Yesterday, I introduced this series on Curve Fitting.
Ever since reading articles years ago about the dangers of curve fitting and how “many a trading career has been shattered” from it, I’ve been skeptical of these claims.
It’s not that I think it’s impossible to lose money from curve fitting your strategy. You absolutely can.
But to lose a ton of money from it, you would have to be incredibly naive, have a reckless appetite for risk, and have a sloppy process for making changes to your strategy.
I haven’t met a trader yet that I would describe that way.
You should have a default stance of skepticism – about the industry in general, anything you see traders say (especially on Twitter), and your trading strategies.
And if you’re reading this, you probably already have it.
When you have a healthy dose of paranoia, a decent process for making changes, and don’t aggressively size up, your downside should already be limited.
Even if you naively apply a rule that’s egregiously curve-fitted and go live with it (YOLO!), a good process will be able to spot it quickly.
When you have a solid process and are reasonably careful, the worst case scenario from curve-fitting a halfway decent strategy is a flat equity curve that you thought would be up and to the right, not the “end of a trading career.”
So work on your process, and the curve-fitting boogeyman should become a lot less scary.
Tomorrow, we have to talk about what curve-fitting is and isn’t.
-Dave
P.S. Curve-fitting is on my mind based on the feedback of several traders who are beta testing a new product I’m working on – a tool that helps you create a trading strategy with edge. The window for beta testing is ending soon, so reply quickly if you’re interested in a sneak peek.
