Author Topic: Testing Actual Spins vs RNG  (Read 8983 times)

Scarface

  • New
  • **
  • Posts: 34
  • Thanked: 30 times
Re: Testing Actual Spins vs RNG
« Reply #15 on: March 11, 2016, 01:26:00 PM »


Yes a real wheel and table will be the true test.   Should get alot of real play over the next few weeks.  I really think the holy grail, if it exists, is more about money management. 
 

Scarface

  • New
  • **
  • Posts: 34
  • Thanked: 30 times
Re: Testing Actual Spins vs RNG
« Reply #16 on: March 11, 2016, 01:30:13 PM »
Been watch a certain casino for a month. Generally il watch for 4weeks playing my bets on paper before im confident with them. Ive noticed with this new one ive been watching. Number will come out and on player stats will replace the previous number out. Lastnight notice 20 odd numbers player stats behide true stats. They r not even warning players of technical difficulties. Plus numbers in stats bar missing, never showing and numbers there which havent even been out yet. Sorely this is illegal? Ive taken tons of videos and photos.

Makes you wonder if the player has an advantage knowing past spins, or why would they display false numbers.  The casino I play shows the last 20 hit, but have always seemed to be accurate.  Could be a technical glitch, but they should warn players about it
 

Scarface

  • New
  • **
  • Posts: 34
  • Thanked: 30 times
Re: Testing Actual Spins vs RNG
« Reply #17 on: March 22, 2016, 09:55:18 PM »
So, I thought running 30+ tests to 200 spins each would be sufficient.  Every test was in the money averaging between 200 to 500 profit.  Had only one loss of 250 (which recovered to profit playing 20 more spins). 

Live play on a real wheel was a different story.  Played the system exactly as I did on RNG, and lost $1100 in less than 200 spins.  In all the testing I did (over 6000 spins) I never seen anything close to this.  I'm beginning to think testing is just a waste of time.   :(
 

Reyth

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3010
  • Thanked: 795 times
Re: Testing Actual Spins vs RNG
« Reply #18 on: March 22, 2016, 10:59:02 PM »
It has been said by some that until a system hits 10000 spins, you won't have a really good idea of what its capable of encountering.
 

kav

  • www.Roulette30.com
  • Administrator
  • Hero Member
  • *****
  • Posts: 1531
  • Thanked: 580 times
  • Gender: Male
Re: Testing Actual Spins vs RNG
« Reply #19 on: March 22, 2016, 11:11:40 PM »
Very interesting post Scarface!

Testing is not exactly a waste of time. But it is not everything either.
Someone is going to tell you that your testing was insufficient. But this is not the real problem. The important thing is the methodology. This is how I do it:
  • First understand the system. What are you doing and why?
  • Then understand the systems weak points. What is the worse case scenario? (you will see that when I present a system I try to point out its strong and weak points)
  • After you have identified the weak point, test it against such unfavorable sequences. In some cases, it is useful to not use any RNG but you manually create some bad sequences and test the system against them.
Strange things happen in roulette: numbers go missing and numbers go hitting well outside their expected average. This is called variance and this is mostly what kills the player, unless he has determination, patience and huge bankroll.
« Last Edit: March 22, 2016, 11:25:24 PM by kav »
 

kav

  • www.Roulette30.com
  • Administrator
  • Hero Member
  • *****
  • Posts: 1531
  • Thanked: 580 times
  • Gender: Male
Re: Testing Actual Spins vs RNG
« Reply #20 on: March 22, 2016, 11:20:35 PM »
It has been said by some that until a system hits 10000 spins, you won't have a really good idea of what its capable of encountering.
Yes, practically speaking, 10K spins is a good test to evaluate a system. But to give you an example, a 9-step Martingale could pass a 10K spin test (betting contra-Last: the different than the last number), yet in your next visit to casino fail miserably.

It is very important, even before you start testing to know what is the weakness, to know what you are looking for.
« Last Edit: March 22, 2016, 11:25:01 PM by kav »
 

Geoffrey

  • New
  • **
  • Posts: 120
  • Thanked: 74 times
Re: Testing Actual Spins vs RNG
« Reply #21 on: March 23, 2016, 12:31:53 AM »
very interesting posts over here (way better than that hex thing) i agree with several things that have been said here, but i'll elaborate further on this tomorrow (its 1.30 in the morning over here so really need to sleep, i'll modify my post when i wake up guys.
 

