Are Strava calories accurate? It depends on where the number came from
Strava calories are as accurate as their source, and there are three possible sources. A ride with a power meter gives the most trustworthy number, because the work you did is measured. A ride without one, or a run recorded in the Strava app, uses Strava’s own estimate from your weight and speed. An activity uploaded from a watch shows the watch’s number, which typically carries an error of 20% or more. Work out which one you are looking at before you decide how much to trust it.
Most confusion about Strava calories comes from treating them as one thing. They aren’t.
Key terms
- Upload partner: the device or app that sent the activity to Strava, such as a Garmin, an Apple Watch or a Whoop.
- Power: the rate of work you put into the pedals, in watts. A power meter measures it directly. Without one, Strava estimates it.
- Work (kJ): power multiplied by time. A head unit’s “kJ” figure is the mechanical work you did.
- Gross efficiency: the share of the energy your body burns that ends up as work on the pedals. The rest leaves as heat.
- Grade-adjusted speed: your running pace corrected for hills, so an uphill kilometre counts as harder than a flat one.
Where does Strava get its calorie number?
Strava’s support article sets the rules:
- If the upload partner sends a calorie value, Strava shows that value. It doesn’t recalculate it. Strava says partners use “a wide range of different algorithms and inputs,” and changing your Strava weight or elevation won’t change those totals.
- If no value is sent, or you record in the Strava app, Strava calculates its own. It does this only for rides, runs, walks and hikes. Every other activity type shows no calories unless the partner sent them.
Strava also says plainly that “calorie calculations are only estimates.” So the useful question isn’t whether Strava is accurate. It is which of these three paths produced the number you are looking at.
Rides with a power meter: the best number you’ll get
For rides, Strava’s article says it “uses your power output and a human efficiency coefficient.” With a power meter, the power part is measured, so most of the uncertainty is gone.
This approach works because, in cycling, energy use rises in a straight line with work rate. A review by Ettema and Lorås (2009) found that about 91% of the variance in energy expenditure during cycling is explained by work rate alone.
The remaining unknown is efficiency, and the arithmetic shows why it matters:
- One kilocalorie is 4.184 kJ.
- If your body turned 24% of its energy into pedal work, burning 4.184 kJ would produce about 1 kJ of work. At that efficiency, kJ on your head unit roughly equals kcal burned. That is the shortcut most cyclists use.
- In a 2017 lab study of 12 amateur road cyclists (Ferrer-Roca et al.) riding at 150-250 W, gross efficiency averaged about 20%. At 20%, every 1 kJ of work costs 5 kJ of energy, which is about 1.2 kcal. The real burn would be about 20% higher than the kJ figure.
Efficiency also isn’t fixed. Ettema and Lorås found it depends strongly on work rate, and that elite riders tend to show higher gross efficiency than lower-level riders, largely because a fixed baseline cost matters less at high power. An easy spin is less efficient than a hard effort.
Strava doesn’t publish the coefficient it uses, so we can’t say whether its numbers land nearer the 24% or the 20% case for you. Either way, a power-meter ride is the most trustworthy calorie number in your training log. Our long-ride calories guide uses the kJ method for a worked example.
Rides without a power meter: a rough approximation
Without a power meter, Strava estimates power from your weight, your bike’s weight, your speed and the elevation change. Strava’s own power estimate article calls the result “a rough approximation.”
Look at what the estimator can and can’t see. It knows how fast you went and how much you climbed. It has no input for wind, drafting in a group or the road surface. A solo ride into a headwind and a sheltered ride in a fast group can look similar on speed and elevation and cost very different amounts of energy. Your bike and body weight also have to be entered correctly, because they feed straight into the estimate.
We found no published study (PubMed, October 2026) that measured how far Strava’s estimated-power calories are from a lab measurement, so we can’t give you an error figure. The method is cruder than a power meter. How much cruder depends on how you ride.
Runs: weight, pace and time
For runs, Strava’s article says: “Calories for runs are a function of speed (grade adjusted), weight, and moving time. Multiply these together with a scaling factor, and that’s the estimate.” You have to enter your weight for it to work.
Heart rate isn’t in that formula. Two runners of the same weight who cover the same route at the same pace get the same calorie figure, whatever their fitness. The formula is simple and transparent. Whether it fits you depends on how close you are to the average runner its scaling factor was built on, and Strava doesn’t publish that factor.
Uploaded from a watch: the watch’s accuracy applies
Many Strava activities arrive from a device, and then the calorie number belongs to the device. That puts the evidence on wrist-worn calorie estimates in charge:
- In a 2022 study of outdoor running (Le et al., Frontiers in Physiology), the Apple Watch Series 6 had a mean absolute percentage error of 24.4% and the Garmin Fenix 6 21.8%. Both overestimated compared with indirect calorimetry.
- A 2020 systematic review of 158 publications (Fuller et al.) concluded that for energy expenditure, “no brand was accurate.”
If your ride or run came from a Whoop, an Oura ring, a Garmin or an Apple Watch, see our pieces on Whoop calorie accuracy, Oura calorie accuracy and Garmin vs Whoop vs Apple Watch.
How much should you trust each Strava number?
| Source of the number | How it is calculated | How far to trust it |
|---|---|---|
| Ride, power meter | Measured power plus an efficiency coefficient | Most trustworthy. Efficiency can shift it by around 20% |
| Ride, no power meter | Power estimated from weight, bike weight, speed and elevation | Rough. Can’t see wind or drafting |
| Run in the Strava app | Grade-adjusted speed, weight, moving time, scaling factor | Rough. Ignores fitness and heart rate |
| Uploaded from a watch | The device’s own model, usually heart rate based | Typically 20% or more error in lab studies |
How to use Strava calories without fooling yourself
- Check the source. Look at which device recorded the activity. That tells you which method produced the number.
- Keep your weights current. Your body and bike weight feed Strava’s own estimates directly.
- Prefer power when you have it. If you ride with a power meter, its kJ figure is the best workout number you have.
- Don’t add Strava calories on top of another estimate of the same session. If your watch, your calorie app and Strava all count the same ride, you have counted it two or three times. Our Whoop vs calorie app explainer shows how quickly that adds up.
- Check your intake against your weight trend. Over 2-3 weeks, the scale tells you whether your overall estimate is too high or too low.
Where Sweatmeal fits
Sweatmeal reads your Strava activities as one input for the day’s target. It doesn’t replace Strava’s number or try to fix it. It counts each session once, keeps it apart from your baseline, and labels which numbers are measured and which are estimated. Over time it adjusts your maintenance from your weight trend. The Strava integration page covers what is read, and the training day calorie calculator shows the arithmetic. The full method is on the methodology page.
Sources
- Strava Support. How Does Strava Calculate Calories? Accessed October 2026.
- Strava Support. How to Get Power for Your Rides. Accessed October 2026.
- Ettema G, Lorås HW. Efficiency in cycling: a review. Eur J Appl Physiol, 2009.
- Ferrer-Roca V et al. Acute effects of small changes in crank length on gross efficiency and pedalling technique during submaximal cycling. J Sports Sci, 2017.
- Le S et al. Validity of three smartwatches in estimating energy expenditure during outdoor walking and running. Front Physiol, 2022.
- Fuller D et al. Reliability and validity of commercially available wearable devices for measuring steps, energy expenditure, and heart rate. JMIR mHealth uHealth, 2020.