Hi @Dan
I'm assuming this is the case.
When I replicate the Algos returns in an excel sheet and create the Sharpe Ratio of actual daily returns, i get a Sharpe of 2.60.
In my excel sheet version, when I use zeros for the returns on all the days that there are no positions then the Sharpe ratio falls to 0.66.
This is the same figure that the Quantopian backtest shows for my Algo.
This is mainly because the average return is dragged down (while also a small reduction in standard deviation of returns which has the opposite effect on the Sharpe Ratio).
I'd argue that including zeros in the returns calculations for days when there are no positions is making the Sharpe Ratio of little value.
Could be I have wrong end of stick!
:-)
Tks