Repair a broken activity file

Your file never leaves your device — conversion happens here in your browser.

Not every bad stretch of GPS data is a single wild point you can point to and delete. Sometimes a device records a run of consecutive points — two, five, a dozen — where the whole segment, not just one point, implies a speed that's physically impossible for the activity, whether that's a cyclist apparently hitting 300 km/h for ten seconds or a runner's pace swinging wildly point to point through a patch of the file. Unlike an isolated spike, there's no clean single point where the track is fine on both sides of it; the bad readings run together, so no single deletion repairs the segment. This tends to follow a stretch of poor signal rather than a momentary glitch — dense tree cover, a long tunnel, an urban canyon between tall buildings, or a device that struggled to reacquire a lock for several seconds and produced a scatter of low-confidence points while it did. FitFileHub's repair tool evaluates implied speed across runs of consecutive points, not just point pairs, so it can tell a sustained bad segment apart from a single outlier. When it finds one, it removes the points needed to eliminate the impossible segment, rejoining the track cleanly on either side. It reports the segment it found — how many points, roughly where in the file — before making any change, the same as every other check it runs. If what you're looking at is one obvious jump-and-return rather than a scattered run, that's the narrower GPS-spike defect, with its own page and a more surgical single-point fix. Everything runs locally in your browser; your file is never uploaded. Free up to 25 MB per file, three files per batch.

Questions

What's the difference between this and a GPS spike?

A spike is a single isolated point with clean track on both sides. Implausible speed is a run of two or more consecutive points where the whole segment is bad, with no single point you can delete to fix it — a distinct defect with a different fix.

What usually causes a sustained bad-speed segment?

Typically a stretch of poor satellite signal — dense tree cover, a long tunnel, or an urban canyon — where the device produces several low-confidence points in a row while it struggles to reacquire a solid lock.

Does fixing this remove more than just the bad points?

No — the tool removes only the points needed to eliminate the impossible segment, then rejoins the track cleanly on either side. It reports what it found before making any change.

Does this check apply to FIT and TCX files, or just GPX?

It applies to any supported format with position and time data, since implied speed is calculated from those two values regardless of file format.