Alright. Here is the explanation for the losses in your simulations:
I used to think the system works in RNGs and real life wheels that are hand spun. What I have come to realize is that in an RNG the system loses because the past spins do not predict or control future spins, as I have been told very, very many times in the past. Here is where my system gets interesting. I bought roulette xtreme about a month ago. I downloaded thousands of spins from a live dealer casino with one of those automated wheels, in which an RNG picks a number, and then the seemingly free-spinning wheel decelerates to force the ball to land on said number. My system cannot beat that. That is true randomness, and that is not real roulette. The system also will not work with online games with RNG, and/or Psuedo RNG. It only works with an actual wheel with actual spin results.
To reassure myself of this I downloaded 10,000 spins from Dublin. It is a giant collection of series of 400 spins. What I noticed immediately was how often I was winning compared to the RNG spins. Let me explain exactly why this was happening, because I have been working on this for a year, now, and the results are exactly what I thought they would be. Here is the explanation:
If you have not heard of the mathematical phenomenon known as the birthday paradox, it is the reason my system works in real life and not on a computer. The birthday paradox shows why in a group of exactly 23 people there is just over a 50% chance that two of those 23 have the same birthday. If you have a graphing calculator the following equation proves this. (365 npr 23)/365^23
That equation will tell you the chance that there IS NOT a match, meaning 1-the answer to the previous equation yields just over 50%. Now that you know this, let me show you how many roulette numbers you would need to have a probability of AT LEAST 50% that at least two of them match each other. The answer, conveniently for both american and european roulette, is 8 numbers. You need at least 8 numbers for there to be a more than likely chance that two match. Here is the equation:
1-((38 npr 8 )/38^8) for american, and 1-((37 npr 8 )/37^8) for european. What this means is that the likelyhood that the last 8 numbers to appear have a match is going to always be over 50%, and its actually more like 55% for american and 56% for european. So from here you would think that if 7 numbers come up in a row that don't match one another there must be a 55% or 56% chance that the next number is one of those 7, right? Wrong. The equation doesn't end there. Here is how it ends.
The 55 or 56% probability of a win is spread evenly among the numbers. The only numbers you are actually concerned about, however, is the next number to appear, or the 8th number, and whatever number matches it from the last 7. So how many ways can 2 of the same number be arranged in 8 different slots? The answer is found with the following equation: 8 ncr 2, which is the same thing as 8 choose 2. The answer is 28. There are 28 different positions the matching numbers can be found in. But how many matter? That is, how many can you profit from? The answer is 7 out of the 28. Here's why. You have 8 different spots, right? Picture the 8th spot. Now picture the last 7 spots. How many can you pair together that involve the 8th spot? 1 and 8, 2 and 8, 3 and 8, 4 and 8, 5 and 8, 6 and 8, 7 and 8. That's all. Just those 7.
The aforementioned 55 or 56% is now multiplied by the 7/28, or exactly .25. This leaves you with around 14 %. So of the 55 or 56% chance that there is a match, provided that the match has not occurred yet in the last 7 numbers, the entire percentage is not applied to the last number, but instead only a quarter of it, which is 14%. So now we want to know if we can still win this bet. Yes, you can still win. Here is why.
You had a 55 or 56% chance of a match in 8. That means there was a 45 or 44% chance of there not being a match, right? Okay so now that only a fourth of your winning percentage is relevant to you, 14%, it is compared to the losing probability, which is 45% or 44%. So here that is. 14/(14 + 44 or 45) = either 23.2 % for american or 23.9% for european. Keep in mind that I rounded the 14 from 13.7 or so, so if you get a different answer just don't round the numbers and try it again. Anyway, so now you have lets go with the 23.2%, just because its lower, and I want to see how much abuse the system takes. So the payout on a straight-up bet is 35 to 1, so if you are betting on 7 numbers, if you win you get 29 dollars, and if you lose you lose 7 dollars. What this means is you can lose 4 times and win only 1 time and still make a profit of one dollar. That would mean you are only winning 20% of the time. Now picture what happens when you win 23.2% of the time. That's right, you would win a lot more.
In the real world, this system does work. But it has to be on a real, hand-spun wheel without any sort of electronic interference. Also, I have found that betting on the last 8 numbers to appear that don't match RATHER than only betting on 7, which was my original idea, actually makes money faster. If you were to do it this way you would have either a 28.49% chance of winning for american, or a 29.45% chance for european. The payout in this case would be 28 dollars for a win and -8 dollars for a loss, which means that in order to profit you would need to be winning at least 25% of the time, or one win for every three losses. I personally recommend the betting on 8 numbers instead of 7, which is why I requested the RX scripting for 8 instead of 7.
Well guys that's about it. I also did some math with how much money you would make doing this. If you were at a wheel that got spun around 60 times per hour you could make around $100 an hour from doing the 8 number betting method, and around the same amount doing the 7 number bet. I recommend the 8 number because it is slightly faster. Also you need a bankroll of around $500 to cover any deviations. Well that was quite the mouthful. By the way I have tested this on my own roulette wheel and it works there too. Have a good one, guys. Any questions just let me know.