This post is directed at the regular posters here who sometimes refer to my web site. First, I do appreciate that and don't stop doing it, the stuff I publish is intended to be seen and used. There will be more work published there soon that will show some very interesting things. To my knowledge nobody has ever published some of the things I'm doing. But I just want to point out what I claim my data shows, and what I have not claimed. That way people who haven't followed the backstory won't have the wrong impression.
- I haven't proven, or ever claimed to have proven, that any online poker site is not rigged. That's an impossible task that will never be accomplished, no matter what analyses are done, no matter what level of regulation and inspection takes place, no matter what country regulates it.
- I have shown to a high level of statistical confidence that flops are influenced by some small card removal effects that are predictable and are caused by preflop player decisions. Those decisions determine which deck stubs are used to deal a flop, and which are not. The biggest of these effects is the rank bias caused by players preferring to play with high cards. I did not discover this effect but I helped to quantify it and confirm it. The second biggest effect is a bias towards paired flops caused by players preferring to play paired hole cards more often than they fold them. This effect on the deck stub is small enough that knowing about it has no exploitable benefit, but is predictable and very significant statistically.
- I have shown to a high level of statistical confidence that flops are dealt randomly from the remaining deck stub after the removal effects that occur preflop. Flops follow the expected distribution for the deck stubs that players select for seeing flops. On the sites I looked at, it is extremely unlikely that "action flops" occur, and if manipulated at all, the frequency is so low as to be pointless.
- I have not yet published similar results for turns and rivers and so I have no conclusions or claims about those streets yet. I do think we can infer from the flop results what later streets are likely to show.
- I have not yet published an all-in analysis, but I've done a lot of work on it with enormous hand samples. When published, it will show another discovery that has never before been quantified, but people have speculated that it exists. Otherwise it will show the expected distribution and no evidence of manipulation. There's no reason to take my word for that at this point, I just mention it because it will be very interesting and possibly controversial. So stay tuned.
So I have a good start on providing information that can give players confidence that the deal is fair, but there is more work to be done. And my primary objective is still to just quantify and discover effects that are not known, not to prove a negative proposition. But once you show that community cards come out as expected, and that all-in confrontations come out as expected, the main thing left is each individual player's distribution of dealt hole cards. And the player himself is the only one who can check those.
spade