Ball type and its effect on scatter

Advanced roulette play
21

That is small sample but large ball lowest amount of hits you have from -1 to -12 that is not the case on the graph.

I have not separated level and tilted wheel… I am not sure you are right about that for scatter.

Am I right or wrong it doesn’t matter?
It is same amount of work to do them separately; if results match you can join them if not you will see the difference.

Fast rotor spins also should be looked separately. With that I do not mean 4 or 5 sec but less than 3 sec. BG has separate part where spins are faster if compared to spins at start of DVD.

Do you mean when the ball is first struck by a fret or when the ball first crosses the line onto the rotor area?

Either way you should get similar graph just shifted.
I look when the ball first time hits the fret. There are some unclear spins so you estimate them or you ignore them.

22

pj you stated the following,

I don’t see that a small degree of tilt will make any noticeable or significant difference… although I can see that if the tilt is taken to extremes (which it is not) then tilt would have an obvious impact on scatter… this is in direct contrast to the effect of a small degree of tilt on ball departure from the track where the effect of tilt is blatantly obvious… (and a delight to see)…

PJ, About how many spins of a particular wheel do you need to ananlyze before you make a determination on whether the wheel is level, semi-tilted or tilted?

Also, when you are out scouting for the best wheel to play on - - - what factors do you personally use to determine if the wheel is level, tilted or semit-tilted?

Forester, and everyone else, Please feel free to give me your feedback on this topic as well.

THANK YOU…

CHEERS… :slight_smile:

23

@Dr Spock

“About how many spins of a particular wheel do you need to ananlyze before you make a determination on whether the wheel is level, semi-tilted or tilted”

I am prepared to consider that a wheel is tilted when after 1 hours observation (about 20-30 spins) and I observe the ball impacting time and time again on either the same vertical diamond or by dropping from the same clock area of the wheel for more than half of the 20-30 spins… this is a purely subjective view and although the sampling is extremely small … it is not practical (for me)in real play to wait beyond 20-30 spins. And further to that I use that same period of time to assess the ball behaviour (speed/distance relationship) and suitability of the rotor speed / consistency of the croupiers etc… generally this 1st hour decides (for me) how and whether I will play… I don’t give a damn what numbers are coming up or whether the croupiers are cross-eyed or stutter… but I need them to be reasonably consistent with their rotor speeds… (and not spinning the wheel too fast)

In answer to your second question refer to my answer above.

In the end we can only do what is practical in the time we have available… which for me is about 5 hours in total…

Cheers
PJ

24

Dr. Spock,
That part is very individual.

I do not spend much time at roulette tables.
It could be in average 2 hours. I prefer to play instead just to watch.
I assume that every wheel is reasonably leveled so I play according to that.
Sometimes I find out that I was wrong specially if prediction doesn’t go well.
During play I observe what is going on. Few times it happened that during play I change the system.
Sometimes I stop and only observe. In usual 4 wheels at once.
If wheel had tilted next time I look only ~3 spins to confirm it.
I am not looking for single dominant diamond but 2 next to each other because it is possible to take advantage of both of them.

On some tables tilt can change when more people come to play.
But strongly tilted wheel may stay that way for long time.
Even casino leveling them it still stays tilted. On the end it doesn’t have to be that it was tilted wheel. Common drop point may be caused because of deformations on the ball track.

I do not know who when and why first started observing scatter from diamond.
If I guess it would be probably because it is the easiest to see the point when the ball hits it. For me separating spin at that point is very wrong and I am trying to change it. Even on “tilted” wheel observing diamond isn’t very good. Because after detecting that ball speed is within particular range and expecting constant amount of rotations to the end. It doesn’t mean that the ball will travel same amount of rotations within same time period. That would cause similar effect I was describing earlier. The diamond would be hit even with wider variety of ball speeds it may be same diamond but rotor position wouldn’t be same.
In previous example it was 9 +4 error, but what if rotor was 2 sec per rotation. It would be 9 +10=19 pockets difference with possibility of anything in between. That is the main reason why I do not like fast rotor. After the diamond hit the ball can stay in the air 0.1s or ~1 s. It is ½ wheel difference.

