Risk-Adjusted Returns Explained: Sharpe Ratio, Sortino Ratio, and How to Evaluate Investment Performance

May 9, 2026 · guides · 14 min read

title: "Risk-Adjusted Returns Explained: Sharpe Ratio, Sortino Ratio, and How to Evaluate Investment Performance" excerpt: "Raw returns tell only half the story. Learn how the Sharpe ratio, Sortino ratio, Calmar ratio, Information ratio, and other risk-adjusted metrics reveal whether a portfolio is genuinely skillful or just taking on extra risk."

Raw returns are the most frequently cited performance number in investing, and they are also one of the most misleading. A fund that earned 25% last year sounds impressive until you discover it did so by concentrating in a single sector that happened to surge, or by running three times the leverage of a comparable benchmark. A fund that delivered 12% in the same period while holding a diversified, low-volatility portfolio may have performed far better on any honest measure of risk-adjusted return.

This guide explains the standard framework for evaluating investment performance beyond raw returns: the Sharpe ratio, Sortino ratio, Calmar ratio, Information ratio, Treynor ratio, Jensen's alpha, and the broader CAPM context that connects them. It also addresses common pitfalls including short track records, time-weighted versus money-weighted returns, and how major fund rating systems incorporate risk adjustment.

Why Raw Returns Are Insufficient

Consider two portfolios over a three-year period. Portfolio A compounds at 18% per year. Portfolio B compounds at 14% per year. On the surface, Portfolio A wins. But suppose Portfolio A achieved those returns by holding a single highly volatile sector, with year-by-year returns of +55%, -20%, and +34%. Portfolio B, by contrast, returned +16%, +12%, and +14% -- low variance, consistent compounding, and far less psychological stress for the investor holding it.

The fundamental problem with evaluating raw returns in isolation is that return is only one dimension of performance. Risk is the other. Without knowing how much risk was accepted to generate a given return, there is no basis for determining whether performance was skillful or merely lucky, diversified or dangerously concentrated, sustainable or dependent on a favorable environment that will not persist.

Three distinct risk-related problems arise when raw returns are used alone:

Survivorship bias. Funds that took excessive risk and blew up do not appear in return comparisons. The sample of observable funds is skewed toward those that survived, which inflates the apparent performance of high-risk strategies.

Benchmark mismatch. Comparing a small-cap growth fund's returns to the S&P 500 is not a meaningful comparison. The fund may have a different risk profile, factor exposure, and volatility profile entirely. Risk-adjusted metrics that account for benchmark tracking correct for this.

Time period sensitivity. A high-octane growth portfolio looks like genius in a bull market and catastrophic in a drawdown. Risk-adjusted metrics smooth this by normalizing return relative to the volatility experienced.

The solution is not to ignore returns but to express them per unit of risk. That is the function of every metric covered in this guide.

The Sharpe Ratio

The Sharpe ratio is the most widely used risk-adjusted performance metric in institutional finance. It was introduced by economist William F. Sharpe in 1966 and remains the standard starting point for evaluating portfolio performance.

Formula

The Sharpe ratio is calculated as:

Sharpe Ratio = (Rp - Rf) / StdDev(Rp)

Where:

The numerator is the excess return -- the amount by which the portfolio outperformed doing nothing but holding risk-free assets. The denominator is the volatility of that return stream. The ratio expresses how much excess return was earned for each unit of volatility accepted.

Interpretation

Below 0.5: The portfolio is delivering less than half a unit of excess return per unit of volatility. This warrants scrutiny. It is not automatically bad -- some defensive strategies accept lower Sharpe ratios in exchange for drawdown protection -- but it suggests the risk-return tradeoff is unfavorable for most investors.

0.5 to 1.0: Acceptable, particularly for funds with low correlation to broader markets. Many well-managed, diversified portfolios fall in this range.

Above 1.0: Generally considered good. The portfolio is earning more than one unit of excess return per unit of volatility.

