Hello,
I'm the author of Cracking Pi.
I've had to re-register since my password was not accepted the second time I tried to log in and an email to rewrite my password was not sent.
I spent several days in Dallas with the author of Zumma's Roulette System Tester. I tested all 15 k trials with the methodology of "action at a distance." This is the methodology of Quantum Mechanics. Einstein called it "spooky action at a distance." You can repeat my experiment and I'm quite certain you will get the same result. That is, a flat bet advantage: .08333 .
There are two matters of note. The first is minor. The original copy of System Tester contained several errors in which strings of numbers were erroneously repeated. When I pointed this out to the author, he patched them. I don't know if he went back to his original data, but my feeling at the time was that he used an RNG. Given that these were less than a hundred numbers total, I don;t think the difference would be particularly statistically significant.
The second matter is very much of significance. The author reported he recorded the data mainly from downtown casinos in Las Vegas. The casinos he named were not particularly mainstream. This may be because he required his data recorders to place bets and the lesser casinos had cheaper minimums. The significance of the less than mainstream casinos is that their dealers were not necessarily required or trained or supervised or encouraged to use a random release. When I questioned him on this point, I was told the dealer's were "throwing from the green all night long." The term refers to a dealer releasing the ball over one of the green house numbers (or a right angle thereto). This has obvious consequences in terms of randomness.
In my own Pi-odds study, I carefully selected dealers who were releasing randomly. From that study, using "action at a distance," I got a flat bet advantage close to what is now expected from a random release: .16666 .
Clearly, the factor of a dealer's release protocol invites deeper inspection.
I'm the author of Cracking Pi.
I've had to re-register since my password was not accepted the second time I tried to log in and an email to rewrite my password was not sent.
I spent several days in Dallas with the author of Zumma's Roulette System Tester. I tested all 15 k trials with the methodology of "action at a distance." This is the methodology of Quantum Mechanics. Einstein called it "spooky action at a distance." You can repeat my experiment and I'm quite certain you will get the same result. That is, a flat bet advantage: .08333 .
There are two matters of note. The first is minor. The original copy of System Tester contained several errors in which strings of numbers were erroneously repeated. When I pointed this out to the author, he patched them. I don't know if he went back to his original data, but my feeling at the time was that he used an RNG. Given that these were less than a hundred numbers total, I don;t think the difference would be particularly statistically significant.
The second matter is very much of significance. The author reported he recorded the data mainly from downtown casinos in Las Vegas. The casinos he named were not particularly mainstream. This may be because he required his data recorders to place bets and the lesser casinos had cheaper minimums. The significance of the less than mainstream casinos is that their dealers were not necessarily required or trained or supervised or encouraged to use a random release. When I questioned him on this point, I was told the dealer's were "throwing from the green all night long." The term refers to a dealer releasing the ball over one of the green house numbers (or a right angle thereto). This has obvious consequences in terms of randomness.
In my own Pi-odds study, I carefully selected dealers who were releasing randomly. From that study, using "action at a distance," I got a flat bet advantage close to what is now expected from a random release: .16666 .
Clearly, the factor of a dealer's release protocol invites deeper inspection.