Relationship Between Specific Chess Skills and ELO

what you're calling luck is a lack of skill      not the random chance involved in poker  one must not count on it yet know to do when your opponent blunders   it certainly feels lucky though

Non-exhaustive list of luck in chess:

> Who you are drawn to play, and with which colour

> Opponent blunders for no reason

> A lot of moves are not humanly possible to evaluate to a clear conclusion. Which way the coin falls on an 'unclear' assessment is basically a matter of luck. The skill is in balancing out the various factors in play and making a judgement call - the better the player, the more likely the judgement to be correct, but even very strong players can be unlucky.

> Arbiter's decisions on controversial matters like the 2-minute rule

> External factors beyond your control, e.g. poor playing conditions, noise etc

> What happens in a mad time scramble when both players are moving very quickly to meet the time control is largely luck

Aside from the element of chance due to human fallibility, there is also the matter of whether a player knows the other's choice of opening (at least in some circumstances). Sometimes one is lucky; sometimes one isn't. Someone who is more skillful (knowledgeable) will do better in the long run, of course, but winning or losing a game can come down to the repertoire of the individuals who happen to be paired. I don't see how this is any less inherently chanceful than what cards one is dealt in a game of poker.

@madhacker and @fburton, what Rick is referring to by the luck factor existent in poker and how it differs from chess is this: In chess - it is up to you to make a good or bad move. If you consistently make the best move from a winning position, you are GUARANTEED to win.

In poker, even if you make the best possible play/move, there is still a chance that the random sequence of cards yet to come will cause your opponent to win. For example, if you AA and KNOW your opponent has 72o before the flop, the correct move would be to get all your money in pre-flop while vastly ahead (something like 80-20 or greater probability of winning). But...he STILL has a chance to suck out with trip 7s, two pair, or a full house for example, depending on the what cards come on the board.

To give an idea of what this would mean in chess -- imagine your opponent has blundered and left his back rank unprotected and you can play Re8#. You know the best move and play Re8. Now, instead of winning outright, we have to roll the dice, where on 2-10 you achieve the desired mate, and on any 11 or 12, the result is reversed and you get mated instead (by your own move).

In the Wikipedia article on the Elo rating system

http://en.wikipedia.org/wiki/Elo_rating_system

it is stated that: "A player's expected score is his probability of winning plus half his probability of drawing."

Why are they talking about probabilities/chances if there is no luck in chess?

But that argument is based on the assumption that it is possible for a human to have the skills to do things perfectly. In essensce, to have superpowers. Where do you draw the line - by that logic, there is no luck in poker either, rather the player was lacking the neccessary x-ray vision skills to see through the opponents cards.

Both x-ray vision and perfect positional evaluation are humanly impossible, hence the luck factor.

@mxiangqi - point taken! The luck in poker comes from a different source - the shuffling of the cards - and is therefore different in nature to the luck in chess. Nevertheless, there are chance elements in chess from both intrinsic (e.g. human imperfection) and extrinsic ("luck of the draw") sources.

What is your authority for making this statement or did you just make it up?  

http://www.chesspublishing.com/content/pokerandchess.htm

http://main.uschess.org/content/view/10007/571/

lot of chatter about skills but not many people answering his question,.

chess has a lot of skill sets generally broken down into openings, middlegames and endgames. Tactics and calculation link all these together. 

endgames are very concrete so probably the easist to categorize as far as require knowledge. Silmans recent endgame book is a great one on this since its organized by level http://www.amazon.com/Silmans-Complete-Endgame-Course-Beginner/dp/1890085103

Openings - pick something simple and straight foward. My advice 

Play the scotch http://www.amazon.com/Starting-Out-Scotch-Everyman-Chess/dp/185744387X/ref=sr_1_2?s=books&ie=UTF8&qid=1334325143&sr=1-2 It is simple to learn and will help guide your play so you can focus on typical plans and ideas.

