A few years ago I wrote an article on a high-tech marathon prediction study which analyzed Strava data from 25,000 runners. They extracted each runner's fastest training segments over distances ranging "> A few years ago I wrote an article on a high-tech marathon prediction study which analyzed Strava data from 25,000 runners. They extracted each runner's fastest training segments over distances ranging ">

What is the most reliable predictor of your marathon time?

“],”renderIntial”:true,”wordCount”:350}”>

A few years ago I wrote an article on a high-tech marathon prediction study which analyzed Strava data from 25,000 runners. They extracted each runner’s fastest training segments over distances ranging from 400 meters to 5 km, plotted the data as a hyperbolic speed/duration curve, used this curve to calculate the runner’s critical speed and used critical speed to predict his marathon time.

If none of this makes sense to you, or you don’t have a GPS watch, or you just can’t be bothered to upload all your workout data into an all-seeing algorithm, then I have another type of marathon prediction study for you. In the European Journal of Applied PhysiologyJapanese researchers led by Akihiko Yamaguchi look at simpler variables like how long and how often you run, and offer general information to keep in mind the next time you tackle 26.2 miles .

Researchers surveyed around 500 runners about their training habits before the Hokkaido Marathon, focusing on monthly training volume, number of running days per week, average running distance and running distance the longest. (According to the newspaper, Japanese runners and racing media generally track their training volume by month, rather than the more common weekly totals in North America.)

Savvy readers will notice that these variables are interconnected: if you know the frequency of running and the average distance covered, you have already specified the monthly training volume. This is what makes this kind of analysis tricky. Many previous studies have attempted to determine which training variables are the best predictors of marathon time. But if, say, total training volume is a good predictor, it’s hard to know if that’s because running every day is the most important thing, or if really long runs are the key. or whether total mileage is what matters, regardless of how you accumulate it.

To circumvent this problem, the researchers divided their runners into subgroups. For example, they created four monthly mileage subgroups: those who drove less than 100,000 kilometers (62 miles) per month; 101 to 150K; 151 to 200K; and over 200K. In each of these groups, monthly mileage had no power to predict who would run the fastest marathon, as everyone ran a similar mileage. Then you can ask which variables do predict the time of the marathon. Is it an operating frequency? Average running distance? Longest running distance? The answer, oddly enough, is that none of them have significant predictive power. For people running a similar overall mileage, the other training variables don’t tell you anything useful.

They followed a similar procedure for training frequency, dividing the subjects into homogeneous groups running once to twice a week, three to four times and five to seven times, then analyzing the effect of the other variables. In this case, the strongest predictor was monthly mileage: for a given running frequency, the more you run, the better. Average running distance was also a predictor, but that adds nothing new: if you run the same number of days per week, those with higher average running distance will also have higher monthly mileage.

Subgrouping the other two variables (average run distance and longest run distance) produced similar results: in each case, total monthly mileage was the best predictor of marathon time in each subgroup. But this relationship only held for people whose average run was at least six miles and whose longest run was at least 12 miles. Below a certain minimum level of training, all predictions are wrong.

So far, it may seem painfully obvious: Those who run more miles run faster marathons. But the analysis of the subgroups allows us to draw more solid conclusions. Most notably, it doesn’t seem to matter how you rack up that mileage: a bunch of short trips or a few long trips produce similar results. which is parallel early summer findings in JAMA Internal Medicine on the health benefits of being a so-called weekend warrior: long-term mortality depends on how much exercise you do, but it doesn’t matter whether you spread your exercise throughout the week or that you integrate it on weekends.

If you dig deeper into the subgroup analyses, you also find that the longest run was a better predictor than the average run. As a result, the researchers conclude that at a given level of mileage, it is better to do one long run and several short ones than to do all of your runs at the same distance. This too fits with marathon orthodoxy which says there is no substitute for long runs.

Compared to the Strava study of 25,000 runners, this one has many shortcomings. It’s very small, the training data is self-reported and (therefore) does not include any speed measurements, the subjects are very lightly trained (on average 93 miles per month, or around 23 miles per week, with an average finishing time of 4h20). If you’re looking to qualify for the next Olympics, or even Boston, look no secrets here: you should rack up some volume and frequency and long runs, without trying to figure out which variables you can neglect.

But there are times in every runner’s life when training slips a few notches on your priority list. In these situations, the rule of thumb in this study seems more useful than the formula for calculating critical speed from your Strava data. The rule is: rack up as many miles as you can, whenever you can, whatever dose you can get. Sometimes races can be shorter or less frequent than you’d like, but on race day it all counts.


For more sweat science, join me on Twitter and Facebookregister at E-mailand check out my book Enduring: Mind, Body, and the Curiously Elastic Limits of Human Performance.

Comments are closed.