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 03-12-2017, 03:31 PM   #1
OFC_OMG
journeyman
 
Join Date: Apr 2013
Posts: 296
HU SNG loses 7/8 All In Pre: right way to calculate short term variance?

small stakes SNG - played ~55 hands HU with villain who folded to most raises, paid off Quads in a big way, folded on 10-high board to my J high 3/4 pot bluff, etc. At one point I had chip lead 16,347 > 1653. IOW I felt I was in control - except for losing 7/8 all ins pre. These were the only hands that got all in pre.

Here are the hand equities. What is the right way to think about the variance of these 8 hands as a set, compared to the average equity of each hand?

QT < K9 42% / 57% (slight dog)
55 < 78 49% / 49% (flip)
K8 < 67 65% / 35% (fav)
2T < AK 34% / 65% (dog)
Q9 < A2 45% / 54% (slight dog)
AT < QQ 29% / 71% (dog)
37 < A4 40% / 59% (slight dog)

Q8 > 76 62% / 37% (fav, only winner)

___________________________________________

Hero avg. equity 46% / Villain avg. equity 47%

Hero loses 7/8 all ins, a 13% win rate

Villain wins 7/8 all ins, an 87% win rate

Hero had > chips in 6/8 hands
OFC_OMG is offline   Reply With Quote
Old 03-14-2017, 08:07 AM   #2
Nixie Bobtie
enthusiast
 
Join Date: Feb 2017
Posts: 73
Re: HU SNG loses 7/8 All In Pre: right way to calculate short term variance?

I've done some quick maths on this and whilst there are some imperfect assumptions I think it should prove robust enough

Simplistically, there are 8 hands and 256 (2^8) combinations of outcome (ignoring ties and the occasions where the bigger stack wins and ends the contest)

The average that you quoted for the villain was low, I calculate it as 53% (vs 46% for Hero)

If each hand is assumed to be dealt out at the average equity (splitting the remaining tie equity across win & lose) then hero could expect to win

0 of 8 hands - 0.7%
1x - 4.8% (for which there are 8 of the 256 possible combinations)
2x - 14.5%
3x - 24.9%
4x - 26.7%
5x - 18.3%
6x - 7.9%
7x - 1.9%
8x - 0.2%

The expectation is for Hero to win c3.7/8 hands, with a standard deviation of +/- 1.4

To have only won once is practically 2 standard deviations from the mean, so a long way from the norm
Nixie Bobtie is offline   Reply With Quote
Old 03-25-2017, 12:09 AM   #3
OFC_OMG
journeyman
 
Join Date: Apr 2013
Posts: 296
Re: HU SNG loses 7/8 All In Pre: right way to calculate short term variance?

Quote:
Originally Posted by Nixie Bobtie View Post
I've done some quick maths on this and whilst there are some imperfect assumptions I think it should prove robust enough

Simplistically, there are 8 hands and 256 (2^8) combinations of outcome (ignoring ties and the occasions where the bigger stack wins and ends the contest)

The average that you quoted for the villain was low, I calculate it as 53% (vs 46% for Hero)

If each hand is assumed to be dealt out at the average equity (splitting the remaining tie equity across win & lose) then hero could expect to win

0 of 8 hands - 0.7%
1x - 4.8% (for which there are 8 of the 256 possible combinations)
2x - 14.5%
3x - 24.9%
4x - 26.7%
5x - 18.3%
6x - 7.9%
7x - 1.9%
8x - 0.2%

The expectation is for Hero to win c3.7/8 hands, with a standard deviation of +/- 1.4

To have only won once is practically 2 standard deviations from the mean, so a long way from the norm
Thank you!
OFC_OMG 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 09:40 AM.


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