John Solitude’s Roulette ABC

Many players are facing a hard time coming to grips with ‘gamblers talk’ about roulette.

Do you want to become a gambling expert?
Click on the below words to find out if you know the right answer to common roulette terms.

More detailed descriptions on any of the terms can be be found in our free John Solitude Roulette Fact and Fiction Guide.

If you have any suggestions for more “difficult” words that need to be explained contact us.


On each spin of the wheel, a number has a 1/37 (French) or 1/38 (American) chance of appearing. ‘Bias’ would indicate that some numbers or sectors have a higher or lower expectancy of appearing due to a malfunction of the wheel or bad maintenance.

Although bias play is still actively being promoted by several roulette strategy sellers, it should be made clear that respectable casinos have installed statistical software as a countermeasure. An example of such a tool can be seen here.

The probability is far higher the casino will notice bias before a player does, because the casino at all times has a larger sample size of outcomes, making the statistical analysis more significant. It is statistically impossible to determine bias on a low sample amount of roulette outcomes, because it would be impossible to distinguish between coincidence or bias.

Two statistical calculations are used to monitor the outcomes of wheels for bias: standard deviation and the chi square test. You can learn how to calculate standard deviation and the chi square test yourself by downloading our automated spreadsheet or our free John Solitude’s Roulette Fact and Fiction guide.

Warning: determining bias in a correct statistical way requires collecting large amounts — think thousands — of outcomes (called ‘clocking‘) otherwise the result would not be statistically significant.

Go back up


‘Clocking’ or sometimes called ‘tracking’ is collecting large outcome samples — think thousands and thousands of spins — of roulette wheels. The goal of ‘clocking’ is generally to find out if a wheel is biased or not. Clocking is preferably done without interruption, otherwise there is no way of knowing if the conditions of the table have changed in the mean time (cleaning, levelling, rotating or switching the pocket seperators).

If the conditions of the table have changed in the mean time, all previous statistical analysis is invalid. It would be the equivalent of running a poll who is about to become president, while the candidates have changed in the mean time.

Go back up

Computer ballistics

Using a (hidden) computer device to predict with margin of error where the ball is more likely to land.

The use of computer devices to predict outcomes is (explicitly) illegal in most venues. Detection can result in law suites and even imprisonment.

Although the idea looks theoretically appealing, it is very hard to successfully use in a real environment. It should also be noted that common DVD-demonstrations of sellers on the internet are not prove if the device is actually working because:

a) the viewer has no way of knowing if the demonstration wheel is biased or not

b) the amount of spins provided is statistically too low to achieve the statistical significance needed, in order to determine if the result is due to coincidence or a real advantage

The only real test would be to observe a demonstration in a real casino environment using a large amount of spins. Again, a short sample would not prove anything, because for every player it is possible to end up with a good result on a low and even medium amount of spins without even using any strategy at all, due to thestandard deviation that can be expected.

You can read more about computer ballistics scams in our ‘scams & fraud‘ section or in our free John Solitude’s Roulette Fact and Fiction Guide.

Go back up

Confidence level

A statistical term, used in the calculation of the Chi Square Test. The confidence level can be manually set and expresses the probability certain outcomes are more likely to be the result of coincidence or bias. You can learn how to set up a Chi Square Test by downloading our automated spreadsheet or our free John Solitude’s Roulette Fact and Fiction Guide.

Casinos use standard deviation calculations in combination with the Chi Square Test to monitor their tables, you can see an example of such a statistical analysis tool over here.

Go back up

Dealer’s signature

Dealer’s signature indicates the idea (not a proven fact) that dealers after a while would subconsciously get in a certain rhythm, launching the ball at a relatively stable velocity (speed).

If this would be the case, and the dealer’s signature is strong enough, a player could predict with margin of error where the next throw is more likely to land, resulting in an advantage.

Although the idea is theoretically appealing, it should be noted collecting low amount of outcomes, is statistically invalid because it would be impossible to distinguish between coincidence or fact.

To determine dealer’s signature one needs to count the distance the ball traveled from number to number between each spin (for instance 7 pockets, 12 pockets, and so on). The average distance traveled would be the ‘dealer’s signature’.