Above 2.0: Excellent. Hedge funds and systematic strategies that consistently achieve Sharpe ratios above 2.0 are considered exceptional. Sustained ratios above 3.0 are rare and should invite scrutiny -- they may reflect genuine skill, favorable conditions, or, in some cases, smoothed or misreported returns.

Worked Example

Suppose a portfolio generated an annualized return of 14% over three years. The annualized standard deviation of monthly returns was 12%. The average 3-month Treasury yield over the same period was 4.5%.

Sharpe Ratio = (14% - 4.5%) / 12% = 9.5% / 12% = 0.79

A Sharpe ratio of 0.79 is acceptable but not exceptional. For every 1% of annualized volatility, the portfolio earned 0.79% of excess return above the risk-free rate.

Now suppose a second portfolio earned only 11% per year but with an annualized standard deviation of 6%.

Sharpe Ratio = (11% - 4.5%) / 6% = 6.5% / 6% = 1.08

Portfolio B, despite earning 3 percentage points less in raw return, has a meaningfully higher Sharpe ratio. On a risk-adjusted basis, it is the superior performer.

Sharpe Ratio Limitations

The Sharpe ratio has two significant structural limitations that every sophisticated investor should understand.

It penalizes upside volatility equally with downside volatility. Standard deviation measures dispersion in both directions. A portfolio that occasionally generates very large positive returns will have higher standard deviation -- and therefore a lower Sharpe ratio -- even if it never produces severe losses. This is a meaningful distortion for strategies with positively skewed return distributions (e.g., trend-following, options strategies with large positive tails).

It assumes normally distributed returns. Standard deviation is a complete risk measure only if returns follow a normal distribution. Most real-world return distributions have fat tails -- extreme events occur more frequently than a normal distribution predicts. The 2008 financial crisis, the March 2020 COVID crash, and the 2022 rate shock were all multi-standard-deviation events that occurred far more frequently than a normal distribution would suggest. The Sharpe ratio systematically underestimates tail risk in fat-tailed distributions.

These limitations motivated the development of the Sortino ratio.

The Sortino Ratio

The Sortino ratio addresses the primary structural weakness of the Sharpe ratio by separating downside volatility from upside volatility. It was developed by Frank Sortino and measures return per unit of downside risk only.

Formula

Sortino Ratio = (Rp - Rf) / Downside Deviation

Downside deviation is the standard deviation of returns that fall below a minimum acceptable return (MAR). In most implementations, the MAR is the risk-free rate, though some practitioners use 0% or a specific benchmark return.

To calculate downside deviation: take only the return periods where the portfolio fell below the MAR, square those shortfalls, average them across all periods (not just the bad periods), and take the square root. This last step -- averaging over all periods rather than just the negative ones -- ensures that a portfolio with many negative periods is penalized more heavily than one with the same magnitude of losses concentrated in a few periods.

Why the Sortino Ratio Is Better for Asymmetric Strategies

For a vanilla long-only equity fund, the Sharpe and Sortino ratios tend to give similar rankings because return distributions are roughly symmetric. But for strategies with asymmetric payoff profiles -- covered calls, protective puts, merger arbitrage, or long-term buy-and-hold compounders that occasionally surge -- the Sortino ratio gives a more accurate picture.

Consider a trend-following strategy that is flat or modestly positive most months but occasionally earns 15-20% in a strong trending month. Its standard deviation is elevated purely because of those large positive months. The Sharpe ratio penalizes the strategy for those gains. The Sortino ratio ignores them, focusing only on the months the strategy fell below the acceptable threshold.

Worked Example

Using the same portfolio from the Sharpe example: annualized return of 14%, risk-free rate of 4.5%. Now suppose that of the portfolio's 36 monthly returns, 10 fell below the risk-free rate hurdle. The downside deviation, calculated using those 10 shortfall months but averaged over all 36 periods, is 7%.

Sortino Ratio = (14% - 4.5%) / 7% = 9.5% / 7% = 1.36

The Sortino ratio of 1.36 is notably higher than the Sharpe ratio of 0.79. This divergence indicates that much of the portfolio's standard deviation came from upside volatility -- favorable for the investor -- not from harmful drawdowns.

