Read in this order
- The period. Does it include a bear market, such as 2008, 2020 or 2022? A test from 2010 to 2021 flatters almost everything.
- Max drawdown. Could you have held on through it?
- Benchmark. Did it beat simply holding an index fund, after costs?
- CAGR. Only now look at returns.
- Risk-adjusted return: Sharpe ratio, Sortino ratio, Calmar ratio.
- Turnover and trade count.
Red flags
| Sign | Why it worries |
|---|---|
| Sharpe above 2 for years | Rare in reality; often Overfitting in investing |
| Very smooth equity curve | Hidden tail risk or look-ahead bias |
| Most gains from a few days | Luck, or one event |
| Turnover over 400% a year | Costs and overfitting risk |
| Short test period | Not enough different markets |
| Many rules and branches | More chances to fit noise |
Compare fairly
Use the same dates, include dividends for both, and match risk: a 3x leveraged strategy should be compared with a leveraged benchmark, or judged on risk-adjusted numbers.
The final question
Can you explain in a sentence why this strategy should keep working? If not, treat the backtest as a coincidence until time proves otherwise.
Try it Backtest a Strategy
See also
- What is backtesting? What backtesting a strategy means, how a backtest is built, the common biases that make backtests look better than reality, and how to use results wisely.
- Overfitting in investing What overfitting is, why strategies tuned to past data fail in live trading, what a large study of community strategies found, and how to measure it.
- Maximum drawdown What maximum drawdown is, how to calculate it, why losses need bigger gains to recover, and historical drawdowns for stocks and leveraged funds.
- Sharpe ratio What the Sharpe ratio measures, how to calculate it from returns and volatility, what counts as a good Sharpe ratio, and where it misleads.
Pages that link here: VIX tier allocation
Last updated September 30, 2026. Education only, not investment advice.