Equally to determining bias, one would need to collect large amounts of spins from the same dealer — think thousands — from different sessions to be able to reach statistical significance.

Go back up

Flat betting

A ‘flat better’ will never raise or lower the stake after each trial, no matter what the result. ‘Visual ballistics players‘ and ‘bias players‘ in general use ‘flat betting.’

Go back up

Gamblers Fallacy

The common players misconception previous results in roulette would have a direct influence on future results. Although probability theory can be used to make estimates what is more or less likely, one should never forget that the amount of pockets left after each trial never diminishes, so by definition one can never rule out the impossible.

Example: although the probability is very small, the ball would finish 5 times in 5 consecutive trials in the same pocket, it can never be ruled out completely. The same goes for any other combination one might prefer to bet.

It is sufficient to be aware of ‘Murphy’s Law': ‘If it can happen, it will happen.’ Because there is not one pocket ever removed from the game, eventually sequences of which the probability is very low (for instance 15 even chances in a row) WILL appear.

Roulette is considered to be a random independent event.

Go back up

House-edge / Negative Expectancy

The mathematical advantage of the house.

When a player bets on roulette the absolute probability to get a single number right on a single spin is 1/37 or 1/38, but the reward is only 35-1. On each spin of the wheel the risk is slightly higher than the reward.

The house-edge in French roulette (single zero) is 2.703% and in American roulette (double zero) it is 5.263%.

In the long run the player is expected to loose the above percentage of the capital he invested. Hence the name “negative expectancy”. Example: if a player would place for 100.000 $ of bets in several years, on average he will loose 2.7% or 5.2% of this amount.

In the short (a couple of hundred spins) or medium run (a couple of thousands spins), the practical result will differ due to the standard deviation. As such a player can be fairly convinced he develop a winning strategy, but the result was only due to luck.

The only way to find out if a certain strategy offers ‘an edge’ is testing a certain strategy using very large amounts of spins, for instance using simulation software.

Go back up

Low profile wheel

Old types of wheels had deep pockets; the result was when a ball hit the rotor it was more likely to end up in a pocket near the impact point, making it less hard to predict the final resting place.

Low profile wheels which are now commonly found in respectable casinos have shallow pockets, increasing the scatter, making it harder to predict where the ball will finally end up.

Go back up

Probability theory

A mathematical theory used to analyze the nature of chance. Originally developed by the French mathematician Pascal Blaise in the 19th century, now widely used in any domain of business occupied with risk, including casinos.

Probability determines the odds and pay-outs of all casino games. The probability formulas needed to analyze roulette outcomes can be found in our free John Solitude’s Roulette Fact and Fiction guide or calculated automatically using our spreadsheet.

Warning: due to the mathematical nature of the game of roulette, it is theoretically impossible to ever achieve absolute certainty of a win, no matter what mathematical system one would use, so there is always a risk factor remaining.

Go back up


A “progression player” will raise (negative progression) or lower (positive progression) the stake, after each bet, depending on the outcome.

“Martingale” is a typical negative progression system (double up after each mistrial on the even chances), while d’Alembert, Fibonacci and Labouchere are combinations of negative and positive progressions.

It should be noted that it is proven beyond any reasonable doubt that NONE of the previous systems will result in profit in the long run.

Go back up


In relation to roulette: unpredictable, chaotic of nature, without structure.

Although roulette outcomes can be analyzed using the mathematical laws of chance (probability theory), the high degree of randomness ensures that it can take a while before any given system will run in to streaks that can not be beaten using probability theory.

The amount of time it will take before previous wins turn into losses is depending on chance and how kind the standard deviation is towards the player.

Go back up

Random independent event

Each spin of the wheel is a completely new event; the absolute probability the ball will end up in any of the 37 or 38 pockets is 1/37 or 1/38 on each new spin. Hence the name “random independent”.

Although probability theory can be used to estimate what is more or less likely when it comes to outcomes of several consecutive spins, because there is not one single pocket removed in the course of the game it is impossible to ever reach certainty.

Blackjack however is a random dependent event: because cards are removed from the deck after each draw, the cards remaining in the deck will always be dependent on the cards which were already removed (hence the name: random dependent event).

