The fastest 24-hour ultramarathoners are from Eastern Europe

Abstract

Ultramarathon running is of increasing popularity, where the time-limited 24-hour run is one of the most popular events. Although we have a high scientific knowledge about different topics for this specific race format, we do not know where the best 24-hour runners originate from and where the fastest races are held. The purpose of the present study was to investigate the origin of these runners and the fastest race locations. A machine learning model based on the XG Boost algorithm was built to predict running speed based on the athlete´s age, gender, country of origin and the country where the race takes place. Model explainability tools were used to investigate how each independent variable would influence the predicted running speed. A sample of 171,358 race records from 63,514 unique runners from 73 countries participating in 24-hour races held in 57 countries between 1807 and 2022 was analyzed. Most of the athletes originated from the USA, France, Germany, Great Britain, Italy, Japan, Russia, Australia, Austria, and Canada. Tunisian athletes achieved the fastest average running speed, followed by runners from Russia, Latvia, Lithuania, Island, Croatia, Slovenia, and Israel. Regarding the country of the event, the ranking looks quite similar to the participation by the athlete, suggesting a high correlation between the country of origin and the country of the event. The fastest 24-hour races are recorded in Israel, Romania, Korea, the Netherlands, Russia, and Taiwan. On average, men were 0.4 km/h faster than women, and the fastest runners belonged to age groups 35-39, 40-44, and 45-49 years. In summary, the 24-hour race format is spread over the world, and the fastest athletes mainly originate from Eastern Europe, while the fastest races were organized in European and Asian countries.

Keywords: Machine learning; Nationality; Origin; Performance; Ultra-endurance.

Fig. 1

Optimal model evaluation metrics and features importance.

Fig. 2

Partial dependence plots (PDP) for gender.

Fig. 3

Partial dependence plots (PDP) for age group.

Fig. 4

Partial dependence plots (PDP) for the country of origin of the athlete.

Fig. 5

Partial dependence plots (PDP) for the country of the country where the race was held.

Fig. 6

Prediction distributions and target value plots for gender.

Fig. 7

Prediction distributions and target value plots for age group.

Fig. 8

Prediction distributions and target value plots for the athlete´s country of origin.

Fig. 9

Prediction distributions and target value plots for the country where the events were held.