Roulette Beating Random by Betting Random

I think researching and finding out what the safest roulette strategy is, play that, then bring out the big gun when the time comes,which only has to be used very briefly.. ..is the way I'm going at the mo

 

Nathan, you say you want proof of this strategy working in the B & M casinos. What's the matter with you? Do you not trust the math geniuses and the computer gurus that insist if it's proven to fail on a computer it cannot possibly win in a casino. They also claim that if it does show positive results after 1,000 spins, it needs to be tested for 10,000 spins and if that should happen to show positive results, it needs to be tested for a million spins, then a billion, then a trillion or until the programmer passes away.

 

I suspect there might be an error with his simulation.
It seems too good to be true.
Imo packing a decent amount of profit in a few million spin cycle is a HG.

 

Mo offence . Give me the REAL thing . I am not a key board jockey but I am a B& M player . accepting winnings and losses with the same grace .


Casino experiences Atlantic City Connecticut casinos , Montreal , Niagara Falls ONT., Las Vegas, Laughlin , Nassau Bahamas, Bad Homburg Germany .




I think I earned my stripes as a recreational player .


;
ND

 

You may be right about him making an error with his simulations. He's done that a couple of times. Read to some of his previous posts. However, the strategy is quite simple, I would be curious to hear what errors he may have made if any. It would also be interesting to see if Mac made the same errors getting similar results. If these two dudes did make errors and achieved those results, then in that case, all we meed to know is what the errors were and play the strategy that way to achieve results like that.

 

Yes, there was a bug. A very small one but it meant that the correct number of busts wasn't counted. Everything else is fine and here is the result of another 30k test :

Number of busts : 11278
Number of +50 sessions : 18722
Ratio of successful sessions to busts = 1.660
Computed Edge (PL / Total Staked) : -0.0264

Previously I only noticed the negative edge so I assumed that the output was correct, but if I'd done a simple check with the arithmetic I would have realized that it was nonsense since even if the busts occurred right from the first bets (ie -128 units down), there would have to be more busts than what was shown to make an overall loss. The computed edge was correct because the losses were being added to the total, it's just the number of busts was wrong.

The error was in line 76 which was if stake > 128. It should have been if stake >= 128. Just one character threw the whole thing off. Well, that's coding, it can drive you crazy sometimes.

Normally I would have created a file which shows more detail and output, so that you could check whether there were any errors, but I didn't have time to do that yesterday. I'll do it at the weekend and also allow for the option of changing the progression length and win target, maybe also an option to use a D'Alembert as well as a Marty.

In fact, maybe I should try Bombus' suggestion of using a die roll to pick the even chance; that would make the selection more random.

 

I think you may have missed the point here. Granted, the number of busts is wrong because of the aforementioned bug, but the positive edge of +2.23% is correct. That doesn't mean the system is a long-term winner, it just shows that in a shorter run (even 231 sessions can be considered a "short" run) you will get negative and positive fluctuations.

 

Jerome, if you may add the option to only play the game where is a positive trend going and avoid the negative parts we would see the nonsense being perpetuated here, that you can somehow outsmart negative variance of random by only playing in an upswing.

 

You claimed different number of played sessions in the past. It was 321, not 231.

 

Yeah that's what I thought - 321 sessions. Looks like I'm not the only one making mistakes.

 

Yep, it's definitely on my to-do list, debunking that nonsense. The trouble is we have to agree what a trend is otherwise the advocates will just come back (after I've shown negative results) and say it's not a "true" trend. There are many ways to define a trend. Or they will say it can't be coded (the ultimate get-out clause).

The only thing which can't be coded is precognition, so I guess that let's you off the hook.

 

Actually Jerome, I appreciate the amount of time and effort you have put into this. I really do, even with your negative conclusions. And of course, it's easy to make mistakes when coding, it can be very confusing and frustrating. The fact that you are willing to put more effort into this shows that you feel that it could possibly have some merit; otherwise, you wouldn't give it an ounce more of attention. A lot of things considered impossible to accomplish in the past started with an idea that wasn't perfect but a good start and close enough to build upon it; and eventually what was considered impossible was defeated. Who knows, you could be the very guy who figures out how to get a real edge over an EC game.

