Bago and Stefanos discussion chart

Roulette Computers
1


Now, i can`t display a 18 or 19 number sector as Stefano mentions but here is a 12 number sector and it also shows the effect that he mentions, just a little bit weaker. Generally, the sample is way too small to make a decision from, because the requirements that i use for my play, comes nowhere near the 3 SD benchmark that i use as a minimum for play.

You will have to make your own decisions on what you think.My own opinion is that neither Stefano, nor Bago can use such small samples as evidence, only as a “opinion maker”, and here they both have a point. Stefano has a 1/2 wheel deviation in the distribution and Bago says it looks random.

You make your own decisions.

NOTE: Most 1/2 wheel deviations will look a little bit like on the 12 number sector. At first sight it looks very visible, but you can imagine how a really biased sector will look like, all you have to do is look at the 1 pocket chart to see that there IS a deviation.

20/33 where the expectation is 26.5/26.5 is not a big thing, some 1.98 SD.

The SD, determines how strong it is, in the 12 pocket case 1.25 SD.

[b]PPS: I did the same thing on 1 of the other charts in the discussion thread, and here we have a

3.13 SD on the 1 pocket chart
3.32 SD on the 5 pocket chart
3.34 SD on the 7 pocket chart
4.17 SD on the 10 pocket chart
3.73 SD on the 12 pocket chart[/b]

2

No, Bago is right 100%. Kelly that is only hits to rotor.

If I make a video to show how I can predict roulette outcome.
If I have no clue how to achieve advantage. I can run 50 or 100 spins and see if whatever I am predicting makes winnings. I can also make video from few parts where I get positive results. It is same as plying in casino and having lucky night.

Lets say I go to casino 20 times, play 50 spins, 25 spins I predict around zero and 25 spins I play opposite around number 10 (it could be any prediction). If once I get final results to match my prediction I save video and show to people.

With such example my winnings will be not related to prediction where the ball really hits rotor. But the ball may stop at my predicted area by hitting rotor and stops instantly by, rolling 2-3 rotations, by jumping 20 pockets, or by jumping 5 pockets. And that is what is really happening on Stefano’s DVD.

If I do reveres engineering on such spins, I will get exactly same as Bago’s graph is displaying.
Only a slight advantage on pockets in front of predicted number where the ball hits to rotor. Because if scatter law says that 40 % of spins the ball will jump most likely lets say 5-12 pockets it is logical after reverse engineering to find slightly more hits 5-12 pockets in front of final results. But it is created by coincidence, or by taking groups of spins and assembling it in to single video.

By that graph Stefano needs more then 1.33 hits to start winning, 48/36
at wide sector from 28-36 and 0-4 he has 1.7
so he has only 0.4 hits extra but that is when the ball is not jumping. To have only 0.4 extra hits to point where the ball hits rotor (without ball jumping) is not result of effective roulette computer. If computer is predicting he will have something as 80% of hits in that area.
Under such condition FF will have 10 times better hit rate. It is needed to be so high, so after randomness of ball jumps system may produce something as 20-30% advantage.
It is imposable to have 1:15 hit rate as Stefano claims where he can’t predict point where the ball will drop on rotor.

Even if results are not as I described based on accuracy of prediction to hits to rotor based on Bago’s graph the best he can have is 1:28 and that is if ball doesn’t jump at all.

To explain it better I took data that Bago’s have from Stefano’s DVD.
Red graph is how well his computer predicts point on the rotor where the ball will drop.
White graph represents how most likely the ball jumps on his wheel.

And the blue graph is a combination, where from any point on rotor the ball scatter law is applied. As you can see final result is good for nothing. We have 8 pick points’ 1-2 pockets wide, which really mean they could show at any place.
[attachimg=1]

3

To avoid high glitches in the graph because there is not enough spin I rounded up result by averaging 3 neighboring results. It allows us to see result more clearly.
Blue graph is always representation for 100 spins.

Based on displayed ball scatter law, if we have 100% accurate prediction to rotor, it will allow us to have close to 100% advantage. Because f that it is just normal that if after we select spins where we have had hit rate as 1:15 that after reverse engineering we will have and some advantage on hits to rotor. Mostly likely it will be as what Bago have found, but far enough from what roulette computer should produce.

Stefano’s results are as this.

[attachimg=1]

And FFA as this. It is far better result because red graph which represents systems accuracy to predict where the ball will drop is much higher. Compare it!

[attachimg=2]

His computer simply can not predict ball failing point. It doesn’t work by physics of the wheel. That is why when we tested his phone with single spin it produces always different result. That Bago tested it on skipping DVD is rubbish, many more people did tested it and got same results, even Stefano’s DVD shows that. That is why he stopped supplying DVD.

Mobile phone is inaccurate for timings; therefore he will never succeed in building useful roulette computer if he uses mobile phone.

FFA uses processor interrupt function; as soon as switch is pressed it registers it. It takes only 1nano second to do it. Mobile phone program is built on top of operating system, it scans what is happening to the switch every 10-20 ms. if phone is checking for SMS or communication with tower it is a priority, so that time can be longer.

4

Basicly i agree. I said Bago has a point in that it looks random, and that Stefano is right about the 1/2 wheel distribution. The numbers for the 1/2 wheel distribution are just not very high. 1.98 SD for 1/2 wheel is not that impressive. On the other hand, the sample is not significant in size. You could easyli find a 53 spin streak like that in a 1000 spin sample with 20% edge. If the expectation is 1:15, i would also expect a little more consistency in the graph.

All i was trying to say, is that mathematicly all these small tests really goes nowhere when comes to making a significant proof.

