Thank you for fast replay. Yes i understand that bot is design for many sites. I know that is just visual "mistake" in real play, but if i run simulate i think is not just visual "mistake". I was doing about billion test numbers in simulation, but when i started with real money, i lost a lot. Then i debugging what is difference between simulation and real play and i see that. If in simulation takes 99.89 as won bet, that is 10% error, and house edge is only 1%, at the end simulation was useless. I am not kind a programming guy, yes i can do simulation with all variables calculate manually (can not use any predefined variable), and all profit etc. But that is extremely slow. I need about month to do 1 billion bet simulations on multi computers, with a lot of CPU power. But if i do all calculations myself (do not use any of predefined variables) then it will take half a year or more.... I rather see that simulation is more strictly (get worse result on simulation) then real betting. It's better to lose many on simulation(because of error ) then tray system on real money and lose than.
But thank you for helping us.
Best Regards