Dr. Schneider: That's a little out of my field—why a casino would cheat. I still don't buy that they do. It would be dangerous—especially in states where the casino industry is regulated. Think of the impact on the public if it were ever proven that the video games are rigged. It would have a devastating effect on legalized gambling.
Frank: Probably. Yet, people still gamble in illegal gambling parlors where the house is composed of known crooks. But I do think you are right. Public sentiment would be a factor in keeping the casinos honest.

Dr. Schneider: That's the motivation for not cheating or programming such things as we've been discussing.
Frank: I've always wondered this—how can one program randomness? Isn't that a contradiction in terms?

Dr. Schneider: Yes, seemingly. Randomness is an interesting study because there are subtle, usually short-lived, patterns in everything. Randomness and probability work hand in hand. You therefore program randomness by making sure that the probabilities over a sufficiently long period of time work themselves out according to theory or as close to theory as to give you a 99 percent confidence rating in the integrity of the program. By the way sometimes they don't. Then, you have a problem. If you give yourself a wide enough range in your programming, millions and millions, or billions of trials, more than likely you will hit an approximation of randomness. That is good enough to assure an honest game.
Frank: Could a programmer program patterns into randomness and not be aware of it? And then it shows up?

Dr. Schneider: Anything is possible. It's unlikely that using the standard randomness programs that such a thing would happen but a chip could be faulty.
Frank: So you think what AP and I observed and what others have reported might be an error and not deliberate tinkering?

Dr. Schneider: It might be an error in either one of two
places. First, it could very well be a programming error. I still would find it hard to believe that a programmer would cheat in such a clumsy way or that the casino industry would risk such a thing. Therefore if that pattern which you observed is real, then I have to think it's a shadow effect, a faulty chip, or something of that nature. Not deliberate cheating. So that's the first place an error might be. However, the second place the error might be is with the observers. You might have noticed a short-term pattern that was ephemeral. Think of how many hands of video poker you played before you started to notice this particular pattern. You might have played hundreds or even thousands of such hands—four-to-a-straight—and never had the situation appear—or, it appeared within a few appearances of its expected frequency—six percent of the time or thereabouts.

Also think of how often that machine was played by others when you weren't there. Did they experience the same thing? So it could be that your observations are at fault, not the programming. The casino or the manufacturer is not cheating you. You are getting the game as it appears to be. The casino's edge is whatever is posted and that's all there is to it. This pattern isn't a long-range pattern, it's just a blip that comes and goes. You happened to catch the blip and recognize a pattern. That's how randomness works. At any point, there seems to be something meaningful, organized, and deliberate happening—except it really isn't and it soon disappears to be replaced by other seeming patterns. The whole is patternless—except that it fits our expectations based on probability theory.

Frank: So the only pattern—if the machine is
programmed properly—is based on long-range probability?