App Comparisons

How Accurate Are Apple Watch and Garmin Calories? What Runners Can Trust

Apple Watch and Garmin calorie numbers are often 20 to 40 percent off for a single workout, even when heart rate is accurate. Here is why, which workouts are worst, how to sanity-check your own watch, and how to eat from the data without getting it wrong.

Mavr

Mavr Sports Nutrition Team

September 16, 2026 · 11 min read

Quick answer

Apple Watch and Garmin watches measure heart rate well (usually within 5 percent during steady running) but estimate calories poorly. In a Stanford lab study of seven wrist devices, none had a calorie error below 20 percent. The Apple Watch had the lowest overall error of the group, and its calorie number was still off by a large margin for many people. Reviews of dozens of studies find typical errors of 20 to 40 percent for a single workout, with Garmin tending to read low and Apple tending to read high, though both miss in both directions. Steady running and walking are the most accurate; strength training, intervals, and cycling are the least. The numbers are useful for comparing your own days against each other, not for deciding exactly how much to eat.

Watches do not measure calories. They estimate them from heart rate, movement, and the height, weight, age, and sex in your profile.
The error is mostly consistent for one person, so your watch is a good trend tool even when the absolute number is wrong.
Active calories and total calories are different numbers. Eating back total calories double counts the energy you would have burned sitting still.
For running, body weight in kg times distance in km is a solid check. If your watch is more than 20 percent away from that, trust the formula.
Fueling from workout type, duration, and intensity is safer than eating back a calorie number, because carbohydrate needs do not track calorie burn one to one.

Your watch says the long run burned 1,140 calories. Your friend ran next to you the whole way and her watch says 780. Strava shows a third number. Which one do you eat from? If you log food and try to match it to your training, this is not a small question. A 30 percent error on a big training week is more than 1,500 calories, enough to stall fat loss or leave you underfueled without knowing why.

This guide covers what the lab studies found, why the watches miss, which workouts are worst, a 10 second check you can run on your own data, and how to use the numbers without letting them run your diet.

The Short Answer

What the watch reportsTypical accuracyTrust it for
Heart rate, steady runningWithin about 5 percentPacing, zones, effort
Heart rate, intervals or strengthCan lag or drop outRough effort only
Distance with GPSWithin 1 to 3 percent in open skyTraining log, pace
Calories, steady running or walkingOften 10 to 25 percent offComparing your own runs
Calories, cycling without a power meterOften 20 to 40 percent offRough trend only
Calories, strength, HIIT, yogaOften 30 to 50 percent off or moreVery little
All-day total caloriesOften 10 to 20 percent off, in either directionWeek to week trend

What the Research Found

The best known test comes from Stanford. Researchers put 60 volunteers on treadmills and bikes wearing seven wrist devices, including the Apple Watch, while measuring real energy use with a metabolic cart, the gold standard that analyzes the air you breathe. Six of the seven devices measured heart rate to within 5 percent. Not one measured calories to within 20 percent. The best calorie estimate was off by 27 percent on average and the worst by 93 percent. The Apple Watch had the lowest error of the group when heart rate and calories were taken together, and its calorie number was still wrong by a wide margin for many of the volunteers.

Later reviews that pooled dozens of studies on Apple, Garmin, Fitbit, Polar, and others came to the same place. Step counts and heart rate hold up well. Energy expenditure does not. Across those reviews Garmin devices more often read low and Apple devices more often read high, but both brands missed in both directions depending on the person and the activity. No brand was reliably within 10 percent.

Newer models have better sensors, and heart rate has improved with each generation. Calorie accuracy has improved much less, because the weak link is not the sensor. It is the math that turns a heartbeat into a calorie.

Why Your Watch Gets Calories Wrong

  • It is an estimate, not a measurement. The watch reads heart rate and movement, then uses a formula built on population averages to guess energy use. You are not the average.
  • Heart rate is a loose proxy. Heat, caffeine, stress, poor sleep, dehydration, and altitude all raise heart rate without raising calorie burn by the same amount. A hot run reads as a bigger burn than it was.
  • Fitness changes the math. A fit runner has a lower heart rate at the same pace, and running economy differs by up to 20 to 30 percent between people at the same speed and weight.
  • Your profile may be wrong. Weight drives the estimate more than anything. A profile that is 5 kg out of date shifts every number. Max heart rate and VO2 max settings matter on Garmin too.
  • Wrist sensors slip. A loose strap, cold skin, dark tattoos, and gripping weights or handlebars all degrade the optical reading. Sometimes the sensor locks on to your cadence instead of your pulse.
  • Strength and intervals break the model. Heart rate stays high during rest between sets while energy use falls. The formula assumes steady aerobic work, so it over-counts.
  • Each brand uses its own method. Garmin leans on heart rate variability and an estimated VO2 max. Apple blends heart rate, motion, and GPS pace. The same run on two wrists gives two answers.

