Apple Watch Gets One Crucial Fitness Metric Wrong, Researchers Say

[MacRumors]

The Apple Watch provides highly accurate measurements of heart rate and step count, but their estimates of calories burned can be significantly off, according to a new peer-reviewed meta-analysis conducted by researchers at the University of Mississippi (via CNET).

The study reviewed 56 previously published studies evaluating the Apple Watch's performance against gold-standard clinical tools in three core areas: heart rate monitoring, step count tracking, and energy expenditure estimation.

The researchers reported low mean absolute percentage errors (a common metric used to assess measurement accuracy) of 4.43% for heart rate and 8.17% for step counts. These fall within the threshold generally considered acceptable for consumer-grade fitness devices. In contrast, the average error for energy expenditure was 27.96%, more than three times the margin considered acceptable for accurate measurement.

The analysis incorporated data from studies spanning multiple Apple Watch models and user groups. The high error margin in calorie estimation was consistent across all cohorts and forms of physical activity.

These devices are great for keeping track of habits and staying motivated. But do not take every number as 100% truth, especially the calories. Think of it as a helpful guide, not a diagnostic tool. It is useful but not perfect.

The findings align with previous independent evaluations that have raised concerns about the reliability of calorie burn estimates from consumer-grade wearable devices like Apple Watches.

While the Apple Watch has undergone continuous generational improvements since its debut in 2015, the researchers noted that even newer models still exhibit considerable error in calorie estimation. However, the study did observe a trend toward improved accuracy in more recent models:

While we cannot say every update is a big leap forward, there is a noticeable trend of gradual improvements over time. It shows that Apple is refining the technology over time.

The researchers emphasized that their analysis is not intended to discredit the utility of wearables, but rather to clarify their limitations and inform both consumers and smartwatch makers.

By showing where the weaknesses are, we can help developers get real feedback. If they know what needs to be fixed, they can design better sensors or algorithms. Our findings can guide improvements and help make these devices more useful for both everyday users and health care providers.

Apple does not publish the algorithms used in Apple Watch fitness tracking, nor does it claim that the device provides clinical-grade energy expenditure measurements. The company has consistently positioned the Apple Watch as a general wellness tool rather than a medical diagnostic device, though it has introduced several advanced health features in recent years, such as body temperature monitoring and sleep apnea detection.

Article Link: Apple Watch Gets One Crucial Fitness Metric Wrong, Researchers Say

 

[DBZmusicboy01]

I trust Mississippi’s Educational Research more than Alabama’s Education 🥲 even if it’s just 10% better.

 

[rightoff]

cool so the one thing i never pay attention to or have need of lol

 

[ChrisMac47]

Is this a random error of 28%, or a deterministic one (i.e. bias), and if the latter, does the watch estimate consistently higher calorie burn, or consistently lower?

 

[tyranne201]

mine is off about 250 calories a day. I count how much I eat and sometimes don't lose that much compared to te watch.

 

[roccobladr]

I'm a registered dietitian who works with clients trying to lose weight. I have many clients with an Apple Watch and stress that these numbers are a fair place to start, but not take it as gospel. That said, there are times where the resting energy and active calories do end up being similar to standard calculations (based on Mifflin St Jeour equations).

 

[hanschien]

The calories burned on the Apple Watch is very inconsistent for me. Every morning for my morning walk, I walk the same path, duration, avg heart rate, and pace. Yet the active calories recorded for these workout session ranges from 300 to 400.

I'm not sure which other metrics Apple use to calculate calories.

 

[tonymillion]

This is the cell signal bars all over again...

"You're wearing it wrong"

 

[iSalameee]

Regardless, if this is accurate or not; Apple needs to make sure its fixed if true and if you are using your watch for fitness and say it is off, its fine, use it for consistency and adjust accordingly.

 

[WarmWinterHat]

This has been pretty well known with in the fitness and Healthcare community, but not with exact figures like this study. It's not specific to Apple, though; all the trackers are off.

