C

Sent: Wednesday, August 19, 1998
Subject: roulette prediction


Please let me now more details about the success of your project, because I am
working on a similar project too.

- What was (is???) your quote of prediction accuracy?
- Do your equations handle with tilt?
- What is your general approach: physical, statistical, ...?
- What experience do you have with casino stuff?

Please contact me on xxxx@xxxxxxx.com perhaps for further conversation about
the topic.

Hope to see you soon :-)

Christian (Germany)


Sent: Monday, January 26, 1998
Subject: Re: Raw Data - Roulette3


In a message dated 98-01-26 18:46:54 EST, you write:

<<
I have no idea how you were able to predict roulette, even if you can count accurately.
>>

Well - given the same equations for a computers predictions, if you can
determine the ball time, the rotor time, and the relative relation between the
two - and you have the calibration coefficients for the wheel, you can predict
the outcome. Same as with a computer.

My first system only required the rotor and ball times, as the ball time began
at the same time the rotor time ended. I created lookup tables for the
predicted pockets. They had ball times vertically and rotor times
horizontally. I was able to reduce the table down to having to memorize 24
4-digit numbers. I had to decode the numbers on the fly, but that wasnt too
bad. The problems were two-fold. First, I still needed an accurate
calibration of the wheel - thus I needed a computer. Second, I had to
rememorize the table for each new wheel. This was tough.

I have a much better approach at this point, but am not inclined to discuss
it. Suffice it to say, it eliminates the need for a computer, and reduces the
memory work involved.

Regards

Dave


Sent: Friday, February 12, 1999
Subject: Roulette data, algorithms, etc.


Gentlemen -

Please indicate the availability of the above and any/all related resources.

I am considering assembling a research team of local college math students to take this work to the next level utilizing Dr. Jeffrey's ground breaking "Neuronal Net" work combined with today's micro-circuitry. As such it should constitute a worthy student project.

When I first worked on a similar project in the '70s with Dr. Fechter at UNLV, our working prototype was the size of a med-size city phone book.

Although such a device would clearly run afoul of legal statute in Nevada, the laws of other venues may be negotiable... to this end my intention is to assemble a second "implementation" team while simultaneously filming a documentary and/or developing a book and movie screenplay based on the actual casino exploits.

Thank you in advance for your prompt reply.

regards, mkd


Sent: Monday, March 08, 1999
Subject: I'd like some help from you guys if possible


I'm completely fascinated by the attempt to try to beat the roulette
wheel. I'm a computer science student, however I'm not Newton. I was
wondering If I could get an e-mail of the actual source code you used in
your computer. From the looks of your website you had some rough
materials to work with. I've read the Eudeamonic Pie, and it has the
concepts, I've read Martin Jensens Secrets of Winning Roulette, and it
clarifies the concepts a bit, but I don't have the physics background,
or the slightest Idea of how to go about formulating the
algorithm. (guess I could see how the space shuttle comes back to earth)
I would never use this in a casino where there is a device law. I may
never leave the house with this technology. Maybe I'd never have it
concealed, I'd probably just work it off of my pc with a toy roulette
wheel at home. Anyhow the subject is fascinating and I'm dying to have
the source code.
Thank you
Sal


Sent: Wednesday, August 04, 1999
Subject: Roulette Data


G'day Rouletters from Sydney, Australia!
Fascinating site of yours!
I'm interested in your raw data to put into neural networks I have designed in the Matlab Neural Network environment.
I currently use them for horse race analysis but with the Sydney casino a few minutes away why not roulette too!!
The roulette prediction idea is an interesting problem in physics.
Is there an equation of motion in terms of Newtonian dynamics for a roulette ball published in terms of relative speed of ball and wheel and frictional forces?
Cheers, Peter


Sent: Monday, September 20, 1999
Subject: Roulette questions


Hello,
I noticed your site about the prediction of roulette play.
You sure have an interesting site, I played Black Jack
professionally myself from 1994-95 but I gave it up due to casino
'heat' and high traveling expenses. I live in Sweden and currently
work as a programmer, but I'm still interested in casino games.