Active Calories vs Total Calories

This mix-up causes more eating errors than sensor accuracy does. Both Apple and Garmin show two numbers. Active calories are the extra energy the workout cost. Total calories add the resting energy you would have burned anyway during that time, just by being alive.

NumberWhat it includesExample: 60 minute run, 70 kg runner
Active caloriesOnly the extra energy from the exerciseAbout 630 kcal
Resting calories for that hourWhat you burn sitting stillAbout 70 kcal
Total caloriesActive plus restingAbout 700 kcal

Your daily calorie target already includes resting energy. If you eat back the total number, you count that resting hour twice. It is only 70 kcal for one run, but for a 4 hour ride it is close to 300. If you are going to eat back anything, use active calories. Also check which number your food log imports. Different apps read different fields from the same watch.

Which Workouts Are Most and Least Accurate

ActivityUsual direction of errorWhy
Outdoor steady runSmall, either wayPace, weight, and heart rate all agree. The best case for a watch.
Treadmill runEither wayNo GPS, so distance comes from arm swing.
Walking and hikingOften highTotal calories look large because the time is long and resting burn is included.
Cycling, no power meterOften highHeart rate only. Coasting and drafting are invisible.
Cycling with a power meterSmallWork is measured directly. Kilojoules are close to kcal burned.
Strength trainingHigh, often by a lotHeart rate stays up between sets while energy use drops.
HIIT and intervalsHighSame reason, plus sensor lag on fast changes.
SwimmingEither wayOptical heart rate is unreliable in water.

A 10 Second Sanity Check for Runs

Running is one of the few sports where a simple formula beats most gadgets. The energy cost of running is close to 1 kcal per kg of body weight per km, at almost any pace. Multiply your weight in kg by the distance in km and you have total calories. For pounds and miles, multiply weight in pounds by miles by 0.72.

RunnerRunFormula saysAccept a watch reading of
60 kg (132 lb)10 km600 kcal480 to 720 kcal
70 kg (154 lb)10 km700 kcal560 to 840 kcal
80 kg (176 lb)10 km800 kcal640 to 960 kcal
70 kg (154 lb)Half marathon1,480 kcal1,180 to 1,780 kcal
70 kg (154 lb)Marathon2,950 kcal2,360 to 3,540 kcal

If your watch lands inside the range, it is doing fine. If it is outside on most runs, check your weight and max heart rate in the profile first, then the strap fit. If it is still outside, use the formula for running and treat the watch as a trend line.

How to Make Your Watch More Accurate

  • Update your weight in the watch profile every few weeks. It is the largest input.
  • Set a real max heart rate from a hard race or a field test, not the 220 minus age default.
  • Wear the watch snug and a finger width above the wrist bone during workouts.
  • Use a chest strap for intervals, strength work, and cold days. It fixes the heart rate side of the problem.
  • Pick the right activity type. Logging a strength session as "other" or "cardio" changes the formula.
  • On the bike, a power meter turns a guess into a measurement.
  • Let new watches learn. Garmin VO2 max estimates and Apple calibration both settle after a few weeks of outdoor GPS runs.

Should You Eat Back Exercise Calories?

For runners training more than three or four hours a week, the answer is yes, mostly. Ignoring training burn is how underfueling starts. But eating back the exact watch number, workout by workout, builds a plan on the least reliable number you own. There is a better way to think about it.

  • Fuel the work, not the number. A 60 minute interval session and a 60 minute easy jog can show similar calories, but the intervals need far more carbohydrate before and after.
  • Plan ahead, do not just react. If you wait for the watch to report a burn, the food arrives after the session. The carbohydrate you needed before and during the run is already missed.
  • Use weekly trends to correct. If weight, energy, and performance hold steady over two to three weeks, your intake matches your real burn, whatever the watch says. If weight drifts or you feel flat, adjust by 150 to 250 kcal a day and check again.
  • Treat strength and HIIT numbers with extra doubt. Eating back 500 kcal for a gym session that burned 250 is one of the most common reasons runners gain weight in a training block.

