Strava is generally accurate enough for training, but GPS conditions and device quality can shift distance, pace, segments, and elevation.
One sharp turn between tall buildings can add a zigzag to the map and turn a solid workout into a suspicious personal record. How accurate Strava is depends less on the app name than on the phone, watch, sensor, satellite view, and calculation behind each metric.
For everyday training, distance and average pace are usually consistent enough to track trends. Race certification, medical decisions, calorie targets, and close leaderboard disputes need a stronger reference than one Strava activity.
Strava Accuracy: What Changes The Result
Strava accuracy begins with the data recorded by your phone, sports watch, bike computer, heart-rate strap, or power meter. Strava can smooth obvious outlier points and recalculate some fields, but it cannot recreate GPS points that were never recorded.
Uploaded activity files may contain a device-calculated distance stream. When that stream is present, Strava generally uses it; when it is absent, Strava connects successive GPS coordinates and adds those short distances together. Bad-data detection can also blend device distance with GPS distance, which explains why the number in Strava may differ slightly from the number saved on the device.
A 2020 instrument-validation study in JMIR mHealth and uHealth tested eight positioning-enabled sports watches in urban, forest, and track settings. Mean absolute percentage error for distance ranged from 3.2% to 6.1% across the tested watches, and open areas produced better results than obstructed areas. Those figures describe the recording devices, not a fixed error rate for every Strava activity.
Device GPS Matters More Than The Strava Logo
The recording device usually has more influence on distance accuracy than Strava’s post-processing. A phone running the Strava mobile app can record a dependable open-road run, while the same phone may struggle in a downtown corridor, under thick tree cover, or deep inside a pocket.
Dedicated sports watches and bike computers can offer better antennas, more satellite systems, dual-frequency reception, barometric elevation, and longer battery life. None of those features makes every track exact. Device fit, placement, firmware, satellite lock, terrain, and weather still shape the file that Strava receives.
Direct sensors deserve more trust than estimates for the field they measure. A calibrated bike power meter is stronger evidence for watts than Strava’s estimated power, and a wheel-speed sensor can maintain distance through a tunnel where satellite reception disappears.
Recording frequency matters too. A device that saves positions less often can draw straight chords across switchbacks, roundabouts, or winding trails, shortening the course. A device that samples noisy positions too aggressively can count every side-to-side jump, lengthening it. The smoothest-looking map is not always the truest one; compare the line with known roads and measured course markers.
Why Can Two People Get Different Strava Results?
Two athletes can complete the same course and still receive different distances, pace figures, elevation totals, and segment matches. Each recorder collects its own sequence of points, applies its own sampling rules, and may send Strava different sensor fields.
- Different hardware: A dual-frequency watch may hold position better in a city than an older phone.
- Different recording intervals: Fewer saved points can cut across bends and shorten distance.
- Different pause rules: One device may pause at a stoplight while another keeps counting moving time.
- Different sensors: A bike wheel sensor, barometric altimeter, chest strap, or power meter can replace or supplement GPS-derived values.
What Does Strava Measure Accurately?
Strava measures location-based fields most consistently when the recorder has a clear sky view and a stable satellite lock. Sensor-based fields are only as dependable as the connected sensor and the data written into the activity file.
| Strava Metric | When It Is Strongest | Main Weak Point |
|---|---|---|
| Route map | Open sky with steady GPS sampling | Drift beside buildings, cliffs, or dense trees |
| Distance | Long outdoor efforts with a good GPS track or wheel sensor | Zigzags add distance; signal gaps cut corners |
| Average pace or speed | Accurate distance paired with consistent pause behavior | Small distance or moving-time errors alter the average |
| Moving time | Consistent manual pauses or steady outdoor motion | Auto-pause and weak GPS can misread short stops |
| Segment time | Clear crossings of both segment endpoints | GPS drift can miss an endpoint or match it late |
| Elevation gain | A calibrated barometric altimeter on covered roads or trails | Weather changes, blocked sensor holes, and sparse basemap data |
| Heart rate | A snug chest strap or well-fitted optical sensor | Motion, poor skin contact, or sensor dropouts |
| Calories and estimated power | Current body and bike data plus direct sensor readings | Model assumptions can be far from individual physiology |
Strava’s current explanation of GPS accuracy limits says a receiver typically needs signals from seven or eight satellites to locate a device within about 10 meters. Buildings, trees, tunnels, mountains, clothing, and the human body can block signals, while reflected signals can make the recorded line jump away from the street or trail.
