The Sharpe ratio is one line of arithmetic: excess return divided by standard deviation. Most candidates compute it correctly the first time they see it.
That is not where the points are. The exam tests interpretation: which measure of risk belongs in the denominator for a given investor, what a ranking actually means, and what you can conclude when the ratio turns negative.
The Sharpe ratio is simple to compute and easy to misuse. The distinction matters.
Definition: The Price of a Unit of Total Risk
Sharpe ratio = (Rp - Rf) / σp
Rp is the portfolio return, Rf the risk-free rate and σp the standard deviation of the portfolio's returns. The numerator is the reward for bearing risk. The denominator is total risk: every source of volatility, diversifiable or not.
In practice, the ratio tells you how much excess return a portfolio delivered per unit of total risk. Graphically, it is the slope of the line from the risk-free asset to the portfolio in return and standard deviation space. The capital market line is that line drawn to the market portfolio, so its slope is the market's Sharpe ratio.
One property matters more than it first appears. Combining a portfolio with the risk-free asset, by lending or borrowing at Rf, changes both return and risk but leaves the Sharpe ratio unchanged. Leverage alone cannot raise it, which is what makes the ratio a fair comparison across portfolios with different risk levels.
Sharpe vs Treynor: Total Risk or Systematic Risk
Treynor ratio = (Rp - Rf) / βp
The only change is the denominator. Treynor uses beta, which captures systematic risk alone. That changes the question the ratio answers.
If the portfolio is the investor's entire holding, total risk is what the investor bears, and Sharpe is the right measure. If the portfolio is one of many in a well-diversified allocation, its unsystematic risk is largely diversified away, and only its beta adds to overall risk. Then Treynor is the appropriate measure.
For well-diversified portfolios the two tend to rank the same way, because there is little unsystematic risk to separate them. For concentrated portfolios they can disagree, and explaining that disagreement is a classic exam task.
M-squared expresses the Sharpe ratio in percentage terms. It scales the portfolio with the risk-free asset until its standard deviation matches the market's, then reports the return of that scaled portfolio:
M² = (Rp - Rf) × (σm / σp) + Rf
Because it is a rescaled Sharpe ratio, M-squared always ranks portfolios in the same order as Sharpe. Its advantage is that the result can be compared directly with the market return. M-squared minus the market return is the risk-adjusted outperformance, sometimes called M-squared alpha.
Worked Example: Two Funds, Two Answers
Illustrative data: the risk-free rate is 3%, and the market returned 9% with a standard deviation of 16%.
| Fund | Return | Standard deviation | Beta | Sharpe | Treynor |
|---|---|---|---|---|---|
| A | 11% | 25% | 0.9 | 8 / 25 = 0.32 | 8% / 0.9 = 8.89% |
| B | 9% | 14% | 0.8 | 6 / 14 = 0.43 | 6% / 0.8 = 7.50% |
| Market | 9% | 16% | 1.0 | 6 / 16 = 0.375 | 6% / 1.0 = 6.00% |
The measures disagree. On Sharpe, Fund B is better, and it is the only fund ahead of the market. On Treynor, Fund A is better, and both funds beat the market.
M-squared confirms the Sharpe view. Fund A scaled to market risk earns 0.32 × 16% + 3% = 8.12%, below the market's 9%. Fund B earns about 0.4286 × 16% + 3% = 9.86%.
The explanation is Fund A's concentration. Its total risk is high relative to its beta, so much of its volatility is unsystematic. For an investor who holds Fund A alone, that risk is real, and Sharpe correctly penalizes it. For an investor who holds Fund A as a small slice of a diversified portfolio, most of it diversifies away, and Treynor correctly ignores it.
On the exam, look for the sentence that tells you which investor you are. "The fund is the client's only investment" points to Sharpe. "One of several managers in a diversified allocation" points to Treynor.
The Negative Sharpe Trap
When excess returns are negative, the ratio's logic inverts. Dividing a negative number by a larger standard deviation produces a value closer to zero, which looks better.
Illustrative case: two funds both returned 2 percentage points less than the risk-free rate. Fund X had a standard deviation of 10%, Fund Y of 20%. Fund X's Sharpe ratio is -0.20 and Fund Y's is -0.10. Ranked mechanically, Fund Y wins, although it delivered the same shortfall with twice the risk.
The practical implication is simple. A negative Sharpe ratio tells you the portfolio failed to beat the risk-free asset. It does not support a reliable ranking of risk-adjusted performance among negative-Sharpe portfolios, so do not read higher as better.
A negative Sharpe ratio answers one question: did the portfolio beat the risk-free rate? It does not tell you which underperformer managed risk better.
Limitations: Non-Normal Returns and Smoothed Data
The Sharpe ratio uses standard deviation as its only measure of risk. That works when returns are roughly symmetric and close to normally distributed. Many strategies do not produce returns like that.
Strategies with negative skewness and fat tails, such as selling options or other insurance-like exposures, can produce steady small gains and rare large losses. Over a period without a loss event, measured volatility is low and the Sharpe ratio looks excellent. The risk is real; it simply has not appeared in the sample.
The second weakness is smoothing. Illiquid assets valued by appraisal or on stale prices report returns that move less than their true economic value. Measured standard deviation falls and the Sharpe ratio rises with no change in underlying risk, and managers with discretion over valuations can produce the same effect deliberately.
Finally, the ratio depends on the measurement period and the return frequency. Compare Sharpe ratios only when they are computed over the same period, from returns of the same frequency.
What to Take Into the Exam
Computing the ratio is the easy mark. Focus on the questions around it: whose risk is being measured, which ratio fits the investor, whether a ranking survives negative excess returns, and whether the return data can be trusted.
This is not about memorization. It is about judgment. The Sharpe ratio is a price, excess return per unit of total risk, and it is only meaningful for an investor who actually bears that total risk. The same logic runs through CAPM and the security market line, where beta replaces standard deviation for the same reason Treynor does.
