Sharpe ratio

Before you rely on it

What to check

Compare only over identical periods and with the same risk-free rate.

The common error

Where people go wrong with sharpe ratio

Ranking funds on Sharpe ratios computed over different windows. The number is not portable between calculations, and the ranking it produces is arbitrary.

Worked through

Sharpe ratio in plain numbers

Fund A returned 9% over the year with a volatility of 12%. Cash paid 3%. The Sharpe ratio is (9% − 3%) ÷ 12%, or 0.50. Fund B returned only 7% with a volatility of 6%. Its ratio is (7% − 3%) ÷ 6%, or 0.67. Fund B earned less but took far less risk for what it earned, so it scores better. If you could lever Fund B up to the volatility of Fund A, you would expect roughly 11% before financing costs. That is the sense of the measure: return per unit of risk. Numbers like these make a poorly diversified high-flyer look different next to a steady performer.

Easily confused

Sharpe ratio and its neighbours

The Sharpe ratio is not the Sortino ratio or the Treynor ratio. Sharpe divides excess return by total volatility, counting upside and downside swings alike. Sortino divides by downside deviation only. Treynor divides by beta, so it fits diversified portfolios. Information ratio compares against a benchmark. Sharpe is also different from return itself: two funds with equal returns and different volatilities rank differently; the result depends on the period and the risk-free rate chosen.

In real decisions

Where sharpe ratio turns up

Analysts and investors use it to compare funds and portfolios on a risk-adjusted basis. Fund fact sheets report it over three and five years; managers use it to check whether extra risk earned extra return; retirees might use it to compare two allocations that differ in bumpiness. A ratio above 1 is often described as good and above 2 as very good, though such thresholds depend on the asset class and period.

Know the edges

Where sharpe ratio stops being reliable

It assumes returns are roughly symmetrical and treats upside volatility as risk. Strategies that sell options or hold illiquid assets can show flattering ratios until a crash; it depends on the time window and how returns are sampled. Past ratios do not predict future results. Use it with drawdown and other measures.

Quick answers

What people ask about sharpe ratio

What is a good Sharpe ratio? Above 1 is often considered good and above 2 very good, but the answer depends on the asset class and period. Compare with similar funds.

Which risk-free rate do I use? A short-term government bill yield for the same period. State the choice so comparisons are consistent.

Can it be negative? Yes, when the return is below the risk-free rate. It then shows that the risk did not pay.

Can I compare Sharpe ratios across periods? Only if the periods use the same frequency and risk-free rate. Annualize consistently, and remember that short histories give noisy figures.