Methodology
How we calculate
Every figure on this site is an estimate produced by the process below. It is written out in full so you can check it, disagree with it, or reproduce it yourself.
1. Finding the two places
When you enter a start and a destination, each is looked up through the Nominatim geocoding service, which is built on OpenStreetMap data. The service returns a latitude and longitude for the best match.
This is the largest single source of error in the calculator, and it is worth understanding. Geocoders return the coordinate of a place's centroid or its principal entrance, not the point you have in mind. For a large site — an airport, a university, a national park — that point can be a considerable distance from where you would actually start walking. Entering a more specific query ("Terminal 5, Heathrow" rather than "Heathrow") produces a materially better result.
2. Measuring the distance
We calculate the great-circle distance between the two coordinates using the haversine formula, which accounts for the curvature of the Earth. For distances under a few hundred kilometres the difference between this and a flat-plane calculation is negligible, but the formula costs nothing and is correct at any scale.
The Earth is treated as a sphere with a mean radius of 6,371.0088 km. The Earth is not a sphere — it is an oblate spheroid, wider at the equator by about 43 km — and a more precise calculation would use Vincenty's formulae on an ellipsoid. For walking distances the difference is well under a tenth of a per cent, which is invisible next to the other approximations here.
3. Adjusting for streets
Nobody walks in a straight line. Buildings, rivers, one-way pedestrian routes and street grids all mean the walked distance exceeds the straight-line distance. We multiply the great-circle distance by a street factor of 1.28.
That figure is a median for urban walking. Research on route directness — the ratio of network distance to straight-line distance, sometimes called the circuity or detour ratio — has generally found values between about 1.15 and 1.45 for pedestrian networks, varying substantially by city form. A regular grid such as Manhattan or Barcelona sits lower; a medieval core such as Rome or Prague sits higher; a place cut by water or a motorway can be far higher still.
This is a genuine limitation. On any specific route the true figure may be well above or below 1.28, which is why the individual route pages on this site quote researched distances for that particular walk rather than the calculator's estimate.
4. Converting distance to steps
Distance is divided by your step length. If you select a height, we estimate step length as 0.414 × height, which is the standard coefficient used for adult walking gait. With no height selected we use 0.75 metres, an adult average.
| Height | Step length | Steps per km | Steps per mile |
|---|---|---|---|
| 1.55 m | 0.64 m | 1,563 | 2,515 |
| 1.65 m | 0.68 m | 1,471 | 2,367 |
| 1.75 m | 0.72 m | 1,389 | 2,235 |
| 1.85 m | 0.77 m | 1,299 | 2,090 |
| 1.95 m | 0.81 m | 1,235 | 1,987 |
Height is a proxy for step length, not a substitute. Individual variation at any given height is considerable, and step length also changes with pace, gradient, surface and load. If you want an accurate figure, measure it — our guide to measuring your stride length takes about five minutes and beats any estimate.
5. Time and energy
Walking time assumes a cadence of 115 steps per minute, which corresponds to a comfortable adult walking pace of roughly 5 km/h. It does not account for road crossings, traffic signals, gradient, or stopping to look at anything.
The energy figure uses approximately 0.045 kcal per step for a 70 kg adult on level ground. Energy cost scales roughly with body mass, so a heavier walker burns proportionally more and a lighter one less. It rises sharply on gradients and on soft ground, and the estimate does not attempt to model either.
What this calculation does not do
- It does not route. We do not query a pedestrian routing engine, so we cannot know about bridges, closures, pedestrian-only zones, or whether a walkable path exists at all.
- It does not know about water. Enter two points on opposite sides of an ocean and it will return a step count for a walk you cannot take.
- It does not account for elevation. A route with 300 metres of climbing returns the same figure as a flat one of the same length.
- It is not a navigation tool. Use a mapping application to actually get somewhere.
Accuracy in one sentence
For an ordinary walk between two well-specified points in a city, expect the estimate to land within roughly 10 to 20 per cent of the real walked distance; for anything involving water, mountains, or a vaguely named location, expect considerably worse.
The route pages are different
The twenty route pages on this site do not use the calculator's estimate. Their distances are researched for each specific walk using published route information and mapping data, and where terrain materially affects the step count — sand at Giza, cobblestones in Rome, stairs in Sydney — the page says so and adjusts. They are more reliable than the general calculator, because they only have to be right about one route.
Corrections
If you have walked one of these routes and measured something different, we want to know. Details are on the contact page, and our approach to corrections is set out in the editorial policy.