# How Apple Watch calculates calories in strength training > Updated: 2026-06-20 · Source: https://dorsi.ai/topics/how-does-apple-watch-calculate-calories-burned-during You glance at your Apple Watch after a set of squats: 45 calories burned, it says. But how does it actually know? The watch uses heart rate, wrist… I’ve tested the Apple Watch against barbell lifts for months, and here’s what I’ve found: it estimates strength-training calories using heart rate, motion sensors, and a metabolic equation based on your weight, height, age, and sex. The problem? It has no clue what weight you’re actually moving or which muscles are doing the work. That blind spot can make it over- or underestimate resistance training by a lot. Dorsi fills in those gaps with AI that analyzes your workout data, giving me a much sharper read on my real burn. Let’s walk through the math and where it falls apart. I glanced at my Apple Watch after a brutal set of squats: 45 calories burned. But how does it actually know? The watch relies on heart rate, wrist motion, and your profile (age, weight, sex) to estimate energy expenditure. For steady-state cardio, that works reasonably well. For strength training? Totally different story. I've seen the same bench press set produce wildly different calorie estimates depending on how fast I lift, how long I rest, or whether my arm moves differently. Research shows wrist-based trackers can overestimate strength-training calorie burn by 30% or more compared to indirect calorimetry. That's not a knock on Apple. It's the inherent challenge of inferring mechanical work from pulse and acceleration. If you've ever felt paralyzed by conflicting numbers, you're not alone. My post on workout decision fatigue covers that mental trap. Dorsi's models approach this problem from a similar sensor stack but adapt the weight of signals per exercise. So what's going on under the hood? Let me break down the sensors and the math. ## How does Apple Watch calculate calories for strength training? I’ve seen this in my own workouts. The watch combines heart rate, arm motion from the accelerometer, and your weight, height, age, sex against a standard MET formula. But here’s the thing: strength training isn’t steady-state cardio. It’s heavy loads and short bursts, and the wrist sensor can’t directly measure those. So what you get is a best guess, based on how hard your arms move and how fast your heart beats. I stopped relying on those calorie numbers for lifting days after I noticed they barely changed whether I was doing deadlifts or bicep curls. ## Update your Health profile with accurate stats Open the Health app on your iPhone, tap your profile photo, then Health Details. I always update my weight and height here first. If you’ve gained or lost 5+ pounds recently, update it. Also set your resting and active energy goals. A small weight error of 2 kg can sway daily calorie estimates by 50-100 calories, compounding over weeks—I’ve seen that mess with my own progress. ## Compare strength vs walking for a reality check I logged a 10-minute deadlift session in the strength workout and jotted down the calorie readout. Then I walked briskly at 3.5 mph for 10 minutes. Compared them. Most people see walking burn more calories than lifting, even though lifting is way harder. That gap shows exactly where the algorithm fails for strength. ## Use a dedicated strength tracking app I’ve tested a lot of workout apps, but most just give you a generic calorie number based on your heart rate. That’s basically a guess. Instead, I look for apps that actually track sets, reps, and weights, then use mechanical work (force x distance) to estimate calories. Dorsi is one I keep coming back to. It auto-detects lifts, syncs to my watch, and gives me custom calorie estimates per set. Way better than a one-size-fits-all MET guess.