What VO₂max actually measures
VO₂max is the ceiling on how much oxygen your body can take in, transport and use per minute, normalised to your body weight and reported in millilitres per kilogram per minute. It is a whole-chain number: lungs get oxygen into blood, the heart’s stroke volume moves the blood, haemoglobin carries it, capillaries deliver it, and mitochondria burn it. Whichever link is weakest sets the ceiling — and in healthy people it is almost always cardiac output, which is why endurance training grows the left ventricle.
It matters for two separate reasons. For athletes it caps sustainable pace: you can be a wonderfully efficient runner, but you cannot spend oxygen you never delivered. For everyone else it is one of the strongest single predictors of all-cause mortality in the epidemiological literature — stronger than smoking status, blood pressure or cholesterol in several large cohorts. Moving from the bottom quintile of fitness to merely below-average is associated with the largest risk reduction of any step on the scale.
The three methods, ranked by accuracy
1. Race result (Daniels/Gilbert VDOT) — best
If you have raced flat out in the last few weeks, this is the estimate to use. Jack Daniels and Jimmy Gilbert fitted two curves: one for the oxygen cost of running at a given speed, and one for the fraction of VO₂max a human can hold for a given duration. Dividing the first by the second gives VDOT — a “VO₂max equivalent” that also folds in running economy.
A 20:00 5K gives a VDOT of about 49.8. Because it is anchored to real performance, it is the most predictive of the three — but note that it rewards efficiency too, so a technically excellent runner will score slightly above their lab VO₂max, and a novice with a low cadence and heavy footstrike slightly below.
2. Cooper 12-minute run — good, and honest
Kenneth Cooper built this in 1968 for the US Air Force by regressing 12-minute run distance against treadmill VO₂max in 115 servicemen, correlation r ≈ 0.90. Run as far as you can in exactly twelve minutes on a track, take the distance, done. In the general population the correlation is closer to r ≈ 0.7–0.8 — the loss comes from pacing skill and willingness to hurt, not from the equation. Go out too hard and you will underestimate yourself by several ml/kg/min.
3. Resting heart rate ratio (Uth–Sørensen) — a rough screen
Uth, Sørensen, Overgaard and Pedersen derived the constant 15.3 (±0.7) from 46 well-trained men in the European Journal of Applied Physiology in 2004. It works because a big stroke volume shows up as a low resting pulse and a wide heart-rate range. But the model was fitted on trained men, and it needs a genuinely measured HRmax — plugging in 208 − 0.7 × age adds its own ±10 bpm of error, which propagates straight into the result. Use it when you have no recent test; do not use it to argue with a race result.
Norm table by age and sex
These are the Cooper Institute normative values in ml/kg/min, the same table reproduced in ACSM’s guidelines. “Good” is the middle-upper band, roughly the 50th to 75th percentile of that cohort.
| Age | Men — good | Men — excellent ≥ | Women — good | Women — excellent ≥ |
|---|---|---|---|---|
| 20–29 | 42.5–46.4 | 46.5 | 33.0–36.9 | 37.0 |
| 30–39 | 41.0–44.9 | 45.0 | 31.5–35.6 | 35.7 |
| 40–49 | 39.0–43.7 | 43.8 | 29.0–32.8 | 32.9 |
| 50–59 | 35.8–40.9 | 41.0 | 27.0–31.4 | 31.5 |
| 60+ | 32.3–36.4 | 36.5 | 24.5–30.2 | 30.3 |
One caveat worth stating plainly: that cohort came from people who presented at a fitness centre for a treadmill test, so it skews fitter than the population at large. If you land in “fair” you are probably around average for your country, not below it.
How to actually raise it
The training answer has been stable for forty years and is boringly simple: a large base of easy aerobic work plus a small dose of very hard work.
- Zone 2, most of the week. Roughly 80% of your training time at a conversational effort — about 60–70% of max heart rate. This is what builds capillaries, mitochondria and stroke volume. Our heart rate zone calculator gives you the exact bpm windows.
- One or two interval sessions. Three to five minutes at 90–100% of max heart rate, four to six times, with equal recovery. This is the stimulus that pushes the ceiling itself.
- Enough volume, for long enough. Untrained people typically gain 15–20% in twelve weeks. Trained athletes fight for 2–3% a year, and mostly improve race times through economy and lactate threshold instead.
- Recover. Intervals only work if you absorb them. Sleep and autonomic recovery gate the whole thing — see our HRV calculator and sleep debt calculator.
Honest words about watch estimates
Apple Watch, Garmin and Polar all estimate VO₂max from the relationship between your pace and your heart rate during outdoor running, calibrated against a population model. For steady running on flat ground the estimate is decent — usually within 10–15% of a lab value, and reliable enough to track direction over months. It degrades when the input data is messy: hill repeats, treadmill running with GPS off, cycling, hot weather, caffeine, or a chest strap swapped for a wrist sensor mid-block.
The practical rule: never compare your watch number to anyone else’s watch number, and never react to a single reading. A 1–2 ml/kg/min move is measurement noise. A steady 3-point climb across two months is real fitness. That distinction — trend against your own baseline rather than one value against a population table — is the same idea Helix applies to recovery and sleep.
Once you have a VO₂max estimate, the useful next step is turning it into pace targets. Feed a race result into our race time predictor for equivalent times at every distance, or use the pace calculator to plan splits for the effort you are actually capable of holding.