Backtesting fundamentals
Trading Strategy Backtesting
Trading strategy backtesting applies explicit entry, exit, sizing, and risk rules to historical market data. Its purpose is to measure how a repeatable process would have behaved—not to predict or guarantee future returns.
A sound backtest starts with explicit rules
Define the market, timeframe, signal calculation, execution timing, position size, exits, and costs before reviewing results. Rules that change after results are known introduce hindsight bias.
Evaluate return and risk together
Total return alone can hide deep losses or unstable behavior. Review drawdown, volatility, trade count, exposure, profit factor, and risk-adjusted measures alongside the equity curve.
Separate research from validation
Use one period to develop the idea and a different out-of-sample period to evaluate it. Walk-forward testing can repeat that process through time, but it still cannot reproduce every live-market condition.
Research checklist
- Use data available at the decision time
- Model fees, spread, and slippage
- Avoid selecting parameters from one favorable period
- Compare results with a relevant benchmark
Questions and answers
What is trading strategy backtesting?
It is the application of predefined trading rules to historical data to estimate past behavior under stated assumptions.
How reliable is backtesting?
It is useful for rejecting weak ideas and understanding behavior, but reliability depends on data quality, realistic costs, unbiased rules, and out-of-sample validation.
How many trades are needed?
There is no universal threshold. More independent observations usually improve confidence, while a small or highly correlated sample leaves greater uncertainty.