The elevation profile tells you more than the map ever will

The first thing most people do when they open a route configurator is turn the map off.

That surprised us, because the map underlay was the part we assumed was the point. Roads, coastline, a bit of context — the things that tell you where you were. We built it first.

Then you sit with it for a while and it stops being surprising. You already know where you went. You were there. The map is telling you something you can recall without help.

What you cannot hold in your head is the shape of the effort.

What a profile actually shows

An elevation profile is a strange kind of drawing. It has no geography in it — a profile of a climb in the Alps and a profile of a climb in Wales can be the same picture. Strip away the place and you are left with the only thing that varies: how hard it was, and when.

That turns out to be the memorable part. The fifteen minutes at the bottom where it looked flat and was not. The descent that arrived twenty kilometres later than you swore it would. The last kick that nobody warns you about because it is only four hundred metres long.

A map cannot show any of that. It shows a line that is equally thin at the top of a col and on a cycle path beside a canal.

Where it does not work

We are not going to pretend the profile always wins.

A flat route has a flat profile, and a flat profile is a horizontal line. If you rode two hundred kilometres across Denmark, the honest chart is a nearly straight stroke with some texture in it. That can be a good print — it is a true picture of what the day was — but it is not a dramatic one, and anyone selling you drama there is drawing something that did not happen.

The routes where the profile earns its place have shape to them: a climb, a traverse, a day with two or three separate efforts stacked into it. If your ride was flat, the track is the better motif, and we will say so.

Why we do not smooth it into a mountain

Elevation data from a GPS is noisy. Not slightly — genuinely, badly noisy, in a way that surprises people who have never looked at the raw numbers. A device sitting still on a table will report a few metres of drift.

That noise is easy to turn into scenery. Take the raw series, plot every point, and a flat Tuesday looks like an alpine stage: hundreds of tiny spikes stacked into something jagged and impressive.

We smooth the curve before we draw it, deliberately, so the shape you see is the shape of the route rather than the shape of the sensor. The number printed next to it does not come from that curve at all — it comes from the activity's own recorded total, which is calculated against a terrain model rather than a barometer.

The distinction matters. A profile is a drawing. A number is a claim. They should not come from the same place, and only one of them is allowed to be approximate.

What this means for the print

If your effort had climbing in it, put the profile on the sheet. It is the part you will point at in five years, and it is the part nobody else's print will have in quite that shape.

If it did not, do not force it. The track alone, one line on paper, is a better object than a chart pretending something happened.