When the Sharpe and Sortino ratios diverge significantly, it usually reveals something important about the shape of the return distribution. A much higher Sortino than Sharpe suggests positive skew (large upside events). A Sortino only modestly better than Sharpe suggests a roughly symmetric or even negatively skewed return stream.

The Calmar Ratio

The Calmar ratio focuses on a different dimension of risk entirely: maximum drawdown. It is calculated as:

Calmar Ratio = Annualized Return / Maximum Drawdown

Maximum drawdown is the peak-to-trough decline over a given period. If a portfolio reached a high of 1,000 and subsequently fell to 750 before recovering, the maximum drawdown is 25%.

A Calmar ratio above 1.0 means the portfolio earned more in annualized returns than its worst peak-to-trough loss. A ratio above 0.5 is generally acceptable for strategies where drawdown tolerance is the binding constraint.

The Calmar ratio is most relevant for practitioners who manage funds with redemption risk -- if clients withdraw capital during a drawdown, the portfolio cannot recover. It is also essential for investors who cannot psychologically or financially withstand deep losses, regardless of eventual recovery.

One caution: the Calmar ratio is highly sensitive to the measurement period chosen. A portfolio with a 5% maximum drawdown over three calm years may have a Calmar ratio of 4.0. Add one year of severe market stress and the drawdown might jump to 30%, collapsing the ratio to 0.5. Always evaluate Calmar ratios across multiple market cycles.

The Information Ratio

The Information ratio is designed for active managers who are explicitly measured against a benchmark. It is defined as:

Information Ratio = Active Return / Tracking Error

Where:

The Information ratio measures how consistently an active manager adds value above their benchmark, not just whether they occasionally beat it. A manager who outperforms by 2% per year with very consistent alpha -- low tracking error -- has a higher Information ratio than a manager who outperforms by 3% per year but with wild swings in relative performance.

As a rule of thumb: Information ratios above 0.5 indicate skilled active management. Ratios above 1.0 are exceptional. Most active managers, net of fees, have negative Information ratios over long periods -- which is a central argument for passive index investing.

The Treynor Ratio

The Treynor ratio substitutes beta for standard deviation in the denominator:

Treynor Ratio = (Rp - Rf) / Beta

Beta measures the sensitivity of a portfolio's returns to movements in the broader market (typically the S&P 500). A beta of 1.0 means the portfolio moves in lockstep with the market. A beta of 1.5 means the portfolio tends to move 50% more than the market in both directions.

The Treynor ratio is most useful when comparing portfolios that are held as components of a larger, well-diversified portfolio -- rather than as standalone investments. In that context, idiosyncratic (non-market) risk is already diversified away at the overall portfolio level, and the only relevant risk is systematic market risk as measured by beta.

For most retail investors who are evaluating a single fund or portfolio against alternatives, the Sharpe ratio (which captures total volatility) is more appropriate than the Treynor ratio (which captures only market-correlated risk).

Alpha, Beta, and the CAPM Framework

The Capital Asset Pricing Model (CAPM) provides the theoretical foundation for much of modern portfolio analysis. CAPM states that the expected return of an asset or portfolio is:

Expected Return = Rf + Beta x (Rm - Rf)

Where:

CAPM defines a baseline: given a portfolio's beta, what return should it have earned? The difference between actual return and CAPM-predicted return is Jensen's alpha.

Jensen's Alpha

Jensen's alpha = Rp - [Rf + Beta x (Rm - Rf)]

A positive Jensen's alpha means the portfolio outperformed what CAPM predicts given its level of market risk. A negative alpha means it underperformed relative to its risk-adjusted benchmark.

