This the test for the tweak I talked about here: link:://:.rouletteforum.cc/index.php?topic=15260.msg131130#msg131130

QuoteThen one thing happens: there is at least one Loss every 10 spins. That's almost guaranteed. Which means there's "always" a single dozen that cycles at least once.

Betting instead for that dozen to hit makes sense. We can use a fibo for that. And then wait for the cycle to end and repeat procedure.

Tracking for cycles of 12 spins is safer.

Method is simple: bet same as last 12th spin, stop on a win. Wait for a new cycle of 12, repeat. It's based on the principle there will be at least a Win every 10 spins or more. Even if there was a Win right on the first spin, we wait for that cycle to end. There must be only one win per cycle. This prevents from going on a string of max 22 losses if there's a hit on the 1st spin and the next only on the 24th. Better be a boring winner than a funny loser.

Attached that session from SBWB that has destroyed a few methods of mine.

Another session. 300+ spins.

Then one thing happens: there is at least one Loss every 10 spins.

psi

Would you expand a bit on the above statement?  Are you saying that in ten spins the last two dozens to come will miss once--that the odd dozen will almost always hit once in ten spins?

Betting the last two dozens/columns to come is a very good method as you miss every sleeper and catch the runs of eight.  In the Zuma tester there are many runs of eight and more.

But I digress.  Please elaborate on the first sentence in Navy.

Thanks for your work.

Sam

Turbo's 2 dozen method consists in betting DTL fourth dozen that hit. Tweaking it to DTL eighth you get lots of wins in a row with occasional long series of losses. Near 100% of the time there will be at least one loss every eight spins. It means one win on a single dozen is almost guaranteed. We can use a progression for dozens and aim for that single win per cycle.
While testing with cycles of 8, only once the win hit after 10 spins, beyond the 8-spin cycle. So I stretched the cycle to 12 spins. The point is not to reach the 12th step of a progression, but to have more (good) variance so the wins hit earlier.
Longest progression so far: 8 steps; there were two zeros. Haven't think about a stop loss yet.

Having good results so far in early testing with this.
Quite a few methods come about by reverse engineering and this looks logical enough.
Just so I know I'm on the right lines, we track 12 dozens then bet that number 13 is same as 1, 14 the same as 2 etc?
I've also looked at columns with similar results.

I like the idea of a Fibo for dozens...glad to see it used in the example...to me it just makes sense to play to win some actual money

When playing dozens I would always have a low stop loss as wins come rapid fire in alot of instances and loss strings can be extensive...don't chase  (I would retreat at something in the two to four losses in a row range) and then either pickup the attack where I left off or drop back a stage or two in the progression.  Normally reset on the win  (BUT... maybe recommence the next attack at a higher starting point depending on your research...this is where testing helps make that call)

Using a single matrix can be a risky deal and it gets riskier using the Fibo.  Don't get me wrong I am not afraid of progressions but we need to use some caution.  If the bet selection is solid it is not always necessary to get to breakeven or profit on a single victory.  There are many ways to maximize profit while still limiting exposure.

If I were looking at using a matrix style bet selection such as this I would look to combine several together and differential bet with the results if you can do the math quick enough (I would always bet on a real wheel)

That's just me and my two cents...take it for what it's worth