I'm considering building a roulette computer myself (I have a couple
of friends who are engineers) and I would like to ask you some
questions.

Were you able to get an advantage even in a real casino
environment, and if so roughly what percentage?
What are the variables I need to collect? (is it enough to know
the speed of the ball and the wheel to make a rough prediction
or should I also clock several laps to calculate the amount the ball
slows down for every lap?)

Regards,
Oskar


Yes, you need to know how much the ball speed decays. (the wheel speed is quite constant for any round of play, in our experience. We did not do too well with this computer, after all. At the time we got it into play, the laws in Nevada were changed to make such items illegal. We gave up.
I think that physical prediction, such as you are considering, a nd as we did, in not necessarily the only way to skin this cat. You may want to investigate neural nets and simulation in addition to physical prediction. Keep us informed of your progress! We have received about 10 or so inquiries to date, and when I heard last from a group in Europe, those people seemed to know what they were doing, but they have not kept us up-to-date about their latest progress.

--NOWSCAPE


Sent: Friday, February 12, 1999
Subject: Roulette data, algorithms, etc.


Gentlemen -

Please indicate the availability of the above and any/all related resources.

I am considering assembling a research team of local college math students to take this work to the next level utilizing Dr. Jeffrey's ground breaking "Neuronal Net" work combined with today's micro-circuitry. As such it should constitute a worthy student project.

When I first worked on a similar project in the '70s with Dr. Fechter at UNLV, our working prototype was the size of a med-size city phone book.

Although such a device would clearly run afoul of legal statute in Nevada, the laws of other venues may be negotiable... to this end my intention is to assemble a second "implementation" team while simultaneously filming a documentary and/or developing a book and movie screenplay based on the actual casino exploits.

Thank you in advance for your prompt reply.

regards, mkd


Sent: Sunday, March 19, 2000
Subject: roulette question


Hi I was wondering if you have read Mathematics of Gambling by Dr. Edward O.
Thorp---he has his own formulas for calculating ball revolutions and i was
wondering how it compared to what you have done. For example for ball
revolutions he would time the first revolution of the ball spinning and put
that time into this equation: 1/(e^(3T/20)-1)-10/3 (if T was .5 then
the answer would be 9.51 revs) I am just guessing that if you did all or
your tests on a different size wheel then Dr. Thorp then these formulas
would be different ---or is there a casino standard size that they all use,
I have seen different size wheels for sale from gaming distributors and I
always wondered if they all bought the same size wheel.
Thanks,
Steve


Sent: Wednesday, March 22, 2000
Subject: roulette timing data


i have recently thought of trying this very thing but i'm a little confused by your methods. It is my contention that all roulette wheels have different values but luckly the casinos don't change the wheels every day.
* --- * But they can easily change the ball... :(

Each wheel can be timed for its own unique timing pattern. However I would really be interested in any assistance I might get from you in the way you mapped your data and the end result in the way of any advantages. Exacly how did the deflectors on the wheel throw the ball?

* ---* The deflectors seemed to have a random effect, according to my recollection. We have that info at http://nowscape.com/blk/roul/skip.htm .

How much of a factor is the deflectors play in the overall result? I would appreciate hearing from you.

My idea is to get a special stop watch with multiple lap data counters. Use the data I receive to establish if there is any pattern in the way the ball as far as timing is concerned. If I can establish that I want to use an infar-red transmitter probably a car starter with long range and connect it to a serial port interface.

*---* I assume you have seem our work shown at http://nowscape.com/blk/roul/roulette.htm . We used a computer to do the timing. First we timed the rotor, and then the ball. One person did this. After the timing, the computer made the predicting, via speech synthesizer over an IR link, built into the shoes. The second person placed the bets.

