meaning before that.

3) Ermm...what I posted was the opposite of a "purely physical model" of chess.

4) ...not even worth responding to. You do not define that makes any reality complete, and while for others if they were being reasonable I might try to argue the point genuinely, in your case I would say just consult your Tarot deck and define any reality you like; it's not worth my time to disabuse you of anything for your own sake. I will chime in if you start misleading other people, though.

megaium
megaium
Show your flair!
Jan 30, 2025
0
#4396
no
DiogenesDue
DiogenesDue
Jan 31, 2025
0
#4397
Kotshmot wrote:

Now, I have already told Koshmot that if he wants to argue that a random move generator can beat Carlsen, and therefore there is luck in chess, then I'll agree that there's about a one in a trillion trillion chance of getting "lucky" -Dio

Just to note, I haven't argued it is realistic that a random move generator could beat Carlsen. I have argued that it demonstrates that luck cannot be excluded in chess and that it manifests in various practical situations. For example, a single "lucky" move is enough, a whole game is not needed for the element to exist.

Chess as a logical construct contains no luck (beyond first move selection). Chess in physical form between two humans does not contain luck either. The "luck" that manifests in all examples here come from the players or the environment. On a practical level, if you attribute luck to player's decisions (mind), that is actually a matter of skill (and a lapse of skill is not luck, nor is falling prey to emotions, lack of focus, etc.). If you attribute luck to the environment (which includes tournaments rules, clocks, what the players ate, weather, lighting, bodily urges, etc.) then that is external to the game entirely and the phrase "in chess" fails. If you attribute luck to what happened before the game (training, pet openings, etc.), that is not part of the game of chess played either.

Look at it this way (anyone that wants to be reductionist and simplify this all down)...if it's not in the PGN or is a purely optional part of the PGN (which is carefully designed to encapsulate all the information needed to recreate a game of chess), it's not part of the game of chess being represented and you cannot apply external examples to it and claim luck. PGN is not a perfect format, though, so you can only use this as a good rule of thumb for trying to decide if something should be considered part of a game of chess.

That's my non-idealist version of things, which I hesitate to bring up since people will seize on anything imprecise and start to argue about what it actually means.

Kotshmot
Kotshmot
Jan 31, 2025
0
#4398
DiogenesDue wrote:

You used to do better on this thread before you fixated on this "chess is like a multiple choice test" idea. The notion has a lot of problems, not the least of which is that the "multiple choices" in chess are not binary, they are graduated, and that there can be multiple right answers in a given chess position.

Meanwhile we wait for Dio's stance on this question (if he'd like to give one), I can address this issue he points out about the choices in chess being gradual.

In chess the choice is made in one position at a time. A single position contains a binary number of answers, one or more of which are "correct" ones. Multiple available right choices would not make a difference on the luck aspect, so I don't see how it is relevant either.

DiogenesDue
DiogenesDue
Jan 31, 2025
0
#4399
Kotshmot wrote:

Meanwhile we wait for Dio's stance on this question (if he'd like to give one), I can address this issue he points out about the choices in chess being gradual.

In chess the choice is made in one position at a time. A single position contains a binary number of answers, one or more of which are "correct" ones. Multiple available right choices would not make a difference on the luck aspect, so I don't see how it is relevant either.

I won't be going down your game show road, because I've already refuted that argument in the past to my satisfaction and it's a rabbit hole of imprecise analogies. You should correct the bolded statement above as it doesn't mean what you probably meant.

Kotshmot
Kotshmot
Jan 31, 2025
0
#4400
DiogenesDue wrote:
Kotshmot wrote:

Meanwhile we wait for Dio's stance on this question (if he'd like to give one), I can address this issue he points out about the choices in chess being gradual.

In chess the choice is made in one position at a time. A single position contains a binary number of answers, one or more of which are "correct" ones. Multiple available right choices would not make a difference on the luck aspect, so I don't see how it is relevant either.

I won't be going down your game show road, because I've already refuted that argument in the past to my satisfaction and it's a rabbit hole of imprecise analogies. You should correct the bolded statement above as it doesn't mean what you probably meant.

Then there is no progress to be made this way, but note that it comes down to your choice. In my view this is a deflection of a simple question, that you deploy because you're not a 100% sure in your position in the potentially ensuing conversation. You can naturally deny that. I will still address your other points.