Take the VHC trade in [LINKSYSTEM_46132063] from 3 days ago. If you roll your mouse over the Risk field, you see MaxDrawdown=~$41k and -17.8% of account equity. Those numbers are correct, as I was watching the trade during that time.
* Why doesn’t the pnl chart show the account drop to 200k before recovering? My guess here is that the chart is updated only once a day and it picks whatever the drawdown is at that time, which I think is a lame excuse for hiding risk.
* Why does the Max peak-to-valley drawdown (historical) in Statistics still show only 11.66%? It’s been 3 days…
Daniel
I find the lack of any feedback puzzling. First, because the issue can be easily fixed. Charts can still be updated only once per day but they should include the intraday account equity Min and Max as a shaded area.
If C2 wants to have some credibility as an auditing platform, issues like this should be addressed or at the very least discussed. You can’t hide behind what’s considered customary in the institutional investing community and say that daily data should be sufficient. No self-respecting institutional investor will give you any money with less than 3 years of trading history. Given that we are allowed and encouraged to invest in systems based on a much smaller time sample, I don’t see why those restrictions should apply to people putting their money with C2 systems. Furthermore, I (and probably others) use the max. drawdown number to set my own stop orders, so I’m interested in the real max drawdown, not sampled at EOD.
But there is of course the possibility that you simply don’t want to show intraday account equity swings. Systems with smooth equity curves sell better, I’d imagine…
Daniel
Daniel:
No one is trying to hide intra-day drawdowns, nor is anyone trying to ignore your previous forum post. You are correct: it is indeed true that an intraday drawdown could possibly be invisible in the equity charts because the charts are updated only periodically (typically more than once per day, but not always).
Increasing the frequency of equity calculations, or altering the methodology of drawdown calculation to incorporate single-trade data, is not a trivial programming job that we can solve easily. It’s a complex project, and the resources required to do it (programming, hardware, CPU) needs to be balanced against other projects and potential customer-helping efforts. You may believe this is the single most pressing issue facing C2. Perhaps you are right, but I’m not sure. There are other things we are working on that could equally help customers trying to evaluate trading systems. Which development project is obviously the right one to throw resources at? It’s not a simple question.
Daniel, I have received your suggestion in the spirit in which I hope it was intended: that you believe this is an important improvement that C2 can and should make (increasing equity calculation frequency to eliminate the possibility of sudden drawdown spikes hiding in more granular data). I agree with you that it is indeed an important project. But I can’t solve it immediately, and as of this writing I can’t commit to a specific schedule for completing it.
And while I agree with you that this is potentially an important limitation, I think that it is not a very common occurrence, and that even modestly increasing equity calculation frequency might make it very unlikely. (Alas, even modestly increasing equity calculation frequency isn’t so simply, either, but I will look into it.)
Please consider your suggestion noted. I’m not ignoring it. I do appreciate it that you took the time to make your concern known.
Respectfully,
Matthew
Thanks for the reply. It’s of course your business how you allocate your resources. I’ve written a lot of software in my time so it’s not obvious to me why this is a major project in any way. The per-trade pnl numbers are updated once every few seconds, so all someone needs to do is add them up per system and keep track of the daily min/max.
Anyway, your call.
Daniel