Random dependent events are by definition more predictable than random independent events, because the probabilities diminish after each trial.

Go back up

Scalloped pockets

Scalloped pockets were originally introduced by John Huxley, one of the major wheel manufacturers situated in London. Scalloped pockets resemble a shallow spoon shape like pocket surface, increasing the amount of scatter and randomness that can be expected.

Go back up

Scamming, scammer

In relation to roulette: a roulette strategy seller who is (deliberately) attracting players by using misleading advertisements, so a player would buy a certain (expensive) system, strategy or device, being dazzled with prospects of great fortune.

The goal of the seller is generally not to get rich by gambling (otherwise if his or her theories would be correct, he or she would be a millionaire already and would have no need to sell devices or strategies) but earning money by becoming a successful scammer.

In the “Scams & Fraud” section you will find the most common scams.

Go back up


When the ball makes it final descent to hit the rotating pocket surface, there are different possibilities, depending on the material, weight of the ball, the angle and the velocity:

– the ball drops dead and finishes in or very near the pocket to where it originally landed

– the ball bounces and ends up pockets away from the original impact point

– the ball climbs on the central cone before ending up in a pocket

Casinos have introduced low profile wheels to increase the amount of scatter, making these wheels far harder to predict as well as by computer or visually than deep pocket wheels.

Although it would be theoretically possible by estimate to calculate by computer (computer ballistics) or mentally (visual ballistics) where the ball might land, neither method can be accurate enough to also calculate the exact spot where the ball might land; it is however the last factor which will determine the amount of scatter.

Go back up

Standard deviation

A statistical calculation, used to determine how far outcomes of a roulette wheel could differ from the expected mean (average).

For the roulette player there are two important standard deviation calculations:

– expected standard deviation (how much can outcomes differ from the mean without getting suspicious), this can be calculated in front before even playing a session

– observed standard deviation (how high are the outcomes differing from the mean on the outcomes you observed)

You can learn how to calculate standard deviation by downloading our automated spreadsheet or our freeJohn Solitude’s Fact and Fiction Guide.

Go back up


Statistics is a sub domain of mathematics, specialized in collecting, ordering and interpreting large amount of data using mathematical calculations.

Statistics are used by the casino to monitor the outcomes of each game offered. In relation to roulette the most important statistical calculations are standard deviation and the chi square test.

Go back up

Statistical significance

A statistical term expressing the figure that is needed before a mathematician can determine if a certain statement is valid or not.

The most common scam of internet roulette strategy sellers is making statements which do not confirm with the figure of statistical significance needed, before one could state a certain statement (for instance “this is a winning system”) is true or not.

Why do you need “statistical significance” before you could state a certain argument is valid or not?

After each spin the amount of mathematical combinations is raised by a multiplication of 37 or 38. So after, two spins the amount of combinations is already 37 * 37 or 38 * 38, for each spin that follows one has to multiply with the amount of probabalities in the game.

Statistical significance is so important because it allows to make a distinction between statements based on mathematical proof rather than superstition.

For instance, in only a couple of hundred spins, the amount of mathematical combinations that appeared are NOTHING compared to all possible combinations the game of roulette could generate in the same amount of spins on another occasion.

So, making a statement based on only a couple of hundred spins is considered to be statistically non significant. Depending on the statement one wants to investigate, it takes thousands or even hundredthousands of spins to reach statistical significance.

Go back up

Visual ballistics

The idea (not fact) players would be able to predict the outcome of a spin, by making mental calculations which involve comparing the wheel rotation speed vs the velocity of the launch vs a reference point.

The John Solitude Project has extensively tested two visual prediction players in a real environment, both failed to produce a positive result in less than 1.000 spins.

Although the idea may be theoretically valid, the enormous concentration needed by the player, may deam it practically impossible.

It should also be noted the introduction of low profile wheels increased the amount of scatter on impact of the ball on the rotating surface. As such, a visual determination player could guess right where the ball is going to land, but he doesn’t have any control over the amount of scatter that will be produced, depending on the material of the ball and the angle of impact.

Go back up