# Marigny de Grilleau system

This post is based upon gamblers fallacy and the outcome is still 50/50 no matter what the past result is. I don’t take any responsibility for any of your action reading this or if you use it in real life.

The claim.
Is all about to find a sequence where some events is overrepresented to a certain degree – when this occur you wait for the underrepresented events to show to a certain degree and try to gain at least +1 unit.

That is a clear and simple explanation and it all boils down to capture correction.

Different events.
There exist different ways to observe singels and series depending on witch method some one decide to use and measuring for a strong imbalance – I will start with cover all does and show you how to calculate and get a 3.0 STD – witch is the bench mark for a strong imbalance.

Series of two contra higher series.
The value for each event is as follows.

Singles has the value of 0
Series of two has the value of 1
Series of three has the value of 0
Series of four has the value of 1
Series of five has the value of 2
Series of six has the value of 3
Series of seven has the value of 4
And so it continues …

Here we skip singles and series of three as we aim to only capture longer series and the overrepresented events has to be series of two and the imbalance has to hit a bench mark of 3.0 STD before we can wait for the underrepresented events to show and try to gain at least +1 unit – using a specific march.

The March.
To make things easy at the beginning we can just argue that we attack after a fictive win to gain +1 with the example above – but it exist other ways.
So after a 3.0 STD we wait for a series of four (or five) to appear and play it will become a series of five (or six).
If a loss we wait for the next series of two to appear and play it will continue and get larger.
If a loss we wait for the next series of two to appear and play it will continue and get larger.
This is one attack with three attempt.

Illustration.
1
2
2 Serie 2 = 1
1
2
1
2
1
1 Serie 2 = 1
2
2 Serie 2 = 1
1
2
1
2
1
1
1 Serie 3 = 0
2
1
1 Serie 2 = 1
2
1
1 Serie 2 = 1
2
2
2
2 Serie 4 = 1
1
1 Serie 2 = 1
2
2 Serie 2 = 1
1
2
2 Serie 2 = 1
1
1 Serie 2 = 1
2
2
2
2 Serie 4 = 1
1
2
2
2 Serie 3 = 0
1
2
1
2
1
1 Serie 2 = 1
2
2 Serie 2 = 1
1
2
2 Serie 2 = 1
1
1 Serie 2 = 1
2
1
1 Serie 2 = 1 /// 3.0 STD
2
2
2
2
1 LOSS -1
1
1 WON +0
1 WON +1
1 WON
1 WON
1 WON
1 WON
2
2
2

Math and calculation:
First I will illustrate this with a simple example.
If you have 14 series of two “events” and 2 series of four “events” present as a sequence then you calculate as follows:

First you have to get the Absolute Std when you calculate.
Then you take 14 – 2 = 12

Now we want to get the statistical std so we continue with…
14 + 2 = 16

Now we take the sqr of 16 = 4

And finally we divide the Absolute Std  whit the sqr

12 sqr 4 = 3,00

The Statistical STD 3,00
Dummy chart

The Law
As every trail is independent with 37 degree of freedom – we can start measuring from any trail or position we like – that means that you can pick any window with a 3.0 std at any place during the distribution.
But note that you should apply a rule witch dictate that only a window with a 3.0 std with in the first 150 trails is valid or else it can become to late to capture change when it come to be present.

That is what we base our observations upon.

Probability:
It dictates that the waves of the distribution after a 3.0 std rapidly or slowly stop to grow and other formation as opposite effect and part of drawdowns of a 3.0 std start to appear and it comes in two different states.
One is that it stop and start to hovering at 3.0 with out any drawdown or grow stronger or you get a rapid drawdown as a direct opposite effect.
Note it can as anything also go back to back to certain degree – well nothing is due so sure it can hit a bench mark of 6.0 std – but that would mean a loss of two or four bets – just to give you a hint.

Read also our collection of 13 Best Roulette Strategies

_______________________________________________________

Marigny de Grilleau
translated from “The gain of one unit on the even money chances at Roulette and Trente et Quarante

One can hear that question in every casino everyday.
The word “new” means according to the definition “which one yet did not see“.
In this sense each day is a new day.
It is quite obvious that people asking this question do not realy mean “new” to express this natural truth.
Their questions is badly formulated and surely they mean “new” in the sense of independent.
Thus they wanted to ask whether each spin is independent of the others, the previous or following spins.
The above question should be asked as follows: ” Are all appearances and are all spins independent?” In this formulation no wordplay and no wrong interpretations are possible.

Grilleau does not hesitate with a clear answer: “No, neither the appearances nor the spins can be independent, because everyone of them is a part of the whole. This whole is arranged and limited in all its movements and is subject to precise laws.”

Each spin, while the ball turns in the wheel, carries in itself a certain quantity of independence and a certain quantity of dependence.

