Did recent world record marathon runners employ optimal pacing strategies?

Research output: Contribution to journalArticleResearchpeer-review

30 Citations (Scopus)


We apply statistical analysis of high frequency (1 km) split data for the most recent two world-record marathon runs: Run 1 (2:03:59, 28 September 2008) and Run 2 (2:03:38, 25 September 2011). Based on studies in the endurance cycling literature, we develop two principles to approximate optimal pacing in the field marathon. By utilising GPS and weather data, we test, and then de-trend, for each athlete s field response to gradient and headwind on course, recovering standardised proxies for power-based pacing traces. The resultant traces were analysed to ascertain if either runner followed optimal pacing principles; and characterise any deviations from optimality. Whereas gradient was insignificant, headwind was a significant factor in running speed variability for both runners, with Runner 2 targeting the (optimal) parallel variation principle, whilst Runner 1 did not. After adjusting for these responses, neither runner followed the (optimal) even power pacing principle, with Runner 2 s macro-pacing strategy fitting a sinusoidal oscillator with exponentially expanding envelope whilst Runner 1 followed a U-shaped, quadratic form. The study suggests that: (a) better pacing strategy could provide elite marathon runners with an economical pathway to significant performance improvements at world-record level; and (b) the data and analysis herein is consistent with a complex-adaptive model of power regulation.
Original languageEnglish
Pages (from-to)31 - 45
Number of pages15
JournalJournal of Sports Sciences
Issue number1
Publication statusPublished - 2014

Cite this