If a portfolio has a beta of 1.2, the risk-free rate is 4.5%, and the market returned 10%, CAPM predicts:

Expected Return = 4.5% + 1.2 x (10% - 4.5%) = 4.5% + 6.6% = 11.1%

If the portfolio actually returned 13%, Jensen's alpha is:

Alpha = 13% - 11.1% = 1.9%

That 1.9% represents return that cannot be explained by systematic market exposure alone. Whether it reflects genuine skill, factor exposure not captured by single-factor CAPM (such as value, size, or momentum tilts), or statistical noise requires further analysis.

Beta in Portfolio Context

Beta is a measure of systematic risk -- the risk that cannot be diversified away. A portfolio of 100 randomly selected U.S. stocks will typically have a beta close to 1.0. A portfolio tilted toward low-volatility defensive names (utilities, consumer staples) may have a beta of 0.7. A portfolio concentrated in high-growth technology names may carry a beta of 1.4 or higher.

Understanding a portfolio's beta is critical for stress-testing. In a year when the market falls 30%, a beta-1.4 portfolio can be expected to fall roughly 42% -- a meaningful practical difference from a beta-0.7 portfolio falling 21%.

Why Most Retail Investors Should Focus on Sharpe and Sortino

Alpha, beta, Jensen's alpha, and the Information ratio are all powerful tools, but they carry a practical complexity that limits their usefulness for most individual investors evaluating their own portfolios or selecting funds.

The core reason is benchmark sensitivity. Jensen's alpha requires selecting an appropriate benchmark. The wrong benchmark produces misleading alpha estimates -- a small-cap value fund benchmarked against the S&P 500 will appear to generate positive or negative alpha based largely on the size and value factor premiums, not manager skill. Proper factor-adjusted alpha estimation requires multi-factor models (Fama-French 3-factor, 5-factor, or Carhart 4-factor) with sufficient data.

The Sharpe ratio and Sortino ratio, by contrast, are straightforward to calculate and interpret without a benchmark. They require only the portfolio's return history, the risk-free rate, and basic statistics. They answer the most practically relevant question for individual investors: given the volatility I experienced, was the return adequate compensation?

The Sortino ratio is generally the more informative of the two for individual portfolio analysis, because it avoids penalizing favorable volatility and better captures the asymmetric risk profile that most investors actually care about -- losing money is the concern, not making money too quickly.

The Challenge of Short Track Records

One of the most pervasive errors in performance evaluation is drawing conclusions from short track records. Two to three years of outperformance -- even risk-adjusted outperformance -- is statistically close to meaningless.

The core problem is that even a completely unskilled manager has a roughly 50% chance of beating a benchmark in any given year by random chance. Over three years, the probability of beating in all three years by chance alone is 12.5% (0.5 cubed). In a universe of hundreds of active funds, dozens post above-benchmark returns for three consecutive years purely by chance, with no skill whatsoever.

Statistically meaningful conclusions about manager skill generally require 10 to 15 years of data -- and even then, fund strategies, teams, and market regimes change enough to complicate interpretation. Most academic research on active management uses rolling 10-year windows at minimum.

Practical implications:

Comparing ETF and Fund Performance on a Risk-Adjusted Basis

When selecting between ETFs or funds in the same category, risk-adjusted metrics provide a more rigorous comparison than raw returns. A practical framework:

Step 1: Define the peer group. Compare only funds with similar mandates, benchmark indices, and factor exposures. Comparing a large-cap blend ETF to a small-cap value ETF using Sharpe ratios is not informative.

Step 2: Calculate or source Sharpe ratios. Most fund data providers (Morningstar, Portfolio Visualizer, fund company fact sheets) publish rolling 3-year and 5-year Sharpe ratios. Use the same time period for all funds being compared.

Step 3: Check the Sortino ratio. If two funds have similar Sharpe ratios but different Sortino ratios, the one with the higher Sortino ratio likely has a more favorable return distribution -- fewer and smaller drawdowns relative to upside.

Step 4: Evaluate the Calmar ratio across a period that includes stress. If the comparison period includes 2022 or 2020, the Calmar ratio reveals how each fund managed drawdown risk.

