if you have time, can you explain my stats please?

[MikeR]

I don't have Pokertracker (I can't afford it right now) but I'm using pokercharts which gives a few stats. I'm having some trouble figuring out what they mean. I especially don't understand the difference between win rate and avg win rate and win rate std. How can your win rate be different than your average win rate? If you have time could you possibly explain? I really respect the posters here and would appreciate any explanations. Thanks guys.
These stats are for 17 sessions at .50/1.00 limit.

Earnings: $36.75
Avg. Return: $2.16
Return Std.: $11.65
Win Rate: $1.34/hr
Avg. Win Rate: $3.32/hr
Win Rate Std.: $13.15/hr
BB/hr: 1.34
Avg. BB/hr: 3.32
BB/hr Std.: 13.15
ROI: 7.21%
Time played: 27h 20m
Avg. Session: 1h 37m
ROI vs. DJIA: 6.93%
Sessions: 17
First: 2005-03-11
Last: 2005-03-15

[Pov]

According to their FAQ:

"The help section of the application provides full definitions and formulas for each of these items."

[MikeR]

Thanks. Here's what I found.
Win Rate - Your overall hourly win rate.
Makes sense.
Average Win Rate - Your average per session hourly win rate.
This is where I'm lost. Should'nt this be the same exact thing?
Win Rate Standard Deviation - The standard deviation of the per session hourly win rate values. In other words, a measure of how your win rates vary from the mean win rate value. Consistent players will have a lower number.
This made my brain hurt, and I think my high number is probably not good.

[Pov]

[ QUOTE ]
Thanks. Here's what I found.
Win Rate - Your overall hourly win rate.
Makes sense.
Average Win Rate - Your average per session hourly win rate.
This is where I'm lost. Should'nt this be the same exact thing?
Win Rate Standard Deviation - The standard deviation of the per session hourly win rate values. In other words, a measure of how your win rates vary from the mean win rate value. Consistent players will have a lower number.
This made my brain hurt, and I think my high number is probably not good.

[/ QUOTE ]

The two win rates are not the same because the sessions you've played are not of equal length in time. Let's say you played:

1 hour at $2.00/hr (won $2)
2 hours at $4.00/hr (won $8)
4 hours at $1.00/hr (won $4)

That's a total of 7 hours for $14 or an overall average of $2/hr. But your average per session win rate is (2 + 4 + 1) / 3 = $2.33/hr.

The standard deviation number is a measure of your win rate's fluctuation around the average. Basically, the lower the standard deviation the more confident you can be that your win rate is somewhat accurate. With your standard deviation of $13.15/hr you know that 68% of the time if you play your normal game for an hour, you will win $1.34 +/- $13.15. You know that 95% of the time you will win $1.34 +/- $26.30. This assumes your actual winrate really is $1.34 / hr. (I'll come back to that)

So all a high standard deviation is telling you is that you will have bigger swings in your bankroll than someone with a lower standard deviation. Having a high standard deviation is not necessarily bad. In fact, if you are aggressive with small edges you will have a signficantly higher standard deviation than someone who plays it safe. However, if you are pushing your edges correctly, you will also have a higher winrate *in the long run*, but you need to have a bigger bankroll in order to prevent yourself from going broke in a downswing.

As for your *actual* winrate . . . statistics can be frustrating. Unfortunately for someone with an "average" standard deviation, it will take well over 100,000 hands to have an average winrate you can be comfortable is correct and not overly influenced by short term winning or losing streaks. Sounds crazy but the math has been shown on these forums many times if you do a few searches for it and care to have your brain hurt a lot worse than it does now.