The Sharpe Ratio: Formula, Worked Example, and When a High Score Is a Warning
Key takeaway
The Sharpe ratio divides an investment's return above the risk-free rate by the standard deviation of those returns, measuring reward per unit of volatility. Above 1 is generally considered good. A very high figure often signals hidden tail risk rather than skill, because volatility and risk are not the same thing.
The Sharpe ratio measures how much return an investment produced per unit of volatility. It answers a question raw performance cannot: whether a 20% gain came from skill or from taking twice the risk of everything it is being compared against.
It is also the most misused number in fund marketing, for a reason built into its construction: it treats upside volatility as a penalty and it can be inflated by strategies that quietly sell tail risk. This guide covers the formula, a worked example, how to read the result, and the specific circumstances in which a high Sharpe ratio is a warning rather than a recommendation.
The Sharpe ratio formula
The ratio subtracts the risk-free rate from the portfolio’s return and divides by the portfolio’s standard deviation:
Sharpe ratio = (Rp − Rf) ÷ σp
| Term | What it is | Where it comes from |
|---|---|---|
| Rp | Return of the portfolio or fund over the period | Fund factsheet or your own records |
| Rf | Risk-free rate over the same period | Short-dated government bill yield in the portfolio’s currency |
| Rp − Rf | Excess return — the part you were not going to get for free | Calculated |
| σp | Standard deviation of the portfolio’s returns | Calculated from the return series, at a consistent frequency |
Two mechanical points that cause most errors in practice. First, the numerator and denominator must cover the same period at the same frequency: monthly returns with monthly volatility, or annual with annual. Second, to annualise from a higher frequency, multiply the ratio by the square root of the number of periods in a year — √12 for monthly data, roughly √252 for daily. A Sharpe ratio quoted without its measurement frequency is not comparable to anything.
A worked example
Two funds, one year, with a risk-free rate of 4%.
| Fund A | Fund B | |
|---|---|---|
| Return | 12% | 8% |
| Volatility (standard deviation) | 15% | 6% |
| Excess return | 12 − 4 = 8% | 8 − 4 = 4% |
| Sharpe ratio | 8 ÷ 15 = 0.53 | 4 ÷ 6 = 0.67 |
Fund A returned half as much again as Fund B and is the worse fund on this measure. It needed 15% volatility to produce 8% of excess return; Fund B produced 4% on 6%. An investor who could borrow or lever Fund B toward Fund A’s risk level would have ended up ahead.
The practical consequence: the Sharpe ratio is a comparison tool, not a score. A value of 0.67 means nothing in isolation — it means something against 0.53 for a comparable fund over an identical period.
How to read the result
| Sharpe ratio | Conventional reading | What to check before believing it |
|---|---|---|
| Below 0 | The investment underperformed cash while taking risk | Whether the period simply contained a bear market |
| 0 to 1 | Ordinary. Most funds and most broad indices sit here over long periods | Nothing unusual |
| 1 to 2 | Good | Whether the measurement window excludes a crisis |
| 2 to 3 | Very good, and uncommon over long horizons | The return distribution and the strategy’s exposure to rare events |
| Above 3 | Exceptional — and the level at which scepticism should increase, not decrease | Whether the strategy is short volatility, illiquid, or marked to model rather than to market |
Broad equity indices have historically produced Sharpe ratios in the region of 0.3 to 0.5 over long periods. Any retail product advertising a sustained figure several times that is describing either a genuinely rare edge or a risk the standard deviation is not capturing.
Where the Sharpe ratio breaks
It penalises upside volatility
Standard deviation measures dispersion in both directions. A fund that occasionally returns +15% in a month is penalised exactly as much as one that occasionally loses 15%. No investor experiences those two events as equally undesirable, which is the ratio’s most fundamental mismatch with how risk is actually felt.
It assumes returns behave normally
The formula treats standard deviation as a sufficient description of risk, which holds when returns follow a normal distribution and fails when they do not. Financial returns have fat tails: extreme moves occur far more often than a normal distribution predicts. For strategies whose returns are skewed, the Sharpe ratio systematically understates the risk.
It can be gamed by selling tail risk
This is the important one. Consider a strategy that repeatedly sells out-of-the-money options and collects small premiums. In calm markets it produces steady, low-volatility gains — a textbook high Sharpe ratio, often above 2. The risk has not been removed; it has been converted from frequent small fluctuations into a rare enormous loss that has not happened yet.
