Notice: 5/8/2017 -- RouletteForum.com is closed to new posting and will remain in read-only mode for 30 days. If you wish to preserve any of your posts, do so soon.

First, I am new here, and joined pretty much for the sole purpose of posting this question. I’ve read through some previous posts and think I might get some info on my question.

I developed a technique for playing roulette that seems to never fail. But, mathematically speaking, that would not be possible. It’s not based on anything stupid like “such and such numbers haven’t come up, or have come up”, or any such nonsense. I based it on actual odds, with an attempt to soften losses over time while waiting for the wins and requires careful bankroll management broken down into very specific levels.

I have tested it on every game imaginable, online, desktop, and even my phone. I limit myself to typical house rules, which most of that software does not do, so that I am not creating an unrealistic event. And I have never failed to turn $100 into a million in a few hours.

I am not going into the details of my technique for two reasons. First, I re-reviewed the math behind it, and it doesn’t make sense why it actually works and so I would rather not get called stupid. Second, I have never used it in real life and so wouldn’t really be relevant anyway.

So what I am asking is basically, is there any substantial difference between these test games that I have played, and actual roulette that I am not aware of? I am a software engineer, and considered writing my own simulator to insure that the randomness is as good as it should be but didn’t see a point in that just yet.

For me, heading to a casino to see if it works is about a 6 hr flight and so to “just test it” without having at least some belief that it may actually behave the same way in real life would be ridiculous. Online isn’t an option because I am in the US.

It’s allways nice to met someone with same hoby, I’m developing a roulette project too
The whole project is based only on matematics, statistics, baysian probability, freqency probability, probability teory and caos teory.
Currently using excel and later if there will be a time I’m willing to create a real worl C++ program that will stand the test of time….
I’m currently working on a forecast modul that is based on mixture of relative frequency and probability of a “series”, so far that works well, I’m not saing it cut’s down a house edge (because that is almost impossible) but alot of time and bankroll can be saved when you have an matematical advantage.

Same as you I’m not going to share my project( until I finish it of course ), same as you I didn’t test it in real world casino (online only) but I will as soon as possible.

Now to answer your question, no the is no difference.
altrought there is probability that house is cheating, anyway house/casino don’t have to cheat their custumers because they allready have and advantage of 2,7% or more.
altrought there is even more probability that online casino cheat, they don’t have to do so because 80% of the players have no clue about probability teory. house will allways make money(with or without cheating).

If I understood your title, you’re saying that you’ve developed a system that don’t harash your bankrol? so that you almost always make some playmoney, and don’t loose.
Well that’s possible. BUT
continue playing and your overal profit will be less then 0$ in a long run.

A simple proove is probability of a distribution.
from begin freqency of winings and loosing is very spread out, longer you play it converges to the limit of 2,7% (european roulete).
Amount of time(spins played) required to proove this is approx. more then 1000 spins.
also assuming you’re playing all the spins in a row(so that RNG doesn’t changes—for online games)

The same rule aplies to real casinos, but we need some money if want to prove that.

I belive you don’t know what to do next to improve your research(same as me), so maybe we can share our researches and help each other to finish that sooner, what say you?

Exactly. The system must lose at some point because it is a requirement of the probabilities involved. Not to mention there is one aspect of it that makes no sense mathematically to me once I examined it closer.

The online testing that I did was in the freebie/non-casino games that you can play online, since I can’t play in online casinos. I was thinking it more likely that the games designed around roulette actually slightly weigh the odds in your favor behind the scenes to keep the game interesting and fun for players.

That is the only explanation that makes any mathematical sense to me at this point. The bankroll aspect defines betting levels and loss acceptance to assist in allowing a sustained loss streak to be acceptable, but turns over wins enough to constantly move forward.

Well, I went ahead and wrote a small program to test my approach. With different results than what I had experienced in games. What I realized was that most of my success in playing has to do with managing my cash reserves. I haven’t fully implemented the personal rules that I follow, so the results are less than impressive when ran through my program. But that actually makes me happy because I hate things like that when they can’t be explained logically. But apparently it has more to do with money management than any improving of the odds, which does make sense to me…

I’m not sure what exactly you mean by money management, I suppose you are talking about controling a progression… because there is a lot of thricks I found by drawing a various progression/payout lines in excel.
in some progressions possible profit goes up and down differently while progression goes as expected or vice versa.
normaly there is no way to relize that at roulette table, means your perception mostly isn’t corret while playing without using some kind of program or sumulator.

In short, I belive that answer to succeseful money management is considering correlation degree between progression and possible payout.
means correlation must always be positive (between 0,01 and 1) to make profit.

