How Accurate Is Apple Watch VO2 Max, Really?
Patricia Bedoya
·
·
6
min read
Cover image generated with AI — © Chubby Studio S.L.
KEY TAKEAWAY
Apple Watch typically reads several mL/kg/min below a laboratory test, and the error has a direction: it flatters unfit people and understates fit ones. Two independent validations found mean underestimation of 4.51 and 6.07 mL/kg/min. Treat the number as a trend line for yourself, not a score to compare against anyone else.
Someone posts their laboratory test result next to their watch's estimate and the two numbers are eight points apart. The thread fills up with people reporting the same gap, in the same direction, and nobody in it can say why.
Here is the honest answer. Against a proper laboratory test, Apple Watch usually reads several mL/kg/min low, and the size of the miss depends on how fit you already are. The error is not random noise. It has a direction, and that is what makes it workable.
How close is the estimate to a lab test?
Two independent validations have now measured this directly, on different protocols and different people, and they agree on the direction.
A 2025 study in PLOS One took 28 adults who had worn an Apple Watch long enough to generate an estimate, then put them through a maximal treadmill test with indirect calorimetry as the reference. The watch underestimated by a mean of 6.07 mL/kg/min, with a mean absolute percentage error of 13.31%. The limits of agreement spanned roughly 6 mL/kg/min below the lab value to 18 above it, which is a wide band for any single reading.
A 2024 study in JMIR Biomedical Engineering tested Apple Watch Series 7 against laboratory cycling in 19 participants. Mean measured VO2 max was 45.88 mL/kg/min and the watch said 41.37, a bias of 4.51 mL/kg/min in the low direction, with a mean absolute percentage error of 15.79%. The intraclass correlation was 0.47, which is poor agreement by any standard you care to apply.
Both are small studies and neither is the final word. But they landed in the same place: the estimate usually reads low, and it is not precise enough to trust as a single number.
Why does the error depend on how fit you are?
This is the part that explains the angry threads. The Series 7 study broke its results down by fitness category and the bias flipped direction. For participants in the poor fitness group the watch overestimated by 3.80 mL/kg/min. For those in the excellent group it underestimated by 12.00.
A 2025 systematic review of 13 wearable validation studies found the same pattern across brands rather than just Apple: devices tend to overestimate at low fitness and underestimate above roughly 55 mL/kg/min. The algorithm pulls everyone towards the middle, which is what any model fitted to a population does when it meets somebody at the edges of that population.
So the practical reading is uncomfortable but simple. If your estimate looks low, it is probably closer to the truth than you would like. If it looks high and you train seriously, it is probably underselling you.
See what your training is actually doing
FitWoody analyses every workout against your own cardiac load and baselines, so you can tell whether a block of training is landing without waiting for one estimate to move. 7-day free trial of Plus.
Download the app
★★★★★
4.5 · 732 ratings on the App Store
What is the watch actually measuring?
Not oxygen. A laboratory test measures the gas you breathe out, which is what VO2 max actually measures. Your watch measures heart rate and pace, and infers everything else.
Apple only generates an estimate from an Outdoor Walk, Outdoor Run or Hiking workout, because it needs GPS pace alongside heart rate to work out how much external work you did for a given cardiac cost. Indoor sessions, cycling and gym machines contribute nothing. It then adjusts for age, sex, height, weight and any declared medications that affect heart rate, and it wants at least 24 hours of wear plus several qualifying workouts before it will show you anything at all.
That is a reasonable way to estimate aerobic capacity, and it runs on the same logic as reading what a given workout costs your heart. It is also why the estimate is only ever as good as the workouts feeding it. A month of flat, easy walks gives the model very little range to fit a curve through.
Does the 14 to 65 range matter?
More than most people realise. Apple states that Apple Watch supports a VO2 max range of 14 to 65 mL/kg/min. Above that ceiling there is simply nothing for it to report.
A well trained endurance athlete in their thirties can sit above 60, and a very good one can clear 65 outright. That single design decision accounts for part of the underestimation the validation studies found at the top end, because the watch cannot tell you a number it was not built to display. If you are near the ceiling, a stubbornly unmoving estimate may be the scale running out rather than your fitness plateauing.
Is the number useless, then?
No, and the direction of the error is exactly why. A measurement that is biased consistently is still a good tracker of change. If your estimate reads four points low today and four points low in six months, the difference between those two readings carries real information even though neither reading is accurate.
So use it as a trend and let go of the absolute value. Compare it against your own history rather than a chart of what is normal for your age, for the same reason that a personal baseline beats a population range on every other metric your watch collects. And give it months rather than weeks, because the easy aerobic work that actually moves VO2 max shows up slowly.
The one reading worth taking closer to face value is the low end. If your watch says your cardio fitness is low, the evidence says it is more likely to be flattering you than insulting you, which is why a low cardio fitness notification is worth acting on rather than arguing with.
What makes your estimate worse?
A handful of things, all of them fixable. Wearing the watch loosely, so heart rate sampling degrades during the exact hard efforts the model needs most. Feeding it nothing but short, flat, easy walks. A height and weight you last updated two years ago, since both go straight into the calculation. And training mostly indoors or on a treadmill, which contributes nothing to the estimate at all.
If you want a better number, the fix is boring: a few outdoor runs or brisk uphill walks with a snug watch, over a couple of weeks. Give the model some range to work with.
Should you get a proper lab test?
If you are training towards something specific, or you want a true baseline before a long block of work, a cardiopulmonary exercise test is the only way to get a real number. One lab test plus your watch trend afterwards is a genuinely strong combination, because the test fixes the level and the watch tracks the direction. For everyone else it is an expensive way to confirm what the trend was already telling you.
What matters more than either is what you do with it. VO2 max is a ceiling, not a performance, and a high ceiling makes nobody fast on its own.
Where FitWoody fits
FitWoody does not try to out-estimate Apple's VO2 max model, and no app on your wrist can. What it does is read the data your watch already collects against your own baselines, and analyse each workout against your individual cardiac load, so the question you are answering becomes whether this week's training is landing rather than whether one estimate is correct to a decimal place.
We shipped custom heart rate zones because the generic ones were wrong for most of the people we looked at, and a zone that is off by a few beats makes every workout built on top of it wrong too.
Co-founder, FitWoody
Frequently asked questions
Why is my Apple Watch VO2 max lower than my lab test?
Because the estimate is systematically biased low for fitter people. One validation found a mean underestimation of 6.07 mL/kg/min, and another found the gap widened to 12 mL/kg/min in the highest fitness category. Apple also caps the reported range at 65 mL/kg/min.
Does cycling or indoor running improve my VO2 max estimate?
No. Apple only generates an estimate from Outdoor Walk, Outdoor Run and Hiking workouts, because it needs GPS pace alongside heart rate. Indoor sessions and cycling do not contribute to the number at all.
How much does my VO2 max estimate need to change before it means anything?
More than you would like. Agreement between watch and lab is poor at the level of a single reading, so treat small month-to-month moves as noise and look instead for a direction sustained over several months.