BGonline.org Forums

OT: Am I the only one here who thinks poker is almost all pure luck?

Posted By: Timothy Chow
Date: Tuesday, 23 September 2014, at 2:49 p.m.

In Response To: OT: Am I the only one here who thinks poker is almost all pure luck? (Bob Koca)

Yes, there are limits to how rigorously one can analyze luck versus skill because both words, in ordinary parlance, have various connotations that are somewhat contradictory.

I personally take the following view:

1. "Luck" is systemic randomness, i.e., randomness in the structure of the game that cannot, even in principle, be controlled by the players.

2. "Skill" refers to how much the choices of the player can affect the outcome of the game, in principle.

Some corollaries of this view are:

a. Skill is a function of the game, not of the population of players. In common parlance, one speaks of "games of skill"; this has the connotation that skill is a function of the game and not the population of players. However, it is also common, as you point out, to regard games in which it is easy to find the best choices, as "having no skill." This implies that skill is a function of the population of players and not just the rules of the game itself. If you want to be mathematically rigorous, you can't have your cake and eat it too, and must make a choice here.

b. Randomness that is associated with the player choices is not referred to as "luck." If I'm a chess grandmaster and my opponent happens to walk into my opening preparation, or had an accident on the way to the tournament and is now badly shaken and unable to concentrate, I don't call that "luck." Also consider the following imaginary game, called "Guess the digits of pi!" A computer picks a random number n between 1 trillion to 2 trillion and you and I have to guess the nth digit of pi. Whoever guesses correctly wins. In my book, you can refer to the computer's random-number generator as a source of "luck," but there is still skill involved because in principle a player could memorize some of the relevant digits and increase the probability of winning. This goes against most people's intuition, who would refer to the game as a game of pure luck, with no skill involved.

Fundamentally, the trouble is that the notion of "control" is vague and philosophical and possibly self-contradictory. "Skill" is supposedly what you can control while "luck" cannot be controlled. But philosophers have debated free will versus determinism for millennia without clear conclusions, and so we cannot expect to draw a sharp mathematical boundary between what a player can and cannot control.

Fortunately, for most real-world games, the blurriness of the distinction between variations in play that are controllable in principle and controllable in practice is not so crucial, because games that are totally controllable in one or the other way tend not to capture people's interest and so get weeded out of the set of games that people actually play. For example, tic-tac-toe is still played, but usually only in populations of children in which drawing strategies are not universally known.

Post Response

Subject:
Message: