In early November, someone in a running community chat asked: what VO2max does your watch show? The answers ranged from 36 to 66. The most telling story came from a participant over 70: the watch showed 36, and a treadmill test at a clinic gave 38. A good match — but is it always like that? Over the past two years, several studies have tested popular watches against a lab test. Their numbers are worth knowing for anyone who tracks their VO₂max from the wrist.

How watches “measure” VO2max in the first place

They don't — they estimate it. True VO₂max is measured in a lab: a person runs a graded test to exhaustion wearing a gas-analyser mask. Watches have no mask. They take your heart rate, GPS speed, age, sex, weight and the maximum heart rate from your profile, and plug them into a model trained on laboratory data. Garmin uses an algorithm from Firstbeat; Apple and Google use their own.

Hence the main rule: accuracy depends on how closely your data resemble the data the model was trained on, and on how correctly your profile is filled in.

What the meta-analysis showed

The international INTERLIVE network (Molina-Garcia and colleagues, Sports Medicine, 2022) pooled 14 validation studies:

devices that estimate VO₂max from resting data overestimated it by an average of 2.17 mL/kg/min, and for an individual the error lay in a band from −13 to +17;
devices that calculate it from exercise are almost unbiased on average (−0.09), but the band for an individual is still wide — about ±10.

The authors' conclusion: such watches are fine for assessing groups of people, but not yet for individual sports or clinical assessment.

Fresh tests by model

Garmin Forerunner 265 (Howard and colleagues, Georgia State University, USA; Journal of Medical Engineering & Technology, October 2026). 20 people spent 12 weeks preparing for a half marathon wearing these watches.

Before the training block, the watch showed 43.95 and the lab test 41.10 mL/kg/min: an overestimate of about 2.9.
After 12 weeks: 45.00 versus 42.43, so the overestimate remained — about 2.6.
Half marathon prediction: 129.97 minutes versus an actual 137.55 — the watch promised a result about 7.5 minutes (around 5.5%) faster than people actually ran.
At the same time, the mean absolute error for both VO₂max and the prediction stayed below 10% — the threshold INTERLIVE considers acceptable for consumer devices.

Garmin fēnix 6 (Carrier and colleagues, University of Nevada, Las Vegas; Sensors, 2025). In 19 participants the mean error was 7.05%, and agreement with the lab was acceptable. The same watch, however, measured blood oxygen saturation (SpO₂) poorly.

Apple Watch (Lambe and colleagues, University College Dublin; PLOS ONE, 2025). 30 people wore the watch for 5–10 days and then did a treadmill test. The watch underestimated VO₂max by an average of 6.07 mL/kg/min. The mean error was 13.3%: in some individuals the watch underestimated VO₂max by as much as 18 units, while it overestimated it by no more than 6.

Google Pixel Watch 3 (the same Dublin group; PLOS ONE, September 2026). 34 participants. The average bias was almost zero — −0.53, with a mean error of 8.65%. But for an individual the watch could be off by about 12 units in either direction (limits of agreement: −12.56 and +11.50). The study was funded by Google, although the authors stress that the company had no influence on the design or analysis.

A review of 13 studies from Ljubljana (Frontiers in Sports and Active Living, 2025) adds an important point: most studies tested Garmin paired with a chest strap, and in 7 of the 13 studies the estimate was judged valid or acceptable. It works for untrained adults and amateurs, but for elite endurance athletes it is questionable.

What “±10” means in practice

A mean error of 7–13% sounds fairly good. But a band of ±10–12 units means that if your true VO₂max is 50, the watch could show 40 or 60, and that would still be “within normal limits” for the device. That is why comparing your number with a friend's from the chat is almost pointless: your errors are different.

What watches do seem to catch is changes in one person. In the Forerunner 265 study, over 12 weeks of training the lab-measured VO₂max rose by about 1.3 units on average across the group, and the watch readings by 1.05: the direction matched. A telling clinical case (Cureus, 2026): in a runner taking tamsulosin, the watch's VO₂max estimate fell from 48.2 to 45.7 over 113 runs and partly recovered after the medication was changed. This is a single observation, not proof, but it shows well that watches notice your own changes, even if the absolute number is inaccurate.