I'm transitioning from an Apple Watch to a Garmin, and on a fairly strenuous trail run on Tuesday (where I was wearing both), they disagreed by about 200kcal. Neither were right, but the Garmin was a bit closer.

The watches agreed on everything else.

 

[imola.zhp]

If I'm on a workout machine that counts calories, the watch displays significantly more calories burned than the workout machine.

 

[Marzzz]

The calories burned as measured by nearly ANY device or fitness machine are wildly inaccurate, and most likely are over estimated by a significant factor. Coincidently (not being OCD here) I have three different devices estimating calories burned while I am cycling, and these measurements are so different as to be just random numbers. As far as actually calories burned, I assume at best 50% of the average.

 

[Killbill2]

I wish they'd get rid of "calories" it's irrelevant.

The continued emphasis on “calories” is not only outdated but fundamentally misleading. Caloric measurement is an oversimplified and reductionist approach to health and nutrition. It fails to reflect the complex regulatory systems of human metabolism, individual differences in energy expenditure, and the profoundly different metabolic effects of macronutrients.

The notion that “a calorie is a calorie” ignores the fact that carbohydrates, fats, and proteins are metabolized through distinct biochemical pathways and exert different hormonal effects—particularly on insulin, satiety, and fat storage. For example, 100 calories of sugar is not metabolically equivalent to 100 calories of protein or fat.

If you’re trying to lose weight, reducing refined carbohydrates and emphasizing protein and healthy fats—while eating to satiety—tends to be far more effective than simply slashing calories. This approach aligns better with how the body actually regulates hunger, energy balance, and fat storage.

It’s time to move beyond the calorie and toward a more nuanced, biologically informed understanding of nutrition.

 

[iAssimilated]

So, the article states "Many Americans rely on their Apple Watches or similar devices..." but then fails to mention any other device. I would like to know how all smart watches match up in these metrics if the author of the study truly wants us to be informed.

 

[bluecoast]

I wish I could turn off the red activity ring as I’ve never seen how it could ever be accurate and this study backs my hunch up.

I get how it is a useful motivator for those who have a largely sedentary lifestyle to get moving.

The issue is that the data it uses looks so precise when it cannot possibly be.

The way the blue ring works seems more useful in this regard.

The green ring id super useful though (however, the coaching seems to be stuck in 2017 or so).

 

[johannnn]

the study did observe a trend toward improved accuracy in more recent models

So what’s the % error on S10 or Ultra2? Study is behind a paywall lol.

 

[nfl_brah]

I wish they'd get rid of "calories" it's irrelevant.

The continued emphasis on “calories” is not only outdated but fundamentally misleading. Caloric measurement is an oversimplified and reductionist approach to health and nutrition. It fails to reflect the complex regulatory systems of human metabolism, individual differences in energy expenditure, and the profoundly different metabolic effects of macronutrients.

The notion that “a calorie is a calorie” ignores the fact that carbohydrates, fats, and proteins are metabolized through distinct biochemical pathways and exert different hormonal effects—particularly on insulin, satiety, and fat storage. For example, 100 calories of sugar is not metabolically equivalent to 100 calories of protein or fat.

If you’re trying to lose weight, reducing refined carbohydrates and emphasizing protein and healthy fats—while eating to satiety—tends to be far more effective than simply slashing calories. This approach aligns better with how the body actually regulates hunger, energy balance, and fat storage.

It’s time to move beyond the calorie and toward a more nuanced, biologically informed understanding of nutrition.

If you’re trying to lose weight, consuming less calories than burning is the most consistent and proven method. Removing foods may result in weight loss, but it’s still due to a caloric deficit. That’s why these tools are useful, knowing TEE and tracking macros works.

 

[tennisproha]

Apple does not publish the algorithms used in Apple Watch fitness tracking, nor does it claim that the device provides clinical-grade energy expenditure measurements. The company has consistently positioned the Apple Watch as a general wellness tool rather than a medical diagnostic device, though it has introduced several advanced health features in recent years, such as body temperature monitoring and sleep apnea detection.