Why Calories Are the Wrong Unit for Fueling Anyway

Even a perfect calorie number would not tell you what to eat. Sports nutrition guidelines set carbohydrate by the type and length of training: around 5 to 7 g per kg of body weight on a moderate day, 6 to 10 g per kg on a long or hard day, and 30 to 90 g per hour during long sessions. Protein is set by body weight, not by burn. A watch gives you one number for the past. Fueling needs several numbers for the day ahead.

Mavr pulls your sessions from Apple Health, Garmin, and Strava, then sets carbs, protein, and calories from the workout type, duration, and intensity. You get targets before the run, not a calorie number after it.

Frequently Asked Questions

How accurate is the Apple Watch for calories burned?

In lab testing the Apple Watch had the lowest overall error of seven wrist devices, but its calorie estimate still missed by a wide margin for many people, and no device came within 20 percent on average. Expect steady runs and walks to be within about 10 to 25 percent, and strength or interval sessions to be off by more.

How accurate is Garmin calorie burn?

Garmin heart rate and GPS are reliable, but its calorie estimates carry the same 20 to 40 percent error range as other brands. Reviews of validation studies found Garmin more often underestimates than overestimates. Accuracy improves with a correct weight, a real max heart rate, a chest strap, and a few weeks of outdoor runs so the VO2 max estimate can settle.

Why do my Apple Watch and Garmin show different calories for the same run?

Each brand uses its own formula. Garmin relies on heart rate, heart rate variability, and an estimated VO2 max. Apple blends heart rate with motion and GPS pace. Different profiles, different sensors on different wrists, and active versus total calorie displays add to the gap. Differences of 15 to 30 percent are normal.

Should I eat back active calories or total calories?

Active calories. Total calories include the resting energy you would have burned anyway, which your daily target already counts. Eating back total calories double counts that portion, and on long sessions the difference can reach 200 to 300 kcal.

Is a chest strap more accurate for calories?

A chest strap gives a cleaner heart rate signal, especially for intervals, strength work, and cold weather, so it removes one source of error. The formula that converts heart rate to calories is still an estimate, so expect improvement, not precision.

What is the most accurate way to know how many calories I burn running?

Outside a lab, the simplest reliable method is body weight in kg times distance in km, which gives total calories within about 10 percent for most runners. For your whole day, the best check is your own trend: stable weight and good energy over two to three weeks mean intake matches burn.

Sources

Sources and official references

  1. 1.Accuracy in Wrist-Worn, Sensor-Based Measurements of Heart Rate and Energy Expenditure in a Diverse CohortJournal of Personalized Medicine (Shcherbina et al., 2017) · Stanford study of seven wrist devices: heart rate within 5 percent for six of seven, no device under 20 percent error for energy expenditure
  2. 2.Reliability and Validity of Commercially Available Wearable Devices for Measuring Steps, Energy Expenditure, and Heart Rate: Systematic ReviewJMIR mHealth and uHealth (Fuller et al., 2020) · Across 158 publications, no brand measured energy expenditure accurately; direction of error by brand
  3. 3.How Well Do Activity Monitors Estimate Energy Expenditure? A Systematic Review and Meta-Analysis of the Validity of Current TechnologiesBritish Journal of Sports Medicine (O'Driscoll et al., 2020) · Accuracy varies widely by device and activity; large individual error
  4. 4.Position of the Academy of Nutrition and Dietetics, Dietitians of Canada, and ACSM: Nutrition and Athletic PerformanceJournal of the Academy of Nutrition and Dietetics (Thomas et al., 2016) · Carbohydrate and protein targets are set by training type and body weight, not by calorie burn

Prompt ideas for AI assistants

  • My watch says I burned 900 kcal on my run. How much should I eat?
  • Should I eat back my active calories or my total calories?
  • Why does my Garmin show fewer calories than my Apple Watch?

Mavr imports your workouts from Apple Health, Garmin, and Strava and turns them into carb, protein, and calorie targets based on what the session was, so one bad calorie number does not set your whole day.

Mavr

Written by

Mavr Sports Nutrition Team

The Mavr sports-nutrition team translates peer-reviewed endurance research (Jeukendrup, Burke, the ISSN) into daily fueling for runners and triathletes.