Risk-Adjusted Return

Framework for comparing investment return with total, downside, systematic, or benchmark-relative risk.

Risk-adjusted return evaluates investment performance in relation to a defined measure of risk. It is an umbrella concept, not one universal formula: different measures use total volatility, downside deviation, market beta, or benchmark-relative risk and can rank the same portfolios differently.

Key Takeaways

  • Raw return does not show how much or what kind of risk accompanied the result.
  • Every risk-adjusted measure embeds a definition of risk and a comparison baseline.
  • Sharpe, Sortino, Treynor, Jensen’s alpha, and Information Ratio answer different questions.
  • A higher historical ratio does not establish future performance, low loss potential, or suitability.
  • Drawdown, tail loss, liquidity, leverage, concentration, fees, and correlation often require separate review.

Why Raw Return Is Incomplete

If two portfolios each return 10%, the outcomes look identical until their paths and exposures are examined. One portfolio may have had larger fluctuations, deeper losses, more leverage, greater illiquidity, or a less favorable benchmark-relative result.

Risk adjustment does not make return objective or risk-free. It makes the analyst state which risk is being used in the comparison.

Main Risk-Adjusted Performance Measures

MeasureReturn comparisonRisk definitionBest suited to
Sharpe RatioReturn above a reference rateTotal standard deviationComparing total return variability
Sortino RatioReturn above a targetDownside deviation below that targetEvaluating failure to meet a stated objective
Treynor RatioReturn above the risk-free rateMarket betaComparing well-diversified portfolios against a common benchmark
Jensen’s AlphaReturn above a CAPM-implied returnBeta within CAPMEstimating model-relative value added
Information RatioReturn above a benchmarkTracking errorEvaluating active return per unit of active risk

These measures are not interchangeable. A denominator that is appropriate for an index-aware active manager may be unsuitable for a standalone portfolio or a strategy with asymmetric losses.

Worked Example: Comparing Four Ratios

Assume two hypothetical portfolios are evaluated over the same period with a 2% risk-free rate and downside target. Their benchmark returned 8%.

InputPortfolio APortfolio B
Portfolio return10.0%9.0%
Total standard deviation12.0%7.0%
Downside deviation below 2% target4.0%5.0%
Beta1.100.70
Tracking error3.0%1.5%

Sharpe Ratio

$$ S_A=\frac{10\%-2\%}{12\%}=0.67 \qquad S_B=\frac{9\%-2\%}{7\%}=1.00 $$

Portfolio B ranks higher because it earned more excess return per unit of total volatility.

Sortino Ratio

$$ So_A=\frac{10\%-2\%}{4\%}=2.00 \qquad So_B=\frac{9\%-2\%}{5\%}=1.40 $$

Portfolio A ranks higher because its measured downside deviation is lower even though its total volatility is higher. Favorable upside variation may be contributing to A’s standard deviation.

Treynor Ratio

$$ T_A=\frac{10\%-2\%}{1.10}=7.27\% \qquad T_B=\frac{9\%-2\%}{0.70}=10.00\% $$

Portfolio B ranks higher per unit of estimated systematic market risk. That comparison is meaningful only if both betas use the same suitable market benchmark and both portfolios are sufficiently diversified.

Information Ratio

$$ IR_A=\frac{10\%-8\%}{3\%}=0.67 \qquad IR_B=\frac{9\%-8\%}{1.5\%}=0.67 $$

The portfolios tie on benchmark-relative return efficiency even though their raw returns and other rankings differ.

The example does not identify one universally better portfolio. It shows that the analyst must select a measure that matches the decision being made.

Choosing the Right Measure

Start with the decision

  • For a standalone allocation, total volatility and portfolio-level drawdown may matter.
  • For a target-based mandate, downside deviation relative to the stated target may be more relevant.
  • For a well-diversified market portfolio, beta-based measures may help isolate systematic risk.
  • For an active manager, benchmark-relative return and tracking error may fit the mandate.

Align the inputs

Return, reference rate, target, benchmark, beta, and risk estimates must use compatible periods and frequencies. Gross returns should not be compared with net returns without explanation, and hedged returns should not be compared casually with unhedged benchmarks.

Use complementary risk measures

Value at Risk and Expected Shortfall are risk measures, not risk-adjusted return ratios. They can supplement performance analysis by describing loss thresholds and tail severity. Drawdown, liquidity, leverage, concentration, and correlation provide other information that a single ratio may omit.

Common Mistakes

  • Calling a portfolio superior because it has one higher ratio.
  • Comparing ratios calculated with different periods, frequencies, fee bases, or annualization methods.
  • Using an unsuitable benchmark or risk-free rate.
  • Treating upside volatility as harmful without considering the purpose of the analysis.
  • Ignoring beta instability, return smoothing, skew, tail losses, and nonlinear exposures.
  • Presenting a backtested ratio as a forecast or guarantee.

Practical Review Checklist

Before relying on a risk-adjusted result, verify:

  1. the return definition and measurement period
  2. the risk denominator and why it fits the decision
  3. the benchmark, target, or reference-rate convention
  4. gross-versus-net fee and transaction-cost treatment
  5. annualization and data-frequency assumptions
  6. sample length, changing exposures, and estimation uncertainty
  7. risks outside the metric, including drawdown, tails, liquidity, leverage, and correlation
  • Rate of Return: Measures the underlying gain or loss before selecting a risk adjustment.
  • Sharpe Ratio: Relates excess return to total return volatility.
  • Sortino Ratio: Relates return above a target to downside deviation.
  • Information Ratio: Relates average active return to benchmark-relative volatility.
  • Benchmark Index: Defines the comparison portfolio for active-return analysis.

Sources

FAQs

Can a lower-return portfolio have a better risk-adjusted return?

Yes. It can rank higher if the selected metric shows that it earned more return per unit of the risk being measured. A different valid risk definition may produce a different ranking.

Is risk-adjusted return the same as the Sharpe Ratio?

No. Sharpe is one risk-adjusted measure. Sortino, Treynor, Jensen’s alpha, and Information Ratio use different comparisons or definitions of risk.

Does the highest risk-adjusted return identify the best investment?

No. The metric may omit material risks, costs, taxes, liquidity needs, correlations, and investor objectives. Historical performance also does not guarantee future results.

This page is for financial education and does not recommend a portfolio, fund, strategy, benchmark, or risk level.

Browse Investing