US scientists have discovered that popular smartwatches frequently overestimate the number of calories burned during exercise. A study published in the journal PLOS One found that Apple, Samsung, and Garmin devices can overestimate calorie burn by an average of 110 calories, or up to 25 percent. Fitbit watches were found to be the least accurate, exhibiting both dramatic overestimations and underestimations, with an average deviation of 128 calories.
Researchers from Florida International University tested the Apple Watch Series 8, Fitbit Sense 2, Samsung Galaxy Watch 5, and Garmin Forerunner 955. In the study, 58 adults completed 60 minutes of moderate to vigorous cycling while wearing the devices. Their calorie expenditure was simultaneously measured using a metabolic analyzer, considered a gold standard for such measurements.
The findings indicate that Garmin and Samsung watches overestimated calorie burn by approximately 25 percent, while the Apple Watch Series 8 was the most accurate among the tested brands, with an overshoot of roughly 12 percent. The Fitbit Sense 2, which costs around £219, demonstrated inconsistent results, sometimes reporting as little as a single calorie for an entire workout or failing to provide a reading altogether.
According to the study's lead author, Jason Kostrna, an associate professor of kinesiology and exercise science at Florida International University, these inaccuracies mean users could be hundreds of calories off each week, potentially leading to a calorie surplus when aiming for a deficit.
Factors Affecting Accuracy
The study also noted that individuals with a higher body fat percentage are more likely to experience greater discrepancies in calorie readings across all tested devices. The authors stated that higher body fat "systematically increased the tendency for devices to deviate from the gold standard," adding an additional 22 to 35 calories of estimation error.
Experts suggest that the inaccuracy stems from how smartwatches estimate calorie expenditure. Unlike direct metabolic analysis, these devices primarily rely on heart rate and movement data. Dr. Cailbhe Doherty, a lecturer at University College Dublin, explained that this method is a significant approximation compared to measuring oxygen consumption. "Essentially, the signal is so far removed from the original biometric rhythm that there's such a high degree of error that can be introduced that way," Doherty said.
While the study highlights significant limitations in the accuracy of calorie-tracking features on smartwatches, the researchers suggest the data can still be useful for tracking personal trends over time rather than focusing on exact daily totals. Kostrna advised users to "track your own trends over time" and to "focus on how you're feeling and use data to get a little better over time" if the data becomes a source of fixation.