A VO2 max figure means almost nothing on its own. Whether 34 mL/kg/min is a problem or a good result depends entirely on who is holding it: for a 28-year-old man it sits below the 25th percentile, and for a 62-year-old woman it sits above the 75th. The useful question is where your number falls in the distribution for your age and sex.
The Verdict
VO2 max chart by age: men
Values are millilitres of oxygen per kilogram of body weight per minute, measured by maximal treadmill cardiopulmonary exercise testing.
| Age | 25th percentile | Median (50th) | 75th percentile |
|---|---|---|---|
| 20 to 29 | 38 | 48 | 57 |
| 30 to 39 | 33 | 44 | 52 |
| 40 to 49 | 29 | 42 | 49 |
| 50 to 59 | 26 | 36 | 43 |
| 60 to 69 | 23 | 32 | 39 |
| 70 to 79 | not shown here | 24 | not shown here |
VO2 max chart by age: women
| Age | 25th percentile | Median (50th) | 75th percentile |
|---|---|---|---|
| 20 to 29 | 31 | 37 | 45 |
| 30 to 39 | 28 | 33 | 40 |
| 40 to 49 | 24 | 31 | 38 |
| 50 to 59 | 21 | 28 | 34 |
| 60 to 69 | 19 | 26 | 31 |
| 70 to 79 | not shown here | 18 | not shown here |
Both tables come from the Fitness Registry and the Importance of Exercise National Database, published by Kaminsky and colleagues in Mayo Clinic Proceedings in 2015 and updated since then. Figures are rounded to whole numbers. The 70 to 79 row carries the median only. Kaminsky and colleagues publish full percentiles for that decade; the quartile figures are omitted here because the secondary reproductions we worked from did not carry them consistently, and we would rather leave a cell blank than print a number we could not check.
How to read your percentile
| Where you land | What it means | Notes |
|---|---|---|
| Below the 25th percentile | Three in four people your age and sex test higher | The band where the association with all-cause mortality is strongest, and also the band where training produces the largest absolute gains. It is a starting point for training rather than a diagnosis. |
| 25th to 50th percentile | Below the median for your age and sex | Ordinary for a sedentary adult. Moving from here upward is the change with the best-documented payoff. |
| Around the 50th percentile | The median of the FRIEND reference population | Note that FRIEND is drawn from people referred for clinical exercise testing rather than a random sample of the population. See the caveat below. |
| 75th percentile and above | Top quarter for your age and sex | Consistent with regular aerobic training. Further gains get smaller and take longer. |
The band worth paying attention to is the bottom quartile. In the Mandsager cohort described in the VO2 max guide, the mortality difference between the least fit group and the next group up was larger than the difference between any two fitter groups. That is an observational association rather than a demonstration that training moves your own risk. The evidence behind that association, and its limits, is covered in what VO2 max is and how it is measured.
Why the decline with age is steeper than it looks
VO2 max falls with age for reasons that are only partly about training. Maximum heart rate declines with age whether or not you exercise, and that caps how much blood the heart can move at peak effort. Muscle mass tends to fall alongside it, reducing the tissue that extracts oxygen from the blood.
The third reason is arithmetic rather than physiology. VO2 max is expressed per kilogram of body weight, so gaining fat mass lowers the number with no change in the heart or the muscle at all. Someone who adds ten kilograms over a decade will see their VO2 max fall even if their absolute oxygen uptake is unchanged. That makes body composition a confounder in any long-run comparison of your own results.
Training does not halt the age-related decline, and it moves the entire curve upward. A trained 60-year-old commonly tests above the median for a sedentary 30-year-old. Improvement at any age is well documented, and the size of the improvement is individual.
Comparing a watch estimate to this chart
A watch estimate does not belong on these tables without a caveat. The tables come from metabolic-cart testing; a watch infers the number from heart rate against GPS pace and never measures oxygen.
If the exact percentile matters to you, the answer is a laboratory test rather than a better watch. Why the two methods disagree, and the conditions that move a wearable figure on their own, are set out in the VO2 max guide.
What a percentile cannot tell you
A percentile places you among people of your age and sex who took the same kind of test. It cannot say why you sit where you do, and a position in a distribution is not an assessment of your health. A result in the bottom quartile alongside a resting heart rate, blood pressure or metabolic panel that is also out of range is worth raising with a clinician, as is any breathlessness or chest discomfort during ordinary exertion whatever the number says. The VO2 max guide lists the situations that warrant medical assessment in full.
Frequently Asked Questions
What is a good VO2 max for my age?
Read your number against the percentile column rather than against a single pass or fail threshold. In the FRIEND treadmill reference data, the median (50th percentile) runs about 48 mL/kg/min for men aged 20 to 29 and about 37 for women the same age, falling to roughly 24 and 18 by ages 70 to 79. Sitting at or above the 50th percentile for your age and sex is a reasonable working target for most people. A raw 35 sits in a different part of the distribution at 25 than it does at 65.
What is the average VO2 max by age?
Average depends on which population you average. The FRIEND registry median falls from roughly 48 mL/kg/min in men aged 20 to 29 to roughly 24 in men aged 70 to 79, and from roughly 37 to roughly 18 in women across the same span. Across those five decades that is a fall of roughly half, or near 13 percent per decade. FRIEND is built from people who underwent clinical cardiopulmonary exercise testing, so it is not a random sample of the general public, and figures from other reference sets differ.
Why does VO2 max decline with age?
Two things fall together. Maximum heart rate declines with age largely independently of fitness, which lowers the amount of blood the heart can move per minute at peak effort. Muscle mass also tends to fall, reducing the tissue extracting oxygen. Because VO2 max is expressed per kilogram of body weight, gaining fat mass lowers the number without anything changing in the heart or muscle. Training does not stop the decline, and it does shift the whole curve upward, so a trained 60-year-old commonly tests above an untrained 30-year-old.
Are VO2 max charts different for men and women?
Yes, and using the wrong one misreads the result by a wide margin. In the FRIEND medians above, female values run roughly 20 to 25 percent below male values at the same age. The usual explanations are lower haemoglobin concentration, smaller heart size and a higher proportion of body fat, which matters because the measure is divided by total body weight. A woman scoring 37 at age 25 is at about the median for her sex and would be near the 25th percentile if read against the male table.
Is the VO2 max on my watch comparable to this chart?
Not reliably. This table comes from maximal cardiopulmonary exercise testing on a treadmill with a metabolic cart measuring expired gas. A watch does not measure oxygen uptake at all; it estimates it from the relationship between your heart rate and your running pace, which is why the estimate degrades if you mostly cycle, walk hills, or train in heat. Wearable estimates track direction over time reasonably well and can be off by several mL/kg/min in absolute terms. Compare a watch figure to your own previous watch figures, and treat its position on this chart as approximate.
Do the FRIEND percentiles apply to everyone?
They are a reference rather than a universal standard, for the reasons set out above the tables. The two practical consequences are that a result from a cycle-ergometer test does not belong on these tables, since cycle values run lower and are published separately, and that the older Cooper Institute tables circulating online use different numbers. That is one reason the same result can be labelled "average" on one site and "good" on another.