We're talking longer than 10 years here. A better estimate might calculate the average 10 year return for every 10 year return period, and the standard deviatio

We're talking longer than 10 years here. A better estimate might calculate the average 10 year return for every 10 year return period, and the standard deviation of that same data set. And then do that calculation for 30 years. I believe you end up with a higher average and smaller deviation at 30 than 10.