MVIIIHealth The MVIII app
← Back

Research · Devices and wearables

Watch VO2 max estimates

A meta-analysis found watch estimates based on exercise data averaged within 0.1 mL/kg/min of laboratory testing, with limits of agreement near 10 either way, while estimates from resting data ran 2.2 high with wider limits. In single studies, average error ranged from about 5 to 16%, and recent Apple Watch models read about 6 mL/kg/min low.

Checked 2026-11-13
Status
Sold in the US as general wellness products. FDA's January 2026 guidance lists measuring aerobic fitness among general wellness claims and states that the agency does not intend to examine low risk general wellness products for compliance with premarket review requirements [1].
Measures
An estimate of maximal oxygen uptake in mL per kg per minute, calculated by each maker's own formula from heart rate with pace during outdoor walks or runs, or from heart rate at rest.
Also called
VO2 max estimate, cardio fitness score, cardiorespiratory fitness estimate, fitness age, Apple Watch VO2 max, Garmin VO2 max, Fitbit cardio fitness score, Polar Fitness Test, smartwatch VO2 max
Used for
VO2 max, cardio fitness, aerobic fitness, endurance, fitness age, training progress, longevity

Strong several reviews agree · Moderate one review or consistent trials · Limited one or a few small studies · Untested no study in people, listed with what it rests on

Matched the reference method3

Reported risks0

No risks sourced yet.

How it compares with the reference method

Matched the reference method3

Limited

VO2 max, Fitbit Charge 2

One study of 60 young adults found the Fitbit Charge 2 fitness score within 1.6 mL/kg/min of laboratory VO2 max on average, with error under 10%.

The numbers and sources
Device
Fitbit Charge 2
Compared with
Laboratory VO2 max testing
Effect
The Fitbit score had a positive bias of 1.59 mL/kg/min against laboratory data averaged over 15 seconds and 0.30 against data averaged over 60 seconds. Tests of equal means and variances did not reject concordance (P = 0.133 and 0.926), and mean absolute percentage error was under 10% for each comparison [5].
Population
65 healthy adults aged 18 to 45 recruited, 60 analyzed, who wore the device for a week and completed a qualifying outdoor run [5].
Sources
[5]
Caveats
The authors limit the result to young, healthy, fit adults who are able to run. Published in 2019 [5].
Limited

VO2 max, Garmin fēnix 6

One study of 19 adults found Garmin fēnix 6 estimates within 7% of laboratory VO2 max on average, meeting the study's preset criteria.

The numbers and sources
Device
Garmin fēnix 6
Compared with
A metabolic system during a graded exercise test
Effect
Against laboratory data averaged over 30 seconds, mean absolute percentage error was 7.05% and concordance correlation 0.73. The study had set validity at an error under 10% and concordance above 0.7 [6].
Population
19 apparently healthy adults, active and sedentary, who completed a graded test and an outdoor run [6].
Sources
[6]
Caveats
Nineteen participants. No external funding, and the authors declared no conflicts of interest [6].
Limited

VO2 max, Garmin Forerunner 265

One study of 20 runners found Garmin Forerunner 265 estimates about 2.7 mL/kg/min above laboratory VO2 max, with error under 10%.

The numbers and sources
Device
Garmin Forerunner 265
Compared with
A laboratory maximal test
Effect
At the start of training the watch estimate averaged 43.95 mL/kg/min against 41.10 in the laboratory (p = 0.009), and after 12 weeks 45.00 against 42.43 (p = .007). Mean absolute percentage error was below the 10% threshold the study had adopted. The watch's half-marathon predictions averaged 129.97 minutes against official times of 137.55 (p = .003) [11].
Population
20 adults in a 12-week half-marathon training program [11].
Sources
[11]
Caveats
Twenty participants. The overestimate was statistically significant at both time points even though it fell inside the study's threshold [11].

