Pacing during an ultramarathon running event in hilly terrain

Purpose. The dynamics of speed selection as a function of distance, or pacing, are used in recreational, competitive, and scientific research situations as an indirect measure of the psycho-physiological status of an individual. The purpose of this study was to determine pacing on level, uphill and downhill sections of participants in a long (~173 km) ultramarathon performed on trails in hilly terrain. Methods. Fifteen ultramarathon runners competed in a ~173 km event (five finished at ~103 km) carrying a Global-Positioning System (GPS) device. Using the GPS data, we determined the speed, relative to average total speed, in level (LEV), uphill (UH) and downhill (DH) gradient categories as a function of total distance, as well as the correlation between overall performance and speed variability, speed loss, and total time stopped. Results. There were no significant differences in normality, variances or means in the relative speed in 173-km and 103-km participants. Relative speed decreased in LEV, UH and DH. The main component of speed loss occurred between 5% and 50% of the event distance in LEV, and between 5% and 95% in UH and DH. There were no significant correlations between overall performance and speed loss, the variability of speed, or total time stopped. Conclusions. Positive pacing was observed at all gradients, with the main component of speed loss occurring earlier (mixed pacing) in LEV compared to UH and DH. A speed reserve (increased speed in the last section) was observed in LEV and UH. The decrease in speed and variability of speed were more important in LEV and DH than in UH. The absence of a significant correlation between overall performance and descriptors of pacing is novel and indicates that pacing in ultramarathons in trails and hilly terrain differs to other types of running events.