Reading results honestly

Once a test is free of lookahead, checked out of sample and charged realistic costs, its results are worth reading. Reading them well means looking past the headline return.

What to look at

• Number of trades. A handful of trades says very little. Results from a small sample can be luck.
• Maximum drawdown: the largest fall from a peak. It shows the worst stretch you would have had to sit through.
• Losing streaks and time to recover.
• Consistency across periods and market conditions, not just the total.
• Sensitivity: whether small changes to the settings change the result a lot. If they do, the result is fragile.

A healthy attitude

The goal of testing is not to find a strategy that looks good. It is to find out what is true about a set of rules, including when they are wrong. A test that shows a strategy does not hold up is a successful test: it has done its job before any money was involved.

Past performance, real or simulated, is not a reliable indicator of future results. Educational content only. Not financial advice.