Backtesting basics and common mistakes

Backtesting means applying a set of trading rules to historical data to see how they would have performed.

What to look at

• Number of trades: a handful of trades says very little.
• Drawdown: the largest fall from a peak. Ask whether you could stay with the plan through it.
• Costs: include brokerage, taxes and slippage.

Common mistakes

• Overfitting: tuning rules until they fit past data perfectly. Such rules often fail on new data.
• Look-ahead bias: using information that was not available at the time of the trade.
• Ignoring change: past results do not guarantee future results.

A sensible habit

Test on one period, then check on data the rules have not seen, and start with small size.

This guide is educational and is not investment advice.