25

@Forester

OOPS! I think I’m in trouble… I started focusing on diamond impact and the resulting scatter… I think what started me on this was the calibration of the FF… please correct me if I am wrong but I was under the impression that during calibration of the FF each ball revolution is timed (switch pressed) until the ball either impacts with a diamond or until it hits the rotor without hiting a diamond… which if this is the case will lead to some very strange timings of ball travel… If this is not the case and the ball is timed only until it has reaches the wheel without interference then calibration could become difficult because of the frequency of diamond impact…? What does everybody else do when calibrating??? :-[

Cheers
PJ

26

Diamond hit or rotor hit. It doesn’t make much difference.
Offset may be different but accuracy of prediction will not change much.
Even if you press the switch by 1 sec late. You will have still accurate system.
You will have problem if rotor speed significantly changes.
For example if you played spins with rotor 10 pockets per sec. and if rotor changes to 13 pockets per sec. you will have offset change by 3 pockets.

If set up spin after first prediction you continue clocking. The system will ignore it for 4 sec (and think bloody another ex Mark’s/Stefano’s player). But after 4 sec if you clocking every rotation it may set to 4-5 sec to the end.
It depends where you set prediction to be you may be wrong by 4-5 sec.
Nothing will happen if rotor is constant but as in precious example you will now have
(13-10 )x4=12
Is not hard to press the switch when ball hits the rotor. It is the best because system is made for that. The time deviation shouldn’t be greater then 0.5 sec so errors are very small. If the ball turns in to spinner simply estimate it, you will be ok. If you are not sure do not press it wait for another spin and try again.

27

Thanks Fellas for the advice,

I’m sure it’ll all come in handy- - - Especially since I’ll be purchasing FF this Friday the 29th. My master plan is now underway. I’ll keep you posted.

Cheers :wink:

28

Cool,
Please use paypal I need to order some tools.
;D ;D ;D ;D ;D ;D ;D ;D ;D ;D ;D ;D ;D ;D

There is many posts in support section to understand system but who can read them all.

:frowning: >:( >:( >:( >:( >:( >:( >:( >:( >:( >:( >:(

Today I found out that I need to buy new superglue for switches.

I buy small bottle for $40, it is amazing glue, very strong. When I join 2 parts it creates boiling temperature.
First switch I made with that glue is more then 6 months old and I am still using it.
But I realized that now glue doesn’t hold good as before.
It must be that after I used it few times it mixed with air and lost strength.

29

[quote=“Forester, post:24, topic:180”]I do not know who when and why first started observing scatter from diamond.
If I guess it would be probably because it is the easiest to see the point when the ball hits it.[/quote]
I’d think that yes, in part it is because it is easy to see. But more profoundly, it has to do with the idea that until diamond hit, the physics of roulette can be mathematically modelled. After diamond hit, the bouncing is chaotic and the only thing one can do is to collect statistics about the scatter distribution. It seems obvious, but it is deceptive.

Forester, you are the only one I’ve seen who has gone a step further and examined the effects of hitting different parts of the diamond. Of course, a hit on the upper part of the diamond implies higher speed of the ball (it hasn’t had so much time to slow down during its spiraling movement) and a bit greater distance form the rotor. Both these factors imply a longer jump and (I assume) a scatter distribution which is both flatter and has its peak further away from the (number under the) diamond impacted. An impact on the lower (actually the middle part) of a diamond can only happen at lower ball speed. And since the distance to the rotor is a bit shorter, the ball will quickly land after the impact and the scatter distribution will likely be more peaky and centered closer to the number under the impacted diamond.

I assume that your focus on rotor hit rather than diamond hit, has to do with you wanting to indirectly include the details of the diamond impact in the modell, rather than leaving it all to the scatter. Even if there isn’t enough precision in ball timing to predict which part of the diamond will be hit, it might be useful for scatter analysis to consider the concept.

After the diamond hit the ball can stay in the air 0.1s or ~1 s. It is ½ wheel difference.

Yes, and you’d like to measure this, right? I would suggest using digital sound recording. It should be easy to manually identify the of the ball hitting a diamond, and also the first after that, when it hits the rotor (or something else). Therefor, on a soundlevel-over-time graph, you can measure the “air time” of the ball with very high precision. In combination with a video frame analysis, the causes and consequences of ball air time could be revealed.

In the best case, the air time of the ball is strictly correlated with what part of the diamond it has hit (longer air time means hitting higher up). This could be verified by measuring ball speed and distance to diamond before impact. Whether it hit (the top of) one diamond, or (the middle of) the next vertical diamond a quarter of a revolution further away, is obvious from video analysis. If lucky, there will be a very peaky and nice scatter distribution for the cases when the ball hits the lower part of the diamond! Most freak scatter events maybe might be defered to the top-of-diamond impact events where the ball spends a lot of time in the air.

