You’re too Young to Stop Running! (Dublin Edition)

The relationship between age and performance based on the data from the Dublin Marathon

barrysmyth
Running with Data
Published in
8 min readOct 15, 2016

--

TLDR;

  • Getting older shouldn’t stop us running and may even have its advantages.
  • We do slow down as we get older but our pacing discipine improves and we hit the wall far less often.
  • And the experience of a few marathons under our belts can take years off our finish-times; experienced older runners can outperform first-timers who are 10 years younger.
  • This is part of a series of posts that I am writing as part of an ongoing analysis of marathon data. So if you are interested in this sort of thing then you will find a growing number of articles here.

Introduction

The wonderful thing about running is that almost anyone can do it. We are born to run and barring particular medical issues we can enjoy this wonderful sport at any age. As an older runner myself I’m encouraged to think (hope?) that my legs may well have another 30 years or more in them. We need to take a little more care in our running and training as we get older, of course, but that’s just about the commonsense of careful maintenance; for more on this topic check out this article on RunnersConnect.

What can we learn about age and running from looking at the data? I think we might all expect and accept that we will start to slow-down as we get older, but it turns out that age has its advantages too when it comes to other aspects of our performance.

Before we dig into this let’s talk about the data. If you have been reading these articles you will know that I have been collecting and analysing marathon running data for a while now. In this article I will, once again, focus on our own Dublin Marathon. I have result data for about 150,000 runners. This data includes age information but it’s not perfect because at best it provides only coarse-grained, 5-year age ranges (e.g. M40 means 40–45 year-old males). Worse still, younger runners in the 20–35 range are classified in the same single age group (seniors). In this analysis I will work with the mid-point of each age group. So seniors (20–35) will have a designated age of 27 and an F/M40 runner will have a designated age of 42.5 (the mid-point between 40 and 45), and so on.

So that’s the data, imperfect as it is, but good enough for our needs, at least when it comes to understanding how age generally impacts on speed, pacing, hitting the wall etc. To form a clearer picture of this we will compare runners based on gender (men vs. women) and experience (first-timers vs. repeat marathoners) as both of these factors also influence peformance. Bear in mind that we are calculating repeat marathoners with respect to the Dublin Marathon 2000–2015 data only as such our estimates of repeats will be conservative and typically underestimates.

The Age of the Dublin Marathon

The first Dublin Marathon took place back in 1980 and so this year’s event will be the 37th in the series. And as our favourite race turns 36 years old, it’s a good opportunity to look the age of its participants. The graph below shows the average age of particpants over the last 16 years, comparing first-timers to repeat marathoners (repeaters). First-timers tend to be a 2–3 years younger than repeaters, which is not so surprising. The average age is fairly stable over the years, hovering around the late 30’s for most of the years, although it is interesting to see a steady increase in average age since 2012; last year (2015) the average Dublin Marathon runner was, for the first time, in their early 40's.

The graph below shows the average percentage of finishers by age, computed across all 16 years. Although many runners are in their 20's and early 30's (about 40–50%) it is clear that there is a long-tail of older runners. Almost 20% of runners are in their mid-to-late 40s, for example, and more than 10% percentage are over 50.

So running marathons is not just a young person’s game. The graph below shows how older runners tend to run more marathons. For example, looking at all runners (that is first-timers and repeaters) we see how the average 55 year-old has run 2 marathons, whereas someone in their 30's has run about 1.5 marathons on average. Men are more likley to repeat that women.

Now that’s a little conservative as a way to look at how age influences marathon experience. A better approach might be to ignore first-timers as in the graph labeled ‘Repeaters.’ Now we see a more significant increase in the number of marathons completed with age. A 55 year-old repeater has run more than 4 marathons. Men still tend to run more marathons than women but the gap narrows with age and when we look a men and women in their 60s, who are repeaters we see that they run an average of about 4.5 marathons each. Woohoo! I don’t know about you but this means I still have 20 years of marathons in me!

