Independent Trials / Bet Selections / Some Thoughts

[Kon-Fu-Sed]

Hi all,


Regarding roulette in general and BET SELECTION METHODS in particular:

I have totally given up on "independent trials"!


But then; what IS "independent trials" - in this context?

Suppose that you are going to test a bet-selection.
You grab some spins from somewhere and begin your test.

How do you use those spins?
Most of us simply use the results as they come:

...9 - 2 - 28 - 32 - 15 - 3 - 32 - 21 - 31... (some Wiesbaden spins)

This is what I call "independent trials". OR rather: we USE them as such.
We don't care about anything but the number-value (incl Red or Column or whatever - it's really all about the number-value categorized in some way).

In this context: What does the "2" and the "28" have in common to qualify them both to be used, in sequence, in a test?
The same question for the second "32" and the "21"...
That they came in sequence at the same table? Does that alone qualify them?

But ARE the results independent of each-other or not?
Well; of course you could apply lots of different math-approaches (incl numerology and such) and claim that it makes them dependent in that context.
You can sort them in columns and rows by some rules, and claim that it makes them dependent in that context.
But it's still looking for number-values...

And number-values cannot be connected in a true sense... Or?

Suppose you are testing a selection method that is number-value based...
At some moment in the test, your selection-method would benefit from a hit on "15".
Beside number "15" on the wheel, you have "32" and "19".
If "32" hits instead of "15" - how would your selection method react?
How are those two related - number-wise - so there can be any conclusion or bet-selection at all when "32" hits?

Through the thousands upon thousands of tests and statistical studies I've performed I haven't found ONE number-value-oriented method to work to my satisfaction.
(Profit in both short- and long-time )

I simply can no longer believe in the way we use spin-results; like chains of independent trials.

So I give it up!
Completely.

...








So I will focus on DEPENDENT TRIALS.




Let's look at the above spin-sequence again:
...9 - 2 - 28 - 32 - 15 - 3 - 32 - 21 - 31...

And I asked: What does the "2" and the "28" have in common to qualify them both to be used, in sequence, in a test?
The same question for the second "32" and the "21"...


Correct anwer (IMHO): NOTHING! They shouldn't be used together!


First of all; Wiesbaden ALTERNATE the directions of the rotor and the ball.

This means that there is a huge difference in how the ball is thrown, from spin to spin.
The "2" and the "28" were not results of similar conditions as the rotor (and the ball) was spun in different directions.
The same for the second "32" and the "21".


Second: There were DEALER-CHANGES in both cases.

This means that one croupier throw the result "2" when the rotor was going in one direction.
Then, another croupier had the "28"-result, when the wheel was going in the opposite direction.
(So what do they have in common?)

And later on the same croupier throw "32" with the rotor going in one direction - the same as the first, actually.
Then there was a dealer-change and the "21" was the first result of the new croupier - with the rotor going in the opposite direction...
(So what do they have in common?)


That's why I asked:
What does the "2" and the "28" have in common to qualify them both to be used, in sequence, in a test?
The same question for the second "32" and the "21"...

Really: WHY should the "2" and "28" be used together and in sequence?

They are not connected in any way - are they?
In my opinion, they are used as independent trials if used together and in sequence.







In my (rather new ) view, we have to deal with the results as if they are "dependent trials".
Because that's what I think they are: Dependent - PHYSICALLY - of at least the rotor/ball directions and the dealer-changes.

We can also add table-tilt, grease on the ball or ball-track... a lot, to be short.
But that can't be seen in a Wiesbaden list...
One has to be there... Unfortunately.


What we DO know and CAN see are, however, some good things...

We know that for the first spin of the day, the rotor is spun clock-wise.
This is good info, as we now can use the results depending on their respective directions.

We can see in the lists when there's a dealer-change.
This is good info, as we know when the results change depending on another croupier.

We also know that the Wiesbaden croupiers (as many in Europe) tries to release the ball when then last hitting number passes the hand.
This is good info, as we have a release-number to use as a measuring point.


But how can that knowledge be used? (Primary to develop a bet-selection method)
How can we see if there's anything to it?
Can Wiesbaden results be used?