Mixed results6

Limited

VO2 max, Apple Watch Series 7

Mixed results: one study of 19 adults found Apple Watch Series 7 estimates averaged 4.5 mL/kg/min below laboratory VO2 max, with a mean error of 15.8%.

The numbers and sources
Device
Apple Watch Series 7
Compared with
Gas analysis during a graded cycle ergometer test to exhaustion
Effect
Laboratory VO2 max averaged 45.88 mL/kg/min (SD 9.42) against 41.37 (SD 6.5) from the watch (P = .01). Mean absolute percentage error was 15.79%, root mean square error 8.85 mL/kg/min and intraclass correlation 0.47 (95% CI 0.06 to 0.75). The authors reported acceptable overall agreement on Bland-Altman analysis alongside low reliability, with overestimation at low fitness and underestimation at high fitness [2].
Population
19 adults (7 women) aged 18 to 63, who wore the watch for at least 2 days and completed an outdoor run [2].
Sources
[2]
Caveats
Nineteen participants. The laboratory test was on a cycle and the watch estimate came from running [2].
Limited

VO2 max, Garmin Forerunner 230 and Polar V800

Mixed results: one study of 44 adults found random errors of 6.6 to 16.4 mL/kg/min for two watches, with the resting-based estimate overestimating in men.

The numbers and sources
Device
Garmin Forerunner 230, which estimates from a 10 minute outdoor run, and Polar V800, which estimates from heart rate at rest
Compared with
A treadmill VO2 max test
Effect
Bland-Altman analysis showed random errors from 6.6 to 16.4 mL/kg/min, larger for the Polar watch. In men, the Polar estimate ran 3.87 mL/kg/min above the Garmin estimate (p = 0.013). The authors concluded the Garmin watch estimated VO2 max in both sexes and the Polar watch in women but not men, and that both carry large random error [7].
Population
22 women and 22 men [7].
Sources
[7]
Caveats
The abstract does not report mean bias against the laboratory for each watch. Published in 2021 [7].
Moderate

VO2 max, Garmin Forerunner 245

Mixed results: three studies found Garmin Forerunner 245 estimates within 6 to 8% of laboratory VO2 max on average, underestimating by 6 mL/kg/min in highly trained athletes.

The numbers and sources
Device
Garmin Forerunner 245, software version 13.00 in one study [8]
Compared with
Respiratory gas analysis during a treadmill ramp test
Effect
In 23 runners, mean absolute percentage error after three outdoor runs was 5.7%, against a 4.0% coefficient of variation for the laboratory test itself. Error was 7.1% at lower fitness and -6.2% at the highest fitness [9]. In 35 endurance athletes, the watch underestimated by 4.73 and 4.05 mL/kg/min over two runs (error 7.9% and 7.2%, intraclass correlation 0.71 and 0.75). Error was 2.8 to 4.1% in moderately trained athletes and 9.4 to 10.4% in highly trained athletes, where the watch read 6.3 mL/kg/min low [10]. In 24 trained youth soccer players, error was 5.58% after one run with an intraclass correlation of 0.37 [8].
Population
23 runners with peak oxygen uptake of 38 to 61 mL/kg/min [9]. 35 endurance athletes, mean age 25.1, mean VO2 max 60.1 [10]. 24 trained male youth soccer players [8].
Sources
[9], [10], [8]
Caveats
Trained runners and athletes, with one research group contributing to all three studies. The authors of one study state laboratory gas analysis remains preferred when precision is required [10].
Limited

VO2 max, Google Pixel Watch 3

Mixed results: one study of 34 adults found Pixel Watch 3 estimates within 0.5 mL/kg/min of laboratory VO2 max on average, with limits of agreement of 12 either way.

