Elias Villiger

Long-distance triathlon has become a major model for studying endurance performance. Research has nonetheless been constrained by distance-specific datasets that separate full-distance from half-distance events and largely omit transition times. Our aim was to assemble, validate, and describe an openly available dataset that removes both constraints. We compiled 2,706,922 athlete-race records from more than 1500 IRONMAN^® and IRONMAN^® 70.3 events held between 2002 and 2026, the final season partial, drawn from the official IRONMAN^® results platform (75.4%) and a third-party aggregator covering event series absent from it (24.6%). Each record contains swimming, first transition (T1), cycling, second transition (T2), running, and overall times in seconds, with finish status, age group, gender, country, and qualification points. The dataset showed close correspondence between sources and high internal consistency. Split times were identical to the second for 98.8% of 4,732,776 athlete–discipline pairs across all 559 race-years present in both sources, summed splits matched reported overall times within one second for 98.3% of records, and separated T1 and T2 times were available for 84.2% and 84.1% of records—to our knowledge, the first population-level provision of this information. Openly deposited with reproducible collection scripts, the dataset supports research on pacing, transition efficiency, sex differences, non-completion, and qualification equity.

Purpose: To identify the best-performing nationalities and race locations for professional IRONMAN® triathletes, and to explore the best environmental conditions. Methods: We collected and analyzed records from 6,848 professional finishers representing 37 countries participating in 54 IRONMAN ^® race locations between 2002 and 2022. Information extracted includes athlete information (sex, country of origin, race location), performance (hh:mm:ss), and environmental conditions (i.e., water and air temperature). A non-linear machine learning predictive model was developed with race finish time as the predicted variable and sex, country of origin, event location, and the locations’ average air and water temperatures as predictors. Several model interpretability tools were used to assess the contribution of each predictor to the model’s predictions. All analyses, modeling, and visualizations were performed using Python. Results: Among the top five, the fastest overall male race times were achieved by athletes from Ukraine (8.71±0.58 h), Portugal (8.79±0.41 h), Poland (8.94±0.68 h), Switzerland (8.98±0.48 h), and Estonia (9.00±0.67 h). The fastest overall female race times were recorded by athletes from Poland (9.37±0.35 h), Finland (9.53±0.64 h), Chile (9.70±0.33 h), Switzerland (9.71±0.55 h), and Denmark (9.78±0.59 h). The fastest overall race times were achieved at IRONMAN® Tallinn (9.01±0.67 h), and IRONMAN ^® Switzerland Thun (9.02±0.38 h). The model identified sex as the most important predictor, followed by the athlete’s country of origin. For air temperature, the model predicted the best race times within a range of 19 to 27 °C. For water temperature, the model favored warmer temperatures for faster race time predictions. Conclusions: The professional IRONMAN ^® performance is influenced by a combination of geographical and environmental factors. European countries and race locations were consistently associated with superior performances, while favorable environmental conditions were characterized by relatively warm water temperatures and moderate air temperatures.

While the dominance of elite East African marathoners is well-established, little is known about the nationality distribution of the fastest age group marathoners. This study examined how the nationality distribution of the fastest marathon runners varies across age groups and sexes in the ‘Berlin Marathon’. We analysed race data from 825,886 finishers (627,962 men and 197,807 women) of the ‘Berlin Marathon’ from 1999 to 2024. Finishers were grouped by five-year age brackets, sex and nationality. We examined performance across five analytical layers: (1) all runners, (2) top 100, (3) top 30, (4) top 10 and (5) the single fastest runner per country and age group. Linear regression models were applied to both the entire dataset and the subset of the top 100 finishers per age group to assess the predictive value of age and nationality on finish time. Most finishers were from Germany, with the USA becoming the second-most represented country after 2017. In younger age groups (20–39 years), the fastest average times for both sexes were achieved by runners from Kenya and Ethiopia. In middle-aged (40–49 years) and older groups (50 + years), top performances shifted toward runners from Germany, the United States, Japan, and Switzerland. In regression models, age group and nationality explained only 4.1% of the variance in male finish times and 2.9% in females across the full dataset. However, for the top 100 runners, explanatory power rose significantly to 52.8% for men and 48.6% for women. The nationality distribution of the fastest age group marathoners shifts with age in the ‘Berlin Marathon’: East African nations dominate at younger ages, while Western Europe and East Asian countries lead among older runners. These results emphasize the global expansion of competitive master running and the role of cultural and demographic factors in shaping endurance performance across the lifespan.

