Skip to main content
Metric

Activity

How accurate your step count, calorie burn, and workout heart rate actually are, and what the research says about the formulas behind them.

Steps, calories burned, and heart rate zones are the numbers that show up after almost every workout, and most people never ask how they're actually calculated. Steps come from an accelerometer inferring motion, not counting footfalls. Calories come from a formula, not a direct measurement of the energy your body used. Your "max heart rate" is usually just your age plugged into an equation. Here's what the research says about how well each of those actually holds up.

What your device actually calls this
Scientific measurementStep count and energy expenditure are inferred from accelerometer motion and heart rate data, not measured directly. Maximal heart rate is typically estimated from an age-based formula (most commonly "220 minus age"), not measured during a real maximal effort.
DeviceWhat they call itWhat it actually measures
Oura"Automatic Activity Detection"Identifies roughly 40 activity types automatically, then reports average heart rate and heart-rate-zone breakdown once the activity is detected. "Active Calories" now factors in exercise intensity, not just movement.
WhoopHeart-rate-based "Strain" and calorie estimateBuilt on a real, named 2005 formula (Keytel et al., University of Cape Town) predicting energy expenditure from heart rate, weight, and age, with proprietary adjustments layered on top.
Garmin"Firstbeat" engineDrives Garmin's calorie and intensity estimates; published accuracy is reported as dependent on wearing a chest strap for best results.
Apple Watch / most wrist wearablesStep count, Active Calories, Heart Rate ZonesStep count from a 3-axis accelerometer inferring wrist-movement patterns, not literal footfalls; accuracy drops sharply at slow walking speeds or when wrist movement is restricted.
FitbitStep count, Active Zone Minutes, calorie estimateThe original consumer step-counting wearable. Step count remains one of the more reliable metrics across this device category; calorie tracking shares the same weakness found across every other brand.

The most surprising finding in the research behind this page: the same wearables that measure heart rate and step count reasonably well are still bad at calories specifically. A systematic review of 65 studies across 72 devices and 29 brands found energy-expenditure error exceeded 30% for every brand tested, even though step count and heart rate came through comparably better in the same review.

Step count & heart rate accuracy: Established · on the Research Desk Calorie accuracy: Uncertain · on the Research Desk 220-minus-age formula accuracy: Uncertain · on the Research Desk
How each device actually shows this
Oura "Automatic Activity Detection" Whoop Heart-rate-based "Strain" and calorie estimate Garmin "Firstbeat" engine Apple Watch Step count, Active Calories, Heart Rate Zones Fitbit Step count, Active Zone Minutes, calorie estimate See every device →

Sources: Oura's own activity-features blog post (ouraring.com), a primary source. Whoop's calorie methodology traces to Keytel LR et al., "Prediction of energy expenditure from heart rate monitoring during submaximal exercise," Journal of Sports Sciences, 2005, doi:10.1080/02640410470001730089. Garmin's Firstbeat detail from Garmin's own technology page (garmin.com). Full study-by-study evidence, including the max-heart-rate-formula literature, lives on the Research Desk linked above.

Research Desk · live file
See the full Activity Research Desk
Every claim rated Established, Emerging, Uncertain, Insufficient, or Unsupported, with a research log reviewed against new studies on a rolling basis.
Open the Research Desk
More on Activity
How Accurate Is the Calorie Count on Your Watch?
People assume a watch inflates calorie burn. Across validation reviews it is the least accurate of the three core metrics, and tends to underestimate.
How Does Your Watch Actually Count Steps?
Wrist counters infer steps from arm-swing acceleration. That holds at ordinary walking pace and degrades with slow, shuffling or otherwise altered gait.
Is 220 Minus Your Age Actually Your Max Heart Rate?
In recreational marathon runners, measured maximum heart rate missed the 220-minus-age formula, running lower than predicted in women and higher in men.
Does Wearing a Fitness Tracker Actually Make You More Active?
Using a tracker is associated with more physical activity than not using one. That is a finding about behavior, not about whether the counting is accurate.
Step Counts Hold Up in Parkinson's, Less Known for Stroke
Fitbit and Garmin step counts fell within 3% of a research monitor in 33 people with Parkinson's. Stroke and brain injury studies rarely report accuracy.
Tracker Data Sorted Depression Scores About 80% Right
Wearable data sorted depression screening results about 80% right in two studies. Only 1 of 42 reviewed studies had outside validation.
Your Smartwatch May Undercount Steps If You Use a Cane or Walker
A rolling walker or oxygen cart threw wrist step counts badly off. A cane, tested separately in 11 people on a different device, did much less damage.
Do Fitness Trackers Accurately Count Steps in People With ALS?
A 2026 study tracked ALS patients for months with a Fitbit and a research-grade accelerometer side by side. Here's how closely they actually agreed.
Do Workplace Wellness Programs and Wearables Actually Work?
Randomized trials in nearly 5,000 and 32,974 US employees found no effect on medical spending, health or absenteeism. What wearable trials, burnout studies and vendor claims add.
Other metrics
Heart Rate VariabilityResting Heart RateSleepVO2 MaxWearable ECGCGMRecoveryActivityBlood Oxygen