Quote from: Bayes on February 14, 2014, 11:41:13 AM

Hi maestro, don't change anything other than the two numbers I indicated. If you do either the program will crash or you'll get unreliable results. I realize this isn't the best solution, but it's better than nothing until I write a javascript version with a GUI.

i get this:


WARNING: temporary files found! Do you want to delete them (y/n)? Exiting...

Quote

hi, using Bayes program ran jan-1 jan-10 spins from Wiesbaden here is some info:


each day has about 300 spins

Attached is a simulation of the 3 in 30 and a 4 in 30 qualifying bet.   There are two separate simulations attached. I used numbers from random.org. It's a very good RNG.  10k trials were used.  Millions of simulations can easily be run, but there's no point to it.  As we run more trials, the results will approach the long term expectation of -2.7% on the single zero wheel.  It's true that there would be the occasional lucky winning 10k spin sample, but the sum of all the samples will still be a net loser.   After all why should we expect anything different when using an RNG?

The horizon is set at a rolling horizon of 30 spins.  There is no stop loss limit, because there's no reason whatsoever to have one if you're going to be using RNG numbers. Using a stop loss limit would produce the same results as not using one. The difference would be that you would not be betting on as many spins.  As your test trial sizes increase, you will discover that this is the case.

On the live wheel, the stop loss should be based on whether or not it's the same dealer.  Since you don't know whether it's the same dealer in your current simulations, and since you don't know if the playing conditions are the same from one spin to the next, the stop loss will not change your long term expectation in the simulations.

QuoteEssence of the WF game is trap as many winners at that peak of the bell curve and minimise  your losses in that endeavour so to work ever more efficiently. As the curve moves, sways in a range of behaviour you need to note its limits but where it is most fertile and attack there, especially when its moving into optimum phase. That is really exciting work and very rewarding. Enjoy the chase and the quest for that perfection. Go for it! -XXVV

XXVV,

The money management part of your method (stop loss) isn't doing anything to help the edge.  Especially on RNG wheels for reasons that are described above.  RNGs do not produce "fertile" moments or a "range of behavior" that you can capitalize on.  So the "chase and the quest for perfection" is nothing more than an exercise in futility.
I suggest you stick to the live wheels, and use some real stop loss indicators, such as the one partially described above.


The result of the 3 in 30 test over 10k trials (8924 bets) was

Edge -3.17%
Loss of 537 units


It's very close to expectation. 




-Xander

Hi Xander


I think you're going off topic again.....WF is not 3 in 30 or 4 in 30....The mini games are either 3 spins long or rarely upto 35 spins before stop loss kicks in at 5th Target....this of course is an example of stats from WF4 which can be a bit more riskier due to targeting a maximum of 4 numbers


I know XXVV mentioned about (3 in 30 and 4 in 30) but that's not the core of the WF method, so I'm confused that you're ethos is to discredit what is a proven basic method of playing Roulette 


For others who think outside the box with their glass always half full, continue to study hard and learn every inch of a strategy so nothing can surprise you and never play except with a positive winning attitude instead of thinking that you're going to lose when you play (Law of Attraction) 


@Biagle....Positive results from your example which are typical with this method 


Gordon

Gordonline,

It's close enough for reasons which I've already described in the post above.

@ Xander,

XXVV is right, and you've had a good run. I'd like to keep this thread on topic and I still have more testing to do. In fact, although there are no bugs in the code, I've just realized that I've left an important rule out of the simulation (so all those who have run the program at compileonline.com, take the results with a pinch of salt).

I know you think that all "hot number" systems are basically the same, but I'd like to keep the thread specific to this particular one, which is the one I'm actually testing.

Thanks for your cooperation.

"However most items you (Xander) have posted seem to want to discredit someone or other, and yet promote your own views" >>> Ohhhh, he's just getting warmed up.

Ken

Here's a short sample from a file generated with the updated program. I've now incorporated VP (meaning virtual play), which means that if you've had no win after the 3rd repeat (a la WF3), play continues "virtually" until a number hits a 3rd time. 