The independence results from the following:
every time the dealer rolls the ball, it is faced with 18 red and 18 black, 18 even and 18 odd as well as 18 high and 18 low pockets. Therefore the ball has the same chance to fall in one of the 36 pockets (we do not consider zero or doublezero this time) since each pocket indicates Red or Black, Even or Odd, High or Low at the same time.

The dependence results from
1. the Law of Deviation (Ecart), STD
2. the Law of Balance (Equilibrium) and
3. the law of the distribution of appearances into different accumulations or clusters and isolated units

Thus the mathematical truth of the independence of the spins is constantly in conflict with the statistic truth of the dependence of the spins.

If between two equivalent appearances none, or only a very small deviation exists, the independence of the two appearances remains retained in their fight against each other.
But if the statistic deviation reaches a certain size, the size of this deviation more or less limits the independece of these appearances and spins.
In this instant the dependence of the appearances on the laws of nature demands again its right, by limiting its freedom for deviation within the statistic average values, of which these never can free itself.

In our opinion neither a single spin nor an appearance can be independ in a roulette permanence of a certain length, for example within 1024 spins.
The dependence of the spins which are affected by chance due to exactly defined laws, is a fact, which the usual gambler does not understand without difficulty. And because of this difficulty the gamblers and also the mathematicians believe in the independence of roulette spins.
In reality each spin and each appearance has its necessary and mandatory function in the whole of a roulette permanence.
Chance does not exist there, because all effects have their visible or hidden causes.

The dependence of the spins on the laws of nature becomes obvious, if we analyze a roulette permanence and classify the developed appearances.
However we do not succeed in each case in determining this dependence, which must be present for all spins, if only small deviations occur, which do not exceed the average statistical Ecart of 1.
We only succeed then, if we determine the partial return to equilibrium after very strong deviations greater than a statistical Ecart of 3.
The roulette ball cannot extract itself from the laws of nature.
These laws force it into the pocket, into which it must fall, so that it can perform the necessary function, which it has to, in the statistic harmony of the whole permanence – like a note in a score.
Chance can let many obvious, strange features develop before our eyes. But nevertheless, statistically seen, chance can not repeat these individual strange things too frequently, like for example a series of 25, which needs approximately 34 million spins to develop once.

It is worth to note  that with the advance of online roulette, such difficult to play strategies are more playable at an online casino.
______________________________________________________
Basic list.
This how you can measuring imbalance and witch I will write about.

1) Series of two contra larger series. [DONE]
2) Series of three contra larger series.
3) Singels contra larger series.
4) Series of two singels contra larger series of singels.
5) Series of three singels contra larger series of singels.
6) Singels contra series.
7) Series contra singels.

Staking plan – Flat betting or progression.

Singels contra larger series.
Singels has the value of 1
Series of two has the value of 0
Series of three has the value of 1
Series of four has the value of 2
Series of five has the value of 3
Series of six has the value of 4
And so it continues …

The Twist
You can combine this with singels contra series.
Singles has the value of 1
Series has the value of 1 no matter length.

The Perfect State
My own development of Marigny where you only allow a state with singels “events” present – witch means no series is allow to chop after each other with in the sequence of 3.00 STD.
This give us the opportunity to get both larger series and series to chop as a significant change due to a strong imbalance.

You calculate that the singels are overrepresented and that the series are underrepresented and they should reach a bench mark of 3.00 STD.
Singels contra larger series is the values I mention above – the twist is that the underrepresented series are only allow to be present as singels events ( not two series that chop after each other ) and the singels contra series has also have to reach a bench mark of 3.00 STD.

Now you have two states into one with the rule above with only singles present and series being underrepresented.

The March.
There exist variations of what kind of march to use – this one capture rapid draw-downs after indication towards present change.

If you revive a series of four you attack once it will become a series of five.
If a loss you attack the first series of two that appears and that it will grow.
Total of two attempts to gain +1 Unit or a loss of -2 Units.

If you revive two series to chop after each other you attack once it will become three series in a row.
If a loss you attack after the next first series appears and that they will chop.
Total of two attempts to gain +1 Unit or a loss of -2 Units.

If you get a mix of both you just bet accordingly towards has been mention above.
I will show and illustrate this with some visual samples.

The Law.
As every trail is independent with 37 degree of freedom – we can start measuring from any trail or position we like – that means that you can pick any window with a 3.0 std at any place during the distribution.
That is what we base our observations upon.

The Probability.
It dictates that the waves of the distribution after a 3.0 std rapidly or slowly stop to grow and other formation as opposite effect and part of draw-downs of a 3.0 std – start to appear and it comes in two different states.
Note, it can as anything also go back to back to certain degree – well nothing is due so sure it can hit a bench mark of 6.0 std – but that would mean a loss of two or four bets – just to give you a hint.

