How to Pace like an Elite in Boston

An analysis of pacing at the Boston Marathon.

TLDR;

  • An analysis of more than 150,000 Boston Marathon race-records during the period 2011–2016 (excluding 2013);
  • How do runners pace the Boston Marathon? Where are the fastest and slowest sections of the race?
  • How does the pacing of elites compare to recreational runners and those running a personal-best?
  • Putting what we learn into practice, we create optimal pacing charts for recreational runners that are tailored for Boston.

Introduction

In a previous post we described how to produce an optimal pacing plan for runners of the London Marathon based on an analysis of the pacing patterns of elites. The leap of faith was that such a pacing pattern may help regular runners to achieve their best times, assuming we can scale the pacing for their goal time.

In this post we present a similar analysis for the Boston course. This time it is based on more than 150,000 individual race-records for Boston runners, from the period 2011–2016, excluding 2013. These race-records include 5km (approx. 3.1 miles) split-times and therefore allow us to analyse pacing at regular intervals.

In total we examined the races of 117,802 unique runners (44% female), including almost 1,300 ‘elites’. For the purpose of this post we define an elite to be a male runner finishing in less than 150 minutes or a female runner finishing in less than 185 minutes; these are based on commonly used thresholds but more stringent definitions are also sometimes applied, either way they are sufficient to distunguish the very fastest runners from the rest of the field.

Pacing the Boston Marathon

To start, the chart below shows the average pacing (minutes per mile) for non-elite (recreational/regular) runners. The ‘ribbon plot’ shows the average pace per race segment (the marked centre line) and the average pacing for men and women. The chart also shows the course profile. As an aside the pacing ribbon has been left-shifted to the halfway point of each race segment for better visual alignment. For example, the first marker, which is plotted at the 2.5km (1.5m) mark actually reflects the pacing for the segment from the startline to the 5km/3.1m point.

On average we can see that the typical Boston marathoner starts at a pace of about 8 minutes and 15 seconds per mile but gradually slows to finish up running at around 10 minutes per mile by the end of the race; most do manage a final ‘sprint’ of sorts to the finish-line. The steep starting descent clearly doesn’t help when it comes to runners controlling their opening pace and fast-starts are known to be problematic later in the race. It’s a hilly course in general but the real hills start around the 15 mile mark and this shows ii the pacing above, with pacing slowing consistently througout the Newton Hills and beyond. Clearly, the average Boston runner starts outs fast and then spends the rest of their race paying for it as they gradually slow, completing a positive split with the second half of the race approximately 14% slower than then first.

A word of caution here, pacing above includes people who hit the dreaded wall in the second half of the race. This means that some of the slow down shown will be exagerated, at least somewhat, by these poor souls.

It’s All Relative

Rather than looking at the actual pacing values in mins/mile we can look at the degree to which a runner completes each segment slower or faster than their own average race-pace. This is shown in the next graph below where the y-axis stands for this relative or percentage pace; we refer to this as a pacing profile. A percentage pace of 90% means that the runner is running 10% faster than their average race-pace, whereas a percentage pace of 110% means that they are running 10% slower, and a percentage pace of 100% means they are running at their average race pace.

We can see how, on average, men and women both tend to start their race about 8% faster (92% percentage pace) than their average race-pace. Then, as the race unfolds, the percentage pace gradually increases, indicating a gradual slowing.

In other marathons, such as London, we have found that women tend to reign in their fast-starts more efficiently than men. They tend to slow-down earlier and, as a consequence, maintain their pacing better than men later in the race. However, there is a very different picture emerging from Boston. We can see above that it is men who tend to slow down more quickly than women; for example, by the 10 mile mark men are running about 2% faster than their average race-pace while women are going just over 3% faster.

This may be one reason why, in the later stages of the race, women slow more than men. This is most obvious around the 20 mile mark, where runners confront the dreaded Heartbreak Hill, and the pacing of men and women diverges significantly after this. For instance, in the final segments of the race men manage to speed up to within about 5% of their average race pace, where as women finish about 10% slower than their race-pace. It is not clear why this pacing pattern is different in Boston compared to other marathons we have looked at, such as London. Perhaps Boston’s qualification standards, and the gender differences they impose, play a role?

Perfect Pacing for Boston

We have found a fairly consistent pattern when it comes to how recreational runners pace Boston. Relative pace varies predictably across a broad field of runners, as it is influenced by growing fatigue, the significant undulations of the course, the noise of the crowd, the excitement of the start etc. It is a consistent pattern but, presumably, it is not a very optimal one. Certainly one lesson seems to be that, on average, runners start fast, probably too fast, The pay-back is a considerable slowing of pace as the race continues. The pacing range for recreational runners is particularly wide, from starts that are about 8–9% faster than average race-pace, to finishes that are 10–12% slower. This makes for an overall pacing range of 20% or more.

This begs the question: what is the optimal pacing profile for Boston? The conventional wisdom is that more even pacing is the way to go and so there is considerable room for improvement for the average runner in this regard. Easier said than done! And surely perfectly even pacing on a hilly course like Boston is not optimal anyway? To arrive an an optimal pacing profile we might start by isolating the pacing profiles of the very best runners, and to do this we will consider two groups of strong runners: those running a personal best (PB), regardless of finish-time, and elites.

