Two Plus Two Newer Archives

Two Plus Two Newer Archives (http://archives1.twoplustwo.com/index.php)
-   Poker Theory (http://archives1.twoplustwo.com/forumdisplay.php?f=17)
-   -   Any good ways to estimate hourly variance? (http://archives1.twoplustwo.com/showthread.php?t=544738)

Foreverastudent 11-12-2007 10:56 PM

Any good ways to estimate hourly variance?
 
I'm sure a post like this has been made before, but I can't seem to find it. I apologize in advance for repeating a topic that has probably already been covered.

I've been tracking my progress at a local casino for a while, and I'm trying to figure out a good way to find a confidence interval for my hourly rate.

Now, finding mean is easy. I've got no problem with that. What I don't know is how to find a good estimate on my variance for each hour played.

When I play, I record the length of my session and my winnings/losses. For example, I can look at my records for yesterday and see that I made $116 over three hours thirty minutes, but I don't go into more detail than that. Obviously, these sessions vary in length, my shortest being about two hours and my longest being about twelve.

Don't worry about being too technical in your explanation, as I have a pretty strong math background. Any help would be greatly appreciated.

JABoyd 11-13-2007 12:46 AM

Re: Any good ways to estimate hourly variance?
 
I have a pretty intense EXCEL workbook I wrote. I can send it to you if you want.

drowski 11-13-2007 03:51 AM

Re: Any good ways to estimate hourly variance?
 
find the standard deviation per hour then use this formula:

2 x SD / square root(total hands/hours played)

this will give a number add/subtract it to your current winrate and that is a 95% confidence interval (ie 95% of the time your true winrate is in that interval)

if you want a 99.7% confidence interval use 3 instead of 2

Foreverastudent 11-13-2007 02:54 PM

Re: Any good ways to estimate hourly variance?
 
Drowski,

What I actually need to know is how to find the standard deviation per hour. Once I have that, I know what to do to find the confidence interval.

What's not clear to me is how to estimate the standard deviation per hour for various sessions of different lengths.

JABoyd 11-13-2007 04:39 PM

Re: Any good ways to estimate hourly variance?
 
Foreverastudent,

I have an excel workbook i created for all of that. Would you like it?

Foreverastudent 11-13-2007 07:17 PM

Re: Any good ways to estimate hourly variance?
 
JABoyd,

I sent you a PM with my email addy. I would like that workbook if you could send it. Thanks.

InTheDark 11-13-2007 08:30 PM

Re: Any good ways to estimate hourly variance?
 
Mason M. details the calculation of SD from those stats you keep in a book I bought 12 years ago. I would really appreciate it if anyone could remember the formula. I know I have the book, and I know where the box containing it is, but...

Albert Moulton 11-18-2007 11:02 PM

Re: Any good ways to estimate hourly variance?
 
I keep the same records. For each session I calculate the average dollars per hour per session. I then use the standard deviation function in EXCEL to get the standard deviation of all sessions' average dollars per hour per session.

Now, this isn't perfect, since a 1-hour session will have the same weight as a 16-hour session, but as long as your sessions are more or less similar in length this should give you a decent measure of variance in your average win/loss rate per hour per session. And it's easy to do with excel given the information that you are keeping.


All times are GMT -4. The time now is 08:17 PM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.