Since 1982 I've written a newsletter, Running Commentary. A new issue appears here each week, and material is archived.

Sat, 02 Dec 2006 05:36:11 -0500

True Miles


How far did you run today? Chances are, you think you know. Chances are even better that you're at least a little off.

The mile is the gold standard of running in the U.S. Runners here talk in terms of minutes per mile and miles per week.

You know exactly what a mile is: 1760 yards, or 5280 feet. But running exactly a mile isn't so simple.

A true mile is hard to find. Distances resist measurement to the degree of accuracy that runners demand.

I can't tell you how long my run was this morning. Outside of races, I don't need to know -- don't even WANT to know, because that would remind me daily how much I've slowed through the years.

Almost 40 years ago I opted out of mile-checking. Instead I ran by time periods -- 30 minutes, an hour or longer -- without knowing or even estimating the distance covered in that time.

This practice, still followed today, greatly simplified my running: no courses to plot or follow. A minute is the same length no matter where it's run, and more easily and accurately measured than a mile.

Despite this preference for by-time running I've stayed interested in how, and how well, courses are measured. That's because I want my race distances to be as correct as possible, and because most runners still run by the mile, and because my students insist on knowing how far they run.

So let's look at the ways runners check their miles, from least to most accurate. This list also traces fairly closely the history of measurement.

-- Wild-guess method. After winning one of my early road races in "world record" 10K time, I guesstimated the true distance at 5-1/4 miles. The race organizer apparently hadn't even quick-checked the course by car.

-- Pedometer method. These devices, which I tried and discarded early on, are hopeless for runners. They're good at counting steps but not at calculating the length of those steps, which vary widely from runner to runner and for you within any run. You could guess distances as closely, which means coming within a mile.

-- Minutes-to-miles method. Here you run a unmeasured course and divide the total time by your typical pace per mile. But that pace differs from day to day, so your estimate could be off by a half-mile in either direction.

-- Car method. This remains the most common way to measure courses. Its weaknesses: car odometers aren't always accurate and aren't calibrated closer than tenth-miles; you can't often drive the shortest possible route that a runner would take, and you can't drive off-road where a runner might go. Cars almost always measure courses longer than they really are.

-- Map method. Using a large-scale map, we once measured with a ruler. Now this method is easier with web devices by the dozen (Google "running routes" for a list). These can be fairly accurate -- if the course involves mostly straight-line running. The more curves, the lower the reliability of plotted distances.

-- Track method. If you don't mind feeling like a caged hamster, run on a track. Standard tracks are four laps to the mile. Unless they're 400-meter track and your laps four laps fall about a dozen yards short of a mile. Or unless you run in lanes other than one, when each lane out from the curb adds several yards per lap.

-- GPS method. It's the current favorite, but runners put more faith in these devices than they deserve. In my marathon training group, three runners wearing the same brand and model of GPS watch can get three different distance readings. The discrepancies aren't huge, a few tenths of a mile over double-digit distances. But if the technology were perfect, they'd all agree to the hundredth of a mile.

-- Bicycle method. This one is better than the one above, but not so good as the one below. You attach a speedometer to your bike, then calibrate that device on a route known to be precise. This is how I measure courses, with a true mile being 1.04 on the dial. GPS wearers in my group question my accuracy, which I trust more than theirs because my device doesn't blank out under trees and bridges. (Walking a measuring wheel, after calibration, works just as well as biking the course but takes much longer.)

-- Certification method. The only recognized way of certifying courses is with a counter mounted on the bike wheel, calibrated against a short course measured to a surveyor's degree of accuracy -- to the inch. The race course must then be ridden carefully, along the shortest possible route. If a course is advertised as "USATF certified," you can trust that the miles are the closest possible to true. But only if you hug the course as measured, which isn't possible in crowds and means you run a bit beyond full distance.

Complicated, isn't it? You can see why I opt for by-time running.

All you do here is punch on the watch at the start and off at the finish, and you know precisely what you ran -- not how far but how much. You can trust today's watches to measure true minutes.
Previous Posts