Author Topic: Interesting Results  (Read 1835 times)

Sheridan44

  • Mature Member
  • ***
  • Posts: 259
  • Thanked: 225 times
  • Gender: Male
Interesting Results
« on: May 20, 2016, 07:10:08 AM »
I've been working on a new project / method which I will present on the forum once I work out the details and narrative. In putting the pieces together, I tested some data that seems very interesting that I wanted to share.

I've added an image of a graph (below) that shows results of some of my test runs using this particular methodology. I ran 4 sessions of 100 spins each = 400 spins total. Each session result is overlaid on the graph, with the blue line being session 1, red being session 2, yellow session 3, and green being session 4. I also created a 5th line representing the average of the 4 - shown in dark purple.

The session lines remained fairly close together up to and around the 24th spin of each. Then they all began to diverge quite wildly. Sessions 1 and 3 drifted above the break even point (5000) for awhile, while sessions 2 and 4 started sinking significantly. Although all went negatively tilted eventually, the short time frame showed relative stability. What really caught my eye was the divergence of each started at nearly the same point (24th spin). You would expect them to scatter, but all at nearly the same point and time? Interesting to say the least. I've got some theories on this - but they would just bounce around in my head.... I'd love to hear from the members and get some input on this. One thing that came to mind is with such a narrow window at the beginning of this divergence, some loss limit triggers (i.e... tipping points where recovery becomes highly unlikely) could be devised from such. The betting method is non-progressive, and does not vary.

I prefer to use and compare multiple shorter term data as opposed to one long graph that stretches to infinity, as they seem more applicable to casino (and real life) situations.


« Last Edit: May 20, 2016, 08:54:04 AM by Sheridan44 »


 
The following users thanked this post: Reyth

Reyth

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3318
  • Thanked: 979 times
Re: Interesting Results
« Reply #1 on: May 20, 2016, 09:47:01 AM »
This is very helpful thank you!

Where did you get your spins from?
 
The following users thanked this post: Sheridan44

Sheridan44

  • Mature Member
  • ***
  • Posts: 259
  • Thanked: 225 times
  • Gender: Male
Re: Interesting Results
« Reply #2 on: May 20, 2016, 11:08:19 AM »
Thank you Reyth.....

I don't have any savvy roulette software (yet), and would be interested in purchasing some soon. Could you suggest a really good one?

This data was from the Roulette (Royal) Panda site, I'm sure they use a RNG and hopefully it's reliable and accurate. I'm going to do the same test with the Wiesbaden Live results. I've also used Random Org occasionally (but not for this yet). I'm going on the assumptions that these outlets are on the level and are spot-on. Are there any other outlets I could pursue to get good numbers?
« Last Edit: May 20, 2016, 11:46:23 AM by Sheridan44 »
 

Sputnik

  • Veteran Member
  • ****
  • Posts: 444
  • Thanked: 362 times
Re: Interesting Results
« Reply #3 on: May 20, 2016, 11:31:03 AM »

Hello, if you can code i would suggest Roulette Extreme from UX Software.
They have a very good tutorial about coding system to the simulation software.
And you can load actual spins from different real casinos or play rng.

Cheers
 
The following users thanked this post: Reyth, Sheridan44

Sheridan44

  • Mature Member
  • ***
  • Posts: 259
  • Thanked: 225 times
  • Gender: Male
Re: Interesting Results
« Reply #4 on: May 20, 2016, 11:34:14 AM »
Thank you Sputnik. I appreciate it!
 

Trilobite

  • Veteran Member
  • ****
  • Posts: 300
  • Thanked: 98 times
Re: Interesting Results
« Reply #5 on: May 20, 2016, 11:56:01 AM »
Hi Sheridan44
Yes, interesting results.
Not knowing anything about the method itself makes it hard to comment on what might be causing the spontaneous scatter.
Can I suggest now that you started with a 4 session example chart, that you repeat the process at least several more times to see if the characteristics remain similar.
If your future tests also show divergence from around spin 24, then another suggestion would be to also repeat simulations over gradually longer spin samples to see if there is any convergence of the graph lines further along the number streams.
 
The following users thanked this post: Sheridan44

Sheridan44

  • Mature Member
  • ***
  • Posts: 259
  • Thanked: 225 times
  • Gender: Male
Re: Interesting Results
« Reply #6 on: May 20, 2016, 12:38:28 PM »
Really good suggestions Tril.. thank you! Yeah I need to expand this thing out to see if there's something to it, or if it's some kind of fluke. But it is initially interesting. I've been tinkering around with some ideas and methods....I've got the crap scattered all over the place, I just want to gather it all, and do a write up on it, then present it here for comments / suggestions and ideas. One thing that the graph does tell me is that if it's gonna happen (profit), it's gonna happen early on in most cases, and the skill is in seizing it when it presents itself. In all probability if one goes into a "grind" mode, they'll get eaten alive. Everyone knows what to do when winning, the mastery is in handling the situation when it all starts going south.
« Last Edit: May 20, 2016, 01:02:20 PM by Sheridan44 »
 

Trilobite

  • Veteran Member
  • ****
  • Posts: 300
  • Thanked: 98 times
Re: Interesting Results
« Reply #7 on: May 20, 2016, 01:44:15 PM »
...the mastery is in handling the situation when it all starts going south.