Yeah, that is close to what I mean. I calculate betting level (amount of bet), bankroll, stash amount, and up/down position to determine next move. Next move could be anything from double, reset, downgrade standard bet level, upgrade it, etc…

It seems that I actually have far more “luck” with using this approach to manage loss, and capitalize on profit. I sincerely doubt now that what I do has anything to do with chosen bet’s beyond the fact that I need to have the potential win within reasonable odds (e.g. betting on a single number or single outside bet would not work for me).

So my success in these games has more to do with tricks I use to handle my cash on hand than it does any “betting” mechanism.

betting on a single number or single outside bet would not work for me).

If you make an array of values in your program where each unit of that array will be assigned to one number on roulette weel and hold relative frequency of that number, then you have answer to this problem :D

that means now you may filter out numbers with lowest frequency from that array using one or more desired criteria.

Also according to probability formula to loose n times in a row, probability of a particular number, split, street bet etc… differs alot!
but this holds only for a serie of looses, means once reference number comes out probability of that number is reseted, and to come out once again in a row is not calculated by the same formula.

for example:
if red comes out 3 times in a row then probability for black to come out next spin is as follows:
1 - (18 / 37) ^ 4 = 86,45 %

but funny (and relistic) thing about this formula is that it never reches 100,00% :D

for example:
if red comes out 3 times in a row then probability for black to come out next spin is as follows:
1 - (18 / 37) ^ 4 = 86,45 %

but funny (and relistic) thing about this formula is that it never reches 100,00% :D

Except that probability doesn’t change. I am not sure where you got that from, but there is no change in probability, regardless of win/loss. The probability for every spin resets on every spin. It is just as likely that 27 will come up 20 times in a row as anything else. And the only reason that people think otherwise is because recognizable patterns are easily spotted. That, and the fact that any wheel tending to show obvious bias would be shut down. But the probability of any recognizable pattern showing up is exactly the same as an unrecognizable one.

The ball doesn’t care where it landed last time. And you could just as easily get 200 straight wins off the same bet as 200 straight losses. Now, in practice, because short runs do not properly show distribution, it may appear otherwise, but “appearing” otherwise, and being otherwise, are two very different things.

On the code comments, outcomes in a machine have no bearing even on one another within a run, let alone on a wheel. I know how to code what you describe in less than a few minutes, and it is an absolute waste of time to do so because it would result in nothing useful.

Except that probability doesn’t change. I am not sure where you got that from, but there is no change in probability, regardless of win/loss. The probability for every spin resets on every spin. It is just as likely that 27 will come up 20 times in a row as anything else.

this is partialy true only because result of formula which I provided never reaches 100% ( as I said ).
I’m trully sory that you don’t understand probability teory :(
And about where did I get this formula from, well, my matematical knowelege comes from my job, experience, scool etc…
I would like to provide you some PDF books so you can see the truth but they are not in english, however here is what I found using google to prove what I said: http://askville.amazon.com/formula-calculate-series-probabilities/AnswerViewer.do?requestId=6567080

escher - 01 April 2012 08:24 PM

And you could just as easily get 200 straight wins off the same bet as 200 straight losses. Now, in practice, because short runs do not properly show distribution, it may appear otherwise, but “appearing” otherwise, and being otherwise, are two very different things.

You are absolutley wrong in this.
probability changes a lot as same number comes out multiple times in a row.
to prove that too here is a link to the most famose “roulette probability” site ( scrool down and you’ll find answer to this ) http://wizardofodds.com/ask-the-wizard/roulette/

escher - 01 April 2012 08:24 PM

On the code comments, outcomes in a machine have no bearing even on one another within a run, let alone on a wheel. I know how to code what you describe in less than a few minutes, and it is an absolute waste of time to do so because it would result in nothing useful.

I don’t dubt you don’t know how to code but you are wrong again by saying that coding this is absolute waste of time.
to prove that too here is another link that shows someting similar to probability of a serie:
A owner of this site is very good in matematics and you can see that probability does changes: http://www.gamblecraft.com/review/roulette/martingale.htm

Many people think same as you but all of those people who think so are wrong.
I’m not bullsheeting, but just saying the facts that many people don’t know.
I can prove that too:

Albert Einstein was first greatest Matemathician who gived most of the time to studdy probability in roulette, and all these formulas and all thse tricky small things have been proven 150 years ago.
However, conclusion from Einstein was that there is no way to beat roulette :D http://www.youtube.com/watch?v=hpufE6xN7OY
but however people are still developing new roulette systems LOL.
are we more smart than Einstein? NO lol.

BUT problem with Einstein was that he didn’t have computer and we do have it!

Math is math, and anyone who says probability changes is either wrong, a liar, or a scammer. You can ask anyone who actually has a math degree. Roulette can’t be beaten, which is why it never has been.

