I have personal experience with this method,wins approximately 55% of the sessions (30 spins),the winning sessions is larger than the losing ones,therefore an overall winner with flat bets.
I have tested for long time and I'd like to add that it's not working at all roulette tables,neither everyday.
Check the attachment for the method's instructions.
System missing a lot to be working.
1. Separation by direction
2. Separation by conditions.
Not all winning numbers are of any interest and not always. Some numbers show nothing at all. Example... ball drop and stop. These numbers you shold exclude from chart. If dealer used to give 4, 4.5 sec rotor and change for 2.5, 3- need other chart.... if you can separate by type of throw- need one more chart as well. Ball change- other chart... ets. Best of all similar aproach is explained on this forum on Mr.Ken topic .
In short , to play like this, need long jumps of ball and stable conditions. If ball stop to produce long jump- betting should stop.
I played this type of aproach during long time and till now , on situable wheels l do it. But to use this method to its full potentiAl, need to understand what makes it working. Doing separate charts by ball behavior and conditions can and will pin point numbers ( sectors) of interest qweaker then any other method. However need to understand wich wheels are situate to such play and wich ones do not qualify.