Upside and Downside Capture Ratio, Explained
Upside and downside capture ratios measure how much of a benchmark's gains and losses a portfolio picked up. Read as a pair, they describe a portfolio's asymmetry.
Upside capture and downside capture measure how much of a benchmark’s gains and how much of its losses a portfolio actually picked up. If a portfolio rose 8 percent in a period when its benchmark rose 10 percent, its upside capture for that period was 80 percent. If it fell 6 percent when the benchmark fell 10 percent, its downside capture was 60 percent. The two numbers are only useful as a pair, because what investors care about is the gap between them.
That gap has a name in plain English: asymmetry. Most people would happily give up some of a rally in exchange for absorbing much less of a fall. Capture ratios are the simplest way to check whether a portfolio has historically behaved that way, or whether it merely looks defensive in conversation.
What the two ratios measure
Both ratios are built the same way, from the same raw material: a series of portfolio returns and a series of benchmark returns over the same periods, usually monthly.
Upside capture takes only the periods when the benchmark was positive. It compounds the portfolio’s returns across those periods, compounds the benchmark’s returns across the same periods, and divides one by the other. In words: of everything the benchmark gained during its good stretches, what fraction did the portfolio gain.
Downside capture does the mirror image. It takes only the periods when the benchmark was negative, compounds the portfolio’s returns over those periods, compounds the benchmark’s, and divides. In words: of everything the benchmark lost during its bad stretches, what fraction did the portfolio lose.
Two mechanical details decide what the number means, and they vary between providers.
The first is the period length. Monthly capture and daily capture can tell noticeably different stories about the same portfolio, because a portfolio that wobbles inside a month but ends flat looks calm monthly and jumpy daily. Neither is wrong. They answer different questions.
The second is compounding versus averaging. Some calculations compound the returns within each bucket, others average them. Compounding is the more common convention and is closer to what an investor experienced, but you should know which one you are looking at before you compare two portfolios from two different sources.
How to read the pair
Read the numbers side by side, never one alone. Four broad shapes come up.
| Upside capture | Downside capture | What it describes |
|---|---|---|
| Around 100 percent | Below 100 percent | Kept up in rallies, lost less in falls. The pattern most investors say they want. |
| Below 100 percent | Below 100 percent | A more defensive profile. Whether it is attractive depends on how much less it fell relative to how much less it rose. |
| Above 100 percent | Above 100 percent | An amplified profile. More of the rally and more of the fall. Often a higher beta or more concentrated portfolio. |
| Below 100 percent | Above 100 percent | Less of the gains, more of the losses. The pattern nobody is aiming for. |
A common shorthand is the capture spread, which is simply upside capture minus downside capture, or the capture ratio, which is upside divided by downside. A portfolio with 90 percent upside and 70 percent downside has a spread of 20 points and a ratio of roughly 1.29. Both compress the pair into a single sortable number, and both lose information doing it. The underlying pair is always worth seeing.
The reason capture ratios resonate is arithmetic, not psychology. A portfolio that falls 40 percent needs about 67 percent to get back to where it started. One that falls 25 percent needs about 33 percent. Losing less is worth more than it first appears.
Where the number comes from matters
Capture ratios are entirely defined by three choices, and each of them can flatter or punish a portfolio.
The benchmark. Capture is measured against something. Compare a mid-cap heavy portfolio against a large-cap index and the capture numbers will mostly describe the size difference rather than anything about the manager. The benchmark has to be a fair representation of the opportunity set the portfolio actually fishes in, otherwise the ratios are measuring a mismatch.
The window. A period containing one long rally and one shallow correction produces very few down months, so downside capture may rest on a handful of observations. A number computed from six negative months is a fragile number. Ask how many periods sat in each bucket before you take either figure seriously.
The return basis. Portfolio returns net of costs against a benchmark price index that excludes dividends is not a like for like comparison. Total return versions of the index are the fairer yardstick, and whether the portfolio series is gross or net of fees changes the answer as well.
What it does not tell you
This is the section that keeps the metric honest.
It does not tell you the total return. Two portfolios can share identical capture ratios and end the decade in very different places, because capture says nothing about the sequence of the periods or about the periods when the benchmark was roughly flat.
It does not tell you about the worst single episode. Capture is an aggregate over all down periods. A portfolio can post a comfortable downside capture and still have suffered one brutal stretch that would have ended an investor’s patience. That is what maximum drawdown is for.
It does not distinguish skill from a tilt. A portfolio persistently underweight the most volatile part of the market will usually show low downside capture. That is a structural characteristic, not evidence of good timing. The same tilt in a different regime can look very different.
It does not account for cash drag, costs or taxes unless the return series already does. A portfolio that was only half invested will show flattering downside capture for a reason that has nothing to do with stock selection.
And it does not predict. Capture ratios are a description of how one portfolio moved against one index over one past window. Regimes change, mandates change, portfolio managers change, and past asymmetry carries no guarantee that the same asymmetry repeats.
Using it sensibly
The practical habits are simple. Look at the two numbers together and at the count of periods behind each. Check the benchmark is fair and the index is a total return version.Look at more than one window, including at least one containing a genuine market fall, and see whether the shape is stable. Then put the pair next to a drawdown figure and a measure of benchmark sensitivity such as beta, because capture describes the average behaviour while drawdown describes the worst of it.
Read that way, capture ratios are one of the more intuitive portfolio statistics available. They translate the abstract idea of risk-adjusted performance into a sentence a non-specialist understands: this portfolio kept up with most of the good and sidestepped some of the bad, or it did not.
Related reading
- Portfolio metrics explained: the hub for every risk, return and portfolio statistic in this series.
- What is maximum drawdown: the worst peak to trough fall, and why it complements capture ratios.
- What is beta in investing: sensitivity to the market, and how it relates to capturing more or less of it.
- What is R squared in investing: how much of a portfolio’s movement the benchmark explains at all.
- Downside deviation explained: another way of measuring only the bad half of volatility.
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.
Frequently asked questions
What is the upside and downside capture ratio?
Upside capture is the share of a benchmark's gains that a portfolio captured during the periods when the benchmark rose. Downside capture is the share of the benchmark's losses it absorbed during the periods when the benchmark fell. Both are usually expressed as percentages, and they are read together rather than alone.
Is a downside capture below 100 always good?
Not on its own. A portfolio can hold a large cash balance and post very low downside capture simply because it was barely invested. That same choice usually drags upside capture down too. The pair only means something when you look at both numbers and at how the portfolio achieved them.
What does a capture ratio not tell you?
It says nothing about total return, about the size of the worst single fall, about costs and taxes, or about whether the pattern was skill or the by-product of a sector or style tilt that happened to suit the period measured. It is a description of past behaviour against one chosen benchmark, not a forecast.