Trading styles

Backtesting

Also calledhistorical testing · walk-forward testing

Backtesting is the process of evaluating a trading strategy, model, or rule set on historical data to see how it would have performed if it had been applied in the past.

What Backtesting means

Backtesting lets a trader or researcher test a strategy against past market data before risking capital. It can show how often the strategy would have traded, how sensitive it is to costs and timing, and how large its drawdowns might have been. The result is only a simulation based on history, not proof that the strategy will work in live trading.

Backtesting is a standard way to screen ideas, compare rules, and look for weaknesses before deployment. It helps identify issues such as look-ahead bias, survivorship bias, and unrealistic assumptions about fills or costs. That matters because a strategy that looks attractive on paper can fail once real spreads, latency, and liquidity are included.

Suppose a strategy buys a currency pair whenever a 10-period moving average crosses above a 30-period average. You apply that rule to two years of past price data, include spread and commission assumptions, and measure the resulting returns and drawdown. That backtest is useful for comparison, but it is still a simplified reconstruction of what might have happened.

Common questions

Does backtesting predict future results?+

No. It estimates how a strategy would have behaved in past conditions, but it does not guarantee future performance.

What is the main risk in backtesting?+

The main risk is overfitting or data mining, where a rule looks good historically but does not generalize well to live markets.

Go to the original material.

01CFA Institute — Backtesting & Simulation02CFA Institute — Standard V(A) Diligence and Reasonable Basis03CFTC — Commodity Trading Systems Sold on the Internet