What a Backtest Is
A backtest applies a set of trading rules to past prices to see how they would have performed. Here is what a backtest can tell you, what it cannot and how to read the results.
Rules meet history
A backtest takes precise rules, such as “own SPY when it closes above its 200-day average,” and runs them through historical data day by day. The output shows how an account following those rules would have grown, how deep its losses got and how often it traded.
Reading the results
| Measure | What it tells you |
|---|---|
| Ending value or yearly return | How much the rules earned over the period |
| Maximum drawdown | The worst peak-to-trough loss you would have lived through |
| Number of trades | How active the rules are, and how much costs matter |
| Win rate and average win vs. loss | The shape of the results |
| Comparison with buy and hold | Whether the rules added anything |
Always compare against a simple benchmark.
A real example
In our backtest on SPY prices from March 2007 to September 2026, owning SPY only when it closed above its 200-day average turned $1 into $3.11, versus $5.30 for buy and hold. The worst drop was about 22%, compared with about 57% for buy and hold. The test excluded dividends and interest on cash, so both results understate real returns.
A real chart: SPY monthly bars, 2006 to 2026. Any backtest over this span must survive 2008, 2020 and 2022. Past performance does not predict future results.
What a backtest cannot tell you
- Whether the future will look like the past.
- How you will feel following the rules through a long losing streak.
- Exact fills, especially in fast markets.
- Whether you found a real edge or just a lucky fit to old data.
Key takeaways
- A backtest runs fixed rules through past prices.
- Look at returns, drawdowns, trades and a benchmark.
- Our 200-day test cut drawdowns but earned less.
- Backtests describe the past, not the future.