Compare two very different patterns
A sequence that advances steadily along a journey is different from one point appearing far away and the next returning near the starting position. The straight-line speed between two coordinates can be large in both cases, so that number alone cannot decide which is real.
Sampling intervals matter too. A long interval can contain real travel, but it can also contain an outage. Treating every far-apart pair as continuous movement would fill gaps with imaginary paths.
Use uncertainty as part of the evidence
Apple supplies a horizontal-accuracy value with location observations. A broad uncertainty region should not be treated as a precise doorstep. Timestamps and speed evidence can add context, while subsequent observations help distinguish a temporary jump from a genuine relocation.
Nowhop can hold suspicious observations in bounded temporary state before letting them affect the current reveal, city or persistent exploration. This is a cautious interpretation step, not a claim that GPS data can always prove where someone was.
Do not solve appearance by rewriting history
Snapping every point to the nearest road could make a line attractive while putting a riverside walk on a road above it. Replacing a tunnel with directions would make the route complete while inventing evidence. Neither is a reliable repair simply because the final map looks cleaner.
The current versioned policy preserves old records and their interpretation. Improvements apply to new observations without silently changing the distances and explored areas users already saved under earlier rules.
Report a reproducible pattern, not a perfect screenshot
When checking a problem, note whether you were stationary or moving, the approximate interval, the transport type and whether the app was backgrounded. Compare a short familiar outdoor journey with an indoor stay. Those observations help separate drift, continuity and permission issues.
Apple’s accuracy definition explains why a coordinate has uncertainty. Apple’s location-efficiency guidance also shows why requesting maximum activity all the time is not a free solution. Nowhop’s filtering tests remain distinct from real-device reliability and battery measurements.
Questions this guide answers
Why not accept everything after removing the speed limit?
Because a bad fix can reveal ground or add distance you never traveled. Removing a simplistic cutoff should improve interpretation, not discard uncertainty checks or turn every jump into a visit.
Why not reject every long interval?
Real journeys can continue between less frequent observations. The policy considers context rather than one elapsed-time number, while still leaving genuinely missing intervals unconnected.
Can the app guarantee that only GPS errors are removed?
No such guarantee is justified. Filtering balances incomplete evidence. The honest goal is better handling of coherent movement and suspect points, with preserved originals and clear limits.
Can a delayed fix move the current city immediately?
Suspicious or stale observations should not immediately move the current reveal or create persistent exploration. Eligibility and coherent follow-up evidence matter before accepting them into the new recording stream.
Can GPS alone identify my exact train line or driving lane?
No such classification is implemented or guaranteed. A plausible recorded journey is not proof of a particular lane or train service, and a known transport network does not fill unobserved underground movement.
Sources & scope
Official pages reviewed for this guide on October 31, 2026. Features and availability can change; check the linked provider before choosing.
Nowhop behavior is based on our implementation specification. Public demos use samples. Read our editorial approach.
See how a day comes together.
Try the Nowhop preview ↗Sample data. No sign-up or location access.