There will have to be two transmitters. One will time the ball by buddy1 who will be responsible for clicking the beaken transmitter as the ball passes an arbitrary spot on the wheel. The computer hooked up to the serial interface will be the timer. Buddy2 will clock the wheel every time he sees "0" pass the same arbitrary spot thus making the computer know where the ball is in reference to the wheel. Bettor at the table will have a small earpiece sown into his jacket at the wrists listening for the column of numbers that will be best to bet on. If this is possible the sky is the limit on the money that can be made at this.

(Jason)


Sent: Monday, May 01, 2000
Subject: roulette computers


hello,

We have spoken before about roulette computers (remember me?--I gave
you the link to mathematics of gambling by Dr. Thorp) I saw on your page on
how you have an article on how casinos found out about blackjack
computers---but that was way back in 1987. Do you know if it is illegal
today to use a computer in casinos? Also i meant to ask you last time did
your roulette computer ever work in actual playing time in the casinos? Did
you ever get the "big payoff" you were hoping for? If you have any stories
about your computer...using it in a casino I would love to hear about them.
Maybe some problems you ran into or things you would do to not get noticed.
thank you in advance
Steve


Hi, Steve,
We gave up on the roulette thing -- my friend wanted to stop, I wanted to continue. We made several (maybe 10) trips to Nevada, each time tweaking our machine a bit more. As I remember, the machine only predicted the correct pocket once. We were close may times, but we only bet on one number, not nearby ones (the expected result is roughly the same). As far as I know it is still illegal in Nevada to use computers -- this was one of my friend's main reasons for stopping. I see no reason why Nevada would set aside this law, once it had it :). If you want to talk, please feel free to call me some evening 801/531-7907, I would enjoy hearing from you. 

We have a casino size roulette wheel, the one shown on my web page, it's for sale:) I think it is 24 in. in diameter. We were disappointed, because the wheel seems to be biased, favoring one sector. We bought it in Reno, NV, from a gambling supply house.

If you have some pics of your machine, I'd be happy to see them.
--  NOWSCAPE

-- Nowscape


Sent: Wednesday, May 03, 2000
Subject: Nathan's Roulette Betting Calculator (1979 Patent document)


Greetings,

First, I wish to complement you on your web site. I found it
fascinating!

I also viewed Nathan's 1979 Patent document and associated figures. I
wish to take a stab at employing a timing device in my roulette play.
Unfortunately, I have only a rudimentary knowledge of electronics. I have checked out
several electronics texts from the local library and am studying them to gain
the necessary background for assembling my own roulette calculator.

Since you have built your own roulette computer, I presume you have an
extensivebackground in electronics. I thus have a request to make. Please feel
free to decline if you wish. My request : can you suggest specific components
(names and numbers) that would be used to construct the Figure 4 device in Nathan's
patent document? I know that with enough study and testing I can construct it
on my own.
But, it would be a great help if you could 'point me in the right direction'.

Again, thanks for providing such a great web site.

kind regards,

Dan_Xxxxx@concentric.net


Sent: Saturday, August 26, 2000
Subject: roulette wheel timing


I'm very interested in your raw data. I have been thinking of this very idea for about a year and would appreciate any info u can give me in the way of prediction formulas. Also i have read that wheels in actual casino play r bent in such a way to skew the results (any comments on this). But most importantly I'm interested in finding out what was the overall advantage u witnessed with your project.


Sent: Monday, November 13, 2000
Subject: data request


Hi from Australia

If possible, I would appreciate a copy of your data set, as I can not get hold of a real roulette wheel here in Perth. I have been dabbling with the whole prediction thing for several years - more of a hobby. I put together a small algorithm based on mean-weighted geometric series. This was more appealing than regression because it could possibly eliminate the set-up time.

Anyway, it's impossible for me to get enough data at the casino, so if you could send me a copy I'd appreciate it. I have no intention of exploring wearable computers - unless I can buy one off the shelf somewhere. I'm more interested in trying to develop a visual technique.

