GPS and multisport

Running watch GPS accuracy, and what has actually been measured

One good study exists. It is six years old, it tested hardware two generations back, and it is still more than anyone else in this category will show you.

By Stephen V.Published July 24, 2026

#ad

We earn a commission when you buy through our Amazon links, at no extra cost to you. It never decides a ranking — the rubric that does is published in full, so you can check us rather than take our word for it. How this works.

The short answer

The best peer-reviewed evidence available found distance errors of 3.2% to 6.1% across eight sport watches measured against a trundle wheel, with only Polar's receivers under 5% overall. On a 400 m track errors fell to 0.9-4.1%. That study is from 2020 and predates dual-frequency receivers, so it describes engineering history rather than any current watch.

The study

Gilgen-Ammann, Schweizer and Wyss published an instrument validation study in JMIR mHealth and uHealth measuring eight positioning-enabled sport watches against a trundle wheel accurate to one centimeter. The watches were an Apple Watch Series 4, a Coros Apex 46 mm, a Garmin fēnix 5X Plus, a Garmin Forerunner 935, a Polar Vantage M, a Polar Vantage V, a Polar V800 and a Suunto 9 Baro. Three environments: an urban city center, uphill and downhill forest, and a 400 m stadium track.

What it found

Overall mean absolute percentage errors from 3.2% to 6.1%.
Only Polar's receivers stayed under 5% error overall.
On the track, errors collapsed to 0.9-4.1%. Urban and forest conditions were markedly worse, with most devices underestimating distance.

That last line is the practically useful one. Your watch is far more accurate on an open track than it is on a wooded trail or between tall buildings — and if you have ever compared a watch against a certified course and been annoyed, environment is the most likely explanation.

Why this study cannot tell you which watch to buy today

We want to be firm about this, because a dated study quoted as though it were current is how misleading claims get made with a citation attached.

Every watch in that study was single-frequency. Dual-frequency (multi-band) consumer receivers arrived afterwards and are now standard across the mid-range and up. The specific failure mode the study measured — error in urban and forest environments — is precisely the failure mode multi-band was designed to reduce.

So the honest summary is: this study establishes that distance error is real, that it is environment-dependent, and that Polar had a good receiver engineering record as of 2020. It does not establish that a Polar Vantage M3 is more accurate than a Forerunner 970 in 2026. Nobody has published that comparison.

What multi-band actually does

A satellite transmits on more than one frequency. A single-band receiver listens to one, so a signal that has bounced off a building arrives looking much like one that came straight down — and the receiver includes the reflection in its position fix. That is what produces a track that wanders across a road or cuts a corner.

A dual-frequency receiver hears the same satellite on L1 and L5. The two frequencies are delayed differently by the atmosphere and by reflections, and the disagreement between them is the diagnostic that lets the receiver identify and discard the bad signal.

This is physics rather than marketing, and it is why we treat multi-band as the one specification worth paying for when GPS drift is your actual complaint. It is also checkable on a spec sheet before you spend anything — see the GPS upgrade path.

How to check your own watch, for free

You do not need us for this. Run a 400 m track, in lane one, for exactly four laps — that is 1,600 m, and a standard track is one of the few precisely measured distances most people have access to.

Under 1% off: your watch is working correctly.
1-3% off: normal, consistent with the published literature.
Over 3% off on a track: something is wrong — usually satellite mode set to power-saving, or a poor-quality fix at the start. Wait for the lock before you begin.

Do the same run on your usual wooded or urban route and compare. If the track is fine and the trail is not, the watch is behaving exactly as the study describes and multi-band is the upgrade that addresses it.

Questions

Frequently asked

How accurate is GPS on a running watch?

The best peer-reviewed measurement found 3.2-6.1% distance error across eight watches, dropping to 0.9-4.1% on a 400 m track. Errors are consistently worse in cities and under tree cover, where reflected signals confuse the receiver. That study is from 2020 and tested single-band hardware, so current multi-band watches should do better — though no one has published a comparable study to confirm it.

Is multi-band GPS worth it?

If you run in cities or under tree cover, yes — it directly targets the reflected-signal problem that causes drift. If you run on open roads or a track, single-band is already performing near its best and multi-band buys you very little.

Why does my watch measure a marathon as 26.4 miles?

Two reasons, and neither is necessarily a fault. Positioning error accumulates, as the published study shows. And race courses are measured along the shortest legal line, which almost nobody actually runs — every time you go wide around a corner or move across the road, you cover real extra distance your watch correctly records.

Which brand has the most accurate GPS?

On the only peer-reviewed multi-watch evidence available, Polar — the only brand under 5% error overall. But that was 2020 and tested watches nobody sells now, so treat it as evidence about Polar's receiver engineering rather than a verdict on any current model.

Show your working

Sources

Every one of these was read in full before it was cited. Where we could not access a paper, we do not quote its numbers.

[1]Peer-reviewed validation2020
Gilgen-Ammann, Schweizer & Wyss, “Accuracy of Distance Recordings in Eight Positioning-Enabled Sport Watches: Instrument Validation Study”, JMIR mHealth and uHealth
DOI 10.2196/17118. Eight watches (Apple Watch Series 4, Coros Apex 46mm, Garmin Fenix 5X Plus, Garmin Forerunner 935, Polar Vantage M, Polar Vantage V, Polar V800, Suunto 9 Baro) measured against a trundle wheel accurate to 1 cm, across urban, forest and 400 m track conditions. IMPORTANT: this predates the dual-frequency receivers now standard, so it describes brand engineering history, not any watch currently sold.
[2]Manufacturer specification2026
Garmin — Forerunner 970 published specifications
Manufacturer-stated battery: up to 15 days smartwatch mode, up to 26 hours GPS mode.
[3]Manufacturer specification2026
COROS — PACE 4 published specifications
Manufacturer-stated: 1.2" AMOLED, 32 g with nylon band, 11.8 mm thick, All-Systems and Dual-Frequency satellite modes, 41 GPS hours, 19 days of daily training and sleep.