Apple’s intention is to market the watch as a medical diagnostic device, with the AirPods getting FDA approval as a hearing aid. In addition to calories, stand counts are also very inaccurate with room for improvement on heart rate as well.

 

[tennisproha]

I wish they'd get rid of "calories" it's irrelevant.

The continued emphasis on “calories” is not only outdated but fundamentally misleading. Caloric measurement is an oversimplified and reductionist approach to health and nutrition. It fails to reflect the complex regulatory systems of human metabolism, individual differences in energy expenditure, and the profoundly different metabolic effects of macronutrients.

The notion that “a calorie is a calorie” ignores the fact that carbohydrates, fats, and proteins are metabolized through distinct biochemical pathways and exert different hormonal effects—particularly on insulin, satiety, and fat storage. For example, 100 calories of sugar is not metabolically equivalent to 100 calories of protein or fat.

If you’re trying to lose weight, reducing refined carbohydrates and emphasizing protein and healthy fats—while eating to satiety—tends to be far more effective than simply slashing calories. This approach aligns better with how the body actually regulates hunger, energy balance, and fat storage.

It’s time to move beyond the calorie and toward a more nuanced, biologically informed understanding of nutrition.

That’s has its own problems since satiety is not highly regulated, especially in people with past weight issues.

 

[0815]

Not surprised ... from my experience, doing the same exercises under same conditions, I get a big variety of calories burned. But still good enough to get a good feeling if an exercise was worth it.

 

[Jean Claude]

Folks who obsess over calories while working out are really doing themselves a disservice. No amount of exercise (while counting calories) is going to help you if your diet isn't on point. You can bike for 100s of miles and watch the calories add up - but if you go home and eat some pizza - you've wasted your entire day. Count carbs and limit sugars, period. Leave the calorie obsession behind....

 

[hanschien]

stand counts are also very inaccurate with room for improvement on heart rate as well.

Yes the stand ring is useless. Apple Watch records standing while laying in bed munching on sunflower seeds binging Youtube/Netflix but if I do some chores around the house, I'll get an alert that it's time to stand...

 

[0815]

So, the article states "Many Americans rely on their Apple Watches or similar devices..." but then fails to mention any other device. I would like to know how all smart watches match up in these metrics if the author of the study truly wants us to be informed.

All devices suck at this ... but calling out Apple produces headlines

 

[sflagel]

I wish they'd get rid of "calories" it's irrelevant.

The continued emphasis on “calories” is not only outdated but fundamentally misleading. Caloric measurement is an oversimplified and reductionist approach to health and nutrition. It fails to reflect the complex regulatory systems of human metabolism, individual differences in energy expenditure, and the profoundly different metabolic effects of macronutrients.

The notion that “a calorie is a calorie” ignores the fact that carbohydrates, fats, and proteins are metabolized through distinct biochemical pathways and exert different hormonal effects—particularly on insulin, satiety, and fat storage. For example, 100 calories of sugar is not metabolically equivalent to 100 calories of protein or fat.

If you’re trying to lose weight, reducing refined carbohydrates and emphasizing protein and healthy fats—while eating to satiety—tends to be far more effective than simply slashing calories. This approach aligns better with how the body actually regulates hunger, energy balance, and fat storage.

It’s time to move beyond the calorie and toward a more nuanced, biologically informed understanding of nutrition.

True for calorific intake, but calorific expenditure, which is what the Apple Watch is meant to measure, should be consistent, no matter what you ate, or not?

 

[sflagel]

Folks who obsess over calories while working out are really doing themselves a disservice. No amount of exercise (while counting calories) is going to help you if your diet isn't on point. You can bike for 100s of miles and watch the calories add up - but if you go home and eat some pizza - you've wasted your entire day. Count carbs and limit sugars, period. Leave the calorie obsession behind....

The Apple Watch tells me I burn over 800 calories in an hour of tennis. No way, not believable; that amount of calories burnt in an hour would help! (In addition to avoiding the pizza).