If you have ever compared two strategies and wondered which one is "better" once you account for bumps along the way, you are already thinking about risk-adjusted performance. The Sharpe ratio is one of the most common ways to put that intuition into a single number.
The idea in one sentence
The Sharpe ratio asks: how much return did you earn per unit of risk? Higher is generally better, because you are getting more reward for each unit of volatility you tolerated.
How it is calculated
For a period you choose (daily, monthly, or annualized):
- Take the strategy's average return over that period.
- Subtract a risk-free rate (often short-term Treasury yields, or zero in simplified backtests).
- Divide by the standard deviation of returns (volatility) over the same period.
In plain terms:
Sharpe ≈ (average return − risk-free rate) / volatility of returns
When annualized, you can compare strategies run on different bar sizes as long as you use the same convention.
How to read the number
There is no universal "passing grade," but rough heuristics help when you are screening ideas:
| Sharpe (annualized) | Typical interpretation |
|---|---|
| Below 0 | Returns did not beat the risk-free benchmark on average |
| 0 to 0.5 | Modest risk-adjusted performance |
| 0.5 to 1.0 | Solid for many systematic strategies |
| Above 1.0 | Strong; verify it holds out-of-sample |
A strategy with a sky-high Sharpe on a short backtest often means overfitting or too little data, not a free lunch. Always stress-test with walk-forward logic, different start dates, and realistic costs.
Sharpe vs raw return
Two strategies can have the same cumulative return with very different Sharpes:
- Smooth equity curve, moderate return → higher Sharpe.
- Lumpy returns, same ending wealth → lower Sharpe.
That is the point: Sharpe penalizes paths that swing wildly even if they occasionally finish strong.
Limits you should know
- Assumes volatility is "bad" symmetrically - large upside moves count against you the same as downside moves.
- Sensitive to outliers - one bad week can drag the ratio down sharply on short samples.
- Not a substitute for max drawdown - you still need to ask how deep the worst peak-to-trough loss was.
For a fuller picture, pair Sharpe with max drawdown, Calmar ratio, and time in market when you evaluate a backtest.
Using Sharpe on Quantly
When you run a backtest or open a tear sheet, Sharpe is computed from the strategy's return series over the period you selected. Use it to compare variants of the same strategy (different parameters, filters, or rebalance rules), not as the only scorecard for unrelated ideas.
Ready to test a hypothesis? Run a demo backtest or join the waitlist for full access.
