Two Plus Two Publishing LLC
Two Plus Two Publishing LLC
 

Go Back   Two Plus Two Poker Forums > >

Notices

Probability Discussions of probability theory

Reply
 
Thread Tools Display Modes
Old 03-05-2020, 10:29 PM   #1
Mig
veteran
 
Join Date: Mar 2005
Location: Flopping like a sucker
Posts: 2,933
-414 Buyins under EV in 336k hands : Odds of this happening

Hi guys,

My stats are beyond rusty. If anyone could help me figure this out

EV Win Rate of = 2.73
Effective Win rate of = -9.60
# of hands played = 336000
std dev / 100 =130

Buy ins under ev = 414



The poker calculator on pokerdope is giving me odds of 0.0000%

So this run is likely to be a 1 in 10 000 000 at this point. I'd like to be able to do the calculation by myself
Mig is offline   Reply With Quote
Old 03-07-2020, 03:50 PM   #2
Yoshi63
old hand
 
Yoshi63's Avatar
 
Join Date: Jan 2007
Location: sandstorm
Posts: 1,940
Re: -414 Buyins under EV in 336k hands : Odds of this happening

I don't know how to mathematically calculate it, but I ran 5000 simulations and found the following:

-Expected bb won (calculated): 2.73*3360 = 9172.8
-Expected bb won (observed): 9374.06
-Standard deviation: 7564.8

The observed data fell within the boundaries as follows:
-One standard deviation (expected 68.27%) = 0.6746
-Two standard deviations (expected 95.45%) = 0.9556
-Three standard deviations (expected 99.73%) = 0.9968

It appears to be normally distributed. As such, a run of 414 buyins under EV would be 414 / 75.68 = 5.47 sigma.

About 1 in 44.5 million chance.
Yoshi63 is offline   Reply With Quote
Old 03-07-2020, 08:36 PM   #3
Mig
veteran
 
Join Date: Mar 2005
Location: Flopping like a sucker
Posts: 2,933
Re: -414 Buyins under EV in 336k hands : Odds of this happening

Quote:
Originally Posted by Yoshi63 View Post
I don't know how to mathematically calculate it, but I ran 5000 simulations and found the following:

-Expected bb won (calculated): 2.73*3360 = 9172.8
-Expected bb won (observed): 9374.06
-Standard deviation: 7564.8

The observed data fell within the boundaries as follows:
-One standard deviation (expected 68.27%) = 0.6746
-Two standard deviations (expected 95.45%) = 0.9556
-Three standard deviations (expected 99.73%) = 0.9968

It appears to be normally distributed. As such, a run of 414 buyins under EV would be 414 / 75.68 = 5.47 sigma.

About 1 in 44.5 million chance.

Thanks for taking the time to do this, really appreciated.

Hadn't thought about this methodology. Might give it a try. Ty
Mig is offline   Reply With Quote
Old 03-11-2020, 09:57 AM   #4
browni3141
Pooh-Bah
 
Join Date: Aug 2015
Location: South Florida
Posts: 3,930
Re: -414 Buyins under EV in 336k hands : Odds of this happening

Which site? This isn't quite proof of rigging but I wouldn't want to play there anymore.

(2.73-(-9.6))/(130/sqrt(3360)) = 5.50 standard deviations. However, AIEV isn't the same as your true win-rate since it only removes one aspect of luck. I'm not sure how this should be accounted for.
browni3141 is offline   Reply With Quote
Old 04-01-2020, 11:34 PM   #5
SligoFella
newbie
 
SligoFella's Avatar
 
Join Date: Jul 2015
Location: Overlooking the Atlantic
Posts: 35
Re: -414 Buyins under EV in 336k hands : Odds of this happening

Hmmmm ... might be the balance for the super user at UB
SligoFella is offline   Reply With Quote
Old 04-10-2020, 01:17 PM   #6
NewOldGuy
Pooh-Bah
 
Join Date: Mar 2009
Location: In the wires
Posts: 5,835
Re: -414 Buyins under EV in 336k hands : Odds of this happening

Quote:
Originally Posted by browni3141 View Post
However, AIEV isn't the same as your true win-rate since it only removes one aspect of luck. I'm not sure how this should be accounted for.

The proper way to account for it would be to use only all-in hands as the denominator, not ALL hands. That greatly inflates the deviation. It isn't close to 5 sigma.

The number of times a player decides to go all in is not random, and so using all hands in the calculation is pure nonsense. The absolute values given by these variance calculators are interesting for what they are, but you can't interpret them the way the OP is trying to do. And the fact that "pokerdope" software builds in such a calculation indicates they also don't understand their own statistic.
NewOldGuy is offline   Reply With Quote
Old 04-10-2020, 05:13 PM   #7
SligoFella
newbie
 
SligoFella's Avatar
 
Join Date: Jul 2015
Location: Overlooking the Atlantic
Posts: 35
Re: -414 Buyins under EV in 336k hands : Odds of this happening

Quote:
Originally Posted by NewOldGuy View Post
The proper way to account for it would be to use only all-in hands as the denominator, not ALL hands. That greatly inflates the deviation. It isn't close to 5 sigma.

The number of times a player decides to go all in is not random, and so using all hands in the calculation is pure nonsense. The absolute values given by these variance calculators are interesting for what they are, but you can't interpret them the way the OP is trying to do. And the fact that "pokerdope" software builds in such a calculation indicates they also don't understand their own statistic.
+1
SligoFella is offline   Reply With Quote

Reply
      

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Forum Jump


All times are GMT -4. The time now is 10:34 PM.


Powered by vBulletin®
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.
Copyright 2008-2017, Two Plus Two Interactive
 
 
Poker Players - Streaming Live Online