Why a climb needs its profile

On most routes the map is the story. On a climb it is not.

A big mountain road drawn from above is a set of hairpins and a line going roughly one direction. It is recognisable if you have ridden it and nearly meaningless if you have not. Everything that made it hard — the two kilometres at eleven percent before the trees end, the false flat that arrives too late to help — is invisible in plan view.

That is what the elevation profile is for.

The profile is the honest half

Put the two together and the print starts arguing with itself in a useful way. The map says *this is where it happened*. The profile says *this is what it cost*. Anyone who has ridden it will look at the profile first.

It also fixes the recognition problem. Alpine hairpins all look broadly alike from above. Nobody confuses one profile with another.

What we do to the numbers before drawing them

Barometric altimeters drift. On a long day the pressure changes, and the recorded elevation wanders with it — one reference ride in our own files fell a hundred metres over twelve hours while finishing exactly where it started.

Drawn raw, that drift tilts the whole landscape. So the trend is removed before the profile is derived: a straight line fitted through the series and subtracted, then a light smoothing, then rebased to zero. A straight line cannot follow a hill, which is precisely why it is safe to subtract one.

The climbing figure beside the profile is a different question, and a harder one. Total ascent depends entirely on how small a rise you are willing to count — the same file gives 1 500 m if you count everything and 460 m if you only count sustained climbs. There is no neutral answer, so the number on your print comes from your activity rather than from our arithmetic.

When to leave the profile off

If the route was flat, leave it off. A horizontal line under a map is not restraint, it is a wasted third of the sheet. We check the file and say so before it prints.