WEARABLE VO2MAX ACCURACY
How accurate is wearable VO2max? What the evidence says
How accurate is wearable VO2max?
The independent evidence here covers eight validation studies of Apple Watch and Garmin models, not every wearable. WHOOP reports an internal lab validation for its current estimate, while Oura documents a different Cardio Capacity method and an experimental automatic mode. Those sources are not interchangeable, so use a wearable value as a personal trend and measure when the number matters.
What a wearable estimate is checked against
VO2max is the most oxygen your body can use in a minute when you exercise as hard as you can. It is written in millilitres of oxygen per kilogram of body weight per minute (ml/kg/min). The reference way to measure it is in a lab: you run or cycle until you cannot go on while a mask and a gas analyser measure the oxygen you breathe in and out. A wearable cannot do that. It estimates VO2max from signals such as heart rate, movement, pace and facts about you, so the question is how close the estimate lands.
Studies report the answer in a few ways. The average gap, also called bias, says whether a wearable tends to read low or high. The mean absolute percentage error, or MAPE, says how far off a single reading is on average, ignoring direction. The limits of agreement are the range that holds about 95 percent of the single-reading gaps, so they show how wide the scatter is. Two wearables can have the same average gap and very different scatter.
This guide reads 8 independent studies: three on Apple Watch and five on Garmin. We use the words the authors use. Where a paper calls a device accurate, that is the authors’ verdict against their own pass mark, not ours. WHOOP and Oura appear in a separate section because their current evidence and methods differ.
How Apple Watch makes its estimate
Apple’s support page says an Apple Watch Series 3 or later estimates VO2max from its heart and motion sensors during an outdoor walk, outdoor run or hiking workout[1]. Indoor workouts do not count[1]. The estimate also takes in your age, sex, weight, height and medications that might affect your heart rate[1]. Apple supports a range of 14 to 65 ml/kg/min[1].
Apple says certain conditions or medications that limit your heart rate might cause an overestimate[1].
- 1. Apple Support: Track your cardio fitness levels checked 30 Oct 2026
Apple Watch against a lab test: three studies
In the first Dublin study, published in 2025, 28 adults were analysed (30 were recruited, and two did not reach their maximum effort). Most wore a Series 9 or an Ultra 2. The watch read 6.07 ml/kg/min below the lab test on average, and a single reading was off by 13.31 percent on average[2]. The limits of agreement ran from 6.11 above the lab value to 18.26 below it[2].
The same group tested the Series 10 and published in 2026. Of 40 adults, 35 were analysed, with an average age of about 23. The watch read 6.25 ml/kg/min below the lab on average, a single reading was off by 13.2 percent on average, and the limits of agreement ran from 18.23 below to 5.73 above[3]. The watch put 5 of the 35 people (14 percent) in their right fitness percentile band and 12 more (34 percent) within one band of it[3]. The authors call the agreement for individuals poor and see use for studying groups[3].
A smaller study of 19 adults aged 18 to 63 tested the Series 7 against a lab test on a cycle ergometer, a stationary bike. The lab average was 45.88 ml/kg/min and the watch average 41.37. A single reading was off by 15.79 percent on average, and the authors rate the reliability as poor[4].
The studies differ in more than the watch. The Dublin team used a treadmill lab test. The Series 7 study used a bike for the lab test and took the watch estimate after an outdoor run[4]. We read no direction into that difference. All three put the watch below the lab on average, but the sizes differ and three studies are few.
| Study | Watch | People analysed | Average gap to the lab | Average size of a single error (MAPE) |
|---|---|---|---|---|
| Lambe et al., 2025 | Mostly Series 9 and Ultra 2 | 28 of 30 adults, average age about 32 | 6.07 ml/kg/min below | 13.31 percent |
| Lambe et al., 2026 | Series 10 | 35 of 40 adults, average age about 23 | 6.25 ml/kg/min below | 13.2 percent |
| Caserman et al., 2024 | Series 7 | 19 adults, aged 18 to 63 | Lab average 45.88, watch average 41.37 ml/kg/min | 15.79 percent |
- 2. Lambe et al., PLoS ONE, 2025: Investigating the accuracy of Apple Watch VO2 max measurements, a validation study checked 30 Oct 2026
- 3. Lambe et al., Mayo Clinic Proceedings: Digital Health, 2026: Accuracy of VO2 max estimates from Apple Watch Series 10 checked 30 Oct 2026
- 4. Caserman et al., JMIR Biomedical Engineering, 2024: Assessing the accuracy of smartwatch-based estimation of maximum oxygen uptake using the Apple Watch Series 7 checked 30 Oct 2026
Who the Apple Watch misses, and what the studies cannot say
Only a few points on this are sourced. Caserman and colleagues conclude that the Apple Watch overestimates VO2max in people with poor fitness and underestimates it in people with excellent fitness[5]. Apple says certain conditions or medications that limit your heart rate might cause an overestimate[6].
