#1
|
|||
|
|||
Winrate Confidence Interval, Can Someone Check My Math
Here are the numbers. Over the past 12,490 hands I made $4,558 in 60.0 hours with a standard deviation of $163.8/100 hands.
I want to find the condidence that I make $50+/hr, so u = 4,558 s = 163.8*sqrt(124.9) = 1830.61 if i make $50/hr then my expected winnings would be 60hrs*$45/hr = $2,700 over 60 hours. I say $45 because I average $5 back/hr from rakeback, which is independant of the quality of my play. So to find the confidence that I make $50/hr or more by playing in this game I do the following calculation: Integral[ 1/(s*SQRT(2*Pi)) * e^(-(x-u)^2/(2*s^2)), x, 2700, +infinity) = 0.845 So what all this means is that if I found 100 people with the exact same stats as me (and they all averaged $5/hr from rakeback), about 85 of them would go on to make $50/hr or more playing in the same game. Is this how I go about doing this? I haven't done something like this in a few months so I am rusty. Thanks for the help |
#2
|
|||
|
|||
Re: Winrate Confidence Interval, Can Someone Check My Math
Just to clarify, in the above post 's' is the standard deviation of the mean win total for 12,490 hands.
|
|
|