Statistical Arguments for the Poker=Skill Debate

[Grasshopp3r]

There are some new software packages that are being used to calculate the "luck vs. skill" results from PT databases. Would assembling large, aggregated databases prove the skill argument?

Here is the software: http://www.pokerevsoftware.com/

[Skallagrim]

Yes. Showing most forms of modern poker end with all but one folding shows that decisions, not chance, determine the majority of hands. Showing that the best starting cards dont win that often also shows the importance of later decsions. Showing that the 2 cards that would have won are often not even playing by the showdown again shows how decisions rather than cards account for most poker results. And finally, showing, if possible, that consistent winners dont consistently get the 'best' cards agains demonstrates the importance of play over cards.

Skallagrim

[schwza]

[ QUOTE ]
Yes. Showing most forms of modern poker end with all but one folding shows that decisions, not chance, determine the majority of hands. Showing that the best starting cards dont win that often also shows the importance of later decsions. Showing that the 2 cards that would have won are often not even playing by the showdown again shows how decisions rather than cards account for most poker results. And finally, showing, if possible, that consistent winners dont consistently get the 'best' cards agains demonstrates the importance of play over cards.

Skallagrim

[/ QUOTE ]

some of these arguments are easily countered by arguing (correctly) that a lot of the luck in poker is not in the starting hands but in what cards come later.

i think that the best argument is that good players win consistently over time. take a big sample, do some statistical analysis, and say "at a 99.9% confidence interval this player is winning at least x amount per hand." the best form of the argument would be to say "judge, we're tracking these 50 players for 2 months" to take care of the counter argument "enough monkeys at enough computers and some of them will look they are consistent winners."

also, the same five guys are at the final table of the world series of poker every year.

[Grasshopp3r]

How would we go about achieving this? Freakanomics ( http://www.freakonomics.com/blog/ ) started a project called pokernomics ( http://www.pokernomics.com/index.htm ), however, they expressed concern about the statistical relevance of the sample as the players who submitted hand histories were winners. Steven Levitt is doing the study. I sent him an email asking for his comments. Steve is also a poker player.

[jschell]

From pokernomics.com:
[ QUOTE ]
For a limited time, we offered a free analysis of people's hand histories.

[/ QUOTE ]
It would be interesting to hear what they told people. Anyone know of someone who did this?

[dragonystic]

i gave them 80,000 hands.
they promised me a free book, and an analysis on my hands.

that was half a year ago
and i still have nothing in return

::

oh, and lastly, i dunno why everyone is trying to come up with a surefire proof that the game is skill. the govt doesnt give a damn, one way or the other, they just want a cut.

[Grasshopp3r]

Coming up with a surefire proof that poker is a game of skill would then place it outside of most definitions of gambling. It would then be legal. That is a big thing.

Also, that is how California legalized its card rooms, through the courts.