Extrapolating a Win Rate
 

Let's assume I have a true win rate of 2 BB's/100 and my standard deviation is 20 BB's/100.

What is the correct way to extrapolate these values for 1, 2 and 3 standard deviation intervals over n hands?

Re: Extrapolating a Win Rate
 

[ QUOTE ]
Let's assume I have a true win rate of 2 BB's/100 and my standard deviation is 20 BB's/100.

What is the correct way to extrapolate these values for 1, 2 and 3 standard deviation intervals over n hands?

[/ QUOTE ]

Mean win = (n/100)*2
ySD = y*20*(n/100)^.5, for y standard deviations

So for a session of n hands, you can expect that roughly two-thirds of the times your results will fall within the range of:

(n/100)*2 +/- 20*(n/100)^.5, or 1SD around the mean

~95% of the time, your results will fall within 2SD, or

(n/100)*2 +/- 2*20*(n/100)^.5

In your example, if you play a 400 hand session, expect to fall in the range of -32 to +48 big bets with two-thirds probability, and -72 to +88 big bets with 95% probability.