I made the same sort of test, that Bago did, on Marks 58 spin sample movie, just with more parameters. I would have made the same thing with Stefanos videos, but it would take 2 - 3 times the effort because the numbers are hard to read properly. It IS possible, there will just be a lot of rewinding, where i could read wheel speed and strike numbers straight away without any rewind on Marks video. This is a couple of years back, he might have released better quality since, i haven`t checked.

I might have a go one day, if i find the motivation.

PS
I will have the motivation if Stefano one day drops a 250 or 500 spin sample in good quality with predictions. I also knows the effort it takes to produce 500 spins on a couple of DVD`s. There WILL be breaks in the video because 500 spins is not something you just do Saturday morning before shopping, so i can imagine the arguing about these breaks afterwards.

Marks data.

I can see that at the time i made the pic, i have left the 4 correlation coloumns out that gives the cross match results. Not to stir up trouble in the community i guess :slight_smile:

5

Kelly you can’t have and claim 1:15 hit rate if you have hits to rotor only 1:28. We all know that ball doesn’t stop instantly and regardless with which accuracy we predict on which number the ball will drop, that same ball will still make jumps which are partly unpredictable. We can only find out by how many pockets the ball will jump most of the time.

I claim with FFA something as 1:12 hits to rotor and 1:24-1:28 as final result. It is logical since scatter on his wheel takes a bit more then 50%.

To have 1:15 hit rate we need that the ball hits rotor with accuracy of 1:6 according to prediction. When I look his red graph, I simply don’t know where would I place bets.

Even if that slight advantage is generated from his mobile phone roulette computer it is not enough for real casino play, where parameters may change and where with such poor prediction we wouldn’t be able to detect it. Success in real game is not linear by defined accuracy of hits, but it goes to exponential curve. It is because when we are more accurate with prediction we can observe things better.

For example exclude ball jumps, if you play and the ball hits 80 , within 6 pockets from predicted number it is more then 3 times better then if the ball hitting 80, within18 pockets. It is because if something happened so the ball starts traveling longer with 6 pockets accuracy we can instantly spot the change, but with 18 pockets it is harder, since we do not know is it coincidence or real change.

For roulette computer to know if it works well or not you do not need many spins.
If you see most of spins hitting rotor as predicted within few pockets of accuracy 100 spins is plenty. Remaining you know that it is scatter law. If Stefano Hourmouzis wants to claim that his computer knows will the ball jump 5, 10, 15, 30,50…etc. pocket by some voodoo magic that is a different story. By physics it doesn’t work, or performance is extremely poor. It is his DVD, Bago only analyzed it so he shouldn’t be upset with him.

Bago is pain in the a… as we all know, but if one day he decides to show FF’s results Stefano can only sh… himself. Regardless is it skipping video as Stefano claimed that he used. But at least he will understand what real roulette computer is.

I did not look Marks video much, from brief look it did show some consistency, but it was assembled from few blocks.

It is same reason why I do not look recent spins which Stefano published. It is because I know that he wouldn’t be so stupid to make same mistake with so poor DVD as he use to supply with his computer, for past 2 years. At that time people did not know what to look for, but now they do, if thy understand 10% of what was writtn.

So what did you find with Mark Howe’s roulette spins?

6

Basicly i agree. There are a deterministic part and a chaotic part in 1 spin. The deterministic part can be calculated by a computer or, to some degree, by VB. The chaotic part is a statistic problem that can be solved in different ways.

And no, based on his diagram, i wouldn`t know where to place the bets either. I would start looking in the major half, but it would still be too early for any decisions, because one could end up between 2 or 3 mountains.

If i had put the last 4 coloumns on the chart from Mark, i would also have taken a stance for or against it because the result speaks for themselfs. There are plenty of ways to cross match the results.
[b]

  1. How does the computer prediction reacts to different wheel speeds ?
  2. How does the prediction correlate with outcome ?
  3. How does the outcome correlate with the strike number and average bounce ?
  4. How does the wheel speed effect the bounce and is this embedded in the prediction ?
  5. Tilt diagram
  6. Scatter diagram.
  7. Wheel speed diagram.
  8. Scatter analysis of each diamond. In Marks case there are a dominant diamond.
  9. Etc.
  10. Etc.
  11. Etc.[/b]

    The conclusion has 14 posts embedded in 4 coloumns and 6 diagrams. But the conclusion remains in the archive, because i don`t wanna take a stance against/for either of the product suppliers on the market. You can call it weak, but im trying to protect my play from people like Bago (publicly posting a pic of a thumper pic that i had send him in confidence) or any other enemy. And i could create a lot of enemys if i wanted to, but thats probaly not gonna help my play and acces possibilitys any good, so i stand back.
7

Kelly, you know him for long time and his big mouth, you shouldn’t share some information if you did not want it to go around.

Do you think I like to make enemies?

You should remember how it started.
We were chatting at GG, it was your post something as “Better roulette predictions” , nobody was talking about Stefano or Bago when they come and claimed that Stefano’s computer can predict even side bets, and that we are idiots.

At that time Bago just bought his computer and he was believing and doing everything what Stefano told him.

Reason why recently I spent some extra time with him was because I really want him to see how much wrong he was, and that I did know much earlier everything about Stefano.

But with Bago nothing helps.
I know that he found out that FF do produce advantage, I know that he even took it to casino, but he has problem with sound. In noisy environment he can’t hear it.
I do not know if it is real problem, as Marcello told me, or he did not wash his ears for long time. Some people took extra powerful IL to get it more laud.

Now Bago is abusing Capello that he cheated on him, because he gave him everything almost for free, but FF has some trouble with sound.
Unreal.

8

@Kelly
Out there are some people you still can trust them, so no need that you put us all in the same basket with Bago. :wink: