Marathon pacing and pack running

The latest pacing study from researcher Brian Hanley, of Leeds Beckett University in Britain, tackles the marathon. Hanley has previously used digitally recorded midrace splits to analyse pacing at the World Cross Country Championships and the World Half Marathon Championships.

The new study, published in the Journal of Sports Sciences,looks at splits from eight World Championship and Olympic marathons between 2001 and 2015.

A couple of things stand out. First, check out the difference between the men’s pacing patterns and the women’s pacing patterns. Here’s the men’s data, broken down into medallists, non-medallists who finished with 5 percent of the winning time, 6 to 10 percent, and so on:

And here’s the comparable women’s data:

After 20 miles, the men get steadily slower at all levels. But the top women (on average) are able to accelerate after 20 miles, and women at all levels are able to sharply accelerate in the final mile or two. This observation matches earlier data from big-city mass-participation marathons, which showed that women appeared to pace themselves more conservatively than men and consequently slowed less in the second half of the race.

As I discussed in a previous blog post, it’s not clear why this happens - there are plausible mechanisms that are physiological, psychological, sociological, etc - but it’s interesting that the same thing is observed in the very best marathoners in the world, without the influence of pacemakers. Hanley suggests that this is something for athletes to bear in mind in their training: elite women win marathons with a kick, men win by hanging on longer than their rivals.

(As an aside, my anecdotal impression is that the pattern is exactly reversed in middle-distance races. Women’s 1500 and 5000-metre championship finals tend to be fast right from the start, while men’s are more likely to be sit-and-kick. Perhaps fodder for a future study...)

The other interesting aspect of Hanley’s analysis is his look at pack running behavior. He divides runners into various categories, based on whether the runner stays with the same pack for the whole race, stays with a pack part way through the race (e.g. up to halfway) then runs alone, moves between various packs throughout the race (what he calls “nomadic” pack running), or mostly runs alone.

The previous half marathon data suggested that running with a pack is beneficial as long as it’s the right pack - i.e. trying to stick with a pack that’s too fast for you is worse than no pack at all. This seems like common sense. The more interesting suggestion was that nomadic packing is the second-best option: you might benefit from making a conscious effort to find people to run with, even after your initial pack breaks up. As I wrote before:

“If you realise that you're stronger than the other runners in your pack, look up the road to see if there's another group you could join. Then, make a decisive move to catch them instead of lingering in no-man's-land. Conversely, if you're getting dropped by your pack, it's very easy to become disheartened. Instead, slow a little, regroup, and prepare mentally to latch onto the next pack that comes up behind you.”

The new data partly supports this contention, with better outcomes for nomadic packers in the women’s races. The pattern is less clear in the men’s data, and that may reflect the harsher cost of blowing up in a marathon compared to shorter distances.

Of course, there are limitations to the conclusions we can draw from this kind of data. Do athletes run well because they stay in packs longer? Or is it the other way around: they’re capable of staying in packs longer when they’re (for whatever other reason) having a good day? I suspect there’s a bit of both going on, but overall it’s an interesting look at what subtle factors distinguish a great day from a good day from a disaster.