Myths and facts of equity curve trading

March 16, 2015 12:00 PM

When does it work?

To determine if and when equity curve trading can be useful, it is instructive to look at some different possibilities for the technique’s performance. This is shown in “Scenario comparison” (below).

For a strategy that goes completely “bad,” with its equity curve nose-diving toward negative territory, the utility of equity curve trading is obvious. This technique can stop you from trading a strategy that is no longer performing at all. This is shown in the top chart. 

For a strategy performing well, most of the time the equity curve will be above its moving average, and therefore you would continue to trade the strategy. This also is desirable. You are trading the strategy because the equity curve technique has confirmed that it is doing well. This is shown in the middle chart.

Now consider the case that is between these two extremes. This is depicted in the bottom chart. This strategy has times where it performs very well, times where it doesn’t and times where the equity has many whipsaws. In this case, the benefit of equity curve trading is not apparent. Sure, it stops trading when the drawdown gets severe, but it also misses out on a lot of the recovery.

Evaluation & goals

The best way to evaluate the equity curve trading concept is not by measuring increased profit, nor by determining the decrease in maximum drawdown. While both are important measurements, they need to be looked at simultaneously for a true evaluation. The best way, then, is to include both metrics by measuring the before and after profit/drawdown ratio.

Depending on the historical backtest trading software a trader uses, the equity curve trading switch may or may not be easy to implement. If your trading software cannot perform this analysis, it is relatively easy to create in Excel (just be careful not to peek at the data by using incorrect formulas).

The goals of using an equity curve switch are really two-fold. First, the switch should turn off a poorly behaving system. Second, a properly working switch will stop trading before deep drawdowns are encountered, and will resume trading before too much profit is given away. Those are ambitious goals for a single switch.

An equity curve switch can accomplish the first goal fairly easily, especially for a poorly performing system. The table “When to switch” (below) shows that a moving average switch does a good job of turning off the system before the maximum drawdown gets too bad, regardless of length. It is reassuring for a trader to know that a there is a clear signal from the switching method to stop trading.

Unfortunately, the second goal—improving the profit/drawdown ratio—is not so easily met. The table below shows, for a sample system, that the equity curve switch doesn’t necessarily increase the net profit, and in fact can make the system worse. The end result all depends on the choice for the moving average length.

Because the moving average length has such an impact, it should be obvious that there are some serious disadvantages with it. Trader A could use a length of 10, Trader B could use a length of 20, and they will have dramatically different results. Of course, that leads many traders to say, “Well, I’ll just optimize it.” But, just like optimizing parameters in a trading system, optimizing the parameter of an equity curve switch is fraught with peril, prone to every type of bias and over-fitting.

Equity curve switching is a popular technique, but it is unclear as to whether it is really beneficial. For decidedly excellent trading systems, and poor performing systems, the approach can be useful. For trading systems in the middle, which make up the majority, the benefits are not so clear. 

While some of the focus has been on turning off or turning down a system after a difficult period, there is some anecdotal logic, particularly in the trend following space, that a system should be turned up after a poor period and perhaps turned down or off after extremely strong performance. 

In the second part of this series, we will examine equity curve trading for some real-life trading systems, and provide some recommendations on how to use the method effectively.

Page 2 of 2
About the Author

Kevin J. Davey has been trading for more than 25 years. Kevin is the author of “Building Winning Algorithmic Trading Systems.”