Roulette: can someone help me analyse this?

**lojo** · 8th October 2007, 09:49 PM

Total bet $500
The second column and second twelve have $70 on them
The 'box of 4s' $40
the double bets $20
each of the 12 numbers has 10

So payouts are:

13 - 930
14 - 1860
15 - 930
16 - 1650
17 - 3660
18 - 1650
19 - 930
20 - 1860
21 - 930
22 - 570
23 - 780
24 - 570
2,5,8,11,26,29,32,35 - 210
all others - 0

I was very lucky on this simulation (starting bankroll $1000, same configuration with much smaller wagers) and only a couple of variations, i.e. covered all black in second column and rode black for 70.27% coverage)

In hundereds and thousands of simulations I haven't been able to determine optimum flat bet size as a percentage of starting bankroll... as in - I am willing to risk ruin, but will stop at a 20% profit -can this be determined?

If not, I would be happy to know how to figure my over all odds on this scheme.

I can roughly imagine how to construct the math, but am not wrapping my head around it.

As far as average ruin vs 120% I'm doing better than 1/5 overall (the only notes I've kept) but if I had a way to construct a model of the odds of some sort maybe it would show it isn't a good idea to go below (arbitrary) 50-70%

I'll bump this now and then if I don't get any help soon

it's a long term curiousity of mine. It feels like high variance roulette

Am I overcomplicating it? Is it as simple as I have twent numbers that pay an average of xtimes bet... or do I need to look at each one as a ratio of my entire bet - the return - the odds of hitting it... then meld them somehow to determine