Reyth

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3010
  • Thanked: 795 times
Re: Testing Actual Spins vs RNG
« Reply #22 on: March 23, 2016, 01:18:16 AM »
Ok, see ya! :D
 

Scarface

  • New
  • **
  • Posts: 34
  • Thanked: 30 times
Re: Testing Actual Spins vs RNG
« Reply #23 on: March 23, 2016, 10:40:48 AM »
Thanks for the input guys.  The last 2 times I've lost in a B&M casino, I tested my system for st least 5000 spins.  Maybe more testing is needed.  Although the systems were different, both focused on cold numbers.  Seems like cold numbers hit more frequently on rng, while hot numbers hit more on a live wheel.
Kav, you definitely hit the nail on the head with  what you said about variance.  The house edge is almost insignificant in less than 200 spins.  Variance is the real killer.  I was thinking if someone was playing the opposite way as me, they would've come out a big winner.
 

kav

  • www.Roulette30.com
  • Administrator
  • Hero Member
  • *****
  • Posts: 1531
  • Thanked: 580 times
  • Gender: Male
Re: Testing Actual Spins vs RNG
« Reply #24 on: March 23, 2016, 11:20:27 AM »
I was thinking if someone was playing the opposite way as me, they would've come out a big winner.
Very good point.

As for cold numbers in B&M casinos, here is what my experience says.

First of all there are two totally different stories if you bet on many numbers (12 numbers or more) or few numbers (5 numbers or less). The appearances, the cost of each lost spin the profit of win, everything is different. So they require a different betting method and approach.

One thing I have observed when betting on few cold numbers is that they usually come in clusters. Say we bet 5 numbers. it is more probable that we will have long periods of misses and a few very close hits, than it is to see the numbers hit, for example, in 8 spin intervals. This is not absolute, and they can come in standard intervals, but most of my experience tells otherwise. This is something I truly believe in and I have tried to take advantage of.
 

Geoffrey

  • New
  • **
  • Posts: 120
  • Thanked: 74 times
Re: Testing Actual Spins vs RNG
« Reply #25 on: March 23, 2016, 04:26:28 PM »
 First off all scarface 23 consecutive wins sure does look promising. Worthy of heaving a closer/ deeper look at it. As far as the RNG vs Live wheel results  (which is better, more trustworthy)I’ll pass on that subject. Personally i haven’t had that much trouble with an RNG, but having heard stories about other players and Rng issues, i admit i approach an Rng with caution. Although i haven’t had many bad runs on it. But i guess more experienced players are better placed to have a clear opinion on it.

And surely like harryj mentioned, i you need help with testing something i do recommend Reyth, testing roulette systems/ strategies seems to turn him on hehe ;)
On the other hand testing is just testing tho it sure gives you an impression of what you can expect, but i agree with scepticus on this one, if your really want to test it out, get to a casino , put money on the table and real wheel.

This puts you in a different setting were you are under pressure, its not only a test of your system, but also a test for yourself, how do you cope when things seems to go the wrong way, can you keep your head cool, when do you decide to quite, do you become careless when on a winning streak, and so on.

I also believe that the holy grail, if any , is more about risk and money management, patience and discipline. I think those are more important than the system and betting selection itself. We’re clearly on the same page on that one.

And also have a closer look at kav has to say. To be honest, that's pretty interesting stuff he is telling.
And if you test something, before you even start, try to identify what kind of gambler are you?, what kind of risk profile you are comfortable with?. Before looking into what system/ strategy you want to use or test, i would first find an anwser to those questions. And work on your money management skills and mental thoughness. I cant speak for everyone, but a loss of 1100 is an amount nobody should be comfortable with, so try to tweak your system a bit with a more defensive approach.

There often is said that the best deffense is a strong offense. But when talking about gambling you could easily turn that around. Imo the best offense is a strong defense.

Most gamblers have a mindset of wanting to win , everybody does, but there is a little difference in approach of the following, and this is purely my believe, not something a heard from other players.
The main objective is not trying to win. The main target should be not to lose anything of your bankroll.

