What Is the Sharpe Ratio? How to Read Risk-Adjusted Return

What does the Sharpe ratio measure?

If two strategies both earned 20% in a year, which one is better? It depends on how bumpy the ride was. The Sharpe ratio shows, in a single number, how much excess return a strategy earned for each unit of risk it took.

How is it calculated?

The formula is simple:

Sharpe = (average return − risk-free return) ÷ standard deviation of returns

• Average return: the mean return over your chosen period (daily, weekly, monthly).
• Risk-free return: the return of a nearly risk-free alternative such as a deposit or a short-term government bill.
• Standard deviation: how much returns fluctuate around their average.

A value calculated from daily data is usually annualised by multiplying by roughly the square root of 252 (trading days). Always note which frequency you used, otherwise the numbers cannot be compared.

What is a good value?

There is no hard threshold, but as a rough guide:

• Below 1: the risk is not being well rewarded.
• Between 1 and 2: generally considered reasonable.
• Above 2: looks excellent, but a value this high usually comes from overfitting or from leaving costs out.

When you see a very high Sharpe, be suspicious before you celebrate, and check the test conditions.

Common mistakes

1. Ignoring commissions and slippage: A Sharpe calculated without costs comes out too high. See Commissions and slippage in backtests.
2. Too little data: A ratio from a few months of data is not reliable. You need a long period covering different market regimes.
3. Looking at one number only: Sharpe treats upside and downside volatility the same. Look at large drawdowns separately.
4. Using survivors only: Leaving delisted stocks out inflates results. See Survivorship bias explained.

Limits of the Sharpe ratio

It assumes returns are roughly normally distributed. For strategies with rare but large losses (tail risk), Sharpe can look good while hiding the real risk. Read it together with measures such as maximum drawdown and losing streaks. For the basics of risk rules, see the Algo trading risk management guide.

When is it useful?

Sharpe is most useful for comparing two strategies tested under the same conditions. On its own it does not say "buy" or "don't buy".

This article is educational and not investment advice. Past results do not guarantee future results; trading carries a risk of loss.