Recently read an article about improvement of system trading incorporating the trading of the system equity curve. Say, set drawdown level at X% and trade the system only if its equity higher and turn trading off if it is lower. Or apply moving average to equity curve and trade only when equity curve above it. Or apply other filters to system equity curves.
Did you tried to trade the systems at C2 based on such approach?
You can actually use this concept in a more advanced manner. Say instead of stop trading if the Equity is above a MA you could scale it down to 50%. So if the run up continues you still earn money. Or you could tie scaling to the raw volatility of the system so you scale according to the changes in system phase. Or you could scale up if the systems is down 1 standard deviation and scale up again if it’s down 2 standard deviations and the contrary on the up side. Or you scale continously according to the actual drawdown.
But these are all concepts that only work with very robust systems and even then you have to be careful with moneymanagement. This is because you basically increase risk on the way down. I wouldn’t use this concept on a trading system I don’t know inside out.
However I do use this concept in a cautious manner in my trading strategy “Volatility Invest IRA” because naturally I know this system inside out.
I computer tested the concept when managing a trading desk where we were running a successful system. We had our coder set up all kinds of tests based on the equity curve and found that only one thing worked. It was similar to AlexanderG’s comment.
What worked was a reversion to the mean method with a 40 period EMA. The period of the EMA might vary from system to system but that is what worked for us. As the equity curve advanced above the EMA you start scaling back the amount traded but never go to Zero. Then as the curve drops below the EMA you start adding on more position. It is easier for stocks and forex than it is for futures since the size of the contracts is so much larger. We traded stocks so it was really easy to scale.
You have to have a disaster stop for the equity curve though. A point where you stop all trading. I would not want to use this on an unproven system. Only use it where you know the system well and have confidence in it. Then it will make considerable more money for you.
Thank you! I was thinking more about setting X% dd for the system, equivalent to average drawdown and then if equity curve goes lower stop trading and wait till return to that X% from below. But your options probably can give better solutions.
Very interesting discussion. I was once part of a platform build myself and we thought about integrating this into the charting system. In the end - I think a feature too complex for the average user, and therefore not a whole lot of return on investment for companies like Collective2 to implement. However, it certainly is cool and more advanced users could find it useful. I think trading the equity curve for most traders is a good idea, but there are exceptions. If the trader isn’t already trading their own equity curve, which they should be unless they are algorithmic / 100% systematic.