lady luck or statistical propensity and diminishing probability

[I have cookies]

lady luck or statistical propensity and diminishing probability
for personal reasons I want to discuss this after reading some topics where Kelly and Snowman comment certain things that I find interesting

past results
repeats
hot numbers

one among the most rare events would be that every number out of 37 would have one show each
the magical number looks to be 5 to 6 past numbers that reduce the house edge to a certain degree - but is still negative

many different ways and sites mention this - like a similar topic at mr Ops site and a topic at GG - even wizard of odds
this is what this topic will be about

did run some actual testing a repeat after six shows and also add repeats with in the next six - the results or waves of hits are amazing winning 25 to 50 sessions in a row before a clean six with out hits occur
that is a interesting topic  

I will discuss one of Pierre Basieux method and also deal with the magical five in general with other similar methods that touch the topic

names I will refer to is what Kelly Snowman Bartow R.D Allison Basieux among others had sad about the topic

Cheers

[I have cookies]

The question is and I refrase the original from GG -    What is the most magnificent tactic to capture hot numbers ?

Here is some qoutes from GG by Kelly, Arte and at the end Snowman- just for inspiration and something to play around with.

Qoute from GG arteinvivo    Posted: 01-Jan-08 22:43    

gizmotron for what it's worth Kelly wrote this sometimes ago but unfortunately I don't understand it maybe you'll be able :

Pierre Basieux actually created a math based system call P & TP Wiederholungen (P and TP goes for Plein and Transversale plein, single number and 3 street and wiederholungen stands for repeats). Positive over 98.000 spins. Negative in the long run which he admits in the beginning of the system. Basis is, that the highest probability for a repeat is at spin 7 so sets up as a experiment a few windows in the area 6, 7 and 8th spin. The single number that you bet on must also at the same time correlate in the dozens probability for the first repeat. As I recall it....

Pre run 6 spins.

Choose the single numbers who haven`t repeated yet.

From these, pick those who stands alone in each of doz 1, 2 or 3 (being ready for the first dozen repeat too)

Bet spin 7 and 8

Short version..............


Qoute from GG Kelly Posted: 02-Jan-08 00:45    

Arte: Example of Pierres play. Pre track 6 spins: 6, 23, 2, 18, 35, 15
No repeats in those 6 spins. The highest probability for a plein hit is at spin 7 nolinks://img184.imageshack.us/img184/2782/firstrepeatprobkg4.jpg so to get a repeat, the plein number must be within those 6 numbers. For safety play, he adds an additional parameter by tracking the same number within the dozens. Looks like this:

Doz. 1 = 6, 2
Doz. 2 = 15,18, 23
Doz. 3 = 35

A repeat is "due" in the third dozen (Doz 1 and 2 has already had repeats) so the single number "35" is backed because there are 2 correlating parameters (doz and plein) calling for a repeat.

As I recall it, betting the appropriate number twice, gave the best results. But don`t hold your breath, long term negative which he admits. In the old days, I sometimes had the play running secondary while tracking a wheel, because it sometimes throws a lot of money off and is easy to track, you don`t have to book, just look at the score board.

Additionally, he recommends booking the numbers on the coloumns as well.

qoute from GG arteinvivo    Posted: 02-Jan-08 10:44    

Thanks Kelly,

I created some stats this morning to see P. Basieux's idea in action, here are some frequencies calculated from 5087 series of at least 24 spins. You'll see that the optimal time in this long series was at spin 8 and not 7. I'll do a simulator and will publish it very soon. Stay tune.

nolinks://i28.servimg.com/u/f28/11/75/07/03/basieu12.gif

What I find interesting is when I do the sum of position 7 & 8 (524+531=1055) and compare it to the other positions above 8 from 9 onward we get this sum : 2249. 2249 / 1055= 2.13 so 1 out of 3 times if we bet 2 times consecutively at spin 7 and 8 we should hit the right target if we back 6 numbers / attempt. There seems to be less dispersion than when we bet on a fixed dozen or column.

qoute from GG Kelly    Posted: 02-Jan-08 11:21    
   
Well, they are pretty close. I think you will find that a different set of samples will show the correct values over the long haul.

Pierre recommends betting twice, spin 7 and 8, for the maximum value which correlates fine with your sample.

At least that was what gave the maximum profit using Baden Baden 1995 (92.000 spins) and Bad Reichenhall 1982 (106.000 spins). Im sure other permanenzes could show different values. If the betting window 6,8 were extended to 5,9, the profit dropped dramaticly turning the the 2 permz negative.

qoute from GG snowman Posted: 01-Jan-08 14:05
note I end this with this qoute just as a general tought   
   
Arte,

I've tested something similar over many thousands of spins.
Basically I was looking at the last number hit, and the last five numbers that have hit.

