Friday, September 14, 2012

20 thoughts on skill vs. chance in poker, part 9: Common intuitive proxies for skill and chance in games

<-- Part 8: How are we measuring skill, anyway?


In the last thought, we've seen that there's not a good basis for developing formal, quantitative measures of comparison of skill and chance between games. Slipping away from the abstract and back into the real world, however, there are plenty of ways to draw practical comparisons between poker and games considered not to be games of chance.

After all, the most important characteristic of poker is that it's a symmetric, closed multiplayer game, which makes it completely distinct from every other activity widely perceived as gambling. Tournament poker (or a fixed-duration cash game) is like every other competitive game or sport; players exercise some skill, some chance elements out of their control can affect the outcome, but the superior player will win more often. Aside from a historical association with casinos and gambling, what differences between poker and other games are shaping its social and legal treatment? Does poker have aspects which are similar to those of other games?

The study by Thomas J. Miles and Freakonomics author Steven D. Levitt, The Role of Skill Versus Luck in Poker: Evidence from the World Series of Poker, hits upon this in a manner that seems to be very accessible for the average person:
... comparison of high skilled and low skilled players entered in each tournament in the WSOP finds that the high skilled player wins 54.9 percent of the match ups. For purposes of comparison, we calculated the regular season win rates for professional sports teams that made the playoffs in the previous season – making the playoffs last year is akin to being a highly skilled player entering the WSOP. Since the year 2007, teams that made the playoffs the previous season win 55.7 percent of their games in Major League Baseball against teams that failed to make the playoffs in the previous year. Thus, in some crude sense, the predictability of outcomes for pairs of players in a poker tournament is similar to that between teams in Major League Baseball. To the extent that baseball would unquestionably be judged a game of skill, the same conclusion might reasonably be applied to poker in light of the data.
In fact, baseball looks to be a great point of comparison in general. David Sklansky noted "Professional baseball is almost all skill even though no team is a 3-1 favorite over another." Baseball is a game where, in terms of batting, "succeeding" anywhere near half of the time over one's career would be a legendary, impossible performance... so why do people get so hung up on the fact that an expert poker player can't win any given hand with 100% probability? There's also a delicious irony in the perception of skill and chance in physical sports that 2+2 poster "Do it Right" noted that I absolutely love and is worthy of mention:
You know what the most ironic part of all of this is? The sample size you get playing poker actually works against us. Imagine poker was played like boxing matches, and we played maybe 5 short sessions per year. There would be players who would be undefeated for years - perhaps even decades. And poker would be widely appreciated as exclusively a game of skill as shown by Joe "Ace" Tyson's 17 year undefeated streak as Poker World Champion.
So what makes the uncertainty in games like baseball and other athletics seem different than that of poker? It is likely primarily due to the fact that poker's randomness is overt and intrinsic, whereas the random elements in sports are non-overt and external. That's worth a lot when it comes to public perception, and it's potentially worth noting in an intellectual discussion, too. In an idealized world, where the physics of weather loses some chaos and yields to the pure rules of a physical sport, one might be able to say that there's no chance in that sport other than strategic uncertainty and the imperfections of human execution. The rules of sports don't demand the external random element, and perhaps would prefer being played in a world without wind. Poker, on the other hand, depends on the randomness its shuffle introduces in a way that could not so easily be fully removed from the game in an idealized world (though I will explore this more later).

The conclusion we can draw from this is that, psychologically, most people find salient randomness only in elements of overt/intrinsic randomness, despite the fact that we've seen that non-overt randomness can contribute just as much or more towards uncertainty in game outcomes (rock-paper-scissors).

The amount of salient randomness in a game is often used as an intuitive proxy for the degree of chance in a game.

This bodes poorly for poker, as it doesn't get much more salient than a game historically framed as a type of gambling where river suckouts blatantly change winners reasonably often. There's not much we can necessarily do about this... though focusing our marketing and television coverage of the game a bit further away from showing only "exciting" preflop races might help.

Coincidentally, intrinsic random elements in games such as poker are a much tamer type of randomness than that of games which are affected by chaotic external forces. Players can always have perfect knowledge of the exact probabilities of every possible intrinsic random component of poker, whereas measuring the effect of weather on sports can only be estimated. Similarly, if we were to stretch our definition of "game" a bit to be able to compare poker to investing, then poker would once again be of a tamer type of randomness than that of market forces. Poker's variances are all finite, bounded, and precisely calculable, whereas much of the world of investing involves potentially-unbounded or fat-tailed randomness. Readers of Taleb's The Black Swan will appreciate this.

Changing our focus to a different class of games, one much closer to poker, we can consider popular mind sports. Since their fundamental structure is so similar, why do games such as Scrabble, bridge, Magic: the Gathering, and other competitive card or video games rarely get considered to be games of primarily chance or gambling? It's not at all clear that any of these games necessarily involves less chance than poker.

The key difference between poker and these other mental games is that poker seems to be a simpler game on the surface. Not only are the rules to poker quite easy to learn, but most Americans have grown up with enough of a loose exposure to the game (or to the often-conflated video poker) to know the hand rankings, to know that bluffing is a thing, but to otherwise think that the game is mostly about "betting on" who will get better cards. If poker takes "a minute to learn, a lifetime to master", the latter part is lost on the common person taking a cursory examination of the game. Meanwhile, Scrabble clearly demands a vast vocabulary, bridge's strategic complexity seems well-ingrained in our culture, and Magic: the Gathering has a huge rulebook and is often downright intimidating in its complexity for new players.

In fact, when the first competitive Magic: the Gathering tour with cash prizes was being launched, the game's lawyers prepared a variety of arguments to present to a court in order to obtain a judgment that Magic is a game of predominantly skill, almost identically to the situation we find poker in. The difference was that, when the lawyers showed up in court to present their arguments, the judge told them not to bother — he had tried to read through and understand the entire rulebook last night, unsuccessfully, and was therefore completely convinced that the game must be primarily skill! Depending on the variants, Magic generally has similar amounts of both skill and chance as poker does. The fact that one game is much easier to learn than another doesn't actually correlate to the depth of skill or influence of chance in a game.

From this, we learn that salient complexity plays a key role in shaping people's impressions of how difficult a game is.

The amount of salient complexity in a game is often used as an intuitive proxy for the degree of skill in a game.

Poker may not be able to completely eliminate river suckouts to bring down its salient randomness, but the other side seems like a front we could improve our arguments upon. Make poker's hidden complexity into salient complexity. Make efforts to highlight the strategic complexity of poker in appropriate settings such as courtrooms. Showing that skilled players win more money than unskilled players with K9o is fine, but it doesn't appeal particularly to the common person's sense of salient complexity. What about showing the court an in-depth hand breakdown, perhaps from a training video or recorded strategy interview with an expert player? I would imagine that, after the first 2 minutes of a poker pro analyzing a single decision within a hand, employing dynamic multi-level strategic interplay and spouting complicated-sounding (but legitimately relevant) numbers to do so, the average person would abandon any misconceptions about poker's simplicity.


Part 10: Games of skill vs. "involving skills" -->

(back to index)

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.

Related Posts Plugin for WordPress, Blogger...