Games of chance are made up of a series of events that are not predictable. If we toss an unbiased coin, the best we can say is that it will land either tails-side-up or heads-side-up and that the two outcomes have the same chance of occurring. Many of the beliefs and “systems” people who gamble develop are based on misconceptions about the nature of random events. So it is worthwhile to examine in more detail the essence of what it means to say that something is random.
Randomness is difficult to define. Random events are unpredictable, erratic, unplanned and independent of each other. However, random events sometimes appear to form a pattern or serve a purpose. For example, there are areas in the night sky, such as Orion’s “belt,” where stars appear to form a straight line. Given enough opportunity, any pattern could form by chance alone. An infinite number of monkeys on typewriters could eventually type out the complete works of Shakespeare.
Although random events appear to happen without a rule or cause, they are in fact the result of material cause (e.g., gravity and friction), but an exact list of forces acting on a random number generator (e.g., dice) may be unknown or impossible to specify precisely.
Sometimes clients believe that there is no such thing as randomness and that it is therefore possible to predict the outcome of games. Other people believe that random events have no cause, they just happen. This can make random events seem rather mysterious. Interestingly many religions used random events as part of their religious ritual to divine the will of the gods (Gabriel, 2003). There is nothing mysterious about random events. All physical events are determined or caused by something. Mechanical randomizers such as bingo balls, roulette wheels and dice use the laws of physics to maximize uncertainty. The basis of all random-like events is a combination of (1) initial uncertainty and (2) complex or non-linear relationships.
Uncertainty simply means that we do not know the exact values of all the variables with absolute precision. Uncertainty is an inherent part of measurement; nothing is ever 100% certain. A car driving at 70 kilometres per hour in cruise control will vary in speed by 1 or 2 kilometres per hour (more on a hilly highway). Thus there is some uncertainty as to the exact speed at any given moment. Orkin (2000) illustrates this problem with the question “How many fish are exactly 12 inches long?” Suppose a type of fish is usually 12 inches long. In all cases 12 inches is only an approximation. If a fish is 12.000001 inches long, it is not exactly 12 inches long. It is not possible to measure something so precisely as to completely eliminate uncertainty.
A complex or non-linear relationship is one in which a small change in the input causes an unpredictable change in the outcome: sometimes a large change, at other times a small one. For example, there is a non-linear relationship between caffeine and performance. Too little caffeine and a person might have trouble staying awake; too much and the person might become agitated and unable to concentrate on what he or she is doing. Suppose a researcher wanted to know how caffeine affected performance on a task. Initial uncertainties in this example would be factors such as how much sleep the research participants had the night before, how many cups of coffee they had that morning and how much coffee they usually drink per day (i.e., their level of tolerance). If the researcher did not make an effort to control for these factors, the uncertainties combined with the non-linear effect of caffeine could produce chaotic test results.
Over the past 30 years, physicists and mathematicians have come to realize that “tiny differences in input can quickly become overwhelming differences in output” (Gleick, 1987, p. 8). Chaos describes the unpredictable effects associated with small changes in a complex system (Gleick, 1987). For example, given the exact same weather conditions, the flapping of a butterfly’s wings might make the difference a week later between a thunderstorm and a sunny day. This is a somewhat romantic exaggeration of chaos, and it is unlikely that a butterfly could actually have such a profound effect. However, when uncertainty is combined with complexity, the results can be completely unpredictable. Although this sounds improbable, physicists have found that small changes to the initial conditions within climate models grow in unpredictable ways because of the complexity of the system. All true random events are the result of chaos, but many chaotic patterns would not be sufficiently random to be used in a casino game.
The problem, from the gambler’s point of view, is that the precision with which the initial conditions would have to be specified in order to predict the outcome is beyond the gambler’s capacity. That is, unless players can control or measure the speed of a roulette ball and the wheel it is rolling around in, they cannot make a precise prediction about where the ball will land (Stewart, 1989). In the late 1970's, a group of engineering and computer science students at Stanford University tried to beat the roulette wheel using a concealed computer (Bass, 1985). Although theoretically possible, their plan ultimately failed because of the practical, legal and safety related difficulties of having to conceal their computer in a shoe. The use of a concealed computer to predict a casino game’s outcome is illegal.
In summary, random events are the result of the chaotic interaction of uncertainty and complexity. This begs the question, “Is anything truly random?” If not, then surely one can predict the outcome of “random” events! On the contrary, the fact is that deterministic chaos does an excellent job of creating random events. The amount of information needed to gain an edge in a game of chance is often extremely large. For example, no dice are perfectly cubed, and this will produce a slight bias. But the bias on a pair of casino dice might not show up until after several thousand bets, and even then would most likely be too small to allow the player to make money. So, while in theory nothing is completely random, in practice many games produce events that are indistinguishable from being purely random.