If you are interested in how my algorithm performs on your data set, I'd be happy to stay in touch.

Cheers

Glen Xxxxxxxxxxx
Perth, Western Australia


Sent: Monday, October 29, 2001
Subject: Roulette prediction

Sir,

I have recently invested in a microcontroller kit and I am wanting to
replicate something similar to your project. I am willing to spend time in
a casino logging data but would be pleased if you could send me your
figures. I hope this is possible. Thankyou for your time.

Regards,

Simon


Sent: Tuesday, May 21, 2002 4:55 AM
Hi Nowscape,

We are two students from England who would like to make our own roulette

prediction device. We are not computer experts but will hopefully find

someone to help us on that side of things. We were just wondering if there

was any data available for our UK/European style roulette wheels (only one

0). Any help with this would be much appreciated. We were also wondering

what sort of equipment you guys would recommend as we couldn't help noticing

that all the examples on the net seem a little dated. Is there any chance

you could put us in touch with others in the UK with similar ideas? (most

notably Mark Howe, who seems particularly advanced). We, like you, would

like to know more about neuro nets as this seems the logical step forward.

Any help would be much appreciated. Thanks for your time.

Rich

P.S. You guys are a real inspiration. Cheers!


We don't have data for a European wheel, but if you can predict the American wheel, you'll be able to predict the European Wheel. You'll have to research what equipment to use -- I've been out of this business for quite some time as you see. As for putting you in touch with others with similar ideas, you haven't said, exactly, what your ideas are. As a matter of principle, I don't reveal the identities of correspondents.

Good luck and best wishes!


Sent: Sunday, June 02, 2002
Subject: POS??


Dear Nowscape, I think the roulette results you have
come up with from your research is fantastic. But on
your website I still can't figure out what you mean by
POSition. How does POSition relate to target
slot / pocket?


POSition is the clock - position, as if the wheel were divided into 12 hours.
Position relates to the target slot via the function of the speed / decay of the rotor.

I hope that helps.


 Date: Wed, 13 Oct 2004
Hi there, i was wondering if you could tell me if there is any possibility of purchasing the mathematical algorithms which predict the game of roulette, especially the European wheel. I believe that with the advancement of technology, today it would be much more worthwhile to set up such an operation than it was, say 20 years ago.

After reading, 'The Eudaemonic Pie' my friends and i were inspired to try and achieve the same goal as we believe in South Africa we have the advantage of the law behind us - it is not illegal to win in the casino by using a device, it is only against company policy!

Hope to hear from you soon

Regards

Nick []


Subject: Anyone out there with a working Casino style wheel?
Date: Wed, 14 Jun 2006 10:45:58 -0600

We are interested in pursuing a project predicting the outcome of roulette
in mid spin. All that we need to make this a reality is some collected data
from a workable Casino style wheel. A graduate from computer science as
well as statistics has written a workable program that can do this with as
little as 100 trials. Here is how it will work...

First we need an accurate timing of the rotor before the ball is spun. It
would be best if from that point on you record each occuring revolution
until the ball is spun and eventually fell into a pocket. The best tool to
accomplish this would be a laser tachometer but budgets being what they are
a good pair of eyes should suffice (one could be purchased on EBay for
$30-$40)

Next we need an accurate timing of the ball. Start the timer as the ball
crosses some arbitrary part of the wheel lets say 00 or 0. Continue to
record from that visual spot when the ball crosses it again in the next
revolution until the final revolution and eventual drop. Keeping as
accurate account of each revolution's split time until it falls. When it
does fall assume that the visual spot you started each revolution at is
12'oclock and tell us the time it fell from there along with all split
times.

When the ball drops stop the rotor timer.

Try to record as accurate as possible the pocket bounce of the ball.