It really does have an effect on the game, but not on every wheel for certain reasons.

It seems to significantly reduce the edge, but it's just not quite there, rendering an edge that's too small or one that's still negative.

You might play around with artificially weighting the last several numbers that have hit at the different intervals. For example, maybe bet the numbers that hit 3 and 4 spins prior slightly more than the numbers that just hit at spins 1 and 2.

As you test this method, you're going to find that you can't really move beyond five to seven spins back.
Five spins back appears to be the magic number of spins.

[gizmotron]

Interesting. I went back to read the original thread posted by arteinvivo. I built a sim that did a single hot number cut out after every 16 spins, that was after enough spins had begun to seek the hottest numbers that is.

It would be far more interesting to build a sim that seeks out the six most active numbers. I've learned that when a number goes really hot it hits in less than 19 spins. It does this many times before 300 spins are completed. I guess by experience that's 14 to 22 times for every hottest number in the 300 spin sequence. So the average hot number hits 18 times out of 300 spins. That averages out to every 16.66 spins per hit. If you take a law of thirds and add it to the 16.66 you get a rounded number of 23. So a progression of say 24 spins applied to 6 hottest numbers should produce interesting results. It's just me but I would deliberately target the six hottest performing numbers in a sequence of about 300 spins. Why does this guy target the coldest numbers?

[I have cookies]

Interesting.

For me i am just curios and wondering a littel about "local attractors in chaos theory" in many different ways.

[gizmotron]

For me it was in finding a logical balance point. So I have slung together a ten step progression for each of the selected 6 hottest numbers.

Bets: 1, 1, 1, 2, 2, 2, 3, 4, 5, 5,

Costs: 6, 12, 18, 30, 42, 54, 72, 96, 126, 156, ( total cost if lost = 156)

Pays: 30, 24, 18, 42, 30, 18, 36, 48, 54, 24,

All you need is any one of the six numbers to hit and you win. So I wonder if 10 steps is close to an average balance point.

[I have cookies]

Cool.

For me is about observations and see it like a wheel signature.
I like the idea using different parameters as mention above as believe there exist other options among does mention above.

Is it possible to prove that gaming decisions can be influenced by past results?

Yes to certain degree if we read what Snowman wrote above and wizard of odds in general say the same thing.
Depending on why the past five numbers would be a better option then any other.

There should be no bias towards or against any of the playable numbers !

Yes fluctuation - but does that equal hot numbers and does hot numbers equal repeats.
What is the conclusion - that all wheels witch is old, dirty and so on has a slight bias.
How come that 5 to 6 past results produce and reduce the house edge to certain degree.

All numbers should resembles a fairly even distribution during 3000 trails.

[I have cookies]

Well one parameter above is to play one number and the other option use a frame with a certain frequency of attack.
So assume we just pick the five past with no repeats and play four times - that would be a window of nine.

No progression - as this is about flat betting.

This is not a method or system - it is just my way to show a different "parameter" witch make things more positive.
Assume that we have scatter with in a high probability area - then a spreed of all does pockets will not hit once each.
Comparison with a dice that has seven sides will not show each once alternating with no repeats.
I assume the probability is around 98% it would not do so.

Now back to our sample of five - is it the first, second, third and so on that is going to repeat from our past five.

1 xx
2 xx
3
4 xxxxx
5 xxx

That is our hot local attractors using our past five hitting in does position for the sequence of attack.
So some one could start out just play three to four number out of our past five.

Below is the hit ratio.
If we would have placed all five we had to win once in 7.
Using three once in 12 and using 4 once in 9.

-4
-4
+1
+1
+1
-4
-4
+1
+1
-4
+1
+1
-4
+1
-4
+1
+1
+1
-4
-4

[I have cookies]

Here is a test playing 2 number.

Local attractors.

1
2 x
3
4
5
6 x

Testing 2 number bet for 18 trails.
Play the first two present.
Probability 1 in 3.

1 L
2 L
3 L
4 L
5 W HIT
6
7
8
9
10
11
12
13
14
15
16
17
18

- - -

1 L
2 W HIT
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18

- - -

1 L
2 L
3 L
4 L
5 L
6 L
7 L
8 L
9 L
10 L
11 L
12 L
13 W HIT
14
15
16
17
18

- - -

1 L
2 L
3 L
4 L
5 L
6 L
7 L
8 L
9 L
10 L
11 L
12 L
13 L
14 L
15 L
16 L
17 L
18 L

Local attractors.

1 X
2
3
4 X
5
6

Testing 2 number bet for 18 trails.
Play the "LAST" first two present.
Probability 1 in 3.

