What this page is, and what it is not
VO2 max is the rate at which your body can take in and use oxygen at maximal effort, in millilitres per kilogram of bodyweight per minute. On this site it is treated as one thing only: an aerobic training metric, useful for setting paces and for watching your own trend over months.
It is not used here as a health measure, a risk score or a “fitness age”. Those readings exist in clinical settings, they depend on a properly measured value rather than a field estimate, and they are somebody's clinical judgement about a specific person rather than arithmetic on a web form. Nothing on this page interprets your number as anything other than a training figure.
The three protocols, written out
Cooper (Kenneth Cooper, 1968) — the 12-minute run. Run as far as you can in twelve minutes at an even effort, then:
VO₂ max = (metres − 504.9) ÷ 44.73
Developed for the US Air Force on a large military sample. Its virtue is that it needs nothing but a track and a watch; its weakness is that it depends entirely on your ability to pace twelve minutes correctly.
Daniels & Gilbert VDOT (1998) — from a race result. Two equations, one for the oxygen cost of a velocity and one for the fraction of maximum you can hold for a given duration:
VO₂ = −4.60 + 0.182258v + 0.000104v² (v in metres per minute)%max = 0.8 + 0.1894393e−0.012778t + 0.2989558e−0.1932605t (t in minutes)VDOT = VO₂ ÷ %max
VDOT is deliberately called a pseudo VO2 max by Daniels himself. It folds running economy into the figure, which is exactly why it predicts training paces better than a laboratory number would, and exactly why it is not the same quantity.
Uth, Sørensen, Overgaard & Pedersen (2004) — from heart rates.
VO₂ max ≈ 15.3 × (HRmax ÷ HRrest)
The cheapest and the crudest. It requires no test at all, and it inherits every bit of error in the maximum heart rate you feed it — which, if that maximum came from an age equation, is about ±10 bpm and moves this estimate by three points or more.
Worked example: 2,400 m in twelve minutes
- Cooper: (2,400 − 504.9) ÷ 44.73 = 1,895.1 ÷ 44.73 = 42.4 ml/kg/min
- The same person running 5K in 25:00: velocity 200 m/min, VO₂ 36.0, %max 0.940, VDOT = 38.3
- The same person with a maximum of 185 and a resting rate of 60: 15.3 × 3.083 = 47.2
Three protocols, the same body, and a spread of 8.9 points — more than a fifth of the headline figure. That is not a fault in any one of them; it is what field estimation of VO2 max is actually like, and it is the reason this page prints all three instead of picking one and going quiet.
Why they disagree
They are measuring different things through different keyholes. Cooper measures how far you can run in twelve minutes, which rewards a runner who paces well and punishes one who does not. VDOT measures a race performance, which includes running economy — two runners with identical oxygen uptake but different technique produce different VDOTs, and the more economical one is genuinely faster. The Uth ratio measures neither: it infers capacity from the range your heart operates over, which correlates with fitness but is also moved by genetics, medication, caffeine and how well you slept.
None of the three is the laboratory quantity. That requires a graded exercise test with a gas analyser, and the field estimates are typically within about ±10–15% of it.
How to use the number without over-reading it
Compare it against your own previous figure, produced by the same protocol, on the same course, in similar conditions. Used that way it is a useful trend line: a genuine improvement in aerobic fitness shows up as a rising Cooper distance or a rising VDOT, and that signal is bigger than the noise.
Compare it against somebody else's number, or against a table of population norms, and you are comparing two estimates whose error bars are wider than almost any difference you would care about. Add that field protocols are usually kinder than laboratory tests, and cross-person comparison stops meaning much at all.
The one genuinely operational use is VDOT: because it is calibrated on race performances, it maps cleanly onto training paces. If you want a number to do something with, use the race-based estimate and turn it into paces, rather than admiring the figure itself.
Limitations, stated plainly
Every protocol here is a field estimate, not a measurement. Cooper assumes even pacing over twelve maximal minutes and a known distance; if you set off too fast, the answer is low. VDOT assumes a hard, recent, evenly-paced race and inherits any error in the course distance. The Uth ratio assumes a real maximum heart rate, and if that came from 208 − 0.7 × age then roughly a third of people are more than ten beats away from it, which moves this estimate by three points or more.
All three were derived on particular populations — military recruits, trained distance runners, and a modest sample of elite and non-elite athletes respectively — and none of them was fitted on you. Absolute values should be treated as approximate; changes in your own value under a fixed protocol are the part worth trusting.
This is a fitness estimate for training purposes. It is not medical advice, it is not a cardiovascular assessment, and it is not a substitute for a professional who can actually see you.