In the 2026 Dublin study the authors split the 35 people afterwards to look for patterns, and they call these splits post hoc, exploratory and hypothesis-generating. The 25 people at or above the 80th fitness percentile were read 7.79 ml/kg/min below the lab on average. The 10 people at or below the 70th percentile were read 2.41 below[7].
The same paper reports a bigger average gap for the 18 women (8.55 below) than for the 17 men (3.82 below). It also notes that 16 of the 18 women and 9 of the 17 men were at or above the 80th fitness percentile, and in its regression neither sex nor fitness level was independently associated with the gap[7]. Sex and fitness cannot be pulled apart in this sample, so it shows no sex difference we can state.
The people tested lean fit. In the 2025 study 21 of the 30 recruited adults were rated Superior or Excellent for fitness, and the authors write that few people with low fitness were willing to do a maximal test[8]. So the studies say little about people with low fitness. In short: in mostly fit adults the Apple Watch read low on average, and the evidence for who it misses most is thin.
- 5. Caserman et al., JMIR Biomedical Engineering, 2024: Assessing the accuracy of smartwatch-based estimation of maximum oxygen uptake using the Apple Watch Series 7 checked 30 Oct 2026
- 6. Apple Support: Track your cardio fitness levels checked 30 Oct 2026
- 7. Lambe et al., Mayo Clinic Proceedings: Digital Health, 2026: Accuracy of VO2 max estimates from Apple Watch Series 10 checked 30 Oct 2026
- 8. Lambe et al., PLoS ONE, 2025: Investigating the accuracy of Apple Watch VO2 max measurements, a validation study checked 30 Oct 2026
Garmin: five studies, and no single direction
Garmin’s page says the Garmin Human Performance Lab engine in your watch estimates VO2max during your run, by placing your pace in the context of how hard your body is working, and that you run as you normally would without a special fitness test[9].
A Forerunner 920XT study of 24 adults found a MAPE of 7.3 percent and a significant underestimate, and the authors found that the devices they tested did not give valid results and were most likely not accurate enough for sport or health care[10]. A fēnix 6 study of 19 adults found a MAPE of 7.05 percent and a concordance score of 0.73 (a score where 1 means perfect agreement), and against its own pass marks of a MAPE under 10 percent and a score above 0.7 the authors called the watch accurate[11].
A Forerunner 230 study of 44 adults, 22 women and 22 men, judged the watch accurate in both sexes, but reported large random errors and advised caution[12]. A Forerunner 265 study of 20 people in half-marathon training found the watch read higher than the test, 43.95 against 41.10 ml/kg/min at the start, a significant difference. Its MAPE was under 10 percent, and the authors say the watch may suit tracking general fitness trends[13].
A Forerunner 245 study of 24 trained young male soccer players found a MAPE of 5.58 percent after one warm-up run and recommended at least two runs[14]. That is a narrow group, so it does not tell you what to expect on your own wrist.
A 2025 review of 13 studies, most of them on Garmin watches running the Firstbeat algorithm, found wearables valid or acceptable for VO2max in 7 of them. Its author calls use in elite endurance sport questionable and says two or more runs could improve validity, but that this needs more study[15].
So the direction changes with the model, the group and the test: one study found an underestimate, one found readings above the test, and two called the watch accurate. None of these studies put a Garmin and an Apple Watch in the same people, and we found no such study, so the numbers on this page are not a ranking of the brands.
