OK, you did a great job of explaining your method/system. I now can relate to it completely. Here is what you are doing. It happens to be another (exact-moment) obfuscation technique. What you are experiencing is a temporary success streak. 33 thousand spins is actually a very small sample when you consider that bets are only placed at 1 or 2 times in the first 15 spins and 1 or 2 times in the next 8 spins. Now if you keep playing you are using the exact moment every 8 or 9 spins with an already chosen exact single dozen bet. This technique will resolve itself to the mathematical expectation of the "house edge" respective of how many zeros there are on the table. There are two things that can be happening with your results. There is a programming error in the spreadsheet or the sample size is too small. I just got done with an obfuscation technique system right here on this board using a deliberately flawed algorithm. Once I corrected the flaw it punched out perfect expectation results for the house's advantage. I suggest you check the spreadsheet for flaws. I put check-able arithmetic results for each bet made and for each spin in the test in locations where I can see if I made a calculation error in writing the software. Last time it was the number of positive results used in a simple addition column of wins. Someone here pointed it out and the baloney festival was over. Sorry no prizes for winning the festival. I know that you do not have a winning system. I'd advise you not to take this into a real casino.