Is the Apple Watch accurate for weight lifting calories?

Not very. In a 2026 lab test, an Apple Watch tracked heart rate during weight lifting almost exactly like an ECG, yet put a 34-minute session at about 300 kcal when a metabolic analyzer measured about 140. The heart rate it reads is accurate, but the formula that turns it into calories doesn’t fit lifting. When you lift, your heart rate climbs far more than your oxygen use, and the watch has no way to know how much weight was on the bar.

What the validation studies found

Apple doesn’t publish how the watch turns its readings into calories. Its support pages say the watch uses your height, weight, sex and age, and that during a workout it picks the inputs that suit the activity, with heart rate one of many (Apple Support). Researchers who test these devices describe smartwatch calorie estimates as models that combine heart rate, motion and your profile (Lee et al. 2026). None of that includes the load you lifted.

Three lab studies put the Apple Watch next to a metabolic analyzer, which measures the oxygen you use, during resistance exercise:

• In Lee et al. 2026, sixty-two young men did a 34-minute session of two sets of 10 reps at their 10-rep max on the bench press, squat, T-bar row and deadlift. The Apple Watch Series 7 matched the ECG’s heart rate to within half a beat per minute on average. Its energy came out at 299 kcal against 141 kcal measured, and the two barely moved together (r = 0.34). None of the other three watches tracked it well either. One got the group’s average about right but missed widely from person to person.
• In Yun et al. 2026, seventy men did the same kind of session with an Apple Watch Series 8 set to Traditional Strength Training. The watch’s heart rate averaged within half a beat of the ECG, but it lagged right after each set. Its energy came out at 296 kcal against 140 kcal measured.
• In Boudreaux et al. 2018, fifty adults wore eight devices, including an Apple Watch Series 2, during cycling and resistance exercise. None of the seven that gave a calorie figure came within the study’s 10 % error limit in either mode.

Why the calorie math fails for lifting

Heart-rate calorie models lean on a relationship worked out in steady activity like running and cycling, where each beat stands for a predictable amount of oxygen, and oxygen is energy. Lifting doesn’t follow that relationship. Straining against a load pushes your heart rate up without a matching rise in the oxygen your muscles use: in one study, heavier squat sets raised heart rate while oxygen use stayed the same (Lässing et al. 2023).

Collins and colleagues measured this in 1991, with men lifting at 40–70 % of their one-rep max: at any given heart rate, lifting used less oxygen than running or cycling typically do. A model built on cardio reads a lifter’s heart rate as more work than was done.

The heart rate also can’t show the load. A rep’s cost rises with the weight on the bar (Robergs et al. 2007), and a wrist sensor has no way to know it.

What a set costs

Lab studies that measured oxygen during and after sets break a session into three parts.

The reps

Robergs and colleagues measured the oxygen cost of bench presses and squats at different loads and built equations from them. A rep costs less than most people expect: about 0.6 kcal for a 60 kg bench press. Scott and colleagues, counting the aerobic and anaerobic cost, found about 0.6–0.7 kcal per rep at 70 % of one-rep max (Scott et al. 2011).

The recovery after each set

After a set, your breathing stays high for a minute or two while the muscle restocks its quick energy. Scott’s group measured that recovery after single bench press sets taken to failure: about 4.7 kcal per set, light or heavy. For heavy sets, the recovery cost more than the aerobic cost of the lifting itself.

The time in between

Most of a session is time between sets: standing, walking, loading plates. Once the recovery is over, that costs about what standing does. Scott’s group took 5 mL of oxygen per kg per minute as a typical standing rest, about half a kilocalorie a minute above resting for a 70 kg person.

Put together for an example 70 kg lifter doing 18 sets of 10–12 reps in an hour, two reps short of failure, our model gives about 100 kcal for the reps, 70 for the recovery and 27 for the 50 minutes between sets, roughly 200 kcal in all. These figures come from our model. The time between sets takes most of the hour but adds little energy.

Why two lifters burn different amounts

Moving your own body is part of the work in squats, lunges and pull-ups, so a heavier lifter’s reps cost more. Your resting metabolism, which an “active” calorie number subtracts, depends on your sex, age, height and weight (Mifflin et al. 1990). And in men and women doing a full resistance session, lean mass helped predict what the session cost (Lytle et al. 2019).