Elevation, Calories, And Power Need More Caution
Strava elevation, calorie, and estimated-power figures involve more assumptions than elapsed time or a direct sensor reading. These fields are useful for comparing your own repeated activities, but small differences should not be treated as laboratory measurements.
Strava prefers recorded barometric elevation from recognized devices. When a file lacks usable barometric data, Strava cross-references GPS points with its elevation basemap and smooths the profile. Coastal roads, bridges, steep terrain, sparse basemap coverage, pressure changes, and dirty altimeter ports can all distort climbing totals.
Calories may come from an upload partner or from Strava’s own model. Running estimates use factors such as weight, grade-adjusted speed, and moving time; cycling estimates can use power, body weight, and bike weight. A wrong profile weight or an estimated wattage value can carry the error into the calorie total.
| What You See | Likely Cause | First Action |
|---|---|---|
| A zigzag beside tall buildings | Reflected satellite signals | Compare the line with the street and disregard the spike |
| A straight line across part of the map | GPS signal loss between two points | Crop only if the bad section is at the start or finish |
| An impossible maximum speed | One or two displaced GPS points | Judge the activity by average speed and the map trace |
| Distance differs from the watch | Strava smoothing or distance recalculation | Check whether the activity offers Correct Distance |
| Elevation differs from a friend’s | Barometer data versus basemap lookup | Use Correct Elevation when the option appears |
| A segment did not match | The track missed a start or endpoint | Inspect the map before asking Strava Support for a review |
| Pace looks too quick after stops | Pause events or moving-time detection | Compare moving time with elapsed time |
Make Strava Readings More Trustworthy
Better Strava data comes from improving the recording before the activity starts. Post-processing can remove obvious errors, but prevention preserves more of the true course.
- Stand outdoors with a broad sky view and wait until the phone or watch reports a GPS lock. A cold receiver may need several minutes.
- Wear the watch snugly and keep a phone near the top of a pocket or pack rather than buried under dense material.
- Use manual pause consistently, or leave it alone and let Strava calculate moving time. Mixing the two creates confusing pace results.
- Calibrate barometric elevation and bike sensors according to the device maker’s instructions.
- Enter current body weight and bike weight before relying on calorie or estimated-power figures.
- After saving, inspect the map before accepting a personal record, segment result, or unusually high speed.
A normal trace follows the street or trail without sudden side jumps, missing blocks, or straight lines across corners. That visual check often reveals bad data faster than the headline numbers do.
The Numbers Worth Trusting Most
Strava is dependable for training trends when the same device records similar activities under similar conditions. Exactness falls as the metric moves farther from a direct clock or sensor reading.
- Trust most: elapsed time, direct heart-rate data, direct power-meter data, and a well-recorded open-sky track.
- Trust with context: distance, average pace, moving time, segment matches, and barometric elevation.
- Treat as estimates: corrected elevation, calories, estimated cycling power, and any result built on a visibly broken GPS line.
For race distance, use the certified course and official timing. For daily workouts, judge Strava by consistency across weeks rather than by whether one activity matches a watch, treadmill, or friend’s file to the last hundredth of a mile.
References & Sources
- Strava Support. “Why Is GPS Data Sometimes Inaccurate?” Explains satellite lock, signal obstruction, warm-up time, and reflected-signal errors.