# Smartwatch calorie estimation accuracy study > Live situation record from CLSTR: https://clstr.news/situations/smartwatch-calorie-estimation-accuracy-study > Updated: 2026-09-08T17:48:12.000Z. Sources: 22. Developments: 3. A study conducted by Florida International University and published in the journal PLOS One has found that popular smartwatches provide inaccurate estimates of calories burned during exercise. Researchers compared the Apple Watch Series 8, Fitbit Sense 2, Samsung Galaxy Watch 5, and Garmin Forerunner 955 against a clinical-grade COSMED K5 metabolic system. The findings indicate an average error margin between 15% and 25%. While the Apple Watch was identified as the most accurate device, it still overestimated energy use. Garmin and Samsung models showed the highest levels of overestimation. The study also noted that Fitbit experienced technical glitches during testing, resulting in missing or implausibly low readings, which led researchers to exclude much of its data from the final analysis. Building on these findings, a study published in the journal Sensors by researchers at the University of Michigan School of Kinesiology suggests that many smartwatch health metrics are algorithmic estimates rather than direct physiological measurements. Assistant professor Adam Lepley noted that while outputs like resting heart rate, step counts, and outdoor pace tend to be more reliable, complex metrics—including calories burned, sleep stages, body composition, and hydration levels—are often generated by combining sensor signals with proprietary algorithms and user assumptions. Accuracy can be further influenced by skin tone, tattoos, sweat, temperature, and device fit. The wearable technology market is currently evolving from simple activity tracking toward advanced health monitoring and AI-driven intervention. Companies like Samsung are integrating BioActive sensors to measure blood oxygen and body composition, while Apple is leveraging Apple Intelligence to provide dynamic fitness recommendations. Additionally, new devices like the Luna Plus, showcased at IFA Berlin, utilize transcutaneous auricular vagus nerve stimulation (taVNS) to actively intervene in physiological states rather than just logging data. ## Timeline ### 2026-09-08: Wearable technology shifts toward AI health monitoring and intervention Smartwatch technology is advancing with AI integration and vagus nerve stimulation, though experts warn that many health metrics are algorithmic estimates best used for tracking long-term trends. 8 sources. https://clstr.news/cluster/smartwatch-health-metrics-may-rely-on-algorithmic-estimates ### 2026-09-04: Smartwatches frequently overestimate calorie burn, study finds A Florida International University study reveals that smartwatches, including models from Apple, Samsung, and Garmin, frequently overestimate calorie burn by 15 to 25 percent during exercise. 7 sources. https://clstr.news/cluster/smartwatches-significantly-overestimate-calorie-expenditure-in-new-study ### 2026-09-02: Smartwatch calorie estimates found to be inaccurate in new study A Florida International University study found that smartwatches like Apple, Samsung, and Garmin can overestimate calorie burn by 15% to 25%, with accuracy decreasing as body fat increases. 7 sources. https://clstr.news/cluster/smartwatch-calorie-estimates-found-to-be-inaccurate-in-new-study --- Cite as: Smartwatch calorie estimation accuracy study. CLSTR, https://clstr.news/situations/smartwatch-calorie-estimation-accuracy-study