With 100 of these trials we can determine the best formula that fits your
wheel and then we will send it to you to experiment with to determine what
advantage if any exist. From there only with the participating partners we
will share our secret on how to best use this to exploit an advantage over
the Casinos. We have found a unique way of possibly getting past the
bringing in of a secret device.


Date: Fri, 17 Jul 2009 19:03:52 -0700 (PDT)
From: Scott <***@yahoo.ca>
Subject: roulette devices

Hello, I stumbled upon your site. I see that it was last updated June 21 of this year. Are you still involved with information and/or the building of roulette computers/tracking devices.

I have been interested in making one - but would much rather buy one or be sold details on how to program and build such a device.

Any and all information you may have would be helpful.

I am "mailing blind" here as I don't even know your name - there are several pics and references to yourself and some former partners but no name for you.

I have contacted a few - likely all - individuals on the net who are selling such devices to date. They all seem to have serious problems with one another.

I am thus confused.

Thank you for any help you may offer,

Scott


Yes, the devices certainly can work. [**] hooked our device to a regulation wheel and tested it, using optical pickups for the ball and wheel to get precise inputs. Ideally, the best thing to do would be to pick one number, but I think it is better to bet on all the numbers on a quadrant or eightth of a wheel. The ball has a tendency to bounce across the diamonds, hit the wheel and skip a few slots, etc. We kept track of where the ball would have hit the wheel, as well as the actual location where the ball stopped, so we could compute an average "bounce" of the ball. The diamonds on the wheel at different angles seemed to me to have little effect on the actual outcome.

Electronic devices are illegal in Nevada, and I understand people have spent time in prison for using them. And the Ritz was very interesting to see. I heard the story as you said, that they were able to keep the money. But I also heard there was some legal action to get it back. I haven't heard anything about it for a couple years or more.

[**] and I are computer geeks, and we spent a lot of time with the programming. Hundreds of hours. If you hire people to do the programming you will be into this 10s of thousands. Your point about concealing the device now days as a PDA is a good one. Maybe they will not detect you right away. But if you start winning, they will pay a lot of attention. Also, if people actually are selling operating units that do win, the casinos will quickly find and deter them. If they are mass-produced, then they will be easy to pick up by looking for a particular frequency.

The video pickup is, in my opinion, bogus. It is just too complicated and would be worthy of NASA if it could be done. It is conceptually easy, but in reality not possible without fixed cameras, massive horsepower, and an astounding amount of programming. Of course, my knowledge is out of date, and I'm sure it is much easier than it would have been 30 years ago. But I think it is not possible with limited resources.

So you're familiar with Thorpe and Eudaemonic. [**] and I have both said we could have written Eudaemonic with our project. We had a broken cable trailing from the pant leg on the casino floor, Apple computer in the hotel room for data analysis, short circuits due to sweating. Not to mention duct tape stuck on the leg for hours that removed all the hair when pulled off. It was fun. But we spent about five years part-time and about $10,000.

Bon Jour,

Rich


From: KRP <nowscAPE@yahoo.com>
To: Scott <***@yahoo.ca>
Sent: Fri, Jul 17, 2009 10:47 pm
Subject: Re: Spam: roulette devices


eh, Scott --
We're out of the roulette game -- as you no doubt noticed the device pictured is quite old. Nowadays Such a device would be built differently but the principle involved -- predicting which slot-pocket-number will be at the intersection (think Time) of the ball and it -- that principle is of course still the same. I'm going to forward your email to my partner, maybe
he'll feel moved to respond. Good luck.

I have seen one such device for sale -- movies and all -- on the 'Net and I don't rust the seller. If he has a good machine then he's be better off keeping it a secret, I think.


Hi again, Scott,

I was not aware of these devices [...].  Do they actually exist, or are these people just claiming they exist?  If they are real they certainly cost much less than what it would take to put a device together yourself. 

