I testing to win 15 to 20% out of 100 using regression strategy ...
Where you some times hit long strings of winning strikes or just cut Short and keep a small profit.

1,5
1,5
1,5
1,5
2.0
1,5
1,0
1,0
1,5
1,5
1,5
4,0
1,5
-9
.
9 - 21,5 = plus 12,5
.
1,5
1,5
3,0
2,0
2,0
3,0
1,5
3,0
-11
.
11 - 17,5 = plus 6,5
.
1,5
1,5
1,5
1,5
1,5
1,5
1,5
4.0
1,5
1,0
1,5
1,5
-10,5
.
10,5 - 20 = plus 9,5
.
3,0
1,5
1,0
1,5
1,5
4,5
-9
.
9 -13 = plus 4
.
3,0
1,5
3,5
1,5
3,0
1,5
1,5
1,5
1,5
1,5
3,0
1,5
-10
.
10 - 24,5 = plus 14,5
.
1,5
1,5
1,5
1,5
1,5
1,5
1,5
1,5
-12
.
12 - 12 = plus 0
.
1,5
1,5
1,5
.
10,5 - 4,5 = minus 6
.

All this is thanks to John Patrick's metodology ...
Total plus 47 units playing 69 sessions ...

 
Now i test one other bet selection and change the overall win targets.
If not a direct win then break even.
If a win push for more, but keep a profit, regression strategy.

Pass 300 trails with testing the 20/100 flaw, that means you just stop at a win target with 20% with out push and pull.
That means you have to win 5 times in a row for each 100, that is the flaw.

We gain more as we catch strikes.

24 sessions and plus 375

45
15
0
15
30
15
0
15
15
0
15
15
45
15
15
45
0
15
0
30
15
15
0
0

 
I will continue with more testing, is just hard to make your mind up when it boils down to how to play ...
I think i will keep the idea of winning or break even or take a loss around 100 ...
Still using the regression strategy ...

 
First 102 sessions using Regression ... with small win target and let it ride ...
Rules:
Flat betting with higer unit value.
At a win lower the bet.
When a win push and pull.
.
Plus 84,5 units
Minus 14 units
Net Gain 70,5 units
.
1,5
1,5
1,5
1,5
2.0
1,5
1,0
1,0
1,5
1,5
1,5
4,0
1,5
-9
.
9 - 21,5 = plus 12,5
.
1,5
1,5
3,0
2,0
2,0
3,0
1,5
3,0
-11
.
11 - 17,5 = plus 6,5
.
1,5
1,5
1,5
1,5
1,5
1,5
1,5
4.0
1,5
1,0
1,5
1,5
-10,5
.
10,5 - 20 = plus 9,5
.
3,0
1,5
1,0
1,5
1,5
4,5
-9
.
9 -13 = plus 4
.
3,0
1,5
3,5
1,5
3,0
1,5
1,5
1,5
1,5
1,5
3,0
1,5
-10
.
10 - 24,5 = plus 14,5
.
1,5
1,5
1,5
1,5
1,5
1,5
1,5
1,5
-12
.
12 - 12 = plus 0
.
1,5
1,5
1,5
.
10,5 - 4,5 = minus 6
.
1,5
1,5
4,5
1,5
1,5
1,5
1,5
1,5
1,5
1,5
4,5
-12
.
12 - 22,5 = plus 10,5
.
1,5
1,5
1,5
1,5
1,5
4,5
4,5
1,5
1,5
1,5
1,5
1,5
3,0
3,0
3,5
1,5
1,0
-9
.
9 - 36 = plus 27
.
1,0
-9
.
9 - 1 = minus 8
.

Now i will compare that with a different rula and test for around 100 sessions.
That would be win or break even, just to see if we could stretch thing more into the postive ....

 
.
It would nice if i could find a selection that let me ride out more dominance, then the regression would really kick in ...

What unit values do you use for your regression bet?

Quote from: MarignyGrilleau on March 07, 2013, 02:32:57 PM
What unit values do you use for your regression bet?

I Place a total of four bets.
2223
If i win i lower my bet.
.
With a real regression plan you would bet 20 10 5 10 15 20 25 30 ... and so on Up and Pull ...
I test to bet 20 and then 5 10 5 10 5 10 5 10 5 10 to capitalize on longer strikes ...
You win when distribution strikes and when it get choppy and it it end up between does two you lose ... depending on how you use your bet selection ...

 
Lets look at the hitting and return ... as you can see to aggressive up and pull destroy winning to keep and hold a profit ...
When 5 10 5 10 5 10 capitalize and hold profits much better ....
.

W 20
L 10  10
.
W 20
L 5  15
.
W 20
W 10
L 5  25
.
W 20
W 5
L 10  15
.
W 20
W 10
W 5
L 10  25
.
W 20
W 5
W 10
L 5  35
.
W 20
W 10
W 5
W 10
L 15  30
.
W 20
W 5
W 10
W 5
L 10  30
.
W 20
W 10
W 5
W 10
W 15
L 20  40
.
W 20
W 5
W 10
W 5
W 10
L 5  45

 
Ok i been doing some further testing and Think will run 20 10 5 10 5 10 5 10 as regression and up and pull ... it capitalize more smooth and hold profits much better.
If loses chop i will aim to break even.
I will run Another 100 sessions ...