The science library Running & gradient

Hills, grade-adjusted pace and running economy

A slope changes what a metre of running costs. What grade-adjusted pace corrects for, and why economy moves your pace independently of VO2max.

4 sources cited Jump to references ↓

Why raw pace lies on hilly ground

Pace records how fast you covered ground. It says nothing about what that ground demanded, and the energy cost of a metre of running changes sharply with the slope, so one steady effort produces splits that look like three different runners ¹.

Running speed is set by two things: the metabolic power you can sustain and the energy each metre costs. Change the second term and pace moves with no change in fitness at all ².

Where terrain makes pace uninterpretable, an honest effort rating for the session multiplied by its duration still gives a usable measure of what the run cost you ³.

The shape of the cost curve

Minetti and colleagues measured the metabolic cost of running across gradients from steep downhill to steep uphill, and the resulting curve is the basis for every grade adjustment in use. Level running costs roughly 3.4 joules per kilogram per metre travelled. Put the same runner on a 10% climb and the cost per metre is about double that; at 20% it is more than triple, and it keeps rising close to linearly on steeper ground ¹.

Downhill running gets cheaper, but only to a point. Cost per metre falls as the slope tips negative and bottoms out somewhere near −20%, where gravity does useful work for you. Past that the braking demand dominates and the cost per metre climbs again, so a very steep descent is expensive rather than free ¹.

What grade-adjusted pace computes

Grade-adjusted pace takes the measured cost of running at the gradient you were on and asks what flat pace would cost the same ¹. A 5:30/km climb whose energy cost matches 4:20/km on the flat is reported as 4:20. The adjustment works because, on level ground, the energy needed to cover a metre is close to independent of how fast you cover it ².

It puts a hilly session and a track session on one axis, so a hard effort is not scored as a bad day because of where you ran it.
It makes week-to-week trends readable when routes are never the same twice.
It works from gradient and speed alone, which is also the source of most of its limits.

Economy is a separate lever from VO2max

Running economy is the oxygen cost of holding a given speed, and it varies by as much as 30% between trained distance runners with similar VO2max values. Within a group that is closely matched aerobically, economy predicts performance better than VO2max does ⁴.

The training that moves economy is not the same training that moves maximal oxygen uptake. Economy responds to accumulated running, to strength and plyometric work, and to biomechanical and neuromuscular factors, so it can improve while VO2max sits still ⁴. Since economy changes the pace you get for a given metabolic power, a gain shows up as pace drifting faster at unchanged effort rather than as a new peak number ².

What GAP will not tell you

The adjustment is a comparison tool, not a claim about experience. It models the energetics of the slope and nothing else: not the muscular damage of a long descent, not broken rhythm on technical footing, not heat, not what the last climb did to the next one ¹.

It is also calibrated on steady running at a fixed gradient, which is exactly what rough terrain is not. Short pitches, switchbacks and noisy GPS grade all degrade it, and on very steep climbs runners walk, which follows a different cost curve that the adjustment cannot see you switch to ¹. On days like that the effort score you logged is the more honest record ³.

References

Numbered in order of first appearance. Every link goes to the original work.