And that's why I'm a grinder.

Just about every system and method eventually ends up a grind anyway, so I figured, why not just make a system that sets out to grind.  :)

 
The following users thanked this post: BlueAngel, Reyth, Sheridan44

ernroo1

  • New
  • **
  • Posts: 51
  • Thanked: 5 times
Re: Interesting Results
« Reply #8 on: May 20, 2016, 02:38:08 PM »
 Nice graph, a few more sessions of the same pattern would be interesting.
« Last Edit: May 20, 2016, 02:40:18 PM by ernroo1 »
 
The following users thanked this post: Sheridan44

Reyth

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3318
  • Thanked: 979 times
Re: Interesting Results
« Reply #9 on: May 20, 2016, 03:50:52 PM »
Hey!  I actually just noticed that universal profitability starts at the 19th spin and so the range there would be spins 19-24! :D
 

Sheridan44

  • Mature Member
  • ***
  • Posts: 259
  • Thanked: 225 times
  • Gender: Male
Re: Interesting Results
« Reply #10 on: May 20, 2016, 06:27:47 PM »
Cool Reyth! Yeah, I don't think 24 is a "golden" number, there's likely a range within that vicinity - and 19 to 24-ish would be right around there. Further testing might narrow it down some. From the results above it seems as though rather definite trends are established by about the 30-35th spin. And a loss of around 50 units is a "point of no return" (I'm using 1 unit bets for mathematical clarity), although Session 4 (green) does make a valiant attempt in the latter stages.
« Last Edit: May 20, 2016, 06:45:53 PM by Sheridan44 »
 

kris

  • New
  • **
  • Posts: 39
  • Thanked: 9 times
  • Gender: Male
Re: Interesting Results
« Reply #11 on: May 20, 2016, 07:11:19 PM »
Sheridan44, thank you for a very interesting and useful post, you made a huge work, respect!
 
The following users thanked this post: Sheridan44

Sheridan44

  • Mature Member
  • ***
  • Posts: 259
  • Thanked: 225 times
  • Gender: Male
Re: Interesting Results
« Reply #12 on: May 20, 2016, 07:40:44 PM »
Thank you Kris! I enjoy and look forward to your posts also. You always raise interesting and thought provoking subjects.   

The easy part of this is gathering the data. Now is the hard part - what the heck to do with it, and how to interpret it - LOL. This is why I'm not shy about sharing my works. There are some sharp people in here that are always very helpful.
 

Sheridan44

  • Mature Member
  • ***
  • Posts: 259
  • Thanked: 225 times
  • Gender: Male
Re: Interesting Results
« Reply #13 on: May 20, 2016, 08:37:13 PM »
Here is a side note to the above data....

I constructed some preliminary stop loss triggers according to the data.

My rules were : (1) A 50 unit stop loss in the initial phases of each session... like 1-24 spins or so. (2) Restricting the stop losses to minus 30 off their highs in the later stages.

Session 1... 100 spin total... High of +112 units at spin #47... Stopped at +88u at spin #53......Balance of -76u upon conclusion of the 100 spins.

Session 2... 100 spin total... High of +42u at spin #18... Stopped at +18u at spin #27....Balance of -172u at 100 spins.

Session 3... 100 spin total... High of +104 at spin #34... Stopped at +74u at spin #43... Balance of -14 at 100 spins.

Session 4... 100 spin total... High of +34 at 2 spins! ... Stopped at -12u at spin #8... Balance of -92 at 100 spins.

All sessions totals.... 400 spins total... +168u using stop losses.... -354u not using stop losses.

This seems to bode well for the old saw... "Take care of the downside, and the upside will take care of itself".
« Last Edit: May 21, 2016, 01:42:45 AM by Sheridan44 »
 
The following users thanked this post: Reyth

Sheridan44

  • Mature Member
  • ***
  • Posts: 259
  • Thanked: 225 times
  • Gender: Male
Re: Interesting Results
« Reply #14 on: May 21, 2016, 02:31:02 AM »
I added 2 more sessions of 100 spins each...bringing the total to 600 spins.

All 6 sessions totals (600 spins)...

+214 Units using stop-loss measures

-536 Units just playing it straight through with no stop loss measures

 
The following users thanked this post: Reyth