Key takeaways
- Calories are estimated from heart rate, movement and your profile; they are never measured directly by a watch.
- Errors are largest for cycling, strength training and other activities where your wrist moves little or your heart rate rises for other reasons.
- “Active” calories exclude your resting burn; “total” calories include it. Mixing them up double-counts.
- Your weight trend over a few weeks is a far better guide to your real TDEE than any device.
On this page
- The short answer
- How a watch estimates calories
- Why the numbers go wrong
- Which activities fool them most
- Active calories vs total calories
- Wrist, chest or research device?
- Get better numbers from your device
- Should you eat back what your watch says?
- How to use your watch well
- Something better than a watch
- Comparing a watch with a TDEE calculator
- Smartwatch calorie questions
The short answer
A watch is a brilliant tool for some things and a poor one for others. Step counts during normal walking and heart rate during steady exercise are usually close to the truth. Calories burned are not. They are a calculated guess layered on top of those measurements, and the guess can be badly off for individual people and activities.
The best-known test of this came from Stanford University in 2017. Sixty volunteers wore seven wrist devices while walking, running and cycling, and their actual energy use was measured in the lab. Heart rate was measured well by most devices. Energy expenditure wasn’t: the most accurate device still had a median error of about 27%, and the least accurate about 93%. A 2020 meta-analysis of 60 studies came to a similar view: energy estimates from wrist and arm devices vary a lot in accuracy depending on the activity.
How a watch estimates calories
No watch can measure the energy you burn. Lab methods do that by analysing the oxygen you breathe in and the carbon dioxide you breathe out, or by tracking special water isotopes over weeks. A watch instead combines:
- your profile: age, sex, height and weight, usually feeding a BMR equation;
- heart rate from the optical sensor on your wrist;
- movement from the accelerometer, and sometimes GPS speed;
- a model, built from lab data on other people, that turns all of that into calories.
The model works best for people who resemble the ones it was built from, doing the activities it was built for. Everyone else gets a rougher estimate.
Why the numbers go wrong
- Heart rate isn’t energy. The link between the two differs between people and with fitness. A fit person may burn more at a given heart rate than an unfit one.
- Other things raise heart rate. Heat, caffeine, stress, dehydration and illness all push heart rate up without burning many extra calories.
- Your wrist doesn’t tell the whole story. On a bike or a rowing machine, or pushing a pram, your wrist moves differently from the rest of you.
- Sensor fit. A loose strap, tattoos, skin tone, sweat and cold can all disturb optical heart rate readings, especially during intervals.
- Your profile. An out-of-date weight or a wrong age changes every estimate.
Which activities fool them most
| Activity | How the estimate tends to behave |
|---|---|
| Steady walking and running | Usually the most reasonable, though errors of 10 to 20% or more are still common |
| Cycling, indoor bike | Variable; the wrist barely moves, so the estimate leans heavily on heart rate |
| Strength training | Often unreliable; heart rate jumps during sets while total movement is small |
| Intervals and HIIT | Unreliable; optical sensors lag behind fast heart rate changes |
| Swimming | Variable; water and arm movement disturb the sensors |
| All-day totals | Depend on the BMR equation and how “active” minutes are counted; can be hundreds of calories out |
Active calories vs total calories
Most devices show two different numbers, and mixing them up is a common source of error:
- Active calories (sometimes called “move” or “exercise” calories) are the estimated calories above your resting burn.
- Total calories include your resting burn as well, so they look much bigger: often 1,500 kcal or more before you have done anything.
If you add total calories from a workout to your TDEE, you count your resting metabolism twice. If you compare a total-calories number with a calculator’s TDEE, you are comparing like with like, but the device’s number is still an estimate with a wide error. Check your device’s settings or help pages to see which one it shows.
Wrist, chest or research device?
Not all trackers are equal. The 2020 meta-analysis found that adding physiological sensors such as heart rate improved energy estimates, and that research-grade devices, the kind scientists strap to the hip or arm in studies, were better for total daily energy than consumer wristbands. Chest-strap heart rate monitors follow fast changes in heart rate more closely than wrist sensors, which helps during intervals.
None of these measure energy directly, though. A better heart rate signal still has to pass through a model of how your heart rate relates to your calories, and that relationship differs from person to person.
Get better numbers from your device
- Enter your current weight, height, age and sex, and update weight every few kilos.
- If your device asks for resting or maximum heart rate, or offers a fitness test, use it; personal values usually improve estimates.