Step 5: Adjust for fees. Risk-adjusted metrics are calculated on net returns in most databases, meaning fees are already reflected. But verify -- some databases use gross-of-fee returns, which flatters active fund performance.

Time-Weighted vs. Money-Weighted Returns

A critical technical distinction that affects how returns are calculated and compared is the difference between time-weighted and money-weighted returns.

Time-weighted return (TWR) measures the compound growth rate of a single unit of currency invested at the beginning of the period, ignoring the timing and magnitude of subsequent cash flows. TWR is the standard for evaluating fund manager performance because it eliminates the distortion caused by investor cash flows in and out of the fund -- which are outside the manager's control.

Money-weighted return (MWR), also called the internal rate of return (IRR), accounts for the timing and size of actual cash flows. It reflects the actual return experience of the investor, not just the fund. If an investor added a large amount of capital just before a drawdown, their MWR will be lower than the fund's TWR. Conversely, if they added capital just before a strong rally, their MWR will exceed the fund's TWR.

For performance evaluation of managers and funds, TWR is the appropriate metric. For evaluating an individual investor's own portfolio experience -- how did my actual invested dollars perform -- MWR is more informative.

When computing Sharpe or Sortino ratios, TWR-based periodic returns (monthly or quarterly) are the standard input. Using MWR-based returns would mix manager skill with investor timing behavior, corrupting the metric.

Why Morningstar Ratings Use Risk-Adjusted Metrics

Morningstar's widely-followed star ratings are not based on raw returns. They are based on a risk-adjusted performance measure that Morningstar calls the Morningstar Risk-Adjusted Return (MRAR).

The MRAR framework is conceptually similar to the Sharpe ratio but uses a utility-function-based approach that more heavily penalizes downside outcomes, reflecting the empirical finding that investors experience losses more acutely than equivalent gains (loss aversion, as described in prospect theory).

Funds are ranked within their Morningstar category based on MRAR. The top 10% receive 5 stars, the next 22.5% receive 4 stars, the middle 35% receive 3 stars, the next 22.5% receive 2 stars, and the bottom 10% receive 1 star.

The practical implication is that a fund can have high raw returns and a mediocre star rating if it achieved those returns through high volatility. Conversely, a fund with modest raw returns but very low volatility and drawdowns may earn 4 or 5 stars. The rating system is explicitly optimized to surface risk-efficient performance, not just raw return-chasing.

One important caveat: Morningstar ratings are backward-looking. Academic research consistently finds that past star ratings have limited predictive power for future performance -- 5-star funds do not systematically continue to outperform. The ratings are useful for identifying historical risk efficiency, not for selecting future winners.

Practical Summary: How to Evaluate Any Portfolio

A rigorous performance review of any portfolio or fund should include at minimum:

  1. Raw annualized return vs. benchmark -- the baseline comparison, with an appropriate benchmark.
  2. Sharpe ratio -- excess return per unit of total volatility, annualized.
  3. Sortino ratio -- excess return per unit of downside volatility only.
  4. Maximum drawdown -- the worst peak-to-trough decline over the evaluation period.
  5. Calmar ratio -- annualized return divided by maximum drawdown.
  6. Beta -- sensitivity to market movements (for equity portfolios).
  7. Track record length -- treat anything under 5 years with appropriate skepticism.

No single metric is complete. A high Sharpe ratio combined with low Sortino ratio suggests the metric is being flattered by a distribution assumption mismatch. A high Calmar ratio over a short period may simply reflect a calm market. The metrics are most useful in combination, and across multiple market regimes.

The purpose of risk-adjusted metrics is not to identify winning funds after the fact. It is to distinguish between return that was earned efficiently -- through genuine skill, diversification, or disciplined risk management -- and return that was simply borrowed from future risk. That distinction is the foundation of serious investment analysis.

This content is for educational purposes only and does not constitute investment advice. Equity Rank does not provide personalized investment recommendations. All metrics, examples, and frameworks described are for illustrative and analytical purposes. Past performance of any strategy or metric relationship does not guarantee future results. Consult a qualified financial professional before making investment decisions.