Backtesting means applying fixed strategy rules to historical prices to see how they would have performed. It turns a hunch into evidence.

Step by step

1. Write the rules down precisely. Entry, stop, target, position size, and when not to trade. If two people would apply your rules differently, they are not precise enough.
2. Choose a sample. At least 50, preferably 100+ trades, across different market conditions: trends, ranges, quiet and volatile periods.
3. Go through the chart candle by candle. Scroll forward so you cannot see what happens next. Record every signal, including the ones you would rather skip.
4. Record each trade: date, direction, entry, stop, exit, result in R (multiples of risk).
5. Calculate: win rate, average win and loss in R, expectancy, largest losing streak and maximum drawdown.

Measuring in R

Recording results as multiples of risk makes them independent of account size. A trade that risked 1R and gained 2.5R is +2.5R whatever the lot size.

Avoiding fooling yourself

• Hindsight bias: "I would have skipped that one." If your rules took it, count it.
• Curve fitting: adding conditions until past results look perfect. A strategy with ten filters that worked last year rarely works next year.
• Ignoring costs: include spread and commission, especially on short-term strategies.

Then forward test

After a positive backtest, trade the rules on a free trial for a few weeks. If live results are in line with the backtest, you have a strategy worth taking into an evaluation.