As black I would play e5 against e4 since it wil help build your pattern recognition for your white opening. http://www.amazon.com/Beating-Open-Games-Mihail-Marin/dp/9197600431/ref=sr_1_3?s=books&ie=UTF8&qid=1334325260&sr=1-3

Against 1 d4 play the tarrasch defense http://www.amazon.com/Grandmaster-Repertoire-10-Tarrasch-Defence/dp/1906552916/ref=sr_1_2?s=books&ie=UTF8&qid=1334325569&sr=1-2

These systems will lead to open games with lots of tactics and traps and mistakes. You will be forced to learn and build your knowledge base

as a note against the sicilian and other openings http://www.amazon.com/Starting-Out-Reliable-Repertoire-Improving/dp/1857444167

this will fill in some gaps for other openings but not be overly complex. 

study tactics a lot here or online . 

keep basic principles in mind

http://www.amazon.com/The-Amateurs-Mind-Turning-Misconceptions/dp/1890085022/ref=pd_cp_b_1

Is probably one of the best in this regard

I also think that

Logical chess move by move - Chernev is one of the best even to this day

http://www.amazon.com/Logical-Chess-Explained-Algebraic-Edition/dp/0713484640/ref=sr_1_1?s=books&ie=UTF8&qid=1334325866&sr=1-1

Silman's recent book is good as well http://www.amazon.com/How-Reassess-Your-Chess-Fourth/dp/1890085138/ref=sr_1_1?s=books&ie=UTF8&qid=1334325919&sr=1-1

I like to think of chess as similar to martial arts. knowing the basics is crucial. Looking at UFC people think "Oh that looks easy I can do that" until they walk into a studio and learn things arent that easy. Mastering basics is key. Start simple and build up from there.

read a lot of game books: alekhine's book of games is amazing

lessons from a grandmaster - gulko 

DO NOT get caught up in memorizing openings

Soltis pawn structure chess is a great book as well.

Some great points have been made in this thread, but I do think there's one skill that correlates very well to chess ability. 

That is the skill of figuring out what your opponent is trying to do and reacting to that. In other words, playing both sides of the board. 

I forget where I read it, but someone did a study with chess players of various skill levels "thinking out loud" about a game (not a position), and it was discovered that amateurs spend about 5% of their time thinking about WHY their opponent made a particular move, and grandmasters spend about 45% of their time. And the more time spent thinking about why one's opponent did something, the higher the person's rating. It was a near perfect correllation. 

Now, obviously lots of other skills come into play, such as tactical ability, opening knowledge, endgame knowledge, and so on, but within those skills a deficiency in one area can be covered up by other skills. What can't be compensated for, however, seems to be awareness of the potential in the opposition's position. 

Gee, thanks for calling recreational poker players like me an idiot.  Much appreciated. ~SMFH~

Nobody is assuming superpowers in chess. The point about making perfect moves was a simplifying assumption to illustrate the nature of luck in poker as opposed to chess.

The point is -- chess is a complete information game. It is possible to make the best possible move every time, assuming you are skilled enough, because the entire position is available to you.

Poker is an incomplete information game. It is not possible to make the best possible move (in a chess sense) every time, only a move that has the best chance of being the best possible move. You only know your hole cards, some tells from opponents, and possibly some of the exposed cards on the board. What cards will come out? What does your opponent REALLY have? You can't know it for absolute sure in poker. Which is why poker players talk about moves that have the highest expectation instead of moves that win or lose.

What you are referring to as "luck" is just skill differential in chess. Yes, external factors such as who your opponent will be, how strong he is relative to you, what openings he plays are variable. But -- once those external parameters are chosen, how you perform is determined by your knowledge and skill only without randomizing factors.

In poker, even if you are the best possible player on the planet facing the worst possible player, although over the longterm you will likely win his money, due to variance/luck, he still has a chance to beat you EVEN if you make the theoretically best moves.

