EC, No Trolling

[Robeenhuut]

Hello Alberto

Generally if d method works you dont need LW registry 2 start betting.  Waiting 4 few virtual loses gives you only a false sense of more security but in reality only reduces your loses and does not affect your strike rate. I just dont see why you would need that especially in trend oriented method 

Regards

[albertojonas]

It is not everything about trending.... You got it sideways.
We want to prevent 9 trails to become 27 trails; Cycle ends as soon as you hit a W and then the bet selection changes.
It is random Vs random and we do not try to catch any trend here.
Filtering LW registry off course reduces number of bets, but intention is reduce overall Losses and take advantage of diminished volatility, thus more equilibrium.
Find me 500 spins without a serie of 8 in any EC .... (7L in a row is what was found in the example so far)
(I have seen more)
So just make a comparison with other Bet selection method. I know Mathematically they are supposed to be all the same, Because they have the same odds.


Thanks for feedback and
Cheers

[Bayes]

Zeros were accounted in play not ignored. They counted as a Loss...
If one puts the LW registry into P&F charting results are even more impressive =)

Cool. 

According to theory, (the so-called law of series) not only should there be half as many series of RR as R, half as many RRR as RR etc (and the same for Black) but this pattern should apply to ANY W/L registry, so half as many WW as W, half as many WWW as WW (and the same for the L series). 

If the theory is correct (and mathematically it is) then there's no way that any bet selection could be better than any other - by symmetry, you will get exactly the same number of streaks (W/L) of the same length on both sides so it follows that no bet selection can make a long term profit flat betting.

However, what I've found is that the way these streaks are distributed can vary, so that some bet selections really do result in lower draw-downs (this is the case for even simple DBL, compared to betting one side continuously).

It would be interesting to do a simulation to see what would happen if the mathematically expected perfect distribution of streaks didn't occur long-term - what would be the advantage? For example, if I was to remove all losing streaks more than say 8, would that be enough to give you a flat-bet profit?, or you could maybe remove a fraction of them and see what the result was. Then you could work on ways to actually eliminate those streaks (or perhaps just cancel them out using some kind of progression).

[iggiv]

who could explain me plainly what it is all about? thanx

[Bettor 27]

Red


L
L
W
W
W
L
W
W
L
W
L
W
W
W
L
L
L
L
L
W
W
L
W
L
W
L
L
W
L
W
L
W
L
L
L
W
L
L
L
L
L
W
L
L
W
L
W
L
W
W
L
L
W
W
W
L
W
W
L
W
L
W
L
W
W
W
L
W
L
L
W
L
L
L
W
W
W
W
L
W
L
W
W
L
L
W
L
W
L
L


WINS = 43
LOSSES = 47


Filtered LW registry (Attack once After a L): Wins =21 / Losses =13

Hi Albert,

When I filter the LW registy (attack once After a L) I get Wins =17 / Losses =11

How do you get W=21 / L =13?

L   
L = L
W   
W   
W   
L   
W = W
W   
L   
W = W
L   
W = W
W   
W   
L   
L = L
L   
L   
L   
W   
W   
L   
W = W
L   
W = W
L   
L = L
W   
L   
W = W
L   
W = W
L   
L = L
L   
W   
L   
L = L
L   
L   
L   
W   
L   
L = L
W   
L   
W = W
L   
W = W
W   
L   
L = L
W   
W   
W   
L   
W = W
W   
L   
W = W
L   
W = W
L   
W = W
W   
W   
L   
W = W
L   
L = L
W   
L   
L = L
L   
W   
W   
W   
W   
L   
W = W
L   
W = W
W   
L   
L = L
W   
L   
W = W
L   
L = L

[TwoCatSam]

OK, I'll expose myself........

I have no idea how to read the chart except for the red/black thing.   Anyone care to clue me in?

Sam

[albertojonas]

Hi Albert,

When I filter the LW registy (attack once After a L) I get Wins =17 / Losses =11

How do you get W=21 / L =13?

