Hey, did I hear you right when you said it was Bayes that discovered a practical limit of 65 EC's in 200 spins?

Also on the

**Pales-Harry Range Technique**, I am wondering if this can be readily applied to other bet selections since 1 unit profit from an EC is a tiny bit tedious. Pales just mentioned a DS as an example. So lets look at that:

The bet cycle here is 5 spins to 1 and the range of interest appears to be in spins 14-27. Maybe spins 15-30 would be convenient since we are dealing groupings of 5 spins. So a range of 15-30 would represent 3 cycles of 5, so in my mind that would be 3 actual bets.

On the trigger, a double street will go missing 15 times about 7% of the time which is maybe once in every 14 spins? .07=100/7=14.29?

That is not too bad right?

Then the idea is to bet it 15 times with an overall chance of hitting at 99.41%. That's not bad and we have a chance at dynamic profit.

I think the betting progression could look like:

1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 4, 5, 6, 7, 8 = 47 units

Now the question becomes, how often will this fail in a row. Like Pales has pointed out, THAT IS THE QUESTION. My guess is that it will not fail more than 4 times in a row as presented (we will definitely beat the 84 expected max loss). If we wish to get that down to 3, we will need to lengthen the progression AND/OR lengthen the trigger, POSSIBLY BOTH.

Well it means we have to run a simulation and I am starting work soon so...

`10 RANDOMIZE TIMER`

20 r = INT(RND * 37): LOCATE 1, 1: PRINT ml

25 IF tr = 15 THEN 50 'process trigger hit

30 IF r < 31 THEN tr = tr + 1: GOTO 20

40 tr = 0: GOTO 20

50 'trigger has hit

60 bc = bc + 1

70 IF r >= 31 THEN bc = 0: tr = 0: GOSUB 200: GOTO 20 'hit achieved within progression

80 IF bc < 15 THEN 20 'miss

90 'progression has failed

100 l = l + 1: tr = 0: bc = 0: GOTO 20

200 'process loss counter

210 IF ml < l THEN ml = l

220 l = 0: RETURN

5 means there is over 75 spins of consecutive loss. It took millions of spins to go from 4 to 5 so I suspect it doesn't move very far into that 6th level but obviously these initial results are unacceptable.

Now its down to some combination of lengthening the progression & the trigger...

21 trigger, 15 progression brings it down to 4 losses...

21 trigger, 20 progression brings it down to 3 losses...

21 trigger, 22 progression brings it down to 2 losses* <=== this is to be desired

This finals at 21 trigger & 22 progression where neither can go any lower and only have 2 expected consecutive losses.

1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 4, 5, 6, 7, 8, 10, 12, 14, 17, 21, 25, 30 = 176 units

36,36,36,36,36,72,72,72,108,108,144,180,216,252,288,360,432,504,612,756,900,1080 = 6336 units

For a total bankroll of 6512 units. This is the very steepest required. We should verify how many hits on the second level actually occur... As I suspected it goes all the way to 22 full bets on the second level. Because these bets are so expensive, it makes sense to attempt to reduce these...

Of course Pales-Harry don't play it this way because like I said I tend to be one-dimensional. Also making a note to self that there is a third level that needs mapping... >.<

Ok continuing on here:

30 trigger and 23 betting range will only fail once, extending 20 bets into level 2 for a total of 43 bets

which appears as:

1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 4, 5, 6, 7, 8, 10, 12, 14, 17, 21, 25, 30,36 = 212 units

36,36,36,36,72,72,72,108,108,144,180,216,252,288,360,432,504,612,756,900 = 5256 units

A 30 trigger is ridiculous for a double street and so the goal is to bring that down.

A 30 betting range and 25 trigger will only fail once, extending 25 bets into level 2 for a total of 55 bets