I've written a simulation to test XXVV's WF3 system. I'm still not 100% sure my code is correct (because the results are so good  ) which is one reason why I've posted a sample here. When I'm sure it's ok, I'll run the sim against a much larger file of spins (both RNG and actuals, to see if there's any difference).

Note that this is a simulation of the WF3 bet only:


Bet on up to 3 numbers which have repeated. A game terminates (start re-tracking) when either a number hits a third time (a win) or a 4th number repeats without any number having hit 3 times (a loss).

UPDATE: Due to a bug in the code which generated the results, they are invalid and should be ignored.

 
Nice work Bayes ...

Cheers

Hello Bayes,


Thanks for sharing the simulated results of XXVV's WF3 system.


Regards

Hello,

I went through the first few games and everything adds up the way it should......

nice job 

It will cut the house edge slightly, but it won't win in the long run.  I've already run such simuations like this one countless times over the years and over several hundreds of thousands of live spins. 

QuoteIt will cut the house edge slightly, but it won't win in the long run.  I've already run such simuations like this one countless times over the years and over several hundreds of thousands of live spins.

Show it to us dear Xander

Short of doing a screen share so you can witness the simulations first hand, I don't know how you could see it.

just put some proofs for your words...

Bayes,

In what program is the simulator written?

I can't seem to open the file.

@ Xander,

Not sure what you mean? It's just a plain text file which you need to unzip before you can read it. The actual simulation is written in a version of BASIC (see below).

@ Bally, thanks for the feedback.

I'm posting the source code because I've just tested the 50,000 spin file and the result was over 25,000 units profit.    My first reaction was that there must be a mistake somewhere, but the code is pretty simple and I'm damned if I can see any problems with it. Anyway, all you programmers please could you look over it. I've included comments so it should be fairly easy to read. Note that this is a simplified version of the program; I've removed some code which marks the end of each game and keeps count of the wins and losses.

Ironically, what I've actually coded isn't what XXVV had in mind but a variation of it. Perhaps I've inadvertently discovered a HG. 

[attachimg=1]

I'm not seeing anything like that.  I'm seeing an occasional profitable run for 30k spins with an edge of up to 7% before it tanks into negative territory on another set of 30k.  The long tem edge is still in negative territory.

The wheel used, and the dealer combinations is going to matter. 

Look deeper for bugs in your simulation.

there gonna be times when it all goes down terribly  . Overall even if you gonna be ahead on different wheels, it won't be too much.

"Ironically, what I've actually coded isn't what XXVV had in mind but a variation of it."

Then shouldn't the code be modified accordingly? To ensure you're testing the  'right' method?

XXVV,

Back in 1984, Frank Barstow published, "Beat the Casino".  Within the book, there was a nearly exact description of this system.  It's nothing new.  Back then, I believe he referred to it as the Pyramid System.  People have been betting "hot numbers" since the invention of the game.  In all fairness, it's one of the best ways to win, since it's a very crude form of biased wheel play, but in the end it's not going to show a profit.
(In short:  Bet on up to 3 numbers which have repeated. A game terminates (start re-tracking) when either a number hits a third time (a win) or a 4th number repeats without any number having hit 3 times (a loss).)

Countless people, including myself have simulated this stuff over a million spins.  And yes, I did say a million.  Some of us have been at for far longer than only ten years.  The overall end result is that it does sometimes work, but in the long run, on the more random wheels it simply won't work.  Your experience is your experience.  My belief is that you've been very lucky, or perhaps you haven't played as many spins as you believe that you have.  Anytime that you're betting just one to a few numbers you will have extended periods of success.


Anytime that you want to see the simulations run live, on my computer using my live spin archives, I'll be happy to show you the results.  I can run the simulations on a screen share so you can see the results first hand.  I'm sure I can find individual wheels where it runs a small profit for 100k spins or so at a time, but not many of them.  Over several different wheels, however, the result is negative.

There's no disputing that playing the hottest numbers is a step in the right direction, and that it's far better than betting on the sleeper nonsense.  But as it is, it's not the holy grail that you believe it is.


-Regards,

-Xander

QuoteI developed my own method for hot numbers called 3 in 30, and prepared variations finally settling on 3 in 35 and used an IBM mainframe computer to test millions of RNG spins with major partnership from a trusted and dear colleague. I did this 20 years ago. -XXVV

XXVV,

Sorry, but my testing results don't show anything of the kind.  I don't want to say that you're exaggerating, but my extensive experience on more spins tested shows that your results are... unique to your testing.  Perhaps there was an error in your program?  If anyone wants to actually see the simulations run, then they can contact me.

There's nothing novel about 3 in 30.  It's simply betting any number that's hit 2 to 3 times above expectation in a horizon of 30 spins.  It's like a very crude biased method that won't quite work well enough to win in the long run.

Also, the Frank Barstow progressions have no strength.  They're only value is for entertainment. 

By the way, if you don't like constructive criticism, then perhaps you shouldn't be posting on a public forum.  Maybe try sticking to a blog.


Best of luck,


-Xander