Roulette, Online Roulette

First, if there are table limits, you can very quickly run up against the table limit. For example, if the table limit is $2,500, you will be close to that limit on the 12th consecutive losing spin and be unable to bet the next time. The odds of you losing 12 consecutive times are not great, but they exist and this will occasionally happen. When it does, you lose ? big. It will take a lot of successive wins to make up for this one loss.

Second, you will very quickly run up into some very big numbers. For example, after the 10th consecutive losing spin, you will be betting $1,024 for the 11th bet. The worst part is, you are betting that amount to win just $1. See the table below:
A variation of the Martingale strategy is to limit your string. For example, you might decide to limit your risk to a series of five losing bets. Your odds of losing that bet are down to 2.96 percent, or said the other way; you have a 97% shot at winning. If you lose again on the fifth spin, take your loss and start over with a $1 bet and work through the sequence again. But the bad news with this is that unless you have a superior bet selection method that is actually managing to put the odds in your favor, even a limited Martingale will result in your losing at the basic rate of the house edge for that particular game over the long term.
Labouchere: The Labouchere betting strategy is sometimes referred to as the "cross-out method". Again, this is for even money bets. It works like this. Begin with a small string of numbers such as 1-2-3. Your bet is the total of the first and last numbers in the series, in this case the sum of 1 and 3 which is $4. If you win that bet, you cross off the 1 and 3 and your next bet is $2 because it is the only number left in the series. If you lose the first bet, you add that total, 4, to the end of the series. Your number series is now 1-2-3-4 and your next bet would be $5. When all the numbers in the series have been crossed out, you will win the total number of units in the original series which in this case would have been $6 (1+2+3).
The advantage of this betting approach is that you only need to win 1/3 of your bets to make a profit. The disadvantage is that it doesn't take a very long losing streak before you are making some pretty big bets.
To counter escalating bets, some advise breaking down a series that contains numbers too large to bet comfortably into two smaller series. For example, the series 4-6-4-8-12 could be broken down into two series of 2-3-2-4-6. That may seem like a good idea on the surface, but the problem is that you now have to win twice as many bets as before in order to cross out the whole series and win your 6 units. You must now win not one out of three bets but two out of three bets until the two series are completed. In other words, now, in order to be successful, you have to play at a much higher win percentage than would normally be expected in order to cancel out both series. Otherwise, the series just grows longer and longer and you have to keep breaking it down again and again until eventually it will get to the point that you would have to win at a very high rate for a very long period of time in order to cancel it out.
Since this method is normally only used for even money bets which you have slightly less than a 50% chance of winning to begin with (because of the zeros on the wheel), you are really in trouble if you break down a series and don't immediately begin to win at a 67% clip. You are destined to fall further and further behind and eventually go bankrupt because of the ever-increasing need for a much higher win percentage than would normally be expected.
D'Alembert. The D'Alembert betting method is simple yet perhaps the most effective of all the approaches described thus far. You simply increase your bets by one chip after every loss and decrease by one chip after every win. So if you begin with a one chip bet and lose, you then bet two chips. If you lose again you will bet three chips. If you win on this spin you will decrease your bet to two chips.
Bets will return to one chip as soon as there is a profit on the series of bets. This method is effective because you always have more money bet when you win than when you lose. And, with a little luck you may even experience several winning sessions.
Sooner or later you will have an extended losing streak and this system will also fail.
Fibonacci. The Fibonacci betting progression is a mathematical pattern that is commonly found in nature. In this progression, the bet called for is always the sum of the two previous bets. For instance in the series 1-2-3, the three is the sum of 1 and 2. The next bet called for would be five because it is the sum of two and three. Then would come 8, 13, 21 and so on, up the line.

A Fibonacci progression generally increases to the next bet in the series after a loss but is often decreased by two levels after a win. For instance in this series, 1-2-3-5-8-13, had the 13-unit bet been won, the following bet would be decreased 2 levels and a five-unit bet would be made. However, some prefer to play more aggressively and only reduce one level after a win. As long as luck holds out, this of course will produce larger wins.
But like all the previous betting approaches described, the Fibonacci, of and by itself, will not produce long-term wins. This system, too, soon has you placing some very large bets.
Law Of Thirds. Many roulette systems are based upon a theory called the "Law of Thirds". The Law of Thirds is a mathematical principle that expresses a tendency of random numbers. The principle is basically this: Over the course of any 36 spins, only 24 of the 36 numbers will be hit. Some of the numbers will be hit more than once. This means that of the 36 numbers, approximately 12 will be multiple hits and twelve of the numbers will not be hit.
These systems use one means or another to identify and bet on these numbers that will be hit multiple times. The problem with the Law of Thirds is that it is not really a mathematical law. It does not work out that way each and every time. There is no real way to predict which numbers will and will not be hit. Remember, one of the eight truisms is that each spin is a separate event and has no relationship to any other spin.