> “I [experimented] until I had a rough outline of a system based on the previous winning numbers,” he told the Sydney Morning Herald in 1969. “If numbers 1, 2, and 3 won the last 3 rounds, [I could determine] what was most likely to win the next 3.”
I'm calling shenanigans here. I can believe that a roulette wheel might have biases, and that you might be able to leverage those to beat the house. I have a very hard time believing that a roulette wheel could have a contingent bias such that the next outcome depended on the previous one. What possible mechanism could cause that? I'm also skeptical that even if there were such a mechanism that you could reliably determine it with the number of data points that one person could plausibly collect and analyze, especially in the 1960s.
In fact, let's do the math: 37*37=1369 so you'd need that many spins at a minimum just to see every possible number pair come up once. Let's suppose that a spin takes a minute. It would take a minimum of 22 hours of watching one roulette wheel to see every combination. More likely you'd need 10-100 times that much data to get a statistically significant result. And that is just for one wheel [UPDATE: and a one-round contingent bias. The claim is a three-round contingent bias. Collecting that much data for even a single wheel would take years of constant effort.]
I'm calling shenanigans here. I can believe that a roulette wheel might have biases, and that you might be able to leverage those to beat the house. I have a very hard time believing that a roulette wheel could have a contingent bias such that the next outcome depended on the previous one. What possible mechanism could cause that? I'm also skeptical that even if there were such a mechanism that you could reliably determine it with the number of data points that one person could plausibly collect and analyze, especially in the 1960s.
In fact, let's do the math: 37*37=1369 so you'd need that many spins at a minimum just to see every possible number pair come up once. Let's suppose that a spin takes a minute. It would take a minimum of 22 hours of watching one roulette wheel to see every combination. More likely you'd need 10-100 times that much data to get a statistically significant result. And that is just for one wheel [UPDATE: and a one-round contingent bias. The claim is a three-round contingent bias. Collecting that much data for even a single wheel would take years of constant effort.]
Nope. Not possible.