  12        +0 
    1        +0 
   26        +0 
   31        +0 
   30        +0 
   24        +0 
    1        +0 
   22        -1 1
   16        -2 1
    9        -3 1
    3        -4 1
    8        -5 1
   21        -6 1
    1       +29   #1 is a winner. Stop game.
      *****
   10       +29 
    6       +29 
   15       +29 
   30       +29 
   12       +29 
    5       +29 
   14       +29 
   21       +29 
   10       +29 
   31       +28 10
   32       +27 10
   17       +26 10
   29       +25 10
    2       +24 10
    8       +23 10
   31       +22 10
   30       +20 10 31
   30       +53           #30 is a winner. Stop game.
      *****
   30       +53 
   27       +53 
   11       +53 
   35       +53 
   24       +53 
   17       +53 
   15       +53 
   34       +53 
    2       +53 
   20       +53 
   27       +53 
   28       +52 27
   18       +51 27
   26       +50 27
   33       +49 27
   11       +48 27
   34       +46 27 11
   20       +43 27 11 34      3 numbers have repeated, go into "virtual". Note bank stays constant.
        VP
   16       +43 27 11 34 20
   30       +43 27 11 34 20
    4       +43 27 11 34 20 30
   32       +43 27 11 34 20 30
   31       +43 27 11 34 20 30
   26       +43 27 11 34 20 30
   19       +43 27 11 34 20 30 26
   30       +43                               #30 has hit a 3rd time, stop game.
      *****
   31       +43 
   18       +43 
    2       +43 
   31       +43 
    3       +42 31
    4       +41 31
   16       +40 31
    7       +39 31
   31       +74            #31 is a winner. Stop game.
      *****
   18       +74 
   19       +74 
   21       +74 
    8       +74 
    8       +74 
    3       +73 8
   33       +72 8
   13       +71 8
   22       +70 8
   28       +69 8
   22       +68 8
    4       +66 8 22
   33       +64 8 22
   10       +61 8 22 33
   34       +58 8 22 33
   31       +55 8 22 33
   35       +52 8 22 33
   30       +49 8 22 33
   22       +82                 #22 is a winner. Stop game.
      *****
   16       +82 
   29       +82 
   28       +82 
   24       +82 
   27       +82 
   27       +82 
   27      +117          #27 is a winner. Stop game.
      *****
   26      +117 
    1      +117 
   24      +117 
   23      +117 
    8      +117 
    0      +117 
   17      +117 
    9      +117 
    8      +117 
   23      +116 8
   36      +114 8 23
   33      +112 8 23
    5      +110 8 23
   24      +108 8 23
   28      +105 8 23 24
   11      +102 8 23 24
    0       +99 8 23 24
        VP                           3 numbers have repeated, go into "virtual" mode.
    4       +99 8 23 24 0
   19       +99 8 23 24 0
   34       +99 8 23 24 0
   28       +99 8 23 24 0
    6       +99 8 23 24 0 28
   20       +99 8 23 24 0 28
   13       +99 8 23 24 0 28
    0       +99                       #0 has hit a 3rd time. Stop game.

Hi XXVV,


Thanks. Hopefully the program now reflects the TRUE method, but I'd still prefer to check it against some known results before running it on a larger spin file. I might get in touch with Gordon about that.

Quote from: XXVV on February 15, 2014, 10:21:19 AM
simply run the 14 completed games on Wiesbaden#3 for January 1 2014. I have a manual check on that and Biagle did an earlier version which differs a little.

Ok, I'll do that first. 

@ All,

I've amended the code in reply #63 where I posted the instructions for compileonline.com, so all those who are testing should start again with the new program. Any problems or if any results don't seem right please let me know. Thanks.

Screenshot:[attach=1]

Last 7 weeks am testing this manually on my 26k Dublinbet live spins/which are not in continuation,but I put
it as a whole for sake of testing/and can report that the final score is;2138+...once it hit about 300+never
went back.....Bravo xxvv....and am now making studies how to applay best way in B&M play.

Hello Bayes,


Thank you very much for the updated program.


Regards @ Best wishes

hats off to Bayes and XXVV  did test on 5000 random spins from random.org

hallo maestro

thank for your
test
question
do you use the ew wf3 with virtual play

is it the result of your test
5000 spis  +170 at the end ?? (you don't use virtual play )

it is winning but there alot up and down ansd +170 after 5000 spin
do you think its enough to say it will continue positiv
it is justt a question
as i just read the topic

i just used source code Bayes done and put spins and run the test it should ply by new rules