When you havent lost anything of your bankroll , you’ve actually have won. Yes it gives the same result eventually , but it simply is a different approach and mindset.
I’m sure you’ll tweak your system a bit further , and come back with something even stronger. I actually like to here from it some more.

Best of luck and let us know how its going.
And remember do have a look on what kav has said. Very important stuff.
 
« Last Edit: March 23, 2016, 08:25:02 PM by Geoffrey »
 
The following users thanked this post: kav, Reyth

Scarface

  • New
  • **
  • Posts: 34
  • Thanked: 30 times
Re: Testing Actual Spins vs RNG
« Reply #26 on: March 23, 2016, 08:07:02 PM »
I was thinking if someone was playing the opposite way as me, they would've come out a big winner.
One thing I have observed when betting on few cold numbers is that they usually come in clusters. Say we bet 5 numbers. it is more probable that we will have long periods of misses and a few very close hits, than it is to see the numbers hit, for example, in 8 spin intervals. This is not absolute, and they can come in standard intervals, but most of my experience tells otherwise. This is something I truly believe in and I have tried to take advantage of.

This is something I have observed many times myself.  I once played number groups that were missing for 20 or more spins (ex. 2, 12, 22, 32).  Seems like once one of them hit, they would start hitting in clusters especially if they've been missing for awhile.
 

Scarface

  • New
  • **
  • Posts: 34
  • Thanked: 30 times
Re: Testing Actual Spins vs RNG
« Reply #27 on: March 23, 2016, 08:12:47 PM »
And also have a closer look at kav has to say. To be honest, that's pretty interesting stuff he is telling.
And if you test something, before you even start, try to identify what kind of gambler are you?, what kind of risk profile you are comfortable with?. Before looking into what system/ strategy you want to use or test, i would first find an anwser to those questions. And work on your money management skills and mental thoughness. I cant speak for everyone, but a loss of 1100 is an amount nobody should be comfortable with, so try to tweak your system a bit with a more defensive approach.
Yes, Kav has some good info on the game.  Really can relate to what he says about Variance vs. House Edge.  Seems everyone is so concerned about losing to the house edge, but variance is what really gets you.  On the flip side, variance can also win you alot of money  :)
 

Scarface

  • New
  • **
  • Posts: 34
  • Thanked: 30 times
Re: Testing Actual Spins vs RNG
« Reply #28 on: March 23, 2016, 08:31:12 PM »
Here's the system I played recently if anyone would like to try it out.
 In practice mode on RNG, I won 27 times and lost once.  The loss wasn't bad and recovered quickly.  Average win was 334 units.  The one loss was -258 units.  Played for 200 spins and stopped.

  • Place 1 unit on 6 splits that has not hit in the last 20 spins. 
  • Once a split hits, move the chip to the unhit single in that split.  For example, if playing 1-4 split and 4 hits, then move the chip to 1.
  • If a single hits, then play a new split that hasn't hit in last 20 spins
  • You will have anywhere between 6 and 12 numbers covered at anytime.
  • Progression:  keep track of new highs.  If you make it to 20 spins without tieing or reaching a new high, add one chip to your bets.  If at anytime, you tie or reach a new high lower your bet back down to one unit and start the count over. 
This is basically how I was playing that worked very well on RNG.  But like I said, I ended up down about 1100 on the live wheel in roughly 170 spins.  Either I was very unlucky playing the worse session from hell, or I did not test long enough, or RNG results differ from wheel results.  I don't know.  If anyone plays this, and loses the amount I did let me know  :D

 
The following users thanked this post: Reyth

scepticus

  • Hero Member
  • ******
  • Posts: 1666
  • Thanked: 307 times
Re: Testing Actual Spins vs RNG
« Reply #29 on: March 23, 2016, 09:20:25 PM »

Jeez !
Am I dreaming ?
Or did I really read that some people actually AGREE with me ?
Scarface thinks " variance " is the roulette player's main enemy !
and Geoffrey agrees with me that the REAL test of a system is betting with REAL money ! 
AND with my often stated view that my main aim is not to lose !
« Last Edit: March 24, 2016, 12:27:15 AM by scepticus »