Let me make some suggestions. You should actually do your coding for Baccarat. It offers better odds than Roulette. If you showed positive results with Roulette, then Baccarat should yield even better results.

You need to determine what would be considered a reasonable long run. As it is, you ran 30,000 sessions. If a person played one session every day of the week, that would amount to 80 years of playing. 30,000 divided by 365 = 82.19. That is way beyond reasonable.

As far as ways of random bet selections, I have used a die also. Even numbers were bet Banker and odd numbers were bet player. That can be reversed, it doesn't matter. I have also used a random number generator for bet selection that produced a random series of 1's and 2's. 1 bet player, 2 bet banker. All that matters here is that the selection is truly random.

I believe you will not find any merit to the D'Alembert progression, or any other for that matter. The main problem I encountered with the D'Alembert was it takes a lot of plays sometimes to recoup back to the last highest gain. I would be placing 250 unit bets at times to recover a 1,500 unit loss. I would eventually recoup, but the downside losses were just to much to make me comfortable. However, it still may be worth a try. Regarding the Martingale, the more steps the better the odds, and of course the higher risk. You may want to figure out what the highest unit value you can use verses the most Martingale steps that can be taken that stay within the maximum table limit.

Bottom line, try everything and anything you can think of. Who knows, you may turn out to be the genius of geniuses

 

I think his long run test is reasonable as it shows the inevitable conclusion of the negative EV.
The way to look at this if the system had been used by 100 players, how many of them would have lost on avarage.

I think modifying the marti progression to just add 1 step between to break even would be interesting.
Basically 1-1-2-4-8-16-32, or a short fibonacci.

 

Here's what needs to be done to satisfy all the questions and possibilities here. Have a hundred million people do a hundred million tests with a hundred million variations of the strategy playing a hundred million sessions over a hundred million years. And if that doesn't do it, then a billion people etc.

 

There’s no need for that I’m not sure why all this testing it’s fun but useless.the maths is there designed for casino s to have the edge in the long run . the more you play the less you win .whales win all the time why do you think the casino keep comping them to get them back playing so they lose even Baccarat you will lose the 5% commission eats your bank roll.not true even money.billion or trillion test useles
It’s the max bet you remove and have a great strategy and a good enough bank roll and wala you beat the even money.casinos figured that 200 years ago

 

Don Johnson beat the hell out of them look him up on YouTube he simply ask them to Change the rules in black jack .
There was a recession in Vegas so stupid casinos were desperate and agreed and look what happen .ihave always said in life Generally he who makes the rules has the upper hand end of story.

 

I looked up Don Johnson on Youtube as you suggested. Very interesting. One thing I learned was he put a lot more effort into learning how yo win than most people will ever be willing to do. He also mentioned that Baccarat gives a player the best odds over all others. I think the reason he preferred Blackjack over Baccarat is because Baccarat is such a brain dead game lacking any kind of skill involved. Plus, I think he could recognize trends better in Blackjack. Yep, the best way to learn how to be a winner is to listen to the winners. Everybody else has nothing much to offer.

 

Well, actually, it was more to do with the fact that you claimed to have won with it over such a long period that got my attention. These "random versus random" kind of systems have been around for a long time, but I expect you knew that already.

Anyway, anyone who wants to experiment with different parameters can play around with the attached software. Nothing fancy, it runs in a Windows console. You can set the length of martingale, win target and number of sessions, and also run the program multiple times.

Here's an example with target of 15 units, maximum stake 32 (5 step marty) over 50 sessions with 4 lots of results generated :

 

thanks for sharing software, but it crashes itself after i hit enter. Yesterdays software also didnt install properly and nothing happened

 

You want to bet.
Change the rules and you’ll see it’s been proven by mr Don Johnson