Hill Grade Adjusted Pace (GAP) Calculator

Running up a 10% gradient takes substantially more energy per kilometre than running on flat ground, yet your GPS watch still shows your speed in minutes per kilometre. That number tells you how fast you are moving, not how hard you are working. Grade adjusted pace closes the gap: it converts your pace on a slope into the equivalent pace you would have run on flat ground at the same level of physiological effort.

The result is a single, comparable number. A 6:30/km on a steep climb and a 4:55/km on a flat road might represent identical aerobic effort - GAP makes that visible. This is especially useful during trail runs, hilly road workouts, or any session where constant changes in gradient make raw pace a poor guide to intensity.

GAP is not a correction for actual speed. It does not tell you how fast you would have run the same route on flat terrain - it tells you the flat-pace equivalent of the effort you produced. The distinction matters when you are trying to set heart rate or perceived exertion targets that transfer across hilly and flat running.

Strava displays a GAP figure on any activity that includes elevation data. The concept is identical to this calculator: it is the flat-pace equivalent of your effort on a given segment or for an entire run. Strava uses GAP to power its relative effort scores, its segment comparisons, and the estimated best-effort times it shows on hilly segments.

There is one important caveat. Strava uses its own proprietary gradient-to-effort model, which is different from the Minetti et al. formula this tool uses. The two models agree closely at moderate gradients (roughly -6% to +8%), but can diverge at steeper angles. If you run the same segment and compare the GAP figures from this calculator and from Strava, small differences are normal and expected - they reflect the difference in the underlying models, not an error in either one.

For training purposes both approaches are useful. Use this calculator when you want to apply the published Minetti model directly, compare efforts across runs, or plan hill workouts with a specific flat-equivalent target pace in mind.

The physiological basis for GAP comes from research by Alberto Minetti and colleagues, who measured the metabolic cost of running at various gradients on a treadmill and published a landmark study in the Journal of Applied Physiology (2002). Their findings showed that the relationship between gradient and energy cost is not linear - it follows a polynomial curve.

Running uphill is expensive. The steeper the climb, the more energy you expend per unit of horizontal distance covered, because you must lift your bodyweight against gravity with each stride. At a 10% gradient the energy cost per kilometre is roughly 87% higher than on flat ground, which is why running hills at flat-road pace is physiologically impossible for any sustained period.

Running downhill is more complicated. Gentle downhills do save energy - your legs do less positive work against gravity. But the saving is not proportional to the cost of the equivalent uphill. Below about -20% gradient the metabolic cost actually starts to rise again, because your muscles must perform substantial eccentric (braking) work to prevent you from accelerating out of control. At very steep downhills, the braking cost can exceed the gravitational saving.

The Minetti model captures this asymmetry. The formula used by this tool - `cost factor = 1 + 6.2g + 25g²`, where g is the decimal gradient - produces a U-shaped cost curve that rises steeply for uphills and flattens gradually for downhills. It is the most widely cited model in the running science literature and forms the basis of several commercial GPS and race-analysis platforms.

The table below shows the expected adjusted pace for a runner whose flat-ground effort corresponds to 5:00/km. The values are computed directly from the Minetti formula used in this calculator. Positive gradients are uphill, negative gradients are downhill.

Notice that a +10% climb adds over four minutes per kilometre to the effort-equivalent pace, while a -10% descent saves under two minutes. The asymmetry is the Minetti curve in action: downhill savings taper off as the braking cost rises, while uphill costs accelerate sharply.

Suppose you are running a 1 km uphill segment at 5% gradient in 6:07 - roughly 367 seconds per kilometre. You want to know what flat-pace effort that represents.

First, calculate the grade factor using the Minetti formula. At g = 0.05: factor = 1 + (6.2 × 0.05) + (25 × 0.05²) = 1 + 0.31 + 0.0625 = 1.3725.

Then divide your actual pace in seconds by the factor: 367 ÷ 1.3725 ≈ 267 seconds per kilometre, which is 4:27/km. So running that 5% climb in 6:07/km represents the same physiological effort as running 4:27/km on flat ground.

You can check this against the table above: the +4% row shows a flat-equivalent effort pace of 6:26/km, and the +6% row shows 7:19/km. A 5% gradient sits between those two rows, so a GAP result in the 4:20-4:30/km range for a 6:07/km actual pace is consistent with the table.

The same logic runs in reverse. If you want to run a hill session at a flat-equivalent effort of 5:00/km and the climb is 8%, multiply 300 seconds by the +8% factor (1.656) to get the target pace on the hill: 497 seconds per kilometre, or 8:17/km. That is the pace you should target on the climb to match your flat-ground effort.

The Minetti model was derived from treadmill running at moderate gradients. It is a strong predictor of energy cost across the range most road and trail runners encounter, but it has limits worth knowing.

At very steep gradients - above roughly +20% or below -20% - the model becomes less accurate. On extreme uphills, most runners shift to walking or power-hiking, which has a different energy-cost profile than running. The formula assumes running gait throughout; if you are power-hiking a 25% climb at a low speed, the GAP output will overestimate the equivalent flat running pace.

Technical terrain adds another layer of uncertainty. Running on loose gravel, roots, rocks, or wet grass requires additional muscular effort for stability and foot placement that does not show up in the gradient figure at all. A rocky 6% trail and a smooth 6% tarmac road will have different actual energy costs even though the calculator would assign them the same GAP.

Fatigue changes the picture as the run progresses. The mechanical efficiency of your stride declines as you tire, meaning the cost of a given gradient rises over the course of a long effort. Early-run GAP figures from this calculator will be more accurate than late-run ones, when cumulative fatigue means your actual effort exceeds the model's prediction.

Use GAP as a guide and a trend indicator, not as a precise measurement. It is most reliable for comparing similar efforts on similar terrain - for example, tracking whether your GAP on your regular hilly training loop is improving over a training block - rather than as an absolute physiological measurement.