I have split all Wiesbaden files I have into CW and CC (directions) sequences for each croupier but unfortunately I couldn't really find anything.
It's not enough data to statistically be secured. Some 1900 days, each having like 25 dealer-changes.
Many of the shifts worked for a very limited number of spins, also.
The number of occurrences for each event were simply too small to make conclusions from.
Also adding similar sequences or parts of sequences gave too small results; maybe up to 500-600 - there are some 40,000 dealers...

In short; nothing showed consistent positive results...









So what to do?

Give it up?


No, no, no!

But I have to find ideas of what to look for.
So I've written a tracker to help me.

It's a very simple program that is only showing what it's tracking - no bet-suggestions or anything.
(It's version 0.001 )

For example, this is the result of the first croupier at Wiesbaden Table #3, 01.02.2007.
The first result comes from a CW rotor direction and this is the sequence:

33 - 10 - 1 - 2 - 3 - 23 - 28 - 6 - 35 - 27 - 8 - 12 - 7 - 34 - 10 - 29 - 29 - 22 - 4

[attachimg=#1]
Do you understand that this dealer has completely different degrees of predictability, depending on the rotor/ball-direction?


This script is only something I'll use to give me ideas what to look for.
Are there common things to find?
Like the above: Is it common that one direction is so predictable while the other is not?

I don't know.
But I can spend an hour or two in the evenings, to try to find out.
And maybe write new versions of the script when the ideas pop up...


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

This part of the post has been removed!

There is a new version of the script posted in the Softare section:

nolinks://vlsroulette.com/other-software-for-roulette/kfs%27-distance-tracker/

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


Best regards,
KFS



(So many words - so little said )

[Boo_Ray]

Hi kfs
great program and nice post

[Kon-Fu-Sed]

Thanks Boo_Ray!



I forgot to mention that because it's run in your browser you can of course save the "graph"'s frame for later reference.

Just REMEMBER to choose a unique name


/KFS

[ChickenDinner]

Great post KFS.

Interesting to me because I've been heading in a similar direction. One thing I have noticed is that toward the end of a dealer's shift, say the last 5-10 spins, sometimes their throw becomes a little more predictable, ie, CW shows a movement of +10, +11, +8, +12, etc. What I put this down to is that the dealer's ball throw becomes warmed up and so more habitual. I have not tested this idea enough yet to be sure that there is anything in it, as it could simply be a coincidence on the actuals I've looked at. But your software will be a great help in my testing so many thanks.[smiley=thumbsup.gif]

Cheers,
CD 

[TwoCatSam]

I have a thought!  (Aren't you amazed!!??)

KFS

Brilliant, ol' bean! 

Have you considered the dealer spins with one hand and then the other?  If a person is right-handed, wouldn't they spin more accurately with their right hand--more consistent?

In my viewing the Microgaming wheel, I cannot determine if the dealer is trying to spin as the last number passes under her hand or not.  I will look closely and maybe put one of my videos on DVD so I can slow it down on the TV and look there.

Again, great work!

Samster

[Boo_Ray]

sam I dont think that microgaming and playtech dealers try to do that.. they just pick up the ball watch around wawe a few times then randomly release ball..

they look.....you know...

[TwoCatSam]

Boo_Ray

Sometimes they wave and flop their arms so much, you'd swear they were in a Broadway production!!  SWAN LAKE or something!!

Sam

[Kon-Fu-Sed]

[highlight]@ ChickenDinner,[/highlight]

What I put this down to is that the dealer's ball throw becomes warmed up and so more habitual.

You know, I have thoughts like that myself.
And it wouldn't surprise me at all if this is a correct observation.
Why not, if you think about it: Throw the ball fifteen times and why wouldn't the next throws be in a routine fashion, more or less.

Rest assured I, also, will keep an eye on that phenomenon.


[highlight]@ Sam,[/highlight]

If a person is right-handed, wouldn't they spin more accurately with their right hand--more consistent?

That's just exactly what I wanted to show in the first image in the post.
The CW throws are spread all over while the CC are within a nine-pockets distance from the release (+12 to +20/-17).
Except the last CC spin...

As a matter of fact, I'm one of those who actually have watched your videos - well; some of them at least. (And I really enjoyed your comments )

And from what I have seen there, I agree with Boo_Ray and yourself - I don't think they throw the ball in that fashion.
And they wave their arms... They really do. Why? Anyone knows?