Limitations of the studies

Small samples — 19–34 people, mostly healthy amateurs.
Different reference protocols (treadmill, different ways of averaging gas-analysis data) — this alone shifts the result by a few percent.
Firmware changes. An algorithm tested in 2025 may work differently a year later.
There are almost no data on trail running, heat and cold — the tests were done in the lab and on steady runs over flat ground.

How to apply this

Watch the trend, not the number. A steady rise over several months of training is a good sign, even if it is only 1–2 units. The absolute value is only a rough guide.
Fill in your profile correctly. Age, weight and especially maximum heart rate have a big effect on the calculation. If your maximum is set “by formula”, the estimate can drift — see our article on maximum heart rate formulas.
Give your watch good data. Steady outdoor runs with GPS, ideally with a chest strap. Intervals, hills, cross-country running and optical heart rate “locked” onto cadence spoil the estimate.
Discount race predictions for optimism. The Forerunner 265 promised a half marathon faster than the real one by 7.5 minutes on average. Check the prediction against your recent race results in the performance index calculator.
Want an accurate number? Take a lab test. A field test such as the Cooper test is also an estimate, but it is based on your actual result over a distance rather than on heart rate during an easy run; you can calculate VO₂max from the result in the VO2max calculator.
Why raise this metric at all — see our article on VO₂max and longevity; how to do it — in our look at VO₂max intervals.

The bottom line

Watches do not measure VO₂max; they estimate it from heart rate, speed and your profile.
A meta-analysis of 14 studies: on average the exercise-based estimate is almost accurate, but for an individual the error is about ±10 units.
The Garmin Forerunner 265 overestimated VO₂max by about 2.5–3 units and predicted a half marathon 7.5 minutes faster than the real one; the Apple Watch underestimated it by 6 on average; the Pixel Watch 3 is accurate on average, but its spread is ±12.
The number on your watch is good for tracking your own trend and poor for comparing yourself with others.

Sources: Howard J., Ellis R., Fortner C.K., Doyle J.A., Otis J. “Validity of the garmin Forerunner 265 to predict VO2max and race performance times”. Journal of Medical Engineering & Technology, 2026. DOI: 10.1080/03091902.2026.2741912 · Lambe R., Schumann M., Hamilton S. et al. “The accuracy of VO2 max, heart rate and energy expenditure measurements from Google Pixel Watch 3”. PLOS ONE, 2026;21(9):e0356808. DOI: 10.1371/journal.pone.0356808 · Lambe R., O'Grady B., Baldwin M., Doherty C. “Investigating the accuracy of Apple Watch VO2 max measurements: A validation study”. PLOS ONE, 2025;20(5):e0323741. DOI: 10.1371/journal.pone.0323741 · Carrier B., Marten Chaves S., Navalta J.W. “Validation of Aerobic Capacity (VO2max) and Pulse Oximetry in Wearable Technology”. Sensors, 2025;25(1):275. DOI: 10.3390/s25010275 · Molina-Garcia P., Notbohm H.L., Schumann M. et al. “Validity of Estimating the Maximal Oxygen Consumption by Consumer Wearables: A Systematic Review with Meta-analysis and Expert Statement of the INTERLIVE Network”. Sports Medicine, 2022;52(7):1577-1597. DOI: 10.1007/s40279-021-01639-y · Železnik Mežan L. “Accuracy of wearables for determining the maximal oxygen uptake and lactate threshold: a qualitative systematic review”. Frontiers in Sports and Active Living, 2025;7:1707991. DOI: 10.3389/fspor.2025.1707991 · Cohn H., Zamani A., Sessine M., Cher M.L. “Reduction in Wearable-Derived Maximal Oxygen Consumption (VO₂max) During Tamsulosin Use in an Endurance Athlete”. Cureus, 2026;18(5):e109685. DOI: 10.7759/cureus.109685