Quote:
Originally Posted by dbcooper279
How about confidence testing? Take millions of small sample sets (which we are doing) and compare them to the larger sample.
Using the multimillion hands sample to identify whether the poker site is rigged is a logical fallacy. I will try to prove why it is a logical fallacy.
Your formal logic statement is the following: the hands that distributed by PS are perfectly fit into the multibillion hands sample, the randomness figures of the sample is not modified by PS distributed hands, therefore PS poker site is not rigged.
Using an analogy, let say there is a group that includes 1,000 humans that defends rigged poker sites on the 2+2 forum, and the task is to identify whether the group includes only naive, voluntary site defender crusaders and verify that there are no paid shills in the group (just like there is a task to identify whether the hands are the product of a legitimate, random business process and not a manipulated software system).
If your only methods are
a) examining the main characteristics of the member of the group such as number of penis/vagina, brain, livers, lung etc, of the human and
b) examining their writings in order to identify whether the human fit into the not shill group, then your method will not indicate how many paid shill in the group as you guys having the same main characteristics, all of you except the unluckiest have one penis, liver, lung, etc. and you are writing the same arguments here regardless whether you are a naïve defender or a paid shill (just like the not randomly distributed hand has the same characteristics from the viewpoint of statistic as the randomly distributed hand).
The point is, the paid shill perfectly fits into the group of the naive, voluntary defenders (just like the not randomly distributed hand perfectly fits into the randomly distributed hand hands sample), and therefore your examination method would be inadequate to identify illegitimate elements of the group. The only sufficient method is examining the characteristics, their writing AND the surrounding system of the humans such as where they work, from where they write, how they get paid in order to identify not legitimate members (shills).
Similarly, in the context of online poker you need a more adequate method to identify rigged systems and avoid such logical fallacy and you need to examine the whole system and not only the multibillion hands sample in order to conclude whether the system is rigged. That sufficient method would be to examine the site as a complex system by recognizing the card distribution as similarly important module of the system as the multibillion hand sample and examining the card distribution. Yes, examining the multibillion hand sample is crucial and definitely a must, but without examining the card distribution and software system components it is not possible to make conclusion about the integrity of the system.
That’s why it is so important having audited, regulated, component level verified software in place, which important factor you guys fail to realize by being hypnotized by the multibillion hand sample.
PS: it has been established during the previous discussions that the distributing winning hands to designated account wold not modify the result of EV statistic, so please do not start with that usual crap that any modifications would show up in the sample. It has been not suggested that PS would distribute more AA by modifying the random data. What was suggested is that PS distrbute the randomly genrated data to designated accounts.