Re: The Feeding the Fish Theory
Actually in statistics you can never TECHNICAL "prove" anything but with large enough sample sizes you can make VERY, VERY, VERY accurate inferences.... For example whether PartyPoker was likely rigging hands and likely NOT rigging hands.....
Also I should call it the "Feed the Fish" hypothesis.
For in science, the highest levels of scientific knowledge are LAWS and THEORIES. To become a theory in science requires some MAJOR MAJOR effort and experimentation and validation.... If a scientist can one day publish a THEORY, then he/she is almost automatically getting a nobel prize...
Anyway......if PartyPoker was rigging hands , Statistics COULD be used to bust them. With the "feed the Fish" hypothesis where only certain strategic hands are rigged....a much larger sample size would be needed.
So by "tougher" I mean a VERY large Sample size would be needed.... I would argue 250K hands would be too small...
***If math bores you then STOP READING****
To estimate the mimimum sample size for Texas Hold'em I like to use the AA versus KK match up.
*The odds of being dealt AA are 220:1
*The odds of an AA/KK matchup is roughly 20:1
*A decent size sample size for measureing the AA/KK matchup is about 100....but of course a larger sample size is better.
So using round numbers 220 x 20 x 100 = 440,000 hands.
Mmmmmmm....I don't know about you, but I don't have that many hands in my Poker Tracker... And I only have about 25,000 hands of Empire in my data base (I only bought PT a couple of months ago)...
|