Backtesting means applying your strategy's rules to historical charts to see how it would have performed. It will not predict the future, but it tells you whether your rules have any edge at all and what kind of losing streaks to expect.
Step 1: write exact rules
You cannot backtest a feeling. Write the entry conditions, stop placement, target, timeframe, markets and trading hours precisely enough that another person would take the same trades.
Step 2: go through the charts
- Scroll back several months on your main market and timeframe.
- Move forward candle by candle, hiding the future.
- Record every trade that matches the rules: date, entry, stop, target, result in R.
Step 3: enough trades
Twenty trades prove nothing. Aim for at least 50, ideally 100, across different market conditions (trending, ranging, volatile).
Step 4: read the numbers
- Win rate and average win/loss in R.
- Expectancy: (win rate × average win) − (loss rate × average loss).
- Longest losing streak: can your risk per trade survive it inside the daily and maximum loss limits?
Biases to avoid
- Hindsight bias: 'I would have skipped that one'. If it matched the rules, it counts.
- Curve fitting: changing rules until the past looks perfect.
- Ignoring costs: include spread and commission.