- Choose the right workout type before you start. Logging a bike ride as a run, or weights as “other”, changes the maths.
- Pair a chest strap for hard intervals and strength sessions if you care about heart rate detail.
- Keep the sensor clean and the strap snug.
These steps narrow the error; they don’t remove it.
Should you eat back what your watch says?
If your calorie target comes from a TDEE calculator with an activity level, no. Your usual exercise is already included, and adding the watch’s number counts it twice. If you deliberately use a sedentary base and add exercise separately, eat back only part of the watch’s estimate, around half, because devices tend to overestimate and you would have burned some of those calories anyway. Our guide on eating back exercise calories covers both methods.
How to use your watch well
- Keep your profile up to date: weight, age and height feed every estimate.
- Wear it snugly, a finger’s width above the wrist bone, especially for workouts. Consider a chest strap for hard intervals; chest straps track heart rate more closely, though calories are still estimated.
- Use trends, not totals. If your weekly active calories go up by 20%, you probably did more. The exact number matters less.
- Rely on its step count. Steps are the most dependable thing it records. Enter your weekly average in the steps calculator for a transparent estimate.
- Don’t chase calorie goals at the expense of recovery. A ring to close every day is motivating, but rest days are part of training.
Something better than a watch
Your scale and your food log, used together, measure your real TDEE more accurately than any wearable. Eat a steady, fully logged amount for three to four weeks while weighing yourself every morning. Then:
Real TDEE = average calories eaten − (weight change in kg × 7,700 ÷ days)
The adaptive TDEE calculator does the maths and fits a trend line through your weigh-ins. The TDEE Calculator app does it automatically from your log, and its GPS workout tracker records distance and pace for walks and runs.
Comparing a watch with a TDEE calculator
If your watch says your daily total is 2,900 kcal and the calculator says 2,500, don’t average them or pick the higher one. Eat at one of them for two to three weeks and watch your weight. If it holds steady, that number is right for you. Many people find the calculator, with an honest activity level, is closer than the device, but your data decides. See how accurate TDEE calculators are for the calculator side of the story.
Smartwatch calorie questions
Which smartwatch is most accurate for calories?
No consumer wrist device has been shown to be reliably accurate for calories across people and activities. Accuracy varies by model, by activity and by person, so treat any device’s calories as a rough estimate.
Should I eat back the calories my watch says I burned?
Not if your target already includes an activity level. If you use a sedentary base on purpose, eat back about half of the watch’s exercise estimate at most.
Why does my watch say I burned over 3,000 calories?
It may be showing total calories, which include your resting burn, or overestimating activity. Some devices count energetic movements generously. Compare with your weight trend before trusting it.
Are step counts from my watch accurate?
Usually close enough during normal walking. They can undercount at very slow speeds or when your arm doesn’t swing, such as pushing a trolley. Steps are the most useful number your watch gives you.
Is a chest strap better than a wrist sensor?
For heart rate during hard or changing exercise, yes. Calories are still estimated from heart rate and your profile, so they remain approximate.
Does my watch include my BMR in calories burned?
The total calories figure usually does; the active calories figure usually doesn’t. Check which one your device is showing before you compare it with a calculator.
Sources
- Shcherbina A, Mattsson CM, Waggott D, et al. Accuracy in wrist-worn, sensor-based measurements of heart rate and energy expenditure in a diverse cohort. J Pers Med. 2017;7(2):3. doi.org/10.3390/jpm7020003
- O’Driscoll R, Turicchi J, Beaulieu K, et al. How well do activity monitors estimate energy expenditure? A systematic review and meta-analysis of the validity of current technologies. Br J Sports Med. 2020;54(6):332-340. doi.org/10.1136/bjsports-2018-099643
- Fuller D, Colwell E, Low J, et al. Reliability and validity of commercially available wearable devices for measuring steps, energy expenditure, and heart rate: systematic review. JMIR Mhealth Uhealth. 2020;8(9):e18694. doi.org/10.2196/18694
- Case MA, Burwick HA, Volpp KG, Patel MS. Accuracy of smartphone applications and wearable devices for tracking physical activity data. JAMA. 2015;313(6):625-626. doi.org/10.1001/jama.2014.17841
- Westerterp KR. Doubly labelled water assessment of energy expenditure: principle, practice, and promise. Eur J Appl Physiol. 2017;117(7):1277-1285. doi.org/10.1007/s00421-017-3641-x