The only thing that can be beaten is our own nature which is why money management during play is so important. That combined with a thorough understanding of the legitimate odds and probabilities can result in more consistent profits.

As far as the program that you mentioned. A buddy of mine did in fact code something similar for lottery numbers. He knew better but his dad talked him into it. And he discovered what he already knew, over the short term, you will see favoritism of numbers (though without significant meaning or repeatability), however, over the long term, you will always see an equalization of distribution. Always, no exceptions.

The code that you mention is literally less than 20 lines of code in my sleep. Don’t you think that if doing such a simple task could predict anything, every programmer out there would have already done it and become a millionaire?

In addition, RNG’s are not random in the least (it is impossible to generate a real random number). They have the appearance of randomness through clever tricks. And roulette numbers are not random either, they are the culmination of a million imperceptible variables based on physics. If the dealer spins 2 seconds sooner, the number will be different. If he spins slightly more forcefully, different number, etc etc.

You want to find meaning where there is none. Although, you have given me an idea. If I wrote a program to do “number tests”, and advertise on forums like this one, I might just make a fortune off of roulette…

One simple question that proves probability changing in roulette:

What is more possible?
1. same number comes 10 times in a row
2. same number does not came 10 times in a row

Well you know the answer, obviously not all the numbers have same probability all the times.
this is because of time (4th dimension according to Einstein) Time is the main factor and time allways changes everything.

————
Anyway, if someones answer to above question is that both situations have same possibility then ask your self how many times could that hapen?
One time, several time or all the time?

Knowing that you gotta ask you self only one question :D

What is more possible?
1. same number comes 10 times in a row
2. same number does not came 10 times in a row

The odds of one number coming up 10 times in a row is 35 to 1, repeated 10 times.
The odds of any other number coming up except that number is 1 to 35, repeated 10 times.

Therefore, comparing one number to “not that number”, unfairly weighs toward any other number of course because it has a significant advantage.

(Not counting 0’s)
You are comparing the odds of 35 numbers to 1 number vs 1 number to 35. It has nothing to do with probability changing over time.

However, if you want to ask what is the likelihood that the number 25 will come up after the number 25 instead of the number 27. 35 to 1 either way. Exactly same odds.

However, this kind of misunderstanding during observation is what leads people to believe as you do. And pretty much makes my point for me.

It was my pleasure chating with you escher, now it’s time to sleep (+1 UTC here LOL)

Every spin is a new event. The odds do not change period. Someone once said you would never get this sequence. 1.2.3.4.5.6.7.8.9 in roulette. Why not? It has the same odds of coming up as any 9 number sequence. Playing roulette blindly….well yes, it is not beatable.

if red comes out 3 times in a row then probability for black to come out next spin is as follows:
1 - (18 / 37) ^ 4 = 86,45 %

Known as The Gamblers’ Fallacy, this logic has been refuted in a number of ways. I will explain my favorite approach, which in logic is known as reducto ad absurdum.

First, I have a problem with your calculation. Grinding your numbers I get 94.4%, not 86.45%, but no matter.

Second, what you are trying to calculate is the probability of getting at least one black number in four spins. Because of the green 0, that is not the same as getting four reds in a row. Hence your formula should be 1 - (19 / 37) ^ 4 = 93.0%.

So, you conclude that if the probability of seeing at least one black in four spins is 93%, and you have had no blacks in three spins, the probability of getting black on the fourth spin is now 93%. But how did you get that probability? By assuming the probability of non-black remained constant at 19 / 37 = 51.35%, and therefore the probability of black remained constant at 48.65%! In other words, you use a constant probability to show that the probability varies. This is a logical absurdity.

Skuler,
Yes you’re right, this is known as gambler’s fallacy.

First thank you for corecting my equation.
I didn’t meant so (tough it looks so), what I meant is playing a serie of let’s say 10 spins on same color or whatever, then this formula is used calculate the probability of wining n-th spin (or loosing n spins in a row).
Yes Red in this example have the same probability of coming out every spin, but in probability teory to calculate n-th same event occuring in x-th trials this formula is used.

That’s what I said about what are the chances that for example number 17 comes 20 times in a row, well obviously this can be very very rarley hapend!, but however this hipotesis is just hipotesis because there is no such thing as 100% statistical significance (or 100% accuracy) in statistics.

Online roulette is also very popular in Australia with many of the internet casino brands directly targeting Australians, resulting in a huge choice of roulette sites available to Australians. CasinoReef recommends Ruby Fortune casino as one of the top online roulette casinos available in AUD. For New Zealanders, CasinoKiwi recommends Jackpot city casino as the most popular choice for new Zealanders playing roulette online.