I.The average of an outing says very little
You finish a 40-km outing: 4h 12min, 850 m of elevation gain, 9.5 km/h average. Three numbers to summarise three hours of relief, surges, moments of heavy legs, fast descents, small climbs that hurt. Statistically, that's correct. As a sporting reading, it says nothing — or rather, it says exactly the same thing as a completely different outing.
Picture two outings that finish with the same three figures. The first is flat for thirty kilometres, then climbs 850 m all at once over the last ten. The second alternates rolling false flats for four hours straight. Same totals, two radically different sessions — one trains endurance then climbing legs, the other smooths the effort over duration. The global average blurs them together.
What you often actually want is to read a portion of the outing. The big climb in the middle. The sprint over the last two kilometres. The technical descent that deserves analysis on its own. And there, the global average is useless: you have to isolate.
II.The traditional answer: the Strava segment
Strava popularised a brilliant idea in 2009: let a user draw a piece of route on the map, give it a name (« Col du Soulor », « Town Square Sprint »), and offer every other Strava user who later passes through it an automatic ranking on that exact stretch. The segment was born. With it came the KOM, the QOM, the leaderboards, the social animation that turned the app into a platform.
For social comparison, it's brilliant. Want to know if you climbed that col faster than last time, or faster than your neighbour? The segment tells you, to the second. That's exactly what Strava does best, and nobody has really challenged it on that ground (our detailed comparison Strava / Garmin / GPXchunk digs into how these roles divide up).
III.The hidden friction of segments
But this system carries a cost that becomes visible the moment you just want to look at your own numbers without sharing anything.
- You need an account. Strava is a social product. Every analysis goes through your profile, your follow list, your activity feed. If you just want to read a track without posting, liking, following, the experience isn't designed for that.
- The segment has to have been drawn. Either by you (manual creation, validation), or by another user who passed through before you. On a quiet trail, on an unlikely portion, on a 200-metre pitch nobody has ever named: no segment, no precise figures.
- The portion is public. Once created, your segment appears in the global leaderboard. If you create it on a discreet path near home, you share it with every Strava user in the area. There's no « private segment for me alone ».
- The bounds are frozen. Want to analyse fifty metres more or less? You have to redraw. Want to test « is the hard part in the first or the second third? » That's three segments to create, three rounds of friction.
For the runner, cyclist or hiker who just wants to read their own track privately, without committing the portion to a public system, the segment is an over-priced answer to the wrong question.
IV.Free isolation: dropping two handles
The other approach fits in one sentence: drop two markers on the track, anywhere, and read the figures between them. No name to give, no validation, no leaderboard. Just two movable bounds on your own file, and every measure recomputed in real time for whatever sits between them.
This is the approach GPXchunk uses. You load your .gpx file in
the browser (nothing is sent to a server), you see the track on the map and
the elevation profile, and you drop two handles — one start, one end.
Everything between them is your portion. Distance, time, elevation, speed,
pace, elevations, net gradient, climbing speed, heart rate: the twelve
measures are recomputed instantly for that portion alone (for what each one
really says, see our guide to the twelve
measures).
Want to try fifty metres more? Slide the handle. Want to analyse the first third, then the last? Reposition the handles, read, reposition again. The portion is never frozen — it's as mobile as the question you ask of it.
V.Three ways to drop the bounds
Depending on what you're looking for, you don't drop the handles in the same place or the same way. GPXchunk offers three, and they complement each other.
On the map. You spot a place visually — a crossroads, a summit, a notable bend — and you click. Useful when you know the terrain and the bound is a precise geographical landmark. This is the intuitive way to analyse « the climb up the col » or « the descent on the way back ».
On the elevation profile. You see the elevation curve at the bottom of the screen, and you spot the climb visually — the moment the curve rises, the summit, the peak, the descent. You click on the profile, not on the map. This is more precise for isolating an effort by its elevation shape: a 300 m climb, a precise descent, a steep pitch. Often, the portion you care about is more visible on the profile than on the map.
To the kilometre. You type a distance — for example « from km 18.5 to km 22.0 ». Useful when comparing with a friend who said « I cracked at km 19 », or when analysing a segment of an exact length (« my last 5 km »). Surgical compared to the other two.
All three are synchronised: clicking on the map updates the profile and the kilometres, and the other way round. You work your portion in whichever view speaks to you, and the others follow.
VI.What you see that you didn't see before
The most striking effect of free isolation is the amount of information that appears — without anyone having to ask for it explicitly. Three concrete examples.
A precise climb. You isolate the ascent of a col within a four-hour outing. Up come, immediately: the exact distance (4.2 km), the time (28 min), the climbing speed (920 m/h), the net gradient (7.3 %), the average heart rate (162 bpm). Five numbers that tell you exactly how you climbed that particular col, independently of the rest of the outing. Compare with the same climb a month from now — you'll know whether you're improving.
A final sprint. You isolate the last 800 metres of a race. Average speed 18.5 km/h, pace 3'15 per km, max heart rate 184 bpm. You see your real closing kick, instead of having it diluted in the average of the previous hour and a half.
A technical descent. You isolate the descent off a col. Distance 6.5 km, time 14 min, negative gain 480 m, average speed 27 km/h. The isolated negative gain tells you how far you came down, the average speed tells the rhythm. On two consecutive descents, you can see which one was faster — or more cautious.
None of these numbers shows up in the global summary. All of them are there the moment you isolate.
VII.When to use which
Strava segments and free isolation aren't competing — they answer two different questions.
If you want to compare yourself — to yourself over time, to other runners, to a leaderboard —, the Strava segment is the tool. Public, permanent, social: that's exactly what it was built for.
If you want to understand yourself — read a precise portion of your own track, try several cuts, analyse an effort without committing it to a public system —, free isolation is more direct. No account, no sharing, no friction.
A Strava segment is public; an isolated portion is private. One is for comparing, the other for understanding.
The two tools don't conflict: a serious runner can perfectly well export their track from Strava, open it in GPXchunk to study a climb in detail, then return to Strava to post the ride and check the leaderboard. Each tool doing what it does best.