We wrote out software in machine code and BASIC, which was a great deal of work.  We also made several iterations of the machine, which caused us to write a lot of versions and create much of our own hardware.  Our communication took place through an infrared transmitter hooked up to a speech synthesizer.  I did the data entry, and [my partner] wore the
earpiece under his long hair. 

Your description of functionality is exactly what we did.  Time the ball and the wheel relative to a fixed point on the outside of the roulette wheel, and perform a fit to a curve to predict where the ball will come down, and where the wheel will be at that time.  Conceptually  straight-forward, but difficult to implement.

We collected data on a wheel, plotted the decay curve of the ball and the wheel, and then loaded the curve data into the machine to go play for real.  We found a number of interesting things along the way:

1.  Wheels are different, as you probably know.  They decay at different rates, have different slopes off the upper edge, and are sometimes not level. 
2.  Balls are different - different materials, different diameters. 
3.  Yes, barometric pressure affects the ball, which we did not take into consideration. 

There are a couple books on computerized roulette.  "The Eudaemonic Pie" is about a group of physicists who approached the problem in the early 80s, and they have some interesting info.  We found after reading the book that we were trying to do the same thing at the same time.  I don't know whether they succeeded, but I think not. 

Also, a book called "Beat the Dealer" (by Thorpe, about card counting.  That's another story . . .) talks briefly about hooking a computer to a roulette wheel in the 1960s, but with no detail. 

Your comment about it being necessary to just input one variable . . . not sure I understand.  I think it is necessary to time both the ball and the wheel separately against another reference point.  If you time the ball and wheel as one input, such as when they intersect, you will not be able to separate out what speed the wheel is traveling vs. what speed the ball is traveling. 

So far as there being 'something in it' for me, I enjoy hearing other people and their approach to the problem.  We have had a few people send queries, but so far as I know no one has actually implemented anything.  We'd appreciate any info you would like to share if you go forward.  Nothing specific or confidential, but perhaps a general note or two.  It would be fun to vicariously be at the wheel.

My suggestion would be to spend the bucks on tuition or an out-in-the-open project.  Electronic devices are illegal in many casinos, and I think the possibility of actually realizing big money from this is small.  But, perhaps you can make it happen.  And it certainly was a lot of fun.

It was just curiosity on the location.  I go to BC / Alta. often, but haven't been in the Canadian east.

Good luck with the project!

R.

Scott,
This page --
http://nowscape.com/blk/roul/explain.htm  -- shows how to interpret the data in the .zip files I'm attaching here [in this email]. We considered the decay of the ball as a second degree polynomial (y = ax^2 + bx +c) ... the equation of a parabola, or of a falling object -- and we considered the equation of the decay of the central part, the rotor, as a straight line (Y = mx + b) even though technically it's a parabola also, but for practical consideration it can be considered as linear as indicated.

The idea that a parabola is the best fit is born out from physics and from our empirical data; a few graphs of which hare posted on our web pages here.

Attached are our data [files from previous experiments], let me know.


From: Scott<**@yahoo.ca>
Subject: Re: : roulette devices
To: KRP <nowscape@yahoo.com>

Thanks for the links and data. I'll take a look at it and log it in my already cramped mind!

I appreciate your efforts and will let you know how I progress if I decide to purchase - or even make - a device.

Scott.


From: *** <**@live.co.uk>
Subject: roulette software
Date: August 18, 2011 6:45:09 AM MDT
To: <nowsc@yahoo.com>

hello there,my name is harry and i was reading some article on google in which u was talking about a roulette sofware calculating the speed of the ball and the speed of spin to predict in vicinity the landing of the ball..i know its a long time since ur work and probably u havent done this project in a long time but i am very interested as many people would be to know what do i need to put something brilliant as this together to work.what type of software would help me with all the calculations.plz if u have any knowledge of how to put this thing together i would be grateful.all i need is orientation and i will find a software programmer to build it for me.thx in advance and sorry to bother u .u must think what person im i to even ask u such a thing lol.