The numbers and sources
Device
Google Pixel Watch 3
Compared with
Indirect calorimetry during a maximal treadmill test
Effect
Bias was -0.53 mL/kg/min with limits of agreement from -12.56 to 11.50 and a mean absolute percentage error of 8.65%. The authors concluded the estimates lack individual-level validity for clinical use while the low bias suggests use for population monitoring. In the same study, heart rate bias was -2.39 beats per minute (error 2.73%) and energy expenditure error 15.02% [13].
Population
34 participants, with watch estimates generated in daily life [13].
Sources
[13]
Caveats
One watch generation and 34 participants [13].
Limited

VO2 max, Polar Fitness Test on Polar Vantage M2

Mixed results: one study of 24 adults with cardiovascular risk factors found the resting Polar Fitness Test within 1 mL/kg/min on average, with limits of 11.4 either way.

The numbers and sources
Device
Polar Fitness Test, a resting test, run on a Polar Vantage M2
Compared with
Cardiopulmonary exercise testing on a treadmill
Effect
Mean bias was -1.0 mL/kg/min with limits of agreement of plus or minus 11.4, a mean absolute percentage error of 13.7% and an intraclass correlation of 0.743. One of five walk-test equations performed similarly (error 12.6%). The authors concluded the test may serve at population level and that its variability limits use for individual clinical decisions [14].
Population
24 adults (9 women), mean age 57.4, in medically supervised exercise training, 7 of them on heart rate-limiting medication [14].
Sources
[14]
Caveats
Twenty-four participants, some on medication that lowers heart rate [14].
Moderate

VO2 max, wearables using exercise data pooled

Mixed results: a meta-analysis found exercise-based watch estimates averaged within 0.1 mL/kg/min of laboratory VO2 max, with limits of agreement of about 10 either way.

The numbers and sources
Device
Consumer wearables that estimate from exercise data, pooled [15]. Mostly Garmin watches paired with chest strap heart rate sensors [16].
Compared with
Laboratory VO2 max testing
Effect
In a meta-analysis of 14 validation studies, wearables using exercise-based data had a bias of -0.09 mL/kg/min with limits of agreement from -9.92 to 9.74. The authors concluded these estimates suit population-level measurement and that error for an individual is large [15]. A later systematic review of 13 studies found the wearable valid or acceptable for VO2 max in 7 of them, with estimates generated from submaximal outdoor runs, and questioned usefulness in elite endurance athletes [16].
Population
14 validation studies [15]. 13 studies of untrained adults, recreational athletes and highly trained athletes [16].
Sources
[15], [16]
Caveats
The journal published a correction to the meta-analysis in 2026, and the figures here are those in the article's abstract [17].

Did not match the reference method4

Limited

VO2 max, Apple Watch Series 9 and Ultra 2

One study of 30 adults found Apple Watch estimates averaged 6.1 mL/kg/min below laboratory VO2 max, with a mean error of 13.3%.

The numbers and sources
Device
Apple Watch on watchOS 10 or later, mostly Series 9 (17 participants) and Ultra 2 (6), with a few Series 5 to 8 and SE 2
Compared with
Indirect calorimetry during a maximal treadmill test
Effect
The watch underestimated VO2 max by a mean 6.07 mL/kg/min (95% CI 3.77 to 8.38), with limits of agreement from -6.11 to 18.26. Mean absolute percentage error was 13.31% (95% CI 10.01 to 16.61) and mean absolute error 6.92 mL/kg/min. The authors concluded the estimates need refinement before clinical use [3].
Population
30 participants who wore a watch for 5 to 10 days to generate an estimate. Most had fitness in the "good" to "superior" categories [3].
Sources
[3]
Caveats
A fit sample, with one participant in the "fair" category and none lower. The authors declared no competing interests [3].
Limited

VO2 max, Apple Watch Series 10

One study of 40 adults found Apple Watch Series 10 estimates averaged 6.3 mL/kg/min below laboratory VO2 max and placed 14% of participants in the right fitness band.

