Baccarat Marigny De Grilleau - Main Topic -

I will start the main topic with Marigny De Grilleua where all hes work will be presented.
Now for the unsure or the punters who find Marigny De Grilleau fuzzy I can show one simple experiment that will describe the power behind math and probability-based solutions from this great person.

How to predict larger series to show in the future!

1. Wat for seven singles and any amount of series of two as one sequence that ends with a series of three.
2. Obesrve is the series of three become a series of four, if yes, end of the sequence and you made a direct win.
3. If you don't get a series of three you watch for the series of three to show in the future until you have seven singles and no series of three, then you have a sequence beyond 3 STDV against you.
4. During your observations you will see the likelihood of at least one or two series of three minimum after a sequence of seven or more singles in a row with an unknown amount of series of two and where it ends with a series of three.
Then the probability dictates that at least a series of three or two series of three will show versus seven more singles.

Observations you need to do is to watch how many winning sequences you get versus 3+ STDV sequences.
And come up with a staking plan.
I would suggest aiming for one series of four - direct win - or one series of three after that.
Don't aim for two series of three, even if the likelihood is high to reach that before seven singles shows.
If you have a staking plan in several layers you might want to aim for both.

Here is the playing model and the values for each event.

Singles have a value of 1
The series of two has a value of 0
Series of three has the value of 1
Series of four has the value of 2
A series of five has the value of 3
And so it continues increasing...

The reasoning behind the selection.
14 singles and 2 series of three is exactly 3 STDV.
So when you have half of that, seven singles (at least) and your expectation is not to get 3+ STDV.
And you wait for one series of three to show, then one more will get you at exactly 3 STDV.
Two more series of three will give you regression and below 3 STDV.
But if you get seven singles and one series of three and another seven singles you are beyond 3+ STDV.
It's like collecting windows or sequences based on probability and math and taking action based on that.

Cheers Patrik From Sweden

EDIT - I got real baccarat results from Ketih (cps10) and will make some observations and post some results.

 

Here is one illustration of one sequence:

pbbppbbpbppbpbppbpbbb - pbbppbppppbpbbbbppbpbbbpb

8 singles and one series of three and after that I let the sequence run until I got seven singles to see how many series i got.
2 series of four and 1 series of three before we hit seven singles - a clear sign of regression and how easy it is to predict larger series in the future.
2 singles before the first series show 4 singles before the second series show and 6 singles before the third series show.

Cheer Patrik From Sweden

 

Here are 10 more sessions to show the LW-Registry
I want to highlight that when you get a direct win with a series of three becoming four, you have the value of 2 equal to two series of three.
So all sessions that end up at exactly 3 STDV or below is a winner and all the losing sessions is beyond 3+ STDV

LWLWLWW
W
W
LLLWW
W
W
W
WLWLWW
W
W

Not bad for a math and probability-based approach.

pbbpbppbppbpbpbppbppbpbbppp - bppbbpppppbpbpppbpb - LWLWLWW

pbpbppbppbppbpbpbbbbp – W

bppbbpbpbpbpbppbpppppb – W

pbpbbpbppbpbpbppbbppbpbbb - pbpbbbbbbppbpbbbbbbpbp – LLLWW

pbbppbpbbpbpbbpbppppb – W

bpbpbpbpbppbppppb – W

pbpbpbppbppbbbbbp – W

pbpbpbpbbppp - bbppbbbpbbpppbpbbppbpppbbbbbbpppbbbbppbbbppbbpppppbppbpb – WLWLWW

bpbpbbpbbpbppppppb – W

bpbpbppbpbbppbpbbbbbp - W

Cheers

 

Hi Patrik

Could you explain where your bets are made to produce this registry? Maybe just that first run at least of LWLWLWW?

Cheers!

 

p
b
b
p
b
p
p
b
p
p
b
p
b
p
b
p
p
b
p
p
b
p
b
b
p
p
p
-
b
p L
p W
b L
b W
p L
p W
p W
p
p
b
p
b
p
p
p
b
p
b
-
LWLWLWW

A series of two makes you break even and restart and a series of three or above produces two wins in a row.
I will test this with the Cinga A Ling variant with a 4 level progression of eight steps each.
Will post the staking plan.

So when you get singles and an unknown amount of series of two as a selection window.
You wait for a series of three to show as the first indication of regression.
That has a value of 1.
Then we bet that this series of three will become four in a row and if we win the session is over.
If we lose we aim for one more series of three with a value of 1.
That is 14 versus 2 and exactly 3.0 STDV.
But the nice thing about this is that most of the time we will get regression and more series of three to show as you can see above and hit below 3.0 STDV.

Cheers

 

So your trigger is the series of 3 and series of 3 only? After a single you are betting another series of 3 is right around the corner and you stop at 3 regardless of if the series could go 4 or more?

 

1) wait for a selection window to end with a series of three
2) bet once that the series of three will become four in a row, if you win, stop, if you lose continue until you hit a series of three and win two in a row.
3) if you get seven singles and no series of three you have a window of 3+ STDV and you lose that session.

