LoopRS driving me LOOPY!
I have a running test, have to increment (or i suppose decrement is proper) each skip by hand, but progress is being made.
so far, the repeats tend to hold positionally, for all 5 WB positions.
I have backtested stand-alone on the red ball to find that the number that was drawn 54 draws ago is most likely to repeat
NOTE most likely means 34 occurrences.. that's only 3.59% of all draws...
I tested the red ball back to 241 draws and 54 back is king.
I am only back 136 draws in the white balls (harder process with the oopr's) but the frontrunners for each position are all positional, with the largest cluster of repeats coming from 10 draws back (WB1, 3 and 4 are all highest @ 10 draws back, keep in mind were talking less than 7%)
it may take weeks but I will carry this out until i can go back no further... at least it will give me time to devise the next system.
The bright side is that I re-adjusted my system criteria, if I can't backtest easily, I drop it (that means the recycle bin for the PROfile monstrosity... actually the thumb drive archives, I'm an excel packrat since the HDD crash! I lost all my stat systems and v-trac sheets in that "incident", along with my big histories of PA 3/4/cash5)
I let the backtest sheet go a few draws ahead, it matched the 16 and 36 in the PB, albeit out of position. picked 19 for the pb, which was 36... oh well, glad I didn't play that pick.
for the RB data, the longest gap was 136 draws, the shortest was 2 draws, the average was 29 draws and the most popular gap was 8 draws which happened 6 times in the 34 hits. by contrast, the most popular PB is #20 with 35 hits... longest gap was 122 draws, shortest was 3, most popular gap for 20 is 20, which happened 3 times... so not far off from playing the hottest number, but even that represents less than 4% of hits.
much work to do...