By way of comparison, assume we randomly pick two chess opponents..one is a GM and one is beginner. Once we have randomly determined these external parameters, who would you pick to win a single game?

The GM will win pretty much every single time. Why? Because his skill and knowledge are vastly superior.

Now, let's perform the same experiment with poker. We randomly pick two opponents for a short poker match of x hands. We pick a fish and Doyle Brunson (insert any other pro here if you like). Who would you pick to get all the money?

Unfortunately, we cannot guarantee Doyle will win virtually every time. If the match is long enough and he can pick his spots, he can attempt to overcome luck and will likely get the money. But no matter the length, there will just be times when Doyle consistently gets his money in as the favorite, but the fish lucks out even if he made the worst possible moves consistently! It is exactly this luck factor that makes poker attractive to and encourages the average fish to go to Vegas and lose his shirt thinking he is better than he is (when in fact he was on the positive side of some poker variance or a "heater" in poker parlance).

And the matter of Elo levels (the original question)...? At what levels should someone embark on doing the exercises and reading the books you listed?

mxiangqi, I still think you are making an artificial distinction between what is possible and what is not. When you say about chess "It is possible to make the best possible move every time, assuming you are skilled enough, because the entire position is available to you.", this isn't true within the confines of the human condition.

This is why I used the analogy of x-ray vision, to try to demonstrate that point. If we accept that you have to go outside the confines of the human condition to make your statement hold true, then we can also say that the entire position is available to a poker player, as we're not constraining ourselves by his lack of ability to see through cards, just as we are not constraining ourselves by the chess player's inability to see through an effectively infinite labyrinth of variations.

The line between skill and luck is quite blurred in real life.

Chess is a complete information game. So what you say is true in theory - but not in practice!

Your GM vs beginner argument indicates the importance of skill in chess. However, exactly the same can be said of poker only to a different degree. The skilled poker play will still win more than the beginner in the long run in spite of the random element.

I am happy to accept that chance is a bigger part of poker compared to chess. What I cannot accept, because it is patently untrue, is that "there is no luck in chess" and that the chance element is absent from that game. I refer back to my post about how Elo is defined in terms of chance.

The use of probability/statistics for ratings is simply to provide a mathematical model which can account for overperformance (improving players) and underperformance (aging players, players who are rusty, etc). It is a convenient way of saying e.g. two equally strong players should generally draw or trade wins, and of quantifying in some way the skill differential.

Also, as you pointed out, there are external factors which may influence the game -- tiredness, weak spots in preparation vs . particular variations that happen to show up in the game, mood, etc. These could loosely be called "luck", but within the framework the game itself, there are no randomizing factors built into chess as in poker.

Yeah, pulled it right out my hat. Keep posting your links, maybe you'll even convince yourself.

fburton, I think I'll leave this debate to you - we're saying the same things at the same times, except I'm using stupid analogies and you're not

@fburton and madhacker --

I agree -- there is luck in both chess and poker. However, the nature and extent of luck is quite different in the two games.

In the game of chess, luck/randomness is not built into the game rules. A move that should win will not through a roll of the dice lose.

In the game of poker, luck/randomness IS built into the game. The "correct" move may lose, due to the random shuffle of the deck.

Therefore, we can say that in chess, luck is mostly an external factor - along the lines of the examples already discussed in the thread. In poker, it is mostly an internal factor.

That is -- in poker the "luck" or randomness comes from the game, in chess from the humans' moods, energy, focus, knowledge, pairing, etc.

The point about degress of involvement of luck is also a good one. To illustrate the difference -- a poker pro once expressed the role of skill in poker as:

"On any one given hand, it might be 99% luck and 1% skill. Over the course of a tournament, it might be 30% skill and 70% luck. Over the course of a month, maybe it’s 30% luck and 70% skill, and over the course of a year maybe it’s 90% skill and 10% luck."

In chess, I would say it is about 99% skill and 1% luck (the same pro actually said chess is 100% skill, but he may not be in a position to say).