Hope that helps

Cheers

 

When betting for a series of four you are basically FTL once the original series of three has ended.

For instance, pppb <- bet for a series of 4 with b

If unsuccessful, and p comes up, you are betting for the next series of 3? So you bet for p to go to three and then stop. At any time you lose you continue to bet for the series of three again unless you lose to 7 singles.

 

Not correct
You get for example seven singles and one serie of three

pbpbbppbpbbpb ppp bet p

So you bet that the first series of three will become four
And after that FTL until two wins or seven singles

Hope that helps

 

It makes perfect sense to me now.
Would it make sense to bet for chops using that same logic? Looking forward to the Ching A Ling progression applied to it.

How I play I would have netted 10.45 units

 

Yes Keith it make perfect sense
Because a series of four has a value of 2
And then that would been the same as getting two series of three as they ave the value of 1

So both is part of regression

Cheers

 

I want to show another experiment dividing bets into groups of three.
Marigny De Grilleau also states that in certain situations should not start to attack until we have a +1 indication of regression.

So a series of three is +1 value and if we start betting once after a series of three and stop, win or lose.
Up to three attempts as that is the minimum expectation of the regression with this selection.


pbbpbppbppbpbpbppbppbpbb (PPP) - bppbb (ppppp) bpbpppbpb - LW

pbpbppbppbppbpbp (bbbb) p – W

bppbbpbpbpbpbppb (ppppp) b – W

pbpbbpbppbpbpbppbbppbp (bbb) - pbp (bbbbbb) ppbpbbbbbbpbp – LW

pbbppbpbbpbpbbpbppppb – W

bpbpbpbpbppbppppb – W

pbpbpbppbppbbbbbp – W

pbpbpbpbb (ppp) - bbpp (bbb) pbb (ppp) bpbbppbpppbbbbbbpppbbbbppbbbppbbpppppbppbpb – LLL

bpbpbbpbbpbppppppb – W

bpbpbppbpbbppbpbbbbbp - W

Here we would aim for one win and use a staking play with several layers.
This is experimental so is important that each one observe and get own statistics.

Cheers

 

Here is a short sample of an LW registry that looks promising.
I rather base my selection on math and probability than use a series of two and bet three times on a rolling basis.
Where regression after imbalance manifests as you can see below.

LW
LLW
LLW
LLL
LW
LLW
W
LW
LW
W
LW
LLW
LW
LW
W
W
LLL
W
W
LW
W
LLW
W
W
W
LW
W
W
W
W
W
W
LW
W

Cheers

 

Again the initial bet would be after 7 singles and then attempt extend a series of 3 to a series of 4. Stop after 1 bet win or lose, then wait for another series of 3, quit after three such losses.

And after a win, you reset waiting for 7 singles again.

 

The first scenario I get +3.80. Flat bet was all I would need there.

The long WL registry: +13 less commissions and high bet = 7 units

 

Interesting. My understanding is that I have to wait for 7 singles, disregard the series of two, and wait for a series of three and bet the three goes four. This gives a StD of 3.0 or less.

So I looked at the last bunch of shoes I happened to take pictures of. Where I play there are four shoes continuously going 24/7. So I just picked the last 4 shoes from my data and continue on as one data stream.

Looks like there's W W then on the 4th Blue shoe FTL stop after 2 LIAR. LWLWWLL. Then on the 6th Blue shoe LWWLL using FTL And stopping at a single.

The stream on the 4th shoe after B 7 is
11123111212111132311111 ( Those last 5 singles continue to the next shoe. But unfortunately, it only goes 6 singles before 3 hits.

For what it's worth, there were 7 wins and 7 losses. Theoretically, shouldn't the strike rate be significantly higher than the expected value?

Maybe Patrik or Keith will check it.

Cheers

 

Theoretically. I will take a look at this as well.

 

Jimske it would be much more easy if you wrote down each sequence as B11212111122211311223131221

I on to one experiment to get a load of data RX has a pattern recognition system coded in the library.
I set it to seven R/B and let it run until the opposite show and then check the next seven results.
Did a small test with 10,000 trials and tested a couple of sequences where you get the majority of 2 to 3 opposites hitting on a regular basis.

That is the same as using the Law Of Series and Marigny Playing Modell - it just speeds up things to see the average hit or strike ratio among sequences.
Before the 3+ STDV shows.

Getting back with more results.

Cheers

 

I did that above for one example. Starts B 7. I just figured it was easy to count the ones for the rest of this small sample.

Cheers

 

The stream on the 4th shoe after B 7 is
11123111212111132311111 ( Those last 5 singles continue to the next shoe. But unfortunately, it only goes 6 singles before 3 hits.

111231112121111323 - LWLWW

That is the LW-Registry for that sequence and who cares if the last events are 5 singles when you already hit regression?
As you say wait for the first series of three and bet once, a win stop, if not continue until the first series of three and stop.

Cheers