Measure Distance on a Map

What is being measured

Each leg of your path is measured as a great-circle distance: the shortest route across the surface of a sphere, which is the arc you would trace if you pulled a string taut over a globe between the two points. It is not the length of the line as drawn on your screen — the map is flat and the Earth is not, so a straight screen line and the true shortest path are different things, and the difference grows with distance and latitude.

For short legs the two are indistinguishable. Over a thousand kilometres they are visibly not, which is why long-haul flight paths arc north on a flat map instead of running straight across it. The number here is always the arc, never the drawn line.

How much the flat map lies

The straight line you could draw across a Mercator map has a proper name — a rhumb line, the course you would steer at one constant compass bearing. It is a genuinely useful thing for a sailor, and it is reliably longer than the shortest path.

Here is the size of the gap on real routes:

The pattern is latitude. Nairobi to Singapore runs along the equator, where the flat map tells the truth and the two figures are identical. The North Atlantic crossings pay a few percent. Anchorage to Oslo — both near 60°N, nearly half a world apart in longitude — pays 36%, which is 2,335 km of entirely imaginary distance. That is the route no flat map can draw sensibly, and the reason polar flights look absurd on the seat-back screen.

Under a few hundred kilometres the distinction stops mattering: Ljubljana to Vienna is 278 km by either method. If everything you measure is local, you can forget this section entirely. It is when people measure continent-scale distances on a flat map that they get answers which are confidently wrong.

Turning it into a real-world distance

A straight-line measurement is a floor. Any actual route is longer, and transport planners describe how much longer with a detour (or circuity) factor — the ratio of route distance to straight-line distance. Multiplying by a rough factor gets you surprisingly close without opening a routing app:

These are rules of thumb, not published statistics — useful for sanity-checking an estimate, not for quoting. A 4 km straight-line measurement across a suburb is realistically a 6 km drive; across open country it is closer to 4.5 km.

Multi-point paths, and how many clicks to use

Adding points gives you the sum of consecutive straight legs, not a smooth curve. That has a practical consequence: the more points you place along something that bends, the closer the total gets to the real length, and it always approaches from below. Every chord cuts the inside of the arc it replaces, so the error runs one way only — you undershoot, never overshoot.

The good news is how quickly it improves. Trace a semicircular bend with four segments and the total comes up 2.5% short. Eight segments cuts that to 0.64%, sixteen to 0.16%. Each doubling of your click count quarters the error, so the gap closes far faster than the effort grows — which is why there is rarely any point going beyond a dozen or so clicks per curve.

So place points wherever the thing you are tracing changes direction, and be generous on curves. If you are measuring something genuinely straight — a fence line, a cable run, a runway — two points are exact and more clicks add nothing.

Results are shown in kilometres or miles. One nautical mile, if you need it, is 1.852 km or about 1.151 statute miles — it was originally defined as one minute of latitude, which is why it turns up on marine and aviation charts rather than on road signs.