How Workout Wizard does it

Workout Wizard estimates a workout’s active energy from the sets you log, so it works without a watch. Each completed set adds its reps, from Robergs’ bench press equation for upper-body exercises and his squat equation for lower-body ones, at the weight you logged. Free-weight leg exercises also add your bodyweight. It adds Scott’s recovery, scaled down when the week’s target leaves reps in reserve. Scott only measured sets to failure, so that scaling is our assumption. The time between sets costs a standing rest. Bodyweight exercises count only the part of the body they lift: a push-up loads your hands with about 72 % of your bodyweight, as measured on a force plate (Suprak et al. 2011).

Without your details, it assumes a 70 kg lifter. Add your sex, age, height and weight in Settings, or let the app read them from Apple Health, and it uses your weight, your resting metabolism and your lean mass. It prefers a body fat from a scale or a scan. Without one, it estimates it from your waist, and failing that from your height, weight, age and sex.

The estimate sits next to the workout clock behind a flame, and it keeps growing while you rest. It also shows on your Lock Screen and in the workout’s history.

If you wear an Apple Watch, the app’s Watch companion shows its number and Apple’s side by side, so you can see the gap yourself. It also sends your heart rate to the iPhone. Once you’ve given your age and Health has your resting heart rate and VO₂ max, the heart rate tailors the cost of the later part of each rest, with only a fraction of it counted, and a cap, because lifting inflates the heart rate. That fraction is our own estimate between what two studies suggest (Collins et al. 1991; Lässing et al. 2023).

Without the Watch, you can also have the app save each workout to Apple Health as strength training, with its estimate. It’s off by default.

So, is it accurate?

An Apple Watch measures your heart rate well during lifting, but its calories for a lifting session came out about double what a lab measured in two 2026 studies. Lifting raises your heart rate more than your oxygen use, and the watch doesn’t know your loads. A better estimate starts from the sets: the reps, the recovery after each set, the time in between, and your body.

References

• Apple Support. Calibrate your Apple Watch for improved Workout and Activity accuracy. support.apple.com/en-us/HT204516 (accessed 2026-11-03).
• Apple Support. Get the most accurate measurements using your Apple Watch. support.apple.com/en-us/105002 (accessed 2026-11-03).
• Boudreaux BD et al. 2018. Validity of wearable activity monitors during cycling and resistance exercise. Med Sci Sports Exerc 50(3):624–633. doi:10.1249/MSS.0000000000001471. PMID 29189666.
• Collins MA, Cureton KJ, Hill DW, Ray CA. 1991. Relationship of heart rate to oxygen uptake during weight lifting exercise. Med Sci Sports Exerc 23(5):636–640. doi:10.1249/00005768-199105000-00018. PMID 2072844.
• Lässing J, Maudrich T, Kenville R, Uyar Z, Bischoff C, Fikenzer S, Busse M, Falz R. 2023. Intensity-dependent cardiopulmonary response during and after strength training. Sci Rep 13:6632. doi:10.1038/s41598-023-33873-x. PMID 37095279.
• Lee T-H, Jun D-U, Bae J-Y, Roh H-T, Cho S-Y. 2026. Comparative validity of smartwatch-derived heart rate and energy expenditure during endurance and resistance exercise. Sensors 26(8):2526. doi:10.3390/s26082526. PMID 42076635.
• Lytle JR et al. 2019. Predicting energy expenditure of an acute resistance exercise bout in men and women. Med Sci Sports Exerc 51(7):1532–1537. doi:10.1249/MSS.0000000000001925. PMID 30768553.
• Mifflin MD et al. 1990. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr 51(2):241–247. doi:10.1093/ajcn/51.2.241. PMID 2305711.
• Robergs RA, Gordon T, Reynolds J, Walker TB. 2007. Energy expenditure during bench press and squat exercises. J Strength Cond Res 21(1):123–130. doi:10.1519/00124278-200702000-00023. PMID 17313290.
• Scott CB, Leighton BH, Ahearn KJ, McManus JJ. 2011. Aerobic, anaerobic, and excess postexercise oxygen consumption energy expenditure of muscular endurance and strength: 1-set of bench press to muscular fatigue. J Strength Cond Res 25(4):903–908. doi:10.1519/JSC.0b013e3181c6a128. PMID 20703175.
• Suprak DN, Dawes J, Stephenson MD. 2011. The effect of position on the percentage of body mass supported during traditional and modified push-up variants. J Strength Cond Res 25(2):497–503. doi:10.1519/JSC.0b013e3181bde2cf. PMID 20179649.
• Yun J-H, Ko J-M, Lee T-H. 2026. Criterion validity of smartwatches for measuring heart rate and energy expenditure during resistance exercise. Bioengineering 13(10):1099. doi:10.3390/bioengineering13101099. PMID 42938956.