L   
L = L
W   
W   
W   
L   
W = W
W   
L   
W = W
L   
W = W
W   
W   
L   
L = L
L   
L  =L
L   
W  =W
W   
L   
W = W
L   
W = W
L   
L = L
W   
L   
W = W
L   
W = W
L   
L = L
L   
W   = W

and so on...

[6th-sense]

Hi albert could i ask a couple of questions as i,m pretty interested in this stuff and my comp won,t open the file....am i right in thinking your using say a random 250 list of results as your betting guide ? I understand the betting registry fine...

[GLC]

Nice one Alberto,


It's good to see you sharing some of your expertise with us again.


I've been testing this with different bet selection methods with equally good if not better results.


You know how much I love even chance methods.


The strike rate is so good, any bet method will work.


GLC

[albertojonas]

Nice one Alberto,


It's good to see you sharing some of your expertise with us again.


I've been testing this with different bet selection methods with equally good if not better results.


You know how much I love even chance methods.


The strike rate is so good, any bet method will work.


GLC

I am just glad you like it =)

[Bettor 27]

As i mentioned before i used to play this with great success.
Ego also suggests Humble progression or Fibbo.
I like to filter LW registry...

Thanks AJ - can you elaborate further regarding how you used to play this?

Regards

B27

[GLC]

Guys,  I'm trying to understand why this works.


We're betting Red.  Ignore zero


If we have 2000 spins, half of the spins will be Red and half will be Black.



After each Red will be 500 Reds or streaks and 500 Blacks or chops.


If we only bet for a chop after a chop, we will have a potential of 250 possible wins.  Losses could be less because we are tracking through streaks which can last multiple lengths.
In other words chops are finite, restricted to lengths of 1 whereas streaks are infinite not being restricted to any length.


Even though I'm not clear on exactly how this works out mathematically, it appears to work logically.


Am I on the right path?
Am I overlooking something?
Are we just being lucky so far?
If this works, shouldn't it work for both Red and Black or L's and W's in the same sequence of numbers?





GLC

[Robeenhuut]

Guys,  I'm trying to understand why this works.


We're betting Red.  Ignore zero


If we have 2000 spins, half of the spins will be Red and half will be Black.



After each Red will be 500 Reds or streaks and 500 Blacks or chops.


If we only bet for a chop after a chop, we will have a potential of 250 possible wins.  Losses could be less because we are tracking through streaks which can last multiple lengths.
In other words chops are finite, restricted to lengths of 1 whereas streaks are infinite not being restricted to any length.


Even though I'm not clear on exactly how this works out mathematically, it appears to work logically.


Am I on the right path?
Am I overlooking something?
Are we just being lucky so far?
If this works, shouldn't it work for both Red and Black or L's and W's in the same sequence of numbers?





GLC

Its rather puzzling George. 2 me its still sort of hit n miss approach. I would rather back up d color so play FTL style using LW registry.  AJ plays differently betting on chops 2 continue but not continuously  so u dont get wiped out by long series of repeats.  Both events - chops n series have of coz d same chance of happening  but if u look at d board u immediately register first series of repeats.... But its flat betting so it wont hurt much f u wrong 

Regards

[albertojonas]

The hardest point to get to apply this with success is a betting selection that produces a Loss / Win registry
with the lowest swings achievable. We all know it is hard to win flat betting Ec's.
With a moderated Bankroll, anything between 6 to 10 Losses in a row is nice to play, filtering the LW registry.


Flat
Humble Fibonacci  progression
Guetting progression


About the progressions, George is The Best expert on the matter, and we know some are more adequate than others relatively to certain LW distributions.


================================================


The success to this is the logic within the bet selection.
What is suggested here is "double filtering".
... If the bet selection suggests that you play virtually until you achieve 2 losses in a row and then attack the game continuously and stop on a win, it constitutes already a form of primary filtering.
Then if you re-filter your "pseudo-real" results, your strike rate should be increased somehow.


The roulette spins endlessly but you choose when to play.


The only thing one can develop to make this even better is in the line of Bayes thinking.
Track a strong ecart and bet for correction after an indication.

[Tomla021]

Sure looks good on some of my WL tabulations--very simple also