Anyway, I don’t think it is the point of impact with the rotor that is of primary interest. It is the air time. The two are different because the ball might make its first landing on something else than the rotor. And air time is best measured acoustically, since the time resolution of digital audio is many times higher than in video analysis.

30

Rollo,

You raised a very interesting idea. However what I am trying to figure out is exactly would a digital sound recorder would fit into the grand scheme of things since all bets will be off atleast a second or 2 before the ball ever hits the first diamond or rotor.

Worse case scenario, is I have seen some dealers call no bets a kind of late in the spin at the precise moment the ball hits the first diamond or rotor - - - and if you’re lucky - – -that would give you time to maybe reach across the table really fast and sneak 1 bet quick bet in - - and over half the time they’ll probably call off that bet off any how- - - especially if they see a trend developing of you always trying to sneak in last second bets spin after spin.

And if you keep it up with you last second bets, all they have to do is stop waiting for the ball the hit a diamond or rotor before they call no more bets - - - AND THAT WILL BE THE END OF THAT. ;D

CHEERS :wink:

31

;D
Yes, it would be after the fact. I only meant it for use in “laboratory” studies. Video frame analysis doesn’t allow for studying directly which part of the diamond was hit, or exactly where the ball lands after hitting the diamond. With a simple digital audio recording, for which there probably is freeware around, the exact diamond impact time and exact first landing time, could be determined. After the fact.

It might help understand why the scatter distributions look like they do. Maybe the traditional scatter distribution could basically be divided into two. One for top of the diamond hits, and one for mid-diamond hit (or rather, a continuous between them). It might turn out to be useful to know, even though I am not sure how… For example, on a tilted wheel the ball should spiral inwards faster (because the wheel leans that way) and the balls should hit diamonds lower. And taking that backwards, maybe a scatter analysis could help identify a tilted wheel? I’m just speculating, though.

If a roulette computer on the field can be used to predict not only which diamond will be hit, but whether it will hit it high or low, and adapt the scatter distribution adjusted prognosis accordingly, well, that is another question…

32

Rollo, I do not need to measure that time, I did it only at home to see how much it is beneficial to us. Since the FF goes for number on the rotor it diamond is hit earlier predicted number will be further distanced but longer ball jump will partly compensate for that.
I wouldn’t say that FF doesn’t have enough accuracy to predict exact diamond hit. It does. The FF will reduce clocking errors most likely from 0 to 10 ms. On a leveled wheel we can say we have 50 ms difference from diamond to diamond. So it may be just enough.
On tilted heel we have 200 ms, therefore it’s easier.

Main reason why it will not be always as the FF predicts is that after the FF predicts there is still significant time of ball traveling which may not be always 100% inline with measured ball parameters. So we can only observe it as an average.

Once when the ball hits rotor is where trouble starts.
Most stable scatter is when the ball hits diamonds middle upper part.
Top part most likely creates 2 scenarios.
One when the ball hits directly to the rotor pocket, creates high impact and stops instantly and the other one when it starts rolling and that can be long.
Middle of dimond also likes to create 2 scenarios. The ball traveling is sharlpy cut. The ball stopps instantly or even goes backwards, and the other one when it roles.

33

This is from almost 2 years ago.

1 diamond hit and ball scatter on tilted wheel

2 what do we really get after defining that the ball is within targeted time window and base prediction on that. The window is divided in to 4 parts.
If you add all 4 together you do not get much. I call it plain prediction

3 Prediction after FF takes adjustment based on ball speed.

34

All

As we have discussed a couple of years ago… “Rollers” must not be included in any considerations of Bounce & Scatter.
They must be considered as just an anomaly at least until we establish what is causing them.

Diamond position strike is interesting and a positive discussion point but again unless we can predict it it is only a research subject.

In my opinion the answer to Bounce & Scatter lies in identifying the average distances involved with different ball types and sizes and especially with the different ball travel direction and checking these on different wheel types. Then modifying the FF to assist in including these in the prediction.

Mike.

35

So you better start thinking about new name for new talking FF.
:smiley: :smiley: :smiley: :smiley: :smiley:

36

How about the “FiF” or the “FIF” - the Forester Intelligent Fibrillator?