The measured volatility is low precisely because the loss is absent from the sample. When it arrives, it arrives all at once. A high Sharpe ratio computed over a period that excludes the strategy’s characteristic disaster is not merely unhelpful — it is most confident exactly when it is most wrong. This is the arithmetic behind the risk of ruin, and why position sizing cannot be delegated to a performance statistic.
It is sensitive to the window
Change the start date by six months and the number changes materially. Funds choose the periods they advertise. Comparing two Sharpe ratios computed over different windows is meaningless, and comparing them over the same window still assumes both strategies face the same risks within it.
The alternatives, and what each fixes
| Measure | What it divides by | The problem it addresses |
|---|---|---|
| Sharpe ratio | Total standard deviation | The baseline: return per unit of total volatility |
| Sortino ratio | Downside deviation only | Stops penalising upside volatility. The natural fix for the first flaw above |
| Calmar ratio | Maximum drawdown | Measures against the worst loss actually experienced rather than average dispersion — harder to hide a tail event from |
| Information ratio | Tracking error against a benchmark | Isolates the manager’s contribution from the market’s, rather than from cash |
| Treynor ratio | Beta | Uses market risk instead of total risk, for a position held inside an already diversified portfolio |
None of these replaces the others. The practical habit is to look at the Sharpe ratio first because it is universally reported, then at the maximum drawdown, because the two together reveal what the average has hidden.
History and origins
William F. Sharpe introduced the measure in a 1966 paper on mutual fund performance, calling it the reward-to-variability ratio; the name Sharpe ratio was applied by others afterwards. Sharpe went on to share the 1990 Nobel Memorial Prize in Economic Sciences for his work on the capital asset pricing model.
He revised the definition in 1994 to compare a portfolio against a chosen benchmark rather than against the risk-free rate alone, generalising it into what is effectively the information ratio. The 1966 form remains the version quoted on factsheets, which is worth knowing when two sources report different figures for the same fund.
Frequently asked questions
What does the Sharpe ratio measure?
Excess return per unit of volatility: how much an investment returned above the risk-free rate, divided by the standard deviation of those returns. It exists to make two investments with different risk levels comparable.
What is a good Sharpe ratio?
Above 1 is generally considered good and above 2 very good, but the figure is only meaningful against a comparable investment over an identical period. Broad equity indices have historically sat around 0.3 to 0.5 over long horizons, so a sustained figure far above that deserves scrutiny of what risk the volatility measure is missing.
What is the Sharpe ratio formula?
(Rp − Rf) ÷ σp, where Rp is the portfolio return, Rf the risk-free rate over the same period, and σp the standard deviation of the portfolio’s returns. To annualise from monthly data, multiply by √12; from daily data, by approximately √252.
Can the Sharpe ratio be negative?
Yes. A negative value means the investment returned less than the risk-free rate over the period, so the investor took risk and was not compensated for it. In a negative reading the ratio’s ordering breaks down — a more volatile fund can show a less negative figure — so it should not be used to rank losing investments.
What is the difference between the Sharpe and Sortino ratios?
The Sharpe ratio divides by total standard deviation, penalising upside and downside movement equally. The Sortino ratio divides by downside deviation only, so a fund is not marked down for occasional large gains. Sortino is generally the more informative of the two for asymmetric strategies.
Why can a high Sharpe ratio be misleading?
Because volatility is not the same thing as risk. Strategies that sell insurance against rare events — short volatility, some credit and some illiquid strategies — produce steady small gains and therefore low measured volatility, right up until a single event erases years of them. The ratio is highest immediately before that event.
Related reading
- Position sizing and the risk of ruin — why a strong risk-adjusted return still does not tell you how much to allocate.
- How to manage investment risk — the wider framework the ratio is one input to.
- Algorithmic trading strategies — where Sharpe ratios are routinely quoted for backtests and routinely overstated.
- Crisis-proof investments — how strategies with attractive ratios behave when the tail arrives.
- Sustainable and green ETFs — comparing funds on risk-adjusted rather than headline returns.
- Portfolio construction in 2026 — the allocation decision the ratio informs but does not make.
This article is general information, not personalised investment advice. It does not take into account the financial situation, objectives or risk tolerance of any individual reader, and the same text is distributed to all readers. Illustrative figures are examples, not forecasts. Capital is at risk and past performance does not indicate future results.
