The Dublin City Marathon, 2017

The Data of the SSE Airtricity Dublin City Marathon


  • We examine the race records of more than 15,800 runners from the 2017 SSE Airtricity Dublin City Marathon.
  • We consider the gender, age and club affiliations of runners and how this impacted on their performance
  • We also compare runners based on their pacing and how often they hit the wall.


This year was the 38th year of the Dublin City Marathon, which began in 1980 with just 2,000 runners. This year the organisers celebrated 20,000 registrations, with the marathon selling out long before the big day, yet again. The timing data lists just under 16,000 finishers however; this suggests that up to 20% of registrations were either no-shows or did not finish.

As usual the Dublin Marathon brought out the crowds for which it has become famous. The day itself was near perfect for supporters and runners alike — ok, perhaps a bit too warm at times for the runners — but dry and calm.

What follows is an initial analsysis of the data generated by this year’s participants, focusing on gender, age, and club affiliations, and how these factors impact performance and pacing.

Who are the Dublin Marathoners?

Let’s begin by getting a sense of who this year’s runners were, at least in terms of their gender and age. The chart above shows the percentage of male and female runners, overall, according to the standard marathon age categories. Over two-thirds of runners are male and approximately 60% of runners, male and female, are between the ages of 35–49.

Everyone’s a Winner

What about finish-times? The chart below shows the number of male (blue, above the line) and female (red, below the line) runners who finished at various finish-times; the height of each line corresponds to the number of runners cross the line at that precise time.

Obviously there were very few finishers crossing the line for the fastest finish-times (shortly after 130 minutes) with a sharp increases in finishers from about 180 minutes onwards, for men, and 240 minutes onwards, for women. The most popular finish-time for men was 240 minutes; we can see large spikes of finishers just before this landmark time. Other landmark times are also easy to identify with obvious spikes corresponding to 180 and 210 minutes too, as well as smaller spikes for intermediate times, most of which are associated with pacers in Dublin. A similar patter is evident for women.

The insert in the top-right of above chart shows an expanded portion of the larger chart, but focusing on the top-30 male and female runners to cross the line. The separation between the lines gives a sense of how competitive the men’s and women’s race was on the day. The male race was hotly contested, with Kenyan Bernard Rotich winning the men’s race in a time of 2 hours 15 minutes, closely followed by Yurii Ruskyuk and Asefa Legese Bekele, only seconds behind.

The women’s race provided a more emphatic win for Nataliya Lehonkova of Ukraine, who finished in a time of 2 hours and 28 minutes, followed by Ashu Kasim some 4 minutes later, and Khapilina Viktoriya a little over a minute after that.

The Wisdom of Age

As we might expect age has a significant influence on finish-times. The chart below shows how, as we get older, we also get slower, but only after our mid- to late-30s. For example, on average, the fastest finish-time for men occurs for the 35–39 year-old group, but men and women both slow in their 40s, 50s, and 60s.

In Dublin this year about 29% of runners listed a club affiliation, slightly more females (33%) than males (27%), and we might wonder how these club runners stack up against those who did not list such an affiliation. The finish-time chart above also shows the average finish-times for these club runners and, as expected, both male and female club runners enjoy faster finish-times compared to non-club runners, and across all age-groups. For example, the average male runner in the 40–44 age-group finished in 240 minutes, whilst a club runner of the same age finished about 15 minutes faster. There is also evidence that this ‘club dividend’ tends to improve with age for men and women, as the finish-time gap between non-club and club runners gradually widens.

Good for Age

It turns out that it‘s not all bad news for older runners, if we calibrate between finish-times and ageby using an age-independent measure of finish-time; this allows us to more fairly compare the finish-times of runners across different age ranges. One way to do this is to use the famous Boston Qualifier (BQ) times. These are gender-based and age-based finish-times, which runners need to achieve, in order to be eligible to qualify for the Boston Marathon. For example, at the time of writing (October 2017) male runners in the 35–39 years old age bracket need to finish no later than 3 hours and 10 minutes (190 minutes) to be eligible for qualification. For similarly aged women the qualification time is 220 minutes.