| Study | Watch | People | Average size of a single error (MAPE) | What the authors report |
|---|---|---|---|---|
| Passler et al., 2019 | Forerunner 920XT | 24 adults | 7.3 percent | Significant underestimate; devices did not give valid results |
| Carrier et al., 2025 | fēnix 6 | 19 adults, active and sedentary | 7.05 percent | Judged accurate against their pass marks |
| Snyder et al., 2021 | Forerunner 230 | 44 adults, 22 women and 22 men | Not given in the abstract | Judged accurate in both sexes; large random error |
| Howard et al., 2026 | Forerunner 265 | 20 people in half-marathon training | Under 10 percent | Read higher than the test: 43.95 against 41.10 ml/kg/min |
| Duking et al., 2024 | Forerunner 245 | 24 trained young male soccer players | 5.58 percent after one run | At least two runs recommended |
- 9. Garmin Technology: VO2 max checked 20 Oct 2026
- 10. Passler et al., International Journal of Environmental Research and Public Health, 2019: Validity of wrist-worn activity trackers for estimating VO2max and energy expenditure checked 30 Oct 2026
- 11. Carrier et al., Sensors, 2025: Validation of aerobic capacity (VO2max) and pulse oximetry in wearable technology checked 30 Oct 2026
- 12. Snyder et al., Journal of Strength and Conditioning Research, 2021: Comparison of the Polar V800 and the Garmin Forerunner 230 to predict VO2max checked 30 Oct 2026
- 13. Howard et al., Journal of Medical Engineering & Technology, 2026: Validity of the Garmin Forerunner 265 to predict VO2max and race performance times checked 30 Oct 2026
- 14. Duking et al., Journal of Sports Science and Medicine, 2024: Assessment of maximum oxygen uptake in elite youth soccer players, smartwatch technology compared with respiratory gas analysis checked 30 Oct 2026
- 15. Zeleznik Mezan, Frontiers in Sports and Active Living, 2025: Accuracy of wearables for determining the maximal oxygen uptake and lactate threshold, a qualitative systematic review checked 30 Oct 2026
WHOOP and Oura: different methods, different evidence
WHOOP says its current VO2max uses three layers: a passive model, a GPS-augmented model and a ground-truth calibrated model. It updates weekly, needs at least 14 recoveries in the last 21 days for the passive estimate, uses a GPS-tracked outdoor run of 15 minutes or more from the last 90 days for the GPS model, and lets members enter a lab result by hand[16]. WHOOP also reports internal cross-validation against 248 maximal treadmill tests with gas-exchange analysis. That is a manufacturer report, not one of the independent peer-reviewed studies in the Apple Watch and Garmin table[16].
Oura calls its general metric Cardio Capacity, an age-adjusted estimate. Its current help page describes an initial estimate from profile data and a six-minute outdoor walking test that uses your phone’s location, distance and heart rate. The test cannot be done on a treadmill, and Oura lets you add a lab or other trusted wearable value by hand[17].
Oura Labs now describes automatic VO2max sensing as an experimental option for enrolled members. It can replace the walking test for that experiment and uses qualifying GPS workouts, so it is a different path from the general Cardio Capacity instructions[18]. We did not add independent accuracy figures for WHOOP or Oura: the PubMed search logged for this guide did not identify a peer-reviewed external validation for either estimate, and a maker’s internal result cannot be carried over to every wearable.
- 16. WHOOP Locker: How accurate is WHOOP VO2 Max? checked 31 Oct 2026
- 17. Oura Help: Cardio Capacity (VO2 Max) checked 31 Oct 2026
- 18. Oura Help: Oura Labs automatic VO2 max checked 31 Oct 2026
How to use a wearable number
This is our advice, not a finding of these studies. Use a wearable VO2max to follow your own number over weeks, and do not hold it against another person’s device or a lab result. The Forerunner 265 authors say the watch may suit tracking general fitness trends[19], and the Dublin authors see the Apple Watch as more useful for studying groups than for single readings[20].
Check the conditions the maker names. Apple counts only outdoor walk, run and hiking workouts, and says conditions or medications that limit your heart rate might cause an overestimate[21].
When the number matters, for a goal or a race plan, measure instead of trusting an estimate: a lab test, or a field test such as the Cooper 12-minute run. The VO2max calculator under the answer turns a Cooper distance into a VO2max, and the guides linked at the end of this page cover the Cooper test and how to raise your number.
If you want to see what a VO2max number says about your age, the fitness age calculator estimates a fitness age from a VO2max you type in, or from a Cooper distance, plus your age and sex. It is a short estimate from a few inputs, not a medical test, and nothing you enter is saved.
