#1 Roulette Forum & Message Board | www.RouletteForum.cc

Roulette-focused => General Discussion => Topic started by: Roulettebeater on Nov 20, 09:32 AM 2019

Title: Probability and expectation
Post by: Roulettebeater on Nov 20, 09:32 AM 2019
So the probability of 13 random numbers is 13/37 =35.1

Now I have a system that giving me the probability of 13 numbers (not picked randomly) = 35.8 %

Is this an advantage ?
Title: Re: Probability and expectation
Post by: Joe on Nov 20, 09:38 AM 2019
Quote from: Roulettebeater on Nov 20, 09:32 AM 2019Is this an advantage ?

I need the sample size before I can give you an answer. ie the number of bets placed from which you got that %.
Title: Re: Probability and expectation
Post by: Roulettebeater on Nov 20, 09:50 AM 2019
Quote from: Joe on Nov 20, 09:38 AM 2019
I need the sample size before I can give you an answer. ie the number of bets placed from which you got that %.

Well, I have 20k spins so sample size is 20k spins.

The probability of hit obtained from betting 13 numbers is 35.8 %

0.7% isn’t enough to win flatbetting
Title: Re: Probability and expectation
Post by: Joe on Nov 20, 11:59 AM 2019
Your results are significant at the 5% level. It doesn't mean you definitely have an edge, but you are 'probably' onto something.

Null hypothesis: population proportion = 0.351
Sample size: n = 20000
Sample proportion = 0.358
Test statistic: z = (0.358 - 0.351)/0.0033749 = 2.07414
Two-tailed p-value = 0.03807
(one-tailed = 0.01903)


The two-tailed p-value is what you're interested in. It means there is a 3.8% chance that you got those results by chance. Because this is a pretty low probability, it suggests that they are not due to chance, but you should really get some more data and test again to confirm it.

(link:s://img.techpowerup.org/191120/screenshot.png)

Title: Re: Probability and expectation
Post by: Roulettebeater on Nov 21, 08:29 AM 2019
Quote from: Joe on Nov 20, 11:59 AM 2019
Your results are significant at the 5% level. It doesn't mean you definitely have an edge, but you are 'probably' onto something.

Null hypothesis: population proportion = 0.351
Sample size: n = 20000
Sample proportion = 0.358
Test statistic: z = (0.358 - 0.351)/0.0033749 = 2.07414
Two-tailed p-value = 0.03807
(one-tailed = 0.01903)


The two-tailed p-value is what you're interested in. It means there is a 3.8% chance that you got those results by chance. Because this is a pretty low probability, it suggests that they are not due to chance, but you should really get some more data and test again to confirm it.

(link:s://img.techpowerup.org/191120/screenshot.png)

Cool !

You know that even with this slight advantage I can’t win flatbet, the idea to apply a progression is also risky

I ran out of ideas !
Title: Re: Probability and expectation
Post by: 2BobBet on Nov 21, 03:35 PM 2019
Can you not use Paroli up as you win ? will this work on that bell curve , I suppose it depends on how far that bell curve reaches before the inevitable down curve!
Title: Re: Probability and expectation
Post by: Roulettebeater on Nov 24, 04:48 AM 2019
Quote from: 2BobBet on Nov 21, 03:35 PM 2019
Can you not use Paroli up as you win ? will this work on that bell curve , I suppose it depends on how far that bell curve reaches before the inevitable down curve!

i never tought of it!

can you give us an example
Title: Re: Probability and expectation
Post by: nottophammer on Nov 24, 05:38 AM 2019
RB; what you are showing for 13’s, is in-line with data for random.org spins.
This is a question from Priyanka.

Notto - One question for you.

Why wait for non hitting 9 spins and waste your precious time. I have done some tests on picking numbers 7, 14, 16, 18, 21, 23, 27, 32, 36, even last night's 55 spins aired and start betting from spin 1 itself. Flat bet and take off the number that hits going 8 numbers, 7 numbers upto 4 numbers. What do you think?


Pry’s answer is on page 2 of ROTT in note pad.
Is your 13 #’s better than 9 #’s ?
Title: Re: Probability and expectation
Post by: Bigbroben on Nov 24, 04:45 PM 2019
Quote from: Roulettebeater on Nov 20, 09:32 AM 2019
So the probability of 13 random numbers is 13/37 =35.1

Now I have a system that giving me the probability of 13 numbers (not picked randomly) = 35.8 %

Is this an advantage ?

You agree: On a win, you get (36-13=23) points.  On a loss: -13 points.
35.8% wins x 23 = 823.4 points
64.2% (100-35.8 ) x 13= 834.6 points.
Difference: 823.4 win- 834.6 loss = -11.2 points in 100 spins.
-11.2 in 1300 (100x 13) units laid = -0.86% edge.

Looks like not enough....
Title: Re: Probability and expectation
Post by: Bigbroben on Nov 24, 05:00 PM 2019
You need 36.115% to break even.
Title: Re: Probability and expectation
Post by: Roulettebeater on Nov 25, 04:46 AM 2019
Quote from: Bigbroben on Nov 24, 05:00 PM 2019
You need 36.115% to break even.
you right!

but dont you think there is a way to come ahead with such bet selection?
Title: Re: Probability and expectation
Post by: Bigbroben on Nov 25, 08:12 AM 2019
I read this somewhere on this site:
Increase the accuracy of your predictions!
Title: Re: Probability and expectation
Post by: Roulettebeater on Nov 25, 08:17 AM 2019
Quote from: Bigbroben on Nov 25, 08:12 AM 2019
I read this somewhere on this site:
Increase the accuracy of your predictions!

its already increased!  :thumbsup:
Title: Re: Probability and expectation
Post by: 2BobBet on Nov 25, 09:43 AM 2019
Sorry for the late reply RB 
Paroli up as you win progression

1 unit if win, next bet 2 units, if win, next bet 3 units ect

There are many ways to play this . You could hold at a unit stage for 2 wins in a row or if you are in a win, loss situation you could bet the same stage twice

Heres an example
bet 1 unit , win    bet 2 units , win  bet 3 units, win  bet 4 units , lose   bet 4 units win

You could work out your 36.6% edge into this progression . Im not sure how but Kelly citeron may help you with this.