Kokodrilo (title of nobility for Big-time Gambler):
The Super Roulette Strategy is presented in great detail on its dedicated page:
The Best-Ever Roulette Strategy, Systems based on mathematics of progressions and free-for-all
There are further details on how I applied my roulette strategy in real casinos:
Why casinos still ban the skilled players, winning gamblers.
Also, there are facts presented in my probability book. Additionally, my book presents the very new and original FFG deviation and how I can apply it to casino gambling, roulette specifically. The FFG deviation has a great advantage over the statistical standard deviation: FFG can be calculated in advance. The two parameters, however, are very close in size.
The facts on how I applied my Super Roulette Strategy in real-life casinos (Atlantic City, New Jersey, USA) are beyond doubt. Otherwise, the casinos I mentioned, their employee names, badge numbers, exact times and dates, would have filed lawsuits against me long ago!
This thread is not the place to republish facts already published. It makes no sense whatsoever! Thus, you can answer your questions… yourself by re-reading the materials I just mentioned.
Perhaps this is the place to talk in a little more detail about the probability of streaks. But, again, you need study more thoroughly my theories freely and clearly expressed at my website. You should understand well how my free software SuperFormula works and what it does. You should also be aware that the number of trials (roulette spins, in this case) is very important. The longer the sessions (number of trials), the longer the losing streaks can get. Granted, your winning streaks get longer as well — but not as long as the L or – (minus sign).
So, you sy you play the double-dozen / double column roulette strategy. A double dozen (like the roulette numbers 1 – 12 and 13 – 24) have a probability p = 24 / 37 (I assume you are European). That is the W probability. The L (loss) probability is the reverse: 1 – W or p = (37 – 24) / 37 = 13/37. You can calculate the probability of streaks even with a hand-held calculator.
What is the probability of a single win in a number of spins N? You describe a single W streak this way: LWL. That is, the W begins immediately after a loss L and ends immediately (followed by another loss). You describe a double-win (WW) streak this way: LWWL. That is, the W begins immediately after a loss L, hits again, and is followed by a loss. And so on …
Let's take a session equal to 37 * 3 roulette spins: a number of trials equal to 111 spins. How many WW double-win streaks can we have in 111 spins? LWWL = 13/37 * 24/37 * 24/37 * 13/37 = 97344/1874161 = 1 in 19.25. If we divide 111 by 19.25 we get 5 or 6 double-win (WW) streaks in 111 roulette spins.
How many triple-losses (LLL) in 111 spins? WLLLW = 24/37 * 24/37 * 13/37 * 13/37 * 13/37 = 1265472/69343957 = 1 in 54.8 or 2 times in 111 spins.
How many triple-wins (WWW) in 111 spins? LWWWL = 13/37 * 13/37 * 24/37 * 24/37 * 24/37 = 2336256/69343957 = 1 in 29.7 or 3 or 4 times in 111 spins.
Of course, you can avoid all that manual labor. You can run my gambling software named Streaks. It belongs in software category 5.6: Scientific Software: Mathematics, Statistics, Probability, Combinatorics, Odds, Algorithms.
Now, don't give me that bull that they throw at me in other gambling forums! That is, they encountered LLL some 20 times in 111 spins or so! That's how they argue against my theories! They even throw their insanity in my inboxes: “Ion, your super strategy brought me into bankruptcy! I lost a dozen times in a row playing a double-dozen! It happened to me numerous times…ye know, the gambler's fallacy is God's punishment for the gambling sin!” On the other hand, you will never hear me say that my theories are 100% guaranteed to win! I see that string in searches referring to my Web site. I am the first human to discover that there is no such thing as absolute certainty. They can take that to the bank, but not to the court!