The numbers and sources
Device
Apple Watch Series 10 running watchOS 11
Compared with
Indirect calorimetry during a maximal treadmill test
Effect
The mean difference was -6.25 mL/kg/min (95% CI -8.28 to -4.23), with limits of agreement from -18.23 to 5.73. Mean absolute error was 6.79 mL/kg/min and mean absolute percentage error 13.2% (95% CI 9.6 to 16.8). Against published fitness percentile bands, the watch placed 5 participants (14%) in the correct band and a further 12 (34%) within one adjacent band. The authors described individual-level agreement as low and suggested use for population-level assessment [4].
Population
40 healthy adults (20 women), tested between October 2024 and January 2025 [4].
Sources
[4]
Caveats
One watch generation and software version. Supported by Ireland's Health Research Board, and the authors reported no competing interests [4].
Limited

VO2 max, Garmin Forerunner 920XT and Polar V800

One study of 24 adults judged VO2 max estimates from two sports watches not valid, with error of 7.3% for one and above 10% for the other.

The numbers and sources
Device
Garmin Forerunner 920XT and Polar V800
Compared with
Indirect calorimetry
Effect
Mean absolute percentage error was 7.3% for the Garmin Forerunner 920XT, which significantly underestimated VO2 max (p = 0.027), and above 10% for the Polar V800. The authors concluded the tested devices did not give valid estimates of VO2 max [12].
Population
24 healthy adults [12].
Sources
[12]
Caveats
Published in 2019. The same study tested three other wrist devices for energy expenditure only [12].
Moderate

VO2 max, wearables using resting data pooled

Not shown to match laboratory VO2 max when based on resting data. A meta-analysis found a 2.2 mL/kg/min overestimate with limits of agreement from 13 below to 17 above.

The numbers and sources
Device
Consumer wearables that estimate from resting data, pooled
Compared with
Laboratory VO2 max testing
Effect
Wearables using resting information significantly overestimated VO2 max, with a bias of 2.17 mL/kg/min and limits of agreement from -13.07 to 17.41. The authors concluded exercise-based estimates were closer to laboratory values than resting-based ones [15].
Population
14 validation studies across both estimate types [15].
Sources
[15]
Caveats
The journal published a correction to the meta-analysis in 2026, and the figures here are those in the article's abstract [17].

Does using one change anything

Shown to help1

Moderate

Cardiorespiratory fitness with tracker use

Shown to raise fitness modestly across systematic reviews of activity tracker trials. One crossover study found a 1.5 mL/kg/min rise in estimated fitness over 8 weeks.

The numbers and sources
Effect
An umbrella review of 39 systematic reviews and meta-analyses found activity trackers improved fitness (SMD 0.3) and physical activity (SMD 0.3 to 0.6) [18]. In 50 inactive adults who followed their activity on a Garmin Vivosmart 4 for 8 weeks after a 4-week control period, estimated cardiorespiratory fitness rose by 1.52 mL/kg/min (95% CI 0.82 to 2.22) and daily steps by 982 (95% CI 492 to 1,471). Blood pressure did not change [19].
Population
163,992 participants across 39 reviews [18]. 62 inactive adults recruited, 50 completed, mean age 48, 84% women [19].
Sources
[18], [19]
Caveats
These trials tested wearing an activity tracker, not seeing a VO2 max number. In the crossover study, participants were also instructed to reach 150 intensity minutes a week, there was no randomized control group, and fitness was reported as an estimate [19].

