REF-824 · Hanley, 2014 · Journal of Sports Sciences, Vol. · Paper · PubMed

Pacing profiles and pack running at the IAAF World Half Marathon Championships

B+moderate evidenceMeasured on elite to world-class athletesWritten before October 2026, being re-checked

The authors analysed finishing and split times for 491 men and 347 women from six IAAF World Half Marathon Championships, computing mean speed over each 5 km segment and the closing 1.1 km and grouping athletes by finishing time and by how they ran in packs.

The fastest men and women held their speed fairly evenly from 5 km to 15 km, while slower athletes lost speed from 5 km onwards. Those who stayed in a pack for the whole race slowed less than those who did not, or who kept pack contact only to 5 km.

Part of the slowdown between 15 and 20 km may have been tactical, saving something for a fast finish, which paid off especially for medallists; athletes who stayed with the same rivals all race accelerated most at the end.

Participants

491 men and 347 women competing at six IAAF World Half Marathon Championships.

Protocol

Finishing and split times collated for six championships.
Mean speed calculated for each intermediate 5 km segment and the final 1.1 km.
Athletes grouped by finishing time, and classified by packing behaviour (in a pack throughout, only to 5 km, or not).

Key Results

The best men and women largely held their split speeds between 5 km and 15 km; slower athletes slowed from 5 km onwards.
Running in a pack throughout went with smaller decreases in pace than running alone or staying in a pack only to 5 km.
Slowing from 15 to 20 km may have been fatigue, or a tactic to set up a fast finish, which benefited medallists in particular.
Athletes who stayed with the same rivals for the whole race accelerated most at the end.
The abstract gives no speeds, percentages or p-values.

Practical Application

In 491 men and 347 women across six World Half Marathon Championships, the best men and women held an even speed from 5 km to 15 km while slower athletes faded from 5 km on, and those who ran in a pack the whole way slowed the least. Run the half evenly from the early kilometres, and find a group running your target pace and stay with it.
The author's own advice: pick out rivals likely to have similar ability and goals, and build running with them into the pre-race plan.
Inference, not tested here: in a mass race, a pace group or a few runners of matched ability plays the role of the pack; staying in a pack only for the first 5 km did not protect against slowing.
The slower final 15-20 km segment was not always fatigue — in medal races it set up the finish. Not from this paper; the base's own practice: a recreational runner should not copy that tactic and should hold pace to the end.

Limitations

Elite championship races only; tactics, pack sizes and motivation differ from mass-participation races.
Packing behaviour was chosen by the athletes (and depends on ability), so pack running is associated with, not shown to cause, smaller slowdowns.
Course, weather and championship year differ; the abstract gives no speeds or effect sizes.
Written from the abstract and record metadata only (full text not read).
How this entry was checked and changed
Why this level

large observational dataset of championship races; descriptive, with packing behaviour self-selected rather than assigned