Speed & Age

I suppose it’s no surprise to hear that we tend to get slower with age but it’s not all bad news, far from it in fact. The graph below shows how finish-times increase with age for men and women, first-timers and repeaters. Men and women add about 10 minutes to their finish-time every 5-years. That doesn’t seem so bad, actually, and better yet there is an antidote to age — experience — because repeat marathoners out-perform first-timers of the same age (and younger).

Look at it this way: a female first-timer in her early 40s runs a 300-minute marathon on average; but a female repeat marathoner of the same age will finish in only 270 minutes. Even better, a female repeat marathoner in her early 50’s (with 3–4 marathons based on our earlier graph) can expect a similar finish-time to a first-timer in her early 40's.

It’s even better for men. A male first-timer in his early 40's will finish in around 250 minutes, but a repeat marathoner of the same age will cross the line in about 235 minutes. A male repeat marathoner in his early 50's, with 4 marathons under his belt, will achieve a finish-time that is all but identical to the male first-timer in his 30's!

In other words, repeat marathon experience can buy us up to 10 years in age-related finish-times! It’s not quite linear because as we get older the benefits of experience tend to reduce, but some benefit persists.

Pacing & Age

Finish-times are just one perspective on performance, and a fairly narrow one at that. In other articles I have discussed pacing and pace variation and the link to hitting the wall. How do these aspects of performance vary with age?

Below is a graph of the average percentage pace variation (second-half versus first-half) for males and females, first-timers and repeaters, as they get older. Less pace variation is considered to be a sign of a more disciplined runner and we can see a pace variation sweet-spot for males in their late-40s to eary-50s, whereas women achieve their best pace variation in their early 40s. Throughout our 50s and 60s our pacing control tends to deteriorate again but it remains below the levels we experienced in our 30s.

Once again, marathon experience has a major impact on pace variation. A 55 year-old male repeat marathoner has far better pacing control than a 30 year-old male first-timer (11% vs 14% pace variation respevtively). And although the data for women is a little more erratic, due to lower numbers of older female repeat marathoners, the basic message is the same: a female repeat marathoner in her late 50's or early 60's has a pace variatuion of approximately 8%, far lower than a 30 year-old female, first-timer.

Hitting the Wall & Age

Finally, let’s look at the dreaded ‘bonk’, or hitting the wall as we say on this side of the Atlantic. As described in an earlier article we can estimate whether a runner hits the wall by looking for a 30% slow-down in the second-half of the race. If pacing control tends to improve with age then we should find that less runners hit the wall as they get older.

Once again the data bears this out. The graph below shows the percentage of runners who hit the wall, comparing men and women, first-timers and repeaters. In another article we saw how, in general, women tend to hit the wall far less than men ,and we can see here that this continues across all ages; simply put, men hit the wall about 3–5 times more frequently than women. Repeat marathons help both genders, but a strong female advantage remains.

Age helps too. Whether a first-timer or a repeat marathoner, older males hit the wall far less often than younger males. For women there is a similar trend but the improvement is less pronounced because their likelihood of hitting the wall was low already.

You’re Never Too Old to Run

I think this is good news for older runners. The evidence will be all around us in a few short-weeks as the streets of Dublin are once again filled with marathoners, young and old. The average age of participants in the Dublin marathon has been increasing of late and the average particpant is now in their 40’s. Although our speed may take a modest hit as we get older, the benefit of the extra years can be seen in our pacing discipline. Simply put we run more even-paced races and hit the wall far less often. And, on top of this, the experience of a few marathon can take years off us!

--

--

Running with Data
Running with Data

Published in Running with Data

A collection of articles at the intersection between data science and endurance running.

barrysmyth
barrysmyth

Written by barrysmyth

Professor of Computer Science at University College Dublin. Focus on AI/ML and data science with applications in e-commerce, media, and health.

No responses yet