What Is the Treynor Ratio? Excess Return per Unit of Market Risk

The Treynor ratio measures how much return a portfolio earned above the risk-free rate for each unit of market risk it carried, where market risk is measured by beta. It is the Sharpe ratio with the denominator swapped: total volatility out, sensitivity to the market in.

The swap encodes an assumption worth being explicit about. Treynor assumes the risk specific to individual holdings has already been diversified away, so the only risk that should count against a portfolio is the part it cannot escape by holding more names: the part that moves with the market.

What it measures

The formula in words:

Treynor ratio = (portfolio return minus the risk-free rate) divided by the portfolio’s beta.

The numerator is the excess return, the same quantity the Sharpe ratio uses. Portfolio return over the window, minus what could have been earned in a near risk-free instrument such as short-dated government treasury bills. It is the part of the return that had to be earned by taking risk.

The denominator is beta. Beta measures how much a portfolio has historically moved for a given move in its benchmark. It is estimated by regressing the portfolio’s returns on the benchmark’s returns over a chosen window, and it is the slope of that relationship. A beta of 1 means the portfolio has tended to move roughly in line with the market. A beta of 1.4 means it has tended to move about 40 percent more than the market in both directions. A beta of 0.7 means it has tended to move less. For the estimation details and the many ways beta is misread, see what is beta in investing.

The two kinds of risk being separated here have standard names. Systematic risk is the part driven by market-wide forces: rates, liquidity, growth, sentiment. No amount of diversification within equities removes it. Unsystematic risk, also called specific or idiosyncratic risk, is the part driven by things particular to individual companies. Holding more names shrinks it.

Sharpe charges a portfolio for both. Treynor charges it only for the first.

How to read it

The output is an excess return per unit of beta, so it is expressed in return units rather than as a pure number. Some illustrative arithmetic: if a portfolio returned 16 percent, the risk-free rate was 6 percent, and its beta was 1.25, the Treynor ratio is 10 divided by 1.25, which is 8. A second portfolio returning 12 percent with a beta of 0.6 has a ratio of 6 divided by 0.6, which is 10. The second delivered less headline return but far more return per unit of market exposure.

A useful reference point: the market itself has a beta of 1 by definition, so the market’s own Treynor ratio is simply its excess return. A portfolio with a higher Treynor ratio than the benchmark produced more excess return per unit of market exposure than the benchmark did over that window.

Reading habits that matter.

• Check the benchmark used to estimate beta. Beta is not a property of a portfolio. It is a property of a portfolio relative to a chosen index. The same portfolio measured against a broad index and against a narrow one will have different betas, and therefore different Treynor ratios. See benchmark selection for portfolios.
• Check the estimation window and frequency. Beta computed from two years of weekly returns and from five years of monthly returns are different estimates, sometimes materially.
• Look at the R-squared alongside it. R-squared tells you how much of the portfolio’s movement the benchmark actually explains. If the benchmark explains very little, the beta estimate is weak and the ratio built on it is weaker. See what is R-squared in investing.
• Be careful when beta is very small. Dividing by a number close to zero produces enormous ratios that carry no information. Some providers suppress the figure in that case; many do not.

Treynor and Sharpe disagree exactly when a portfolio carries a lot of risk that is not market risk. That disagreement is a measurement of concentration.

When Treynor is the right lens

The choice between Treynor and Sharpe is really a question about context, and it has a clean answer.

Use Treynor when the portfolio is a component. If a fund or sleeve sits inside a larger, already diversified portfolio, then its stock-specific risk mostly washes out against everything else held alongside it. What it genuinely contributes to the total is its market exposure. Charging it for specific risk that the wider portfolio has already neutralised overstates its cost. This is the situation Treynor was designed for.

Use Sharpe when the portfolio is the whole thing. If this portfolio holds all of the capital in question, then the investor bears every kind of risk in it, diversifiable or not. Total volatility is then the honest denominator.

The practical implication is that a concentrated portfolio will often look better on Treynor than on Sharpe, because Treynor simply does not charge it for its concentration. That is appropriate when the concentration is diversified away elsewhere and misleading when it is not.

What it does not tell you

It assumes diversification that may not exist. This is the central caveat. Treynor’s entire justification is that specific risk does not matter because it has been diversified away. Applied to a ten-stock portfolio held on its own, that assumption is simply false, and the ratio will flatter the portfolio for exactly the risk that is most likely to hurt.

Beta is a historical estimate, not a constant. Beta drifts. It changes with the portfolio’s holdings, with the market regime, and with the estimation window. A ratio built on a single point estimate of beta carries all of that instability, silently.

Beta assumes a straight-line relationship. The regression that produces beta fits one slope through the whole sample. If a portfolio behaves differently in falling markets than in rising ones, a single beta averages the two together and describes neither. Upside and downside capture ratios exist to expose that asymmetry.

It is silent on depth. Like Sharpe, Treynor says nothing about the worst peak-to-trough fall. A portfolio with a strong Treynor ratio can still have put its holders through a severe drawdown. That requires maximum drawdown and the Calmar ratio.

It breaks down for negative betas and near-zero betas. The arithmetic still runs, but the resulting numbers are not interpretable in the usual way, and comparisons become nonsense.

It says nothing about skill against a benchmark. Treynor measures return per unit of market exposure, not return earned beyond what the benchmark delivered. That question belongs to alpha and to the information ratio, which divides active return by tracking error.

It depends on the benchmark being a fair one. If the index chosen does not represent what the portfolio actually does, every number downstream of it, beta included, is measuring the wrong relationship.

Using it in practice

Treynor is most useful as a comparison tool inside a defined universe: several funds measured against the same benchmark, over the same window, with the same return frequency, each intended to occupy a similar slot in a larger allocation. In that setting it answers a genuinely important question, which is which manager delivered more for the market exposure they consumed.

It is least useful as a standalone headline. Read it beside the beta itself, the R-squared, a drawdown measure, and at least one total-risk measure. When Treynor is strong and Sharpe is weak, the portfolio is carrying a lot of non-market risk, and the next question is whether anything in the wider allocation actually offsets it.

Related reading

• Portfolio metrics explained: the hub that maps every risk and return metric to the question it answers.
• What is beta in investing: the denominator on its own, how it is estimated and how it is misread.
• What is the Sharpe ratio: the same numerator with total volatility in the denominator.
• What is the information ratio: the benchmark-relative measure of active skill.
• Sharpe vs Sortino vs Calmar: how the main risk-adjusted ratios differ in what they charge for.

This article is educational. Altys Labs is not a registered research analyst or investment adviser, and nothing here is investment advice or a recommendation to buy, sell, or hold any security.