atrader.ai

Reading the scorecard

Backtest Metrics

Backtest metrics summarize how a ruleset behaved over historical data. They are useful for comparison, but they are not guarantees of future results.

CAGRmax drawdownSharpewin ratetradesbuy and hold

Key Ideas

  • Total return measures the full percent gain or loss over the selected period.
  • CAGR annualizes that return, making periods of different lengths easier to compare.
  • Maximum drawdown shows the worst peak-to-trough decline during the backtest.
  • Sharpe estimates return per unit of daily volatility, annualized with a 252-trading-day convention.

Common Uses

Strategy pages compare the ruleset against buy-and-hold benchmarks like SPY, TQQQ, or SOXL.
Replay pages show daily signals so users can inspect where trades came from instead of only reading the summary.
Portfolio Builder aggregates multiple strategy sleeves and reports combined return, drawdown, Sharpe, and cash allocation.

Examples

Max drawdown of -35%

At some point in the period, the account was 35% below its prior high before recovering or continuing lower.

Strategy versus buy and hold

The benchmark answers whether the extra rules improved the ride compared with simply owning the underlying asset.

Watch For

  • A great backtest can still be overfit or depend on a market environment that will not repeat.
  • Changing date ranges can dramatically change CAGR, drawdown, and win rate.
  • Trade count matters because frequent switching can increase slippage and tax complexity outside the backtest.