Sharpe/Sortino ratio

Hi team,

and thank you for constantly developing and upgrading such a great alternative to MT (as I don't believe in the control schemata and principles that software is offering to their brokers).

As your backtesting/optimization tab comes along with some trade statistic numbers, I kept an eye on the -Sharpe- and -Sortino- ratio recently. I know exactly how both of them work.

According to the numbers that come out of cAlgo at the moment I am sure you are using daily values for return-rate and standard-deviation (that are part of the formula)  and annualize the results at the end to calculate the ratios.

The problem is that in case that you are not a scalper but trying to backtest a "trendspotting" robot you are facing lots of ups and downs on almost a daily basis. Which (I think) is pretty normal for this sort of a robot as you can hardly forsee more than 70% of the right directions traditional trades are going to head (talking of position holding times of about 6 to 96h).

Due to this fact and the leverage that is involved I keep getting rather bad results for sharpe and sortino ratio on this sort of trading algorithms. (< 0,5)

But when looking of the overall results of 36 months or more the history doesn't seem to be that super volatile and risky as the sharpe ratio actually states with its low value. It is only the daily reference that pulls it down so much.

Can you please add a setting to change sharpe and sortino ratio reference to a weekly and/or monthly time period? (or maybe showing them all at the same time like - "d/w/m" - )

I see that going back to a weekly or monthly period would be some kind of a step backwards as you miss out some information that are actually there anyways, but on the other hand the values would be more competitive and comparable with e.g. fonds or other financial products.

I don't think the statistics would then be "faked" or incorrect.

I am sure I would be able to get way better results for bots with such trading times, being able to optimise a bit more efficiently.

As you arleady have a function for sharpe and sortino, all you needed to do is add another instance to it that doesn't refer to the daily values but weekly and monthly. So it wouldn't even be much of a deal I suppose.

Sorry about this terrible long explanation but I didnt't know how to put this more clearly :/

Hope I didn't overlook something or have a major fallacy here.

Looking forward to hearing from you soon.

Best regards,

Nexus