Great article. Historical standard deviation is important of course but don't you guys also do a forecast standard deviation? Here is an example of a forecast standard deviation estimation. Current price = $10. Suppose there are three forecast states of the market next period - state#1 - probability = 20%, price=$13, returns = 30%, state#2 - probability = 30%, price=$12, returns = 20%, state#3 - probability = 50%, price = $9, returns=-10%. Forecast returns = 0.2*0.3+0.3*0.2-0.5*0.1 = 0.06+0.06-0.05 = 0.07 = 7%, variance = 0.2*(0.07-0.06)^2 + 0.3*(0.07-0.06)^2+0.5*(0.07-0.05)^2 = 0.00025 and so the standard deviation is the square root of 0.00025 = 0.0158 = 1.58 percentage points which means that the 95% confidence interval of expected returns = 0.07-1.96*0.0158 to 0.07+1.96*0.0158 = or approx 4% to 10%. Hope I got the numbers right.