Backtesting
Backtesting runs a strategy against historical data to measure its edge before risking live money - garbage in, garbage out.
Backtesting runs a strategy on historical data to measure its edge before risking money. Its quality depends on realistic costs and honest fills, and overfitting to the past is the classic failure.
Backtesting is the process of testing a trading strategy against historical data to measure its performance: win rate, expectancy, drawdown and behaviour across market regimes. A backtest is only as good as its assumptions: realistic spreads and slippage, honest execution fills, and data that covers different market conditions. Overfitting is the classic failure, where a strategy is tuned so precisely to the past that it fails live. A good backtest is the cheapest education a trader can buy; a bad one is the most expensive.
How It Works
- Apply entry and exit rules to historical price data
- Measure win rate, expectancy, drawdown and regime behaviour
- Forward testing and small live size validate the backtest
Trading Tips
Include spread, commission and realistic slippage or the result is fiction
Test across trending and ranging periods, not just the good ones
If a strategy looks too good, it is overfitted
Backtesting Example
Say a strategy backtests 35% annually over ten years of data with realistic spreads, then paper-trades flat for three months. The backtest measured the past honestly. The flat quarter says regimes changed. Both results are true and both matter.
How Traders Use Backtesting
Backtest with costs, slippage and out-of-sample data or do not bother. One great backtest earns a demo trial, never live size. Strategies graduate on forward results, not historical curves.
Related Terms
Put Your Knowledge Into Practice
Compare regulated brokers and find the best one for your trading style.