What I understand is, that they do this in the bigger casinos in Germany and France.
(Not wave their arms - throw the ball)
Anyone knows about other countries or casinos?


[highlight]@ Mr Chips,[/highlight]

I have successfully used what you call "chains of independent trials"

Good for you!
Good luck.

Really.



[highlight]@ All,[/highlight]

If you like my ideas and my program: Please use it and post your experiences. I'd like that.

I already have found something I have to do add to the script, btw:
The possibility to load spins and auto-run them.
A simple copy-and-paste facility, like in the other trackers/alerts I've written.

Will be done soon...



I won't be able to log-in for a few days so [highlight]have fun[/highlight] in the mean-time!
KFS

[mistarlupo]

Kon-Fu-Sed,
Great effort, mate! Thank you. Much appreciated!

Quick question. What if there is not difference in how the ball is thrown,
from spin to spin? I mean, it is either CW or CCW. I think KFS's program is able
to manage with that...

A little tweak. Let's suppose we need the CW graph only.
Firstly, we select the direction the rotor is usually spun (CW). Then we start
with filling the numbers (rls and hit)... But after each spin we'll need to
enter some numbers for the CCW direction, right? So we enter some non-sense
data in its place, just to have something for the program. Finally, we just ignore
the CCW graph, which has absolutely nothing to do with our tracking.

Hopefully, I am not missing something.
Anyway, thanks again. Good luck all!

Regards,
m

[Kingpin]

KFS

Thank you very much, it is a very well done and professional looking program you have made here. I think it's the best program i've seen for tracking dealer signature. If there is such a thing. (I think there is).
I look forward to trying this out.

THANK YOU! [smiley=tekst-toppie.gif]

Best Regards
Kingpin

[Kingpin]

I have a newbie question...

I forgot to mention that because it's run in your browser you can of course save the "graph"'s frame for later reference.

How exactly do i save it?

Kingpin

[Kingpin]

Wow this program is really nice 

Just tried it out on some old spins from dublin bet. It seems like on of the graphs (cw or cc) often has a pattern, while the other one is much more random. Very interesting in deed.

I'm really looking forward to your next version of this KFS it would really be great if you could run spins in it automatic.

I have a suggestion: It would be nice if you could make a small column in the bottum of the graph, showing what sides of the zero (or which sectors of 9 maybe better), had the most hits in total.

Thanks again!

Best Regards
Kingpin

[Kon-Fu-Sed]

Hi all,

Thanks for your kind words!
I'm glad you like the script.

The new version is posted here (in the Software section):
nolinks://vlsroulette.com/other-software-for-roulette/kfs%27-distance-tracker/


For mistarlupo there's now an option for one-directional spins.

For Kingpin there's an instruction on how to save (at least how I do it in my browser).
The other matter...? I'm not sure what you mean but isn't it just counting the black squares?



Anyway; Let's discuss the PROGRAM in the software section.


Regards,
KFS

[Kingpin]

Yeah KFS you're right, i can just count them no problem, it was just my lazyness speaking i guess...
I look forward to try the new version. Thanks!

Kingpin 

[iboba]

Dear chaps,your subject was very interesting to me,mr. Kon-Fu-Sed,espec.after your remark;throw the ball 15 times,after it will be throwing in routine fashion.Do you seriously think so?---You believe in dealers signatures????You might as well do,my dear chaps,even thought, no any croupier will addmit anything of the sort.Now when Im retired,and far from the place where I enter the game of such a skill/27yrs.as croupier/--can just give you a hint.every exp.croupier can hit a section arround number he aims,at least 6 out of 12 times,even more/that was the norm at my apprentice days,after months of excersise,and if one couldnt curry out the average.6 out of 12,he would seek other job.Now questions............ How many times did you see a table full of chips,except one number,and it hits exactly uncovered number????Do you think its accidentaly???How many times did you witness a gay plays big units,on small ser. hand after a hand it hits arf.and oppos.serie,untill he walks out broke????Did you ever see  one or two players/alone at the table/playing for big units,beat the croupier????And now opposite question,for your thoughts;How often did you see a player,playing for big units,hiting certain ser.consecutively 10 or 12 times????Could  tell you many more similar examples,but think that you all get the point.Cheers,mates,and careful

[iboba]

-