Convert mi to m

A mile is 1,760 yards — a number that comes from a Roman thousand paces, adjusted by an Elizabethan statute to fit the furlong. It is exactly 1,609.344 metres.

The metre has been defined since 1983 as the distance light travels in 1/299,792,458 of a second. It is not a physical object any more; the platinum bar in Paris was retired.

The mile has been defined in metric terms since the international yard and pound agreement of 1959, and in this page's units that comes out at 1609.344 m to one. The definition is exact by stipulation — it was not measured, and it cannot be refined by measuring it better. The decimal above is the whole of it, with nothing rounded away at the end.

Before that agreement the US and the British versions differed slightly, and old survey figures still carry the older definition. For anything outside land surveying the difference is far below the precision anybody is working to.

A standard outdoor track measures 400 m in lane one, so four laps is 1,600 m and a mile is 1,609.344 — nine metres and 344 millimetres further. Every track that hosts a mile race has a separate start line painted for it, set back from the common one, and that painted line is the most concrete demonstration available that the conversion is exact rather than convenient.

This is also why the 1500 m exists as a distinct event rather than as a metric mile. Three and three-quarter laps is 1,500 m, which is 109.344 m short of a mile — about fifteen seconds for a good runner, which is far too much to treat the two as versions of the same race. The mile survived metrication in athletics precisely because it could not be rounded into one.

Road course measurement is done in metres regardless of what the race is advertised as, using a calibrated counter on a bicycle wheel over the shortest legal path. A course sold as a mile is laid out to 1,609.344 m and then a short course prevention factor of one part in a thousand is added, so the measured length is about 1,610.95 m. The asymmetry is intentional: a course that turns out short voids every time set on it, and a course a metre long costs nobody anything.

That factor is worth knowing before comparing a certified distance against a personal recording. A watch reporting 1.00 miles on a certified mile has under-measured by a metre or two, which is well inside its own error, and a watch reporting 1.01 has probably done nothing wrong either. Neither reading is evidence about the course.

The marathon is defined in imperial units and quoted in metric ones. The international federation fixed it in 1921 at 26 miles 385 yards, which multiplies out to exactly 42,194.988 m — 26 × 1,609.344 plus 385 × 0.9144. The 42.195 km on every race entry is that figure rounded up by twelve millimetres, which is why the metric number has three decimal places and still is not exact.

The consequence for anyone marking a course is that the distance has to be taken from the imperial definition and not from the rounded metric one. Twelve millimetres over 42 kilometres is nothing on its own, but it is the sort of detail that reveals whether a distance was derived or copied — and the same applies to the half marathon at 21,097.494 m, which is a halved marathon rather than a distance anybody chose.

A device that says one mile is doing arithmetic on a path it inferred from position fixes taken a second or so apart. Consumer receivers carry several metres of horizontal error at the best of times, and the sampling turns curves into chords, which shortens the path, while the noise between fixes adds spurious wander, which lengthens it. The two errors do not cancel reliably, and one to two per cent — sixteen to thirty-two metres in a mile — is normal.

Under tree cover, between tall buildings or on a track with tight bends the error grows and its sign becomes unpredictable. This is why converting a GPS mile to 1,609.344 m and treating the result as a measurement is a category error: the conversion added no uncertainty and removed none, and the figure it is applied to was never good to better than tens of metres.

The mile is Roman in name and English in length. The Roman mille passus was a thousand double paces, somewhere around 1,480 m, and the English mile drifted away from it over centuries of local usage. A statute of 1593 fixed it at eight furlongs, and a furlong was already 220 yards — the length of a ploughed strip — so the mile became 1,760 yards to stay compatible with the land measures rather than to be a round number.

That inheritance is why every subdivision of the mile is untidy in metres and tidy in the units it came from. 1,760 yards is 5,280 feet is 63,360 inches; a furlong is 201.168 m, a chain is 20.1168 m and a rod is 5.0292 m. Each of them terminates, because they all descend from the same exact yard, and each of them looks arbitrary because none of them was ever chosen against a metre.

For anything but a marked course, a mile is 1,600 m and the error is 0.58 per cent. Three miles is 4,800 against a true 4,828; ten miles is 16,000 against 16,093. The shortfall grows to about a hundred metres over ten miles, which matters to a course designer and does not matter to somebody working out how far away something is.

The corrections worth carrying are 5 km to 3.107 miles and 5 miles to 8.047 km, because they bracket the range most road distances live in. If a figure has to be right to the metre it should come off the definition rather than an estimate: 1,609.344 m to the mile, multiplied out in full, with the rounding applied once at the end rather than to the factor.

Technical and scientific documents take metres because metres are the SI base unit, and a mile figure entering one should be converted at full precision and rounded once, at the point of writing. Converting 3.7 miles to 5,954.5728 m and then rounding to 5,955 is correct; rounding the mile to 3.7 first and then to the nearest ten metres compounds two roundings that were never independent.

The precision question runs the other way from the arithmetic. A distance given as "about three miles" is uncertain by hundreds of metres, and writing it as 4,828.032 m makes a claim the source cannot support. Carry the digits through the calculation and drop them at the end, to the precision the original had — which for a mile stated to the nearest tenth is the nearest hundred metres, and no finer.

A big-city marathon is measured and marked in metres and then given a second set of markers for the runners who pace in miles, and the two sets interleave without ever agreeing. Mile 1 falls at 1,609 m, between the 1 km and 2 km boards; mile 20 at 32,187 m, a third of the way from 32 km to 33. Only the start and finish coincide, and even the finish only does so because the metric distance was derived from the imperial one.

For a runner this is a pacing problem rather than a measurement one. A watch beeping every mile and a course marker every kilometre give split times that cannot be compared without conversion, and the drift accumulates: by 20 miles the two counts are twelve markers apart. Pick one unit for the whole race and read only its markers, because the arithmetic that reconciles them is the last thing anybody wants to be doing at 30 km.