1 L
2 L
3 L
4 L
5 L
6 L
7 L
8 L
9 L
10 W HIT
11
12
13
14
15
16
17
18

- - -

1 L
2 L
3 L
4 L
5 L
6 L
7 W
8
9
10
11
12
13
14
15
16
17
18

- - -

1 L
2 L
3 L
4 L
5 L
6 L
7 L
8 L
9 L
10 L
11 L
12 L
13 L
14 L
15 L
16 W
17
18

- - -

1 L
2 L
3 L
4 W
5
6
7
8
9
10
11
12
13
14
15
16
17
18

- - -

1 L
2 L
3 L
4 L
5 L
6 W
7
8
9
10
11
12
13
14
15
16
17
18

- - -

1 L
2 L
3 L
4 L
5 L
6 L
7 L
8 L
9 L
10 L
11 W
12
13
14
15
16
17
18

This depend on that one side out of three will sleep for 18 times and secound last present will sleep for 18 times.

[toby]

Interesting. I went back to read the original thread posted by arteinvivo. I built a sim that did a single hot number cut out after every 16 spins, that was after enough spins had begun to seek the hottest numbers that is.

It would be far more interesting to build a sim that seeks out the six most active numbers. I've learned that when a number goes really hot it hits in less than 19 spins. It does this many times before 300 spins are completed. I guess by experience that's 14 to 22 times for every hottest number in the 300 spin sequence. So the average hot number hits 18 times out of 300 spins. That averages out to every 16.66 spins per hit. If you take a law of thirds and add it to the 16.66 you get a rounded number of 23. So a progression of say 24 spins applied to 6 hottest numbers should produce interesting results. It's just me but I would deliberately target the six hottest performing numbers in a sequence of about 300 spins. Why does this guy target the coldest numbers?

What does not win flatbetting cannot win win a progression.

[toby]

Nice analisis I have cookies, we must not fall prey of "gambler´s fallacy".

We all have studied in the past that the best scenario to chase for a repeat is from spin 7 to 10, we know thw law of the thrird.

I have cookies wrote:

"Cool.

For me is about observations and see it like a wheel signature.
I like the idea using different parameters as mention above as believe there exist other options among does mention above.

Is it possible to prove that gaming decisions can be influenced by past results?

Yes to certain degree if we read what Snowman wrote above and wizard of odds in general say the same thing.
Depending on why the past five numbers would be a better option then any other.

There should be no bias towards or against any of the playable numbers !

Yes fluctuation - but does that equal hot numbers and does hot numbers equal repeats.
What is the conclusion - that all wheels witch is old, dirty and so on has a slight bias.
How come that 5 to 6 past results produce and reduce the house edge to certain degree"

Good idea to pay attention of what Snow says.

When you analize what can happen next you are trying to predict what is unpredictable(random).

You should use tools that are not math related to apply to your analisis to easy your predictions.

BS.

[gizmotron]

When you analize what can happen next you are trying to predict what is unpredictable(random).

You should use tools that are not math related to apply to your analisis to easy your predictions.

You should start with a balanced method. In EC's it's so simple a caveman can do it. ( ON AVERAGE ) That's the math that matters, ON AVERAGE . Of course random events have variance. But ON AVERAGE a method can have a balance to it. If you flat bet two sets of dozens (24) numbers and you leave the other 13, 14 numbers exposed then you need two wins for every one loss to balance.  You can find a 50/50 balance point with regards to the payback of a progression that targets the 6 hottest numbers. I've already determined that the balance point is at 24 steps of the progression. That means you need to have balanced risk displaced all along each step of that progression. I prefer  a progression that pays me 1 for step one, 2 for step two, 3 for step three, etc...

[gizmotron]

Nathan, what an interesting idea. I can select two numbers from each of the three dozens based on the coolest six numbers recently occurring and combine it with the strongest likely dozen with a backup bet of 5 units on the second likely dozen. If I were to double your values I could bet 10 for a weakest dozen, or 10 straight up on the best ten from a secondary dozen. I would bet all 12 numbers on the best likely dozen. I could also base bet selections from the best dozen and second best dozen where I'm leaving out the cold numbers at the same time. I'm going to look into this. If the zeros are hot you can add them in and still play a dozens game. ... Just thinking out loud here. Thanks for the idea. It looks like a good one.

[I have cookies]

@toby and all ...

I just notice that using the past five does still being negative does to certain degree reduce the house edge - that is just fun and i am not so serious about it.

But i do find it interesting and sure like the idea or twist regarding P. Basieux.
Local attractors is like take andvantage of the most frequent or last and most present hits with in the high probability area.
Same as having one high probability area using visual ballistics and with scatter use the peaks or neighboors among the peaks witch also gets donations.

