Why does a GPS track show you walking through a lake or a building?
Because the line and the lake are drawn by two different systems that never actually check each other's work. Your track is a string of coordinates the GPS receiver logged as you walked. The lake, the building outline, the treeline â all of that is separate map artwork, drawn from a completely different dataset, and simply placed underneath your track when the page renders. A GPS fix that lands a few metres off to one side is ordinary, near-invisible error almost everywhere on the map. It only becomes something you notice when those few metres happen to land on the far side of a hard line like a shoreline or a wall.
The short answer
Every GPS position carries a small margin of error, typically a handful of metres, that is present on every kind of ground and shows up on every kind of walk. Most of the time that margin is invisible, because open grass, a wide trail, or a car park does not have a line on the map for a stray point to cross. Water and buildings are different: they are drawn as sharp, filled shapes with a clear edge. The same few metres of ordinary drift that would go completely unnoticed in a field becomes an obvious problem the moment it crosses a shoreline or a wall, because now there is a hard boundary on the map for it to be visibly wrong about.
Two systems, laid on top of each other
A recorded track is nothing more than a list of coordinates and the altitude and time that go with each one. It has no idea what a lake is, what a building is, or where either one sits â it is just points. The map underneath it, meanwhile, comes from an entirely separate source: satellite imagery or surveyed map data, built and maintained independently of anything a phone's GPS receiver ever measured. When a route is displayed, the app draws your track as a line and drops it on top of that map. The two never talk to each other, which is exactly why the track cannot notice it has crossed a boundary the map defines. It was never told the boundary was there.
Why the effect concentrates near buildings specifically
Open water tends to produce a clean, if occasionally crooked, line along a shore â ordinary sideways drift, nothing more. Built-up areas are noisier for a real reason: a GPS signal can bounce off a wall or a window before reaching the receiver, arriving a fraction of a second later than a signal that came straight from the satellite. That delayed, reflected signal gets read as if it travelled in a straight line, which places the calculated position slightly off from where you actually stood â sometimes far enough to land inside the footprint of the building that caused the reflection in the first place. This effect is strongest exactly where tall, hard surfaces sit close to the path, which is also exactly where a map is most likely to have a building outline waiting to be walked through.
What this does â and does not â mean for the numbers
A line that clips a corner of a lake or cuts through a rooftop is a rendering artefact, not a sign that the recording failed. The same handful of metres of error that produced it is already baked into every GPS track, everywhere, all the time â it is just usually landing somewhere with no line to cross. It does not meaningfully change distance, pace, or elevation gain for a walk of any real length, for the same reason a few metres of drift never does: the error is small, roughly consistent in size throughout, and swamped by however far you actually walked. It is worth a glance and nothing more.
Things that trip people up
- Assuming the app lost the GPS signal entirely because the line briefly crosses water â losing signal produces a very different pattern, usually a long straight jump or a gap, not a small dip across a shoreline.
- Trying to manually correct or retrace a route to avoid a visual glitch like this â the underlying position data is fine; only the pixel where it happens to sit relative to a map shape looks off.
- Assuming this means GPS accuracy is generally worse near cities or buildings than it is elsewhere â it can be, for the reflection reasons above, but the visible "walked through a wall" moment is usually a single stray point, not a sign the whole stretch of track is unreliable.
- Expecting a different app or a different phone to draw the same walk without ever touching a lake or building outline â every GPS receiver carries the same kind of ordinary position error, so the same walk recorded twice can clip a different shape, or the same shape, on either recording.
Where TrailCam fits in
TrailCam draws your recorded route as a line on the map exactly as the GPS receiver logged it, the same way any track is drawn, with your photos, video clips and voice notes pinned to the point on that line where you captured them. TrailCam is free to download.
Questions
- Does a track crossing a lake or building mean the GPS signal was bad?
- Not necessarily. It usually just means one or two points landed a few metres off from where you actually stood, which is ordinary GPS error. A genuinely lost signal tends to look different â a long straight jump or a gap in the track, rather than a small dip across a boundary.
- Why does this happen near buildings more than in open fields?
- Signals can bounce off walls and windows before reaching the phone, which shifts the calculated position slightly. A field has no hard edge on the map for that shift to visibly cross, so the same amount of error simply goes unnoticed there.
- Should I try to edit or fix a track that clips through a lake or building?
- There is usually no need. The distance, pace and elevation numbers for the walk are barely affected by a handful of metres of drift at one point â it is a cosmetic wrinkle in how the line sits on the map, not a flaw in the recording itself.
Recording your own routes
TrailCam is an iPhone app that records your route with GPS, keeps video clips, photos and voice notes pinned to the spot where you took them, and exports the result as GPX 1.1 or CSV. It is free to download.