- 19. Howard et al., Journal of Medical Engineering & Technology, 2026: Validity of the Garmin Forerunner 265 to predict VO2max and race performance times checked 30 Oct 2026
- 20. Lambe et al., Mayo Clinic Proceedings: Digital Health, 2026: Accuracy of VO2 max estimates from Apple Watch Series 10 checked 30 Oct 2026
- 21. Apple Support: Track your cardio fitness levels checked 30 Oct 2026
How Complete treats a wearable number
Complete starts from your long-term goals: build muscle, lose fat, improve cardio, get stronger, general health or become an athlete. VO2max is one of the numbers it follows. It takes a reading from a lab test you import, a Complete test you run, a value you type in, or Apple Health.
An Apple Health VO2max sample can arrive when HealthKit has one. The Add Reading screen lets you type a value and choose WHOOP, Oura, Garmin, COROS or another source. Connecting WHOOP or Oura for recovery does not automatically import their VO2max into this screen.
The Complete treadmill test starts with ten minutes of warm-up, then the treadmill gets 1 km/h faster every two minutes until you cannot hold the speed. If you have a compatible device reading from the last 90 days, from Apple Watch, WHOOP, COROS or Garmin, Complete compares it with your test result. If the two numbers are within 7.0 ml/kg/min of each other, Complete labels them consistent. If they are further apart, Complete saves your test result and suggests a repeat test or a lab test before you treat the difference as settled. It never averages the two numbers, and an older reading is not compared.
For the VO2max Complete shows as your current number, it looks at readings from the last 12 months and ranks them by how they were made: a lab report first, then a Complete test you have applied, then the newest reading from any other source, such as a wearable. That is a product rule that puts a test you took above an estimate. It is not an accuracy claim. We make no accuracy claim for any VO2max number in Complete, and the 7.0 band is a trigger to check again, not a measure of how accurate either number is.
Frequently asked questions
How accurate is Apple Watch VO2max?
In three studies against a lab test, the Apple Watch average was lower than the lab average each time. In the two larger Dublin studies, mostly of fit adults, it read about 6 ml/kg/min low on average, and a single reading was off by about 13 percent on average. A smaller study of 19 adults found a 15.79 percent average error.
Does Apple Watch overestimate or underestimate VO2max?
In these studies it read low on average. The authors of one study conclude it overestimates in people with poor fitness and underestimates in people with excellent fitness. Apple says certain conditions or medications that limit your heart rate might cause an overestimate.
How accurate is Garmin VO2max?
The studies disagree. A Forerunner 920XT study found a significant underestimate with a MAPE of 7.3 percent. A fēnix 6 study found a MAPE of 7.05 percent and called the watch accurate. A Forerunner 265 study found readings above the test. Each used a different watch and group.
Is WHOOP VO2max accurate?
WHOOP reports an internal cross-validation against 248 maximal treadmill tests with gas-exchange analysis in its current article. Our PubMed search did not identify a peer-reviewed external validation, so we do not compare that manufacturer report with the Apple Watch and Garmin studies or treat it as a guarantee for every member.
How does Oura estimate Cardio Capacity?
Oura’s general Cardio Capacity page describes an age-adjusted estimate, an initial profile-based value and a six-minute outdoor walking test using phone location, distance and heart rate. Oura Labs also describes an experimental automatic VO2max mode based on qualifying GPS workouts, so the method depends on which Oura path you use.
Which is more accurate for VO2max, Apple Watch or Garmin?
We cannot say. We found no study that tested both in the same people, and the studies we read differ in watch model, group and lab test. So this page ranks neither.
Why does my wearable VO2max differ from a lab test?
A wearable estimates from signals such as heart rate, movement, pace and personal details during its qualifying activities, while a lab test measures the oxygen you use at your limit. Apple counts only outdoor walks, runs and hikes. Oura’s general walking test uses phone location and cannot run on a treadmill. The studies found gaps on average and wide scatter for single readings.
What does Complete do when my test and my wearable disagree?
If you take the Complete treadmill test and have a compatible Apple Watch, WHOOP, COROS or Garmin reading from the last 90 days, Complete compares them. Within 7.0 ml/kg/min it labels them consistent. Further apart, it saves your test result and suggests a repeat test or a lab test. It never averages the two numbers.
Can I add my Garmin, WHOOP or Oura VO2max to Complete?
You can type a value from Garmin, WHOOP, Oura, COROS or another source on the Add Reading screen. An Apple Health VO2max sample can arrive when HealthKit has one. Connecting WHOOP or Oura for recovery does not automatically add their VO2max.