Adrian Reid explains why equity-curve filters that turn systems on and off based on recent performance backfire. A filter that triggers only a handful of times in testing has too little data to prove it works, making any improvement just curve-fitting. Build simple systems where every rule fires often enough to matter.
I wanna give you a warning about using equity curve filters to turn your systems on and off depending on how good or badly they're performing. Imagine this, you've got a trading system that's in an all-time largest drawdown. It's in a 30% drawdown. And before this drawdown, the maximum drawdown it experienced was only 15%. So it's now double that. So what do you do? So when that happens, it's easy to think, all right, let me just add some rules, some filters, and fix this system to get rid of this nasty drawdown we're experiencing currently so I can keep trading the system. and this is where equity curve filters are really alluring. And you put an equity curve filter over it. You go, wow, look, the 30% drawdown the system's currently in is now only a 15% drawdown. I fixed the system. What's the problem? Well, the problem, the problem is we put an equity curve filter on, which probably only turns on and off 5 or 10 times in the entire backtest. We've got no real sound basis for choosing which parameter value to control that equity curve filter. So it might appear in the current backtest as though the system is better, but with very few data points, we don't have a high degree of certainty that that will continue to work in the future. I feel like it's a trap. You know, that's like, oh, let's try and make a bad system good by adding complexity. But how often does complexity help us in trading? Not a lot. So equity curve trading sounds like a really great idea, but unless your rule comes into play many, many times in your backtest, it's almost impossible to do it without overfitting or curve fitting to past data. So you've got to be really careful. You're far better off to design simple, robust systems where every single rule comes into play many, many times over the course of the backtest. That way you can have confidence that you've chosen the parameter values correctly. So if you want to learn more about systematic trading, go to enlightenedstocktrading.com/free to get your free Trader Acceleration Bundle. There's a link in the description below. And make sure you give this video a like and subscribe so you don't miss the next one.
