Backtesting

Backtesting is the first line of empirical validation for any systematic trading strategy. By applying a strategy's entry, exit, and risk rules to historical data and recording the hypothetical trade-by-trade results, a trader can assess key performance metrics - net profit, maximum drawdown, Sharpe ratio, win rate, profit factor, and average trade duration - before risking real capital.

The quality of a backtest depends critically on data quality and the realism of execution assumptions. Common sources of overly optimistic backtest results include look-ahead bias (using data that would not have been available at the time of the signal), survivorship bias (testing only on instruments that are still active today), and insufficient attention to transaction costs. Proper backtests model the full spread or commission, realistic slippage, and swap costs for overnight trades.

MetaTrader's Strategy Tester is the most widely used backtesting tool for retail traders, offering tick-by-tick simulation (the highest quality, using millisecond data to model intrabar price movement) alongside bar-by-bar simulation. Third-party platforms such as TradingView's strategy engine, Quantopian-style Python environments, and dedicated backtesting libraries offer additional flexibility and data access.

A clean backtest is a necessary but not sufficient condition for live viability. Out-of-sample testing (reserving a period of historical data not used in strategy development), walk-forward analysis (periodically re-optimising parameters on a rolling window), and demo-account forward testing provide additional layers of validation. The gap between backtest results and live performance - caused by market impact, changing correlations, and regime shifts - is called the implementation shortfall and should always be anticipated.