The Rules.
The playing modell using strict rules as follows.
You have to find a window within the first 100 or 150 trails with a 3.00 STD – depending on witch march you use to have time to capture drawdowns that not has to come rapidly as my example above.

The window of The Perfect State as the example above has with both singels and series be singels events and both has to be measuring to hit at least 3.00 std within minimum 16 events or at most 50 events – probability indicates less and tighter is better.

Team Play.
I assume the tracking of three even money postions during a short period of time every day would be 3×100 to 150 = 300
More wheels more opportunities.

Staking plan.
Flat betting or humble progression.
Well i have run some tests and reach 3.0 STD flat betting and others has done the same – but its up to you to run your own simulations to verify or skip it – as i don’t force any one to use it or study Marigny.

The House Edge.
You have to have a La Partage Rule and put a % on zero.
If you would flat betting you should use a staking plan like 731 where you lower your bets after +1 and have a new free ride for two attempts to capitalize on larger strings of wins.
After +1 you can not lose as two bets using 3 and 3 equals 6 and 1 and 1 equals 2.

Three states
If you use singles and series as two opposite sides of a coin 1/2 then we will have three states.
If singels chop one side get stronger and grow and if series chop the other side get stronger and grow – simple.
If none grow and get stronger they hovering with one of each present.

Definition of correction – my opinion.
If I measuring singles and only allow single series being present among does – then if they stop to get stronger and grow – then there is only two things that can happen.
The singles and the series can start to chop once each – hovering at zero point where the STD does not get weaker or stronger – this part is part of correction and it can come with a small, medium or large wave as anything else. When series start to chop as opposite effect we have a raw drawdown of correction and the STD get weaker.

The three states – waves – can come in any combination – but one common observation after many samples of 100 000s is that the state of hovering and raw draw-downs comes together witch is our tendency towards to capture does and gain at least +1 unit.

March
One thing I like is when and why some one would place hes first bet to capture a tendency towards a small, medium or large correction.
There exist three ways you can use as tendency.
One is to capture a small quick draw-down after 3.0 or 3.5 STD as example above using singels one could pick the first series that appears after a strong imbalance to gain +1 – if a loss then some one would wait for a significant change witch would be two series that chop or that at least three or four then wait for all formation of two to appear and attack.
One other way is the reverse of above where you wait for a significant change first and make one attack and if a loss then you attack after the first singels series appear.
And last some one following one of does two principals but does not place any bet after obtain one fictive win.
Other ways is to wait for a significant change and attack and if a loss you have to get a more significant change then the previous one before you are allow to attack again.
________________

## TEST

Random org 2009 03 31
Sample 10.000 trails.

TPS = The Perfect State / Hybrid [My own development based upon Marigny]

Flat betting with 4 attempts.

Won 16
Loss 0

R/B

TPS 3.0 STD Result +1
TPS 3.0 STD Result +1
TPS 3.0 STD Result +1
TPS 3.0 STD Result +1
TPS 3.0 STD Result +1

E/O

TPS 3.0 STD Result +1
TPS 3.0 STD Result +1
TPS 3.0 STD Result +1

H/L

TPS 3.0 STD Result +1
TPS 3.0 STD Result +1
TPS 3.0 STD Result +1
TPS 3.0 STD Result +1
TPS 3.0 STD Result +1
TPS 3.0 STD Result +1
TPS 3.0 STD Result +1
TPS 3.0 STD Result +1

Next I will show how to run the file and practice the march and verify what I did and write down the rules.

(Try to run FTL or DBL and show me the same results…)
__________________

14 singels has the value 14 as each singels has the value of 1
A series of 4 has the value of 2
A series of 5 has the value of 3

14/7 = 1.52 STD

Singels contra series.

14 singels has the value 14 as each singels has the value of 1
A series has the value 1 and you have three series witch is a value of 3.

14/3 = 2.66 STD
_______________________

Flat betting.

The number of won units is to be divided by the root from the placed bets.

0.5 Keep looking
1.0 Keep looking
1.5 Hint
2.0 Interesting
2.5 Good /
3.0 Very good
3.5 Super
4.0 Dream on
4.5 You wish
5.0 Forget it
5.5 Never in an life time
6.0 The end of roulette as we know it
________________________

We are using the standard deviation in regard to the number of wins in the last X spins. So if you only get 5 reds in the last 20 spins, the number of standard deviations from the mean (or the so-called “z-score”) is about -2 with respect to red. The idea is that the number of standard deviations from the mean won’t swing too far from +/- 3.0. According to the theory this covers 99.7% of the area under the bell curve.

Like I said, I’m sceptical that it will work, but this Marigny de Grilleau character seems to have had a certain reputation. He may have been a crackpot, but let’s find out.
_____________________