Sources

  1. US Food and Drug Administration. General Wellness: Policy for Low Risk Devices. Guidance for Industry and Food and Drug Administration Staff. January 6, 2026. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/general-wellness-policy-low-risk-devices
  2. Caserman P, Yum S, Göbel S, et al. Assessing the Accuracy of Smartwatch-Based Estimation of Maximum Oxygen Uptake Using the Apple Watch Series 7: Validation Study. JMIR Biomedical Engineering. 2024. https://doi.org/10.2196/59459
  3. Lambe R, O'Grady B, Baldwin M, Doherty C. Investigating the accuracy of Apple Watch VO2 max measurements: A validation study. PLOS One. 2025. https://doi.org/10.1371/journal.pone.0323741
  4. Lambe R, Schumann M, Donnelly L, et al. Accuracy of VO2 max Estimates From Apple Watch Series 10. Mayo Clinic Proceedings: Digital Health. 2026. https://doi.org/10.1016/j.mcpdig.2026.100357
  5. Klepin K, Wing D, Higgins M, et al. Validity of Cardiorespiratory Fitness Measured with Fitbit Compared to VO2max. Medicine and Science in Sports and Exercise. 2019. https://doi.org/10.1249/MSS.0000000000002041
  6. Carrier B, Marten Chaves S, Navalta JW. Validation of Aerobic Capacity (VO2max) and Pulse Oximetry in Wearable Technology. Sensors. 2025. https://doi.org/10.3390/s25010275
  7. Snyder NC, Willoughby CA, Smith BK. Comparison of the Polar V800 and the Garmin Forerunner 230 to Predict VO2max. Journal of Strength and Conditioning Research. 2021. https://doi.org/10.1519/JSC.0000000000002931
  8. Düking P, Ruf L, Altmann S, et al. Assessment of Maximum Oxygen Uptake in Elite Youth Soccer Players: A Comparative Analysis of Smartwatch Technology, Yoyo Intermittent Recovery Test 2, and Respiratory Gas Analysis. Journal of Sports Science and Medicine. 2024. https://doi.org/10.52082/jssm.2024.351
  9. Düking P, Van Hooren B, Sperlich B. Assessment of Peak Oxygen Uptake with a Smartwatch and its Usefulness for Training of Runners. International Journal of Sports Medicine. 2022. https://doi.org/10.1055/a-1686-9068
  10. Engel FA, Masur L, Sperlich B, Düking P. Validity of VO2max estimates from the forerunner 245 smartwatch in highly vs. moderately trained endurance athletes. European Journal of Applied Physiology. 2026. https://doi.org/10.1007/s00421-025-05923-x
  11. Howard J, Ellis R, Fortner CK, et al. Validity of the garmin Forerunner 265 to predict VO2max and race performance times. Journal of Medical Engineering and Technology. 2026. https://doi.org/10.1080/03091902.2026.2741912
  12. Passler S, Bohrer J, Blöchinger L, Senner V. Validity of Wrist-Worn Activity Trackers for Estimating VO2max and Energy Expenditure. International Journal of Environmental Research and Public Health. 2019. https://doi.org/10.3390/ijerph16173037
  13. 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. https://doi.org/10.1371/journal.pone.0356808
  14. Neudorfer M, Ötzlinger L, Kumar D, et al. Validation of the Polar Fitness Test for Estimation of Maximal Oxygen Consumption at Rest in Medically Supervised Exercise Training: Comparison with CPET and the 6-Minute Walk Test. Sensors. 2025. https://doi.org/10.3390/s25185649
  15. Molina-Garcia P, Notbohm HL, 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. https://doi.org/10.1007/s40279-021-01639-y
  16. 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. https://doi.org/10.3389/fspor.2025.1707991
  17. Molina-Garcia P, Notbohm HL, Schumann M, et al. Correction: 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. 2026. https://doi.org/10.1007/s40279-026-02547-9
  18. Ferguson T, Olds T, Curtis R, et al. Effectiveness of wearable activity trackers to increase physical activity and improve health: a systematic review of systematic reviews and meta-analyses. The Lancet Digital Health. 2022. https://doi.org/10.1016/S2589-7500(22)00111-X
  19. Larsen LH, Lauritzen MH, Sinkjaer M, Kjaer TW. The Effect of Wearable Tracking Devices on Cardiorespiratory Fitness Among Inactive Adults: Crossover Study. JMIR Cardio. 2022. https://doi.org/10.2196/31501

Everything in the library