Pacing for a Personal-Best

Our dataset includes runners who have run more than one marathon and for these runners we can identify PB times from their set of times; essentially we treat their fastest time among their set of races as their PB. For the purpose of this study we will consider those runners who have run at least 3 Boston marathons. There are almost 10,000 such runners in our dataset and they have run an average of just over 4 Boston Marathons each. For each of these runners we can identify their ‘personal-best’ as their fastest Boston time. Of course this might not be their true personal best, since it depends entirely on the window of data that we have access too, but it at least counts as a good time for these runners at Boston.

Next, we can compute the pacing profiles for these PB runners as before to the get the chart below. On the face of it, we see a similar shape to the pacing profile of the regular runners we looked at earlier: PBs start faster and finish slower than their average race pace, just like regular runners. However, the key difference is that, compared to regular runners, those running a PB start a little slower and finish a little faster. For example, the average PB runner starts only 4% faster than their average race pace, holds this pace for the first 9 miles or so and then starts to skow. But they don’t slow as much as regular runners. Yes, we can still see the effect of the Newton Hills but they slow to only about 5% slower than their average race-pace. In other words, the pace variation for runners completing a PB tends to fall within a range of just 9% (from -4% at the start to 5% at the end) compared to more than 20% from regular runners.

It is also interesting to note the lack of any real difference between the relative pacing of men and women, compared to the very real differences we saw amongst regular runners. This hints that perhaps those running a PB have managed to find a good pacing profile for Boston. Remember too that these pacing profiles are independent of finish-times, and included in the above graph are people PB’ing right across the spectrum of finish-times.

The Optimal Pacing of Elites

What about elite runners? How do their pacing profiles compare? Our Boston dataset contains 1,291 elites, defined here, somewhat loosely, as men and women finishing within 150 and 185 mins, respectively. The graph below shows the pacing profile for these runners.

This time we see a more even pacing profile again, with elites operating within +/- 3% or so of their average race-pace. As with recreational and PB runners, elites start out faster than their average race-pace, but only 3% faster, and they finish slower than their average race-pace, but only 3% slower. Throughout the race elites run closer to their average race-pace than either recreational or PB runners. Again there is virtually no difference between the pacing profiles of elite men and elite women, despite their differing speeds, at least until the very end of the race where men seem to have an additional kick.

For the purpose of comparison, the chart above shows the average pacing profiles for the regular runners, PBs, and the elites, side by side. We can see more clearly now how the PBs and elites produce increasingly even pacing profiles, but not perfectly even ones. And this is the essential point of this work. Running a course like Boston with perfectly even pacing cannot be optimal because it ignores the very significant terrain on offer. The PBs and elites recognise this and marshall their early pacing to ensure that they are ready for the hills when they hit later in the race. By doing this well, the PBs, who after all are just regular runners, manage to run a personal-best (or close to it).

An Optimal Pacing Chart for Bos

Does that mean, if we ran a similar profile that we would enjoy a new PB? Perhaps. Does it mean that if we managed our pacing to be more like that of an elite — much slower of course but in proportion — that we could improve our times further? Probably.

To test this we can run an experiment. If we take the leap of faith that elite pacing profiles are close to optimal for a course like Boston then, in theory, we should be able to map these pacing profiles to any chosen goal-time and produce a pacing profile to achieve that time which is optimally tailored for Boston.

To do this for a given target finish-time of say 4 hours (240 minutes), we first calculate the average race-pace, in this case about 9 minutes and 9 seconds per mile. Next we map our elite pacing profile from Boston to this target time and pace. Elites run the first 5k/3.1 miles just under 3% faster than their average pace and so we adjust this 4-hour pacing to recommend a start pace of 8 minutes and 53 seconds (2.9% faster than the 9 minute, 9 seconds pacing).

Doing this for each of the race segments gives the tailored pacing plan above. It is designed to produce a 240-minute finish in a manner that matches the relative pacing of elite runners. It is not an even pacing plan, because even elites don’t run in this way; by the time you hit the Newton hills you will be running your miles in 9 minutes and 29 seconds, for example.

At the end of this post I include pacing tables for a wide range of goal-times, from 150 minutes to 360 minutes in 5 minute goal-time intervals. To use them you simply need to locate your goal-time in minutes and then read off your paces from that row. Each pace corresponds to the average pace for a given segment of the race as per the column headings, from the first 3.1 miles (5kms) and so on in 5km intervals.

Conclusions

Of course the notion that elites pace perfectly in every race is probably too strong an assumption to make. In any given race there will be many factors that influence their pacing: course conditions, the competition, how they feel on the day etc. However, on average, when pacing is analysed over many runners and a number of years, a strong pattern emerges and it is this pattern that might be important when it comes to informing the pacing of other runners.

The elite pacing profile for Boston is different from London, because the Boston marathon is different from the London marathon. And so we can reasonably conclude that the elite pacing profile for a given race likely corresponds to a highly optimised profile that is tailored to a particular course. The hope is that the pacing tables produced from this profile may allow recreational runners to replicate this ‘perfect’ pacing profile in their own races and for their running ability and target times. Doing so may help to produce better performance on the day.

If these pacing charts help you during your race please let me know!

Appendix — Pacing Charts by Goal Time

150–180 Minutes

180–240 Minutes

240–300 Minutes

300–360 Minutes