To determine whether sex differences in marathon pacing strategies and susceptibility to catastrophic deceleration ("hitting the wall") persist in a massive, high-performance dataset, and to quantify this risk across different performance levels. We analyzed 873,334 finishers from the Berlin Marathon (1999-2025). Pacing stability was quantified using the percentage slowdown in the second half relative to the first. "Hitting the wall" was operationally defined as a deceleration of ≥ 20%. A stratified analysis was performed to compare risks between sexes across five performance categories (Competitive to Casual). Multivariable logistic regression adjusted for age and performance category, sensitivity analyses (deduplicated subset; alternative thresholds), and fine-grained pacing metrics from 5 km splits were also conducted. Male runners exhibited significantly greater mean deceleration (10.73% ± 11.41%) compared with female runners (8.34% ± 8.91%, p < 0.001). The prevalence of "hitting the wall" was nearly double in men (17.63%) compared with women (9.66%), corresponding to a crude Odds Ratio of 2.00 (95% CI 1.97 to 2.03); after adjustment for age and performance category, the disparity strengthened (adjusted OR = 3.88, 95% CI 3.81 to 3.94). The risk disparity widened among the fastest runners: in the Competitive (< 3 h) category, male runners were approximately six times more likely to experience catastrophic deceleration than their female counterparts (1.42% vs 0.23%). The gap was stable across the 27-year archive (Mann-Kendall τ = 0.14, p = 0.33). Despite faster finish times, men demonstrate significantly less stable pacing strategies and a twofold higher crude risk of catastrophic deceleration compared with women, with the disparity most pronounced among the fastest runners. These findings are consistent with the hypothesis that behavioral and strategic factors contribute alongside physiological determinants to sex differences in marathon outcomes, and provide a quantitative basis for further investigation of the underlying mechanisms.

It is well known that the fastest elite marathon runners come from East African countries such as Ethiopia and Kenya. However, to date, there is no information available on the origin of the fastest age group (master) marathoners. This study aimed to determine the countries of origin of the fastest age group marathoners who have participated in the ‘New York City Marathon’ over the past several decades. Race data from 1,009,839 runners (626,183 male and 383,656 female finishers) who completed the ‘New York City Marathon’ between 1999 and 2024 were analyzed. Participants were categorized into five-year age groups: <20, 20–24, 25–29, 30–34, 35–39, 40–44, 45–49, 50–54, 55–59, 60–64, 65–69, 70–74, and 75 + years. The data were stratified by sex (male and female) and country of origin. The dataset was organized into five performance-based subgroups: (i) the entire dataset, including all finishers by age group and nationality; (ii) the top 100 finishers per age group; (iii) the top 30 finishers per age group; (iv) the top 10 finishers per age group; and (v) the top individual from each country within each age group. Regression analyses were conducted to explore demographic predictors of marathon performance. Participation generally increased over the study period, with temporary declines during the COVID-19 pandemic; male participation consistently outnumbered female participation, the 40–44 years age group was the most represented for both sexes, and participation was lowest in the youngest (< 20 years) and oldest (75 + years) age groups. Crucially, analyses focusing on the fastest age-group marathoners revealed clear nationality-based performance patterns. In younger adult age groups (20–39 years), the fastest average race times were predominantly achieved by female and male runners from Kenya and Ethiopia. The < 20 years age category showed comparatively stronger performances from European runners, including those from Poland, Switzerland and Italy. In the 50 years and older age groups, the best average times were increasingly recorded by runners from the United States of America, Japan, Germany and Switzerland. This shift highlights a regional transition in peak marathon performance with increasing age, from East African to European, North American, and East Asian dominance.

Individual factors related to performance in age group triathletes competing in different race distances have been explored in scientific literature. Howev

Background: Most studies investigated the pacing strategy in marathon running races considering intervals of 5 km (i.e., ~3.1 miles). However, a pacing analysis considering shorter intervals (e.g. 1 mile) might provide more insights to understand marathon pacing strategy. The present study analyzed the tracking of pacing in marathon runners considering one-mile intervals from the 4th to the 26th race mile. Methods: The data comprised male and female finishers competing in the ‘New York City Marathon’ from 2011 to 2019. Available information included the overall ranking position, average running speed, and ranking position by age, race time, year of competition, running pace, and split times per mile from the 4th (first point of measurement) to the