Two Plus Two Older Archives

Two Plus Two Older Archives (http://archives2.twoplustwo.com/index.php)
-   Probability (http://archives2.twoplustwo.com/forumdisplay.php?f=23)
-   -   Std q; true win rate? (http://archives2.twoplustwo.com/showthread.php?t=368800)

BobboFitos 10-31-2005 08:24 AM

Std q; true win rate?
 
As one plays more hand, is it intuitive that your std of your winrate will decrease? Or just more accurately resemble your true variance?

I cant figure this out.

Also, is std a good measurement of "running good," or is this simply votality of how I'm playing?

thanks.

DoomSlice 10-31-2005 11:51 AM

Re: Std q; true win rate?
 
The standard deviation will approach its true value as n increases. If it were decreasing, as you said, then eventually you would have a standard deviation approaching 0, which doesn't make much sense here.

BruceZ 10-31-2005 12:50 PM

Re: Std q; true win rate?
 
[ QUOTE ]
The standard deviation will approach its true value as n increases. If it were decreasing, as you said, then eventually you would have a standard deviation approaching 0, which doesn't make much sense here.

[/ QUOTE ]

The standard deviation of the win RATE (called the standard error) does approach zero as the number of hours N approaches infinity. It is the true standard deviation of the of the winnings for 1 hour divided by sqrt(N), and the fact that it goes to zero reflects the fact that that the average win approaches the true win rate as N becomes large. This is not to be confused with standard deviation of the winnings for 1 hour which is what we normally refer to as the standard deviation, and this approaches its true value as N becomes large.


All times are GMT -4. The time now is 11:44 PM.

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