Been playing around with an idea - experimenting past few weeks - some success, need a more thorough way of tracking to properly test - what do the veterans think, does this have any bearing or would it be just a waste of time
If this could be automated through a program would be great, but would probably be quite a task to do
The idea
Start off by tracking say 6 spins for a particular dealer, marking the results onto a diagram of the wheel. After the six spins, overlay different shapes onto the wheel to determine if a pattern could be established for the dealers spins, by placing a point of the shape on the number before last to determine if the next number out fell on another point of the shape or its neighbors (using the past six spins to determine which shape would have produced the highest hit rate). Not every point of the shape and its neighbors need to be bet, as i have found there are times certain points can be ignored based on temporary patterns that emerge, Hot areas of the wheel, and minimum repeats.
Shapes that I have been testing with are (square, kite, 5 point star, and 6 point star). So far in testing have found following the square produces descent results with minimum number bets (been using it as a nuke by betting both points either side of last and their neighbors,) . Been betting 3 times for a hit, and then re-tracking. Having a program that could calculate the best shape (varying different angles to more appropriately trace patterns, would provide faster more accurate tracking, whether or not this would be possible, dont know.
Just a thought.