# Chapter Ten: Chaos Theory Fact and Fiction

In mathematical science an exciting form of modeling has developed known as Chaos Theory. Not too surprisingly, I have seen a number of roulette "specialists" out there selling systems that claim to be informed by chaos theory. I have been shown several of these systems and asked for my assessment.

Unfortunately, everything I have seen so far is utter nonsense that has nothing to do with chaos theory. To explain this, I will begin by describing what chaos theory is and is not, then discussing ways it can and cannot be used with respect to roulette.

Chaos theory is simply a way to mathematically model dynamic and deterministic systems. Both parts are important: A dynamic Roulette system is one that is constantly changing, like the weather. So far it sounds like chaos theory can fit roulette. But the second part is also crucial: the system must also be deterministic. That is, there is a causal relationship between one event and the next. That causal relationship might be fluid and dynamic (that is where chaos theory comes in handy), but there still must be a causal relationship between events. In roulette, the result of each spin is totally independent of the next spin. There is no causal relationship between one spin and the next. Thus, chaos theory simply does not apply. The numbers generated by spinning a roulette wheel are independent and thus the you do not have a deterministic system.

Chaotic systems are not random, whereas the results of roulette spins are. Now, I should note, chaos theory might be used to help a computer model where the spinning ball will land. The ball spinning around the roulette wheel is an excellent example of a dynamic (constantly changing) but deterministic system. By deterministic in this case I mean that there are physical forces (velocity, direction, the frets, air resistance, etc.) that causally determine where the ball will land. Because the system is so dynamic, it is hard to predict where, but computers have been programmed to help do exactly that, and chaos theory could enhance the models used by those computers.

On the other hand, the results of the spins--that is, the number and its properties like even/odd, 1-18/19-36, red/black--are random. The results of one spin have absolutely no causal deterministic relationship to the next.

As it is often said, the wheel has no memory. Because the results are random and nondeterministic, chaos theory is irrelevant and cannot help predict results in any way. It is sheer nonsense and someone claiming otherwise is trying to use the prestige and mystery associated with chaos theory to sell something.

One system supposedly based on chaos theory show little diagrams that show the "patterns" that are generated by spins. So the system tracks trends like R-R-R-B-B-B and claims that chaos theory can help predict these trends.

Utter nonsense. In one sense, of course there are patterns. The most annoying thing that some unhelpful "systems" do is claim that they see "patterns" in the results. If 32 hits after a 22 they will not knowingly and say, "see? 32 often follows a 22." Of course they also would have said that if 32 followed a 23 (inverse numbers), or a 32 followed a 16 (16 x 2 = 32). What is happening here is a reflection of the fact that we humans are very good at seeing patterns. You can "see" some sort of pattern in just about any sequence! The only patterns that are important are the ones that we can track statistically that have implications in terms of probability theory. The rest is nonsense.

If such diagrams were mapping wind currents or how the ball bounces, then the diagrams would be heuristically helpful. But as they are merely reconstructions of the random results, they are meaningless and misleading.

Such systems, like almost all systems based on "patterns," is based on the well-known Gambler's Fallacy. The Gambler's Fallacy refers to the belief that one can predict a given spin based on a set of preceding spins. It is a fallacy because each spin is an independent event that cannot "cause" a future event to occur. The wheel does not know what color a number is and it has no memory of what number just appeared.

The patterns most systems identify as betting opportunities can start or stop at any time whatsoever. They are random. I have charts that take 100,000 spins and codes them into streaks of red/black. You will find that the streaks are unpredictably random (except at extremes - see below): a long streak may be followed by a short one or an even longer one. The patterns supposedly chaos theory systems try to track and simply not reliable.

There are only two times when the gambler's fallacy is not a fallacy: The first is when you have reached an extreme sequence that can be described in very specific probability terms as a statistically-anomalous sequence. That is not what I have seen any chaos theory salesperson do. Their sequences are too short for that.

The second is if you can identify a physical cause to the pattern, such as a biased wheel. This may or may not be possible. The verdict is not in yet concerning dealer signatures, for example. But the point is that those who market systems based on the numbers that appear and claim to be guided by chaos theory are simply wrong. Their systems are totally useless unless they have to do with the physical causes involved and are used in conjunction with a computer designed to predict where the ball will land.