What is being estimated
Push intensity up far enough and oxygen consumption climbs with it, until it stops climbing and flattens. That plateau is VO2max, recorded as millilitres of oxygen per kilogram of body weight per minute. It is the product of how much blood the heart moves per beat, how much oxygen that blood carries, and how much of it the muscles extract, which is why it is used as a single summary of aerobic fitness.
One MET is 3.5 ml/kg/min, the rate of sitting still. A VO2max of 42 is therefore 12 METs, the same unit the activity comparison uses, which makes the two directly comparable.
The four tests, and what each one costs you
| Method | Formula | Error band |
|---|---|---|
| Cooper 12-minute run | (metres − 504.9) / 44.73 | SEE about 3 to 5 ml/kg/min, roughly ±10 to 15%. Published 1968 on military personnel. |
| 1.5 mile run | 483 / minutes + 3.5 | Similar, about 3 to 5 ml/kg/min. Fixing distance rather than time removes the pacing problem. |
| Rockport 1 mile walk | 132.853 − 0.0769·lb − 0.3877·age + 6.315(male) − 3.2649·min − 0.1565·bpm | SEE about 5 ml/kg/min. Validated on untrained to moderately active adults; underestimates trained ones. |
| Heart rate ratio | 15.3 × (HRmax / HRrest) | SEE around 5 to 6 ml/kg/min in the original elite-runner sample, worse elsewhere. The widest of the four. |
The two running tests share a weakness: they both assume genuine maximal effort, so a badly paced or conservatively run test reports low fitness rather than a bad test. The Rockport walk avoids that by being submaximal, and pays for it by having no way to read anyone whose walking pace never raises their heart rate much. The heart rate ratio needs no exercise at all, which is exactly why it is the least informative.
Reading the grading table honestly
The bands come from Cooper Institute percentile data for US adults, simplified. They step down with age because VO2max declines roughly 5 to 10% per decade from the mid-twenties onward, so 40 ml/kg/min at 55 is a considerably better result than 40 at 35.
Two caveats belong with the table. It is derived from a US population, and cut-offs published elsewhere differ by a few units either way depending on the sample and the edition. And it is a percentile ranking, not a health threshold — landing on a boundary means either adjacent grade is a defensible description of you.
How much training moves it
Someone starting from sedentary and training aerobically with any consistency commonly gains 15 to 20% within three to six months. Someone already trained may struggle to find a few percent, and a substantial share of the variation between people is heritable: identical programmes produce large responders and near non-responders in the same study.
A stubborn ceiling does not make the training pointless. Race results depend more on what fraction of the ceiling you can hold — the lactate threshold — and on how little oxygen a given pace costs you, which is running economy. Both respond to training far better than the ceiling does.
Why your watch disagrees
Wrist devices back-calculate VO2max from the relationship between pace and heart rate during ordinary runs. That relationship moves with hills, wind, heat and how your heart rate happens to respond that day, and every manufacturer uses a different algorithm, so the same person gets different numbers from different devices. The field tests here carry their own ±10 to 15%. Neither is a measurement. Pick one method, hold it constant, and read the direction of travel.
Before you run the test
The Cooper and 1.5-mile tests are maximal efforts and place real load on the cardiovascular system. A history of heart disease, chest tightness or dizziness during exertion, or several years away from exercise are all reasons to speak to a doctor before attempting either. Uncontrolled hypertension or diabetes, and any acute infection, are reasons to postpone. If you take a beta blocker or another drug that suppresses heart rate, the Rockport and heart-rate-ratio methods will produce numbers that mean nothing — use one of the running tests, subject to the same clearance.
Questions people ask
Which of the four is most accurate?
For anyone who can run hard, the 1.5-mile and Cooper tests, with standard errors around 3 to 5 ml/kg/min against laboratory measurement. Both assume maximal effort, so a test where you paced badly or left something in reserve reads low. The Rockport walk is safe and appropriate when running is not, but underestimates trained people because the intensity never reaches the range the equation was fitted on. The heart rate ratio is the widest and belongs in the fallback category. Treat all four as ±10 to 15%.
My smartwatch says something different. Which do I trust?
Neither, as an absolute figure. The watch infers VO2max from pace against heart rate on ordinary runs, which drifts with terrain, weather and daily heart rate variation, and each manufacturer computes it differently. The field tests here have their own error band of similar size. The useful discipline is to pick one method, repeat it every three to six months under matched conditions, and read the trend. A watch that shows you gaining and a Cooper test that shows the same are worth more together than either number alone.
I scored low. Is something wrong with me?
Low cardiorespiratory fitness is associated with higher risk across a range of chronic conditions in population studies, but a single field estimate says very little about an individual. Sleep, heat, pacing and how recently you ate can each move the result by a meaningful margin. If you are concerned, symptoms are a better prompt than a number: unusual breathlessness on stairs, or a clear decline in what you used to manage comfortably, are worth a medical appointment in a way that a table cell is not.
How much can I raise it?
From sedentary, 15 to 20% over three to six months of consistent aerobic training is a common result. From already trained, a few percent is hard work, and genetics explain a large share of who responds and who does not. This is less discouraging than it sounds, because the ceiling is not what determines race times. Lactate threshold and running economy both keep improving long after VO2max has stalled, and they are what actually move your finishing times.
Does it inevitably fall with age?
The general pattern is a decline of roughly 5 to 10% per decade from the mid-twenties, which is why the grading table shifts down with age. The rate is not fixed, though: people who keep training show roughly half the decline of those who stop, and by the sixties the gap between the two groups is large. Ageing sets the direction; training sets the slope.