Beating the Roulette Odds by Computer Prediction

Devising a program the wrong way

french_rouletteThe majority of programmers will work out the physics and motion dynamics at first, in order to come up with a mathematical model of a perfect, in theory, roulette wheel. Such a model will usually require basic data, such as position, velocity and deceleration at a given moment in time of both the roulette ball and the wheelhead. Regardless of what method of software prediction may be used, the correct mathematical relationship between the position of the wheelhead and the position of the ball at any moment needs to be known.

After having devised a good mathematical model, some programmers, then, make a mistake – they create a roulette prediction program, which is based entirely on that model. Other experts in the field claim that should not be the case, because, in reality, roulette wheels are not theoretically perfect. Some programmers, however, need more time to come to that realization. They may require more time to understand that the concept itself is the major issue. Those analysts continue following it, as at times the program will appear to work and in other cases it will simply not. Some of them may even believe that the program will only work at the start of the day, when they are alert and their timing seems to be more accurate and vice versa. They may believe the mechanical timing switches are quite sensitive or not sufficiently sensitive, the mathematical model has issues, or the program itself shows intermittent bugs.

Other analysts came to the realization that wheelheads of individual roulette wheels tend to decelerate differently and that depends on the type and condition of the bearings. The same applied to deceleration of roulette balls, as it depends more on the type and the condition of the ball track and less on the size and weight of the ball itself. However, there was another issue – to accurately project the drop off point of the ball, which is a key requirement for a successful outcome. Programmers usually assumed in the mathematical model, that whenever the ball decelerates to a particular minimum speed, it will fall off the ball track. On real roulette wheels, however, this was not always the case. At times, when the ball reached that minimum speed, it might remain in the track for another half revolution, as the track itself might be a bit warped. In other cases, the roulette ball might fall off half a revolution earlier due to a tiny bump or a wear groove in the track. Therefore, the conclusion was no mathematical model was able to project when and where the ball might drop. And to determine with accuracy the ball drop off point was a key factor behind the formulation of a prediction!

Roulette wheel and a winning number

Devising a program the correct way

Eventually, analysts found a way to deal with the flaws of the mathematical model. Whenever a player chooses a specific roulette wheel for play, he/she should characterize it first, so that any unique features the wheel may have can be used as a database for the program.

The typical hardware setup included a timing input device (a small push-button switch operated manually), which was connected via a thin and flexible cable to a pocket-sized computer. The output signal from that computer was usually taken from its internal speaker line and fed into a miniscule radio transmitter. Such a setup was operated by a team of two – a bettor and a timing person. The timing person clicked the push-button switch, which was concealed in his/her pocket, in order to submit timing pulses to the computer, which was hidden under his/her clothes. Based on the timing information, the program devised a prediction, which was then sent to the bettor who was standing at the far end of the roulette table (he/she would hear it in an earphone). The prediction was usually in the form of a string of long and short beeps or a computer-synthesized voice. The time period between the latest timing click and the prediction’s transmission to the bettor was less than a second. Once the bettor received that information, he/she placed bets based on it.

In order to characterize a wheel, a player may do so by using a rudimentary data collection program. It does not make any calculations, but instead collects and stores a time pulse each time the player clicks the push-button switch. The timing data is usually stored in a program array or in an individual data file. To collect data, which simply shows the deceleration of the wheelhead, requires not that much of an effort. Having positioned in proximity to the roulette wheel, the timing person clicks the pocket switch once per wheelhead revolution, as it gradually slows. He/she usually focuses on a fixed reference point on the far side of the apron and clicks the switch each time the single zero or the double zero crosses that point. The timing data will determine a deceleration curve for the wheelhead. To collect data on the features of the roulette ball is a more complex matter. The timing person again clicks the pocket switch each time the ball crosses the reference point. What should be noted is that the latest click is the precise moment when the roulette ball falls off the track and into the wheelhead. Such information is sufficient in order to create a database, which will present the relative time to the ball drop off point for any speed of the ball.

What the timing person needs to do next is to record the location of the drop point. To do this, he/she mentally divides the track periphery into 8 or 16 sections by using the ball deflectors as delineators. Whether to use 8 or 16 sections depends on how accurate the player wants his/her program to be. These sections are then numbered from 1 to 8 or from 1 to 16, as the timing person marks the reference point as the start and continues to count clockwise. Once the ball falls off the track, he/she clicks the section number into the computer. The timing person will usually have enough time to do this before the dealer gives the ball another spin. After all this information has been added to the ball database, the wheel section at any drop point will be known as well.

Now that the information related to wheel characteristics has been collected, it needs to be evaluated, adjusted, refined and reformatted for the program. For that purpose, two data processing programs are usually used – one for the ball and one for the wheelhead.

Roulette wheel and winning number

Tweet about this on TwitterShare on FacebookShare on Google+Share on LinkedInShare on Reddit