I've been working on my own for a while looking at different methods of bet play and
You have, betting next to last.http://betselection.cc/dozencolumn-7/dozen-repeater-iii/http://www.rouletteforum.cc/index.php?topic=15831.0
And these kinds of tracking methods.
I like D'alenbert for betting. I just always have, even though I know it can
be a disaster depending on how it goes.
But a variation of looking at one side of a thing is to switch betting sides
and follow the result that is happening.
For "next to last", on a loss, you switch to "opposite of last".
For the methods looking for 2 EC's or 2 dozen, the same thing.
I have been watching this for a while now. I went to do some paper play today,
and, like a lot of my recent data and zumma tests, it seems to hold up.
One specific thing I was looking at playing the "next to last" was to pick a number
like 5 and if you lose that bet, then you quit and start over. Take your 15 unit loss,
and play sessions based on that. It seems to have a good result.
If this gets off the ground then I'll post some actual runs of live data, like what
I got today. The betting is different between EC's and dozens. But the results always
seem to cluster enough. When I play dozens (+1 -1) and there is a loss I go to playing
the "opposite two" flat, and back again. Two different games, so to speak.
But with the right betting scheme applied to the right bets, this idea seems to hold up.