A small sample can make a weak strategy look profitable or a good strategy look poor. Results should be reviewed across enough trades and different market conditions to understand uncertainty.
How large is your backtesting sample, and which metrics do you track?
To properly test a trading strategy, a minimum of 100 to 200 trades is generally recommended to get a statistically meaningful sample across different market conditions. Smaller samples often look deceptively good due to curve fitting or luck.
When backtesting, it is crucial to track metrics like maximum drawdown, profit factor, win rate, and risk-reward ratio. This sample depth is especially vital when trading with prop firms, where strict daily loss limits and consistency rules leave little room for error if a strategy hits an unexpected cluster of losses.
How large is your backtesting sample, and what metrics do you rely on most?
