How GPS Calculates Distance (And Why It Is Not Always Right)

I was standing on a windy overlook arguing with my phone

I was on this cliffside pull-off outside Sedona, the kind with a tiny parking lot and a view that makes you forget you’re dehydrated. My friend swore the trail to the vortex was “like, a mile.” My phone said 0.6. And then the sign at the trailhead said 1.2. So I did what any map-nerd traveler does: I opened the calculator app, started poking at coordinates, and immediately realized I didn’t actually know what my GPS was doing under the hood.

I nodded like I understood. I didn’t.

So if you’re planning a trip, or you’re trying to settle a “no, it’s definitely farther than that” debate, here’s the deal: GPS doesn’t measure distance the way your brain does. It measures distance the way math does, and then it tries to patch that math onto a messy planet with buildings, trees, tunnels, and your phone stuffed in a cupholder.

And yeah… that’s why it’s not always right!

What GPS is actually measuring (and what it’s not)

GPS gives you a position: latitude, longitude, and (sort of) altitude. Then distance is computed between two positions. That’s it. It’s not reading the path you walked unless you’re using a mapping app that’s snapping your movement to roads or trails. Raw GPS distance is basically “as the crow flies,” except the crow is flying over a squishy, slightly lumpy sphere and your phone is guessing where you are within a fuzzy circle of uncertainty.

So when you see “0.6 miles away,” that’s usually straight-line distance from your current coordinate to the destination coordinate. Not the switchbacks. Not the “oops we took the wrong fork.” Not the part where you wandered 200 feet to take a photo of a lizard (I’m not judging, I’ve done worse).

And then there’s the other kind of distance you see: route distance. That’s not GPS measuring distance; that’s a map algorithm measuring the length of a route on a road/trail network. Different thing.

Here’s a quick cheat sheet, because this is where people talk past each other:

So why does everyone get this wrong? Because we use the same word—distance—for like four different measurements.

The math: how two GPS points become a distance

I used to think GPS distance was some kind of “satellite tape measure.” It’s not. Your phone figures out where it is, then it uses a formula to compute the shortest path over the Earth’s surface between point A and point B. That shortest surface path is called a great-circle distance.

Most apps use a version of the haversine formula because it behaves nicely for small distances and doesn’t freak out near the poles. You don’t need to memorize it, but seeing it once makes the whole thing feel less like magic and more like… okay, fine, it’s trigonometry.

So you take two lat/long pairs, convert degrees to radians, run the formula, and out pops a distance along the Earth’s surface. Not your driving route. Not your hiking trail. The shortest surface arc.

If you want to play with this without spreadsheet pain, I built a few quick calculators that do the conversions and the math the way I wish travel apps explained it:

• GPS distance calculator (straight-line between two coordinates)
• Haversine distance if you want the classic formula specifically
• Lat/long to distance for quick “how far apart are these points?” checks
• Coordinates to miles when you’re thinking in miles and don’t want extra steps
• km to miles converter because half the world writes trip distances in km
• miles to km converter for the other half of your group chat

And just to make it concrete, here’s a worked example with rounded numbers (because nobody wants 12 decimal places while they’re packing).

Example: You’re in Paris near the Louvre (48.8606, 2.3376) and you want to know the straight-line distance to the Eiffel Tower (48.8584, 2.2945). The latitude difference is tiny, the longitude difference is small, and the result comes out in the ballpark of a few kilometers. If your phone says something like 3 to 4 km straight-line, that’s totally believable. If it says 12 km, something’s off (wrong pin, bad fix, or you picked the wrong “Eiffel Tower” in a list, which happens more than you’d think).

So far so good. But here’s where it gets weird.

Why GPS distance is “wrong” (even when the math is right)

But the math can be perfect and you can still get a distance that feels wrong, because the inputs—your coordinates—aren’t perfect. And also because you might be comparing different kinds of distance without realizing it.

So let me run through the big culprits I see when people argue about this on trips.

1) Your position is a guess inside a bubble.
Your phone usually shows a little blue dot, and sometimes a faint circle around it. That circle is basically “I’m pretty sure you’re somewhere in here.” If that circle is wide—say you’re in a city canyon between tall buildings—your coordinate can jump around. And if you’re recording a track (a hike, a run), those jumps add up like tiny fake zigzags. You didn’t actually walk that extra distance, but your track log thinks you did.

2) Straight-line vs route distance (the classic mix-up).
I mean, this is the Sedona argument in one sentence. Your friend is thinking “trail distance.” Your phone’s quick glance number is thinking “straight-line.” The sign at the trailhead is thinking “official trail length,” which might include the recommended loop and not the shortcut people actually take. Three distances, one argument.

3) Sampling rate: how often your phone takes a point.
If your tracker logs a point every second, it captures curves better but also captures more jitter. If it logs every 30 seconds, it smooths jitter but cuts corners on switchbacks (literally). So you can undercount or overcount depending on settings and signal. And you’ll see two people do the same hike and end up with distances that differ by, say, 0.2 to 0.6 miles, and both swear they’re right.

4) Bad altitude (and ignoring altitude).
Most consumer GPS distance calculations are 2D: they assume you’re moving on a flat surface. If you’re hiking steep terrain, the true path length is a little longer than the 2D map distance. Not usually a massive difference on normal trails, but it’s real. And altitude from GPS can be noisy, so some apps avoid using it for distance because it can make things worse.

5) Multipath and signal blockage.
This is the “downtown problem.” Signals bounce off buildings, your phone hears a delayed version, and your computed position shifts. Forest canopy can do a similar thing, just more subtle. If you’ve ever watched your dot drift across a river you’re definitely not swimming in, that’s this.

6) Map data isn’t the territory.
Route distance depends on the map’s geometry. If a trail is drawn with too few points, it becomes a simplified line and the measured length is shorter than the real winding path. Or the trail is mapped wrong entirely. I’ve seen a “trail” that was actually a maintenance road, and the route distance looked clean and efficient… because it wasn’t the route.

And here’s the one that always makes me laugh: sometimes people compare a car odometer to GPS distance and assume one must be lying. They’re just measuring different things with different error sources.

So yeah, GPS isn’t “bad.” It’s just honest about a world that’s messy.

How to get a distance you can actually trust on a trip

So what do you do with all this when you’re trying to plan tomorrow’s drive or decide if you’ve got time for “one more viewpoint” before sunset?

I use a little checklist, and it’s boring, but it saves arguments.

• Decide which distance you mean. Straight-line, route, or recorded track. Say it out loud. It sounds silly, but it fixes 80 percent of disagreements.
• For trip planning, trust route distance, not the blue-dot number. Use the directions route length. That’s what your car will experience (more or less).
• For hikes, trust the official trail length… but sanity check it. Trail signs can be for a loop, an out-and-back, or a “to the viewpoint” segment. If it’s unclear, compare a couple sources.
• For track logs, smooth the noise. If your app has a “snap to trail/road” option, try it. If it has a filtering setting, use it. Otherwise you’ll get the excessiveness of jitter distance.
• Check your accuracy circle before you declare victory. If it’s huge, your distance will be wobbly. Step into a clearing, wait 20 seconds, and you’ll often see it tighten up.

And if you’re doing coordinate math (maybe you’re comparing two campsites, or you’re trying to figure out how far that island is from shore), use a straight-line calculator and keep your expectations in check: it’s a baseline, not a promise.

But honestly, once you know which distance you’re looking at, GPS feels a lot less “wrong” and a lot more “oh, that’s what you meant.”

FAQ