So if some one use the wheel layout and not the table layout and if repeats does not become a direct hit - then we get donations in the same sector of three pockets or same sector of 12 pockets witch also is donation and some kind of local attractors - there is many ways to observe statistical propensity and diminishing probability.


I like the idea but the other way around as P. Basieux mention it that if one sector is due that is the number we bet from a sequence of previous one as we have a very high probability that at least one repeat will occur.
But the option could also be for example place 4 numbers of does sectors that is most frequent and skip the sector that is due and this way not need to play 5 numbers and still get the same frequency of hits with less.

This parameters are new for me and i can see many options using does in with different variations.
I have no tools for this - i do have does for bias but this is a different topic.

To dictate that past five have influense on the game to certan degree is the same as saying that some past results matters as not being due to gamblers fallacy as its proven that it does reduce the house edge to certan degree.

Statistical propensity and diminishing probability .

[I have cookies]

Note I  see this topic pop up in different ways in many different places and can mention with out going into details that even Laurance mention it that local attractors do have potentials when he was refereeing to visual ballistic play using certain high probability areas that place certain bets with a higher value on the most frequent and last present as they do strike.
But I prefer to use the name local attractors and not hot or could numbers as it boils down to simple probability as if we have 6 pockets hitting 6 times it would be around 98,5 % that previous ones will get more donation then that all would hit once.

One simple conclusion is that if some one can find a high hit ratio then some one should not play all does number within that sector or spreed of numbers as we have the option to pin point out local attractors - that is my 2 cents.

It goes two ways same or donations around same witch many different topics deals with.
Example how Arte wrote about repeats using a step latter and certain frequency of attacks just aim for one singel number also work as aim for a sector of three as donation appears next to witch would be the main aim.

@toby as your weird ds with 7 std witch is not random with sector of 14 numbers i am sure the principal of using local attractors would dramatically reduce the high variance when does appears as down swings even when there is a postive expectation but also as laurance sad that due to the high spreed of numbers it could alternate witch would make local attractors weary appealing to use.

[I have cookies]

Just to make a simple comparison towards what I wrote above and get this topic going I will quote Arte from GG with the twist using one singel number of sector of three to show donations as local attractors with in same high probability area.

arteinvivo    Posted: 01-Jan-08 11:30    
   
System: Trojan's Ladder

1. wait for 3 numbers
2. bet on those 3 numbers for 3 spins.
3. if win, start again, if lose...

1. bet on last 4 numbers shown for 4 spins.
2. if win, start at 3 numbers again, if lose...

1. bet on last 5 numbers shown for 5 spins.
2. if win, start at 3 numbers again, if lose...

1. bet on last 6 numbers shown for 6 spins.
2. if win, start again at 3 numbers, if lose...

...increase to 7, then 8 etc.

if there is a repeat number in there amongst your numbers to bet on, wait for another spin so that you have the correct amount of unique numbers to bet on.

arteinvivo    Posted: 01-Jan-08 11:36    

Here are my instructions :
The most advanced system to play on repeat numbers.
Download link:
nolinks://nolinks13.brinkster.com/arteinvivo/systems/most_recent_repeat_wt_n
eibours.htm


The following system is a variation of the system which was sent to me by a french author in order to test the claims of a programmer who pretended having made 1,000,000 units in 10,000,000 spins.

To be frank, I was never able to replicate this result but the betting procedure is for sure one of the best in our galaxy.

As many among you have asked how this system was played, I decided to release one of my most advanced simulator on repeat numbers.

How does it work ?

THE TRACKING PROCEDURE
You must track and play the most recent repeat number.

Ex.: 11,(2),3,6,8,23,25,26,(2) < 2 is the most recent repeat number

You must play this number for A MAXIMUM OF 5 TIMES.

MY VARIATION
When you find a repeat number, add its neigbours. For example, if 0 is the most recent repeat number then add 26 & 32. This way you ALWAYS BET ON THREE NUMBERS.

For a single zero wheel we have this arrangement of numbers around the wheel:
5, 24, 16, 33, 1, 20, 14, 31, 9, 22, 18, 29, 7, 28, 12, 35, 3, 26, 0, 32, 15, 19, 4, 21, 2, 25, 17, 34, 6, 27, 13, 36, 11, 30, 8, 23, 10.

If you refer to this list then it will be easy for you to find the neigbours of any numbers.

My own NOTE.

This is not necessary apply as main strategy as this could apply with in high probability area as in local attractors with any spreed of numbers.
Also that the last three hits with last two or the last three with the last 5 to 6 two:s using same step latter would cut it with larger spreed of numbers as main target.

[I have cookies]

-