37

Nooo, it will not be Fibrillator ;D

38

Forester, when you made that top/middle/bottom hit analysis, how did you measure where on the diamond the ball hit? You cannot find a video frame of the event. I assume that you extrapolated it from measured speed of the ball.

I think that air time is easy to measure with very high precision. It could be done in the field as well since a microphone is very descrete. Of course it needs looking at manually at home afterwards, it is not for real time prediction. It is a research topic, and it is fo course secondary to the all important quest of predicting which diamond will be hit, but I think it can be of real benefit. The ball/diamond impact event is not chaotic. It is the same physics that is repeated in the same way every time. (What happens after the ball lands on the rotor however is chaotic).

We all know that before one can start predicting a new wheel, one needs to reconaitre the behavioral characteristiscs of that specific wheel/ball system. One of the issues to clarify, is what the scatter looks like. Measuring the time the ball spends in the air might help.

The scatter should depend on two variables we can measure: One is of course the rotor speed, which represents the force with which the ball is kicked once it has landed. The other variable is the force with which the ball lands. This is well approximated by the time it has spent in the air since diamond impact (which in turn depends on if it hits the diamond high or low).

The traditional scatter diagram could be developed into a scatter function, the scatter as a function of rotor speed and ball air time. Rotor speed can be very well predicted of course. Air time is more difficult, but what will be certain after some sampling, is how often the ball has long and short air time. From that information, a more reliable scatter distribution might be reconstructed with less observation time.

Predicting where on the diamond the ball will hit, must be difficult since a hit on the top of a diamond, is very close to a hit at the bottom of the next diamond, if the ball misses the top and spirals further inwards for a quarter of a revolution. One might separate likely middle hits from other hits. As prediction precision evolves over time (with practice and equipment upgrades) this might get increased importance. There’s alot to gain from narrowing down the scatter.

39
Forester, when you made that top/middle/bottom hit analysis, how did you measure where on the diamond the ball hit?

I did video in slow motion, it can be estimated, and the ball doesn’t drop much within 40ms difference in between video frames.

The ball/diamond impact event is not chaotic. It is the same physics that is repeated in the same way every time. (What happens after the ball lands on the rotor however is chaotic).

Yes ball/diamond impact event is not chaotic, but I do not see point of measuring time by audio if for each measurement we do not get exact point on diamond. We will end up with spectrum of different times. But in slow motion we can estimate how much linear is “position on diamond” (proportional to ball speed) and distance until the ball hit s rotor. With ff we observe prediction €“rotor hit, there is no need to separate from that diamond-rotor because we can’t do much about it anyway. We will get results prediction-rotor and after rotor hit as you said it is chaotic by some %.
Results produced this way are better then if we predict to diamond and from there look as scatter.

The scatter should depend on two variables we can measure: One is of course the rotor speed, which represents the force with which the ball is kicked once it has landed. The other variable is the force with which the ball lands. This is well approximated by the time it has spent in the air since diamond impact (which in turn depends on if it hits the diamond high or low).

Part of that is included with FF’s prediction because we predicting point on rotor, but from there scatter may be influenced by rotor speed. It would be very complicated and unpractical to try to get full adjustments for that. Even if we have all possibilities in settings we simply wouldn’t know which one to set. Significant rotor speed change can cause many things. From ball traveling shorter to ball traveling longer, but also completely different behavior across rotor pockets. All what FF does after ball hits rotor is that it adds few more pockets to prediction proportionally to increased rotor speed? That may be correct in many cases but in some it may create opposite result. So reality is that with any significant change we must observe what is really happening on the wheel. As additional tool for observation, the last zap when ball is due to hit rotor is very useful.

40

I think with the challenge we face in trying to ‘beat the house’ we can become a little too preoccupied with analysis and can sometimes be unwittingly drawn into the fatal mistake of ‘over analysing’… the FF is already able to identify the point of rotor impact and thereby predict a likely result somewhere in the 18 numbers following the point of rotor impact (or even diamond impact)… the fact is that we now know that the result will occur 75%+ in the 18 numbers following this point… which FF has brilliantly identified… JUST HOW GOOD IS THAT??? … so by placing our 6 bets in that sector we know that we can reliably get a hit 33% of the time on 75% of our plays… the remaining 25% are simply sacrificed to camoflage… so in effect we are being paid 5 to 1 on a 2 to 1 chance (75% of the time)… THIS IS BRILLIANT…!!!

Cheers
PJ