The chart above shows the percentage of male and female runners who achieved their BQ times across the various age groups. Generally speaking, men and women perform similarly by this measure. For example 10% of men in the 50–54 age group achieved their BQ time this year; for women it was slightly higher at 12%. Interestingly, we can see that the likelihood of achieving a BQ time tends to increase with age, at least up to runners in their 60’s, which suggests that experience is helping older runners to improve their age-related performance.

The chart also includes the BQ rates for club runners, showing very significant performance benefits, compared to regular runners. For example, in the 50–54 age group almost 25% of men achieved their BQ times, more than twice the BQ rate for regular runners. A similar benefit is noted for women, albeit not as great as that noted for men. Club runners in their 60s enjoy especially high BQ rates (~40%).

Perfect Pacing

For Dublin we have access to timing information at the 10km, halfway (21.1km), 30km marks, and the finish, so we can look how well runners pace these different race segments. Admitedly these are fairly coarse grained segments, and therefore limit the type of analysis we can carry out, but it’s what we have to work with.

To make it easier to compare runners with different finish-times, it is natural to look at relative paces for these segments; each relative pace (RP) denotes the degree to which the runner is running faster or slower than their average race-pace. For example, a relative pace of 0.9 for the first 10km means that the runner has run this segment 10% faster than their average pace (for the race as a whole).

The chart above shows the relative paces for male and female runners for each of these segments; it also includes pacing for club runners, and those achieving their BQ times. For all three cohorts (all runners, club runners, and BQs) we can see men and women tend to start fast and gradually slow as the race unfolds. The fast-start phenomenon is clearly evident, with men and women running the first 10km of the marathon about 5–6% faster than their mean race pace, and men starting faster than women; we have seen a similar pattern in other races around the world as runners get carried away with the excitement of the start-line.

Generally speaking, women reign in their early pacing more effectively than men and, by the end of the race, suffer a more modest slow-down compared to men; for example, the average male runner completes the final segment about 9% slower than their mean race pace, compared to just over 5% slower for women.

As we might expect, club runners and Boston qualifiers do a better job when it comes to controlling their pacing. BQ runners, in particular, manage a very even pacing profile: they start only 1–2% faster, and finish only 3–5% slower, than their average race pace. Club runners are not quite as accomplished pacers as BQ runners, but they too start slower, and finish faster than regular runners.

What Wall?

There are signs just before the Dublin Marathon’s own ‘Heartbreak Hill’ asking “What Wall?” as runners confront the long drag up Roebuck Road around the 34km mark. For me this was the toughest section of the race and judging from what I saw, it is here that many runners become all too familiar with the wall that these signs are referring to. But how many runners succumb to this most feared of marathon hazards?

For the purpose of this analysis we deem any runner who slows by at least 33% in the second half of the race, compared to the first, to have hit the wall. For example, a runner with an average pace of 5 minute per km in the first half of the race, and on track for a 3:30 finish, would have to slow to 6.65 mins/km for the second half, to be classified as hitting the wall; if they did they would like finish in more than 4 hours instead of 3:30.

When we apply this definition of the wall to Dublin’s runners, and compare the percentage of men and women hitting the wall, based on their finish-times then we get the above result. Faster runners rarely hit the wall but it becomes increasingly likely for slower runners. For example, 15% of men finishing in 5 hours will have slowed by at least 33% in the second half of the race. Women fare much better, and hit the wall only a small fraction as often as men across all finish-times; for example, only about 2–3% of 5-hour female finishers hit the wall.

Indeed this 33% slow-down may be somewhat conservative. Some runners may hit the wall towards the very end of the race, so that their average second-half pace suffers a more modest slow-down. In this case the above results are likely too conservative and underestimate the trune number of runners who hit the wall. On the other hand, it does help us to distinguish between natural fatigue, and those runners who slow by <33%, and truely hitting the wall, which represents a more catestrophic pacing crisis.


That’s it for another year; indeed as I write this my legs feel like it might take the full year to recover! There is no doubt that Dublin has cracked the magic formula when it comes to hosting a truly inspirational event and when the weather behaves, as it did this year, there is nothing like it. And hopefully, this analysis will be of interest to those how participated, whether a supported or a runner.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.