What is a good VO₂max for your age? Two sets of norms, and why they disagree

VO₂max is the most oxygen your body can use per minute, per kilogram of body weight. The cleanest way to get it is in a lab: you run to exhaustion on a treadmill while a mask analyses your breath. The FRIEND registry collected 7,783 tests of exactly that kind from 8 US labs in 2014–15, in adults aged 20–79 without heart disease.

Its medians (the 50th percentile, the value half of people sit above), in mL/kg/min:

• Men: 48.0 at 20–29, 37.8 at 40–49, 32.6 at 50–59, 28.2 at 60–69, 24.4 at 70–79
• Women: 37.6 at 20–29, 26.7 at 40–49, 23.4 at 50–59, 20.0 at 60–69, 18.3 at 70–79

The spread around the median is wide. For men aged 50–59, the 25th percentile is 27.1, the 75th is 39.7 and the 90th is 45.6. For women of the same age the figures are 19.9, 27.6 and 32.0.

FRIEND has real limits. It is a convenience sample of people sent to labs, only 137 men and 98 women were in their 70s, and the 2015 table has no band for 80 and over.

The other widely used table comes from the Cooper Clinic in Dallas. Its values are not measured with gas analysis. They are predicted from how long a person lasts on a Balke treadmill test, or from the work rate they reach. The FRIEND paper prints both tables side by side, which is how the gap is known.

In young adults the two are close: the median for men in their 20s is 43.9 in Cooper and 48.0 in FRIEND. With age they pull apart. For men aged 70–79 the Cooper median is 30.9 against FRIEND’s 24.4. For women aged 60–69 it is 28.8 against 20.0, and for women aged 70–79, 27.6 against 18.3.

That changes the answer a person gets. A man of 72 with a VO₂max of 30 sits at about the 44th percentile against Cooper and the 73rd against FRIEND. A woman of 65 with 28 sits at about the 44th against Cooper and above the 90th against FRIEND.

Neither table is simply wrong. Cooper Clinic patients are a mostly well-educated, health-screened group, and their values are predictions from treadmill time. FRIEND is lab-measured but drawn from referrals. They answer slightly different questions, and a percentile means little until you know which question it answers.

The FRIEND authors describe the drop across age groups as about 10% per decade. Their 2022 update, with 22,379 tests in adults aged 20–89, puts the average treadmill decline at 13.5% (4.0 mL/kg/min) per decade, and 16.4% on a cycle. Its new treadmill standards are 1.5–4.6 mL/kg/min lower than the 2015 ones. Cycle tests read lower than treadmill tests, so the test mode matters when you compare a result with a table.

We have not been able to read the 2022 percentile table itself, so this site does not quote it.

Both FRIEND figures compare different people of different ages at one moment. Following the same people over time gives a different shape. In the Baltimore Longitudinal Study of Aging, 810 adults aged 21–87 had repeated treadmill tests over a median of 7.9 years. Their decline was 3–6% per decade in their 20s and 30s, and more than 20% per decade from their 70s on.

The decline was similar across every quarter of self-reported activity. That does not mean training cannot change the curve; the study was not built to test that. What it does show is that a flat 10% a decade understates how fast the last decades go.

There is no single cut-off. Three reference points are worth knowing:

• A floor. Apple’s Low Cardio Fitness alerts use FRIEND’s lowest fifth for ages 20–59, and 18 mL/kg/min for men or 15 for women from age 60.
• Independence at 85. One Canadian study estimated the minimum compatible with living independently at 85 at about 18 (men) and 15 (women). It is an estimate from one sample.
• The top. In a Cleveland Clinic study, the elite group for men aged 50–59 started at 14.0 METs and for women at 13.0. At 3.5 mL/kg/min per MET, that is 49.0 and 45.5. Those METs were estimated from treadmill speed and grade, not measured.

Between the floor and the top, being above the FRIEND median for your age and sex is a defensible meaning of good. The mortality studies found risk kept falling well above it, which is covered in what "five times the risk" means.

Longstanding scores VO₂max against the Cooper Institute table as Garmin republishes it, not against FRIEND. At older ages that is the stricter of the two: the same value ranks lower. Below the 40th percentile the source table has no points, so the app extrapolates. Its 80+ band is extrapolated by about 10% per decade from the 70s, which the Baltimore data and FRIEND 2022 suggest is generous.

Apple scores against FRIEND. The same watch value can therefore read as low fitness in one place and fine in the other. The app names its table on every result so you can tell which you are looking at.

The practical rule is to compare yourself with yourself: the same test, scored against the same table, each quarter. You can see where a number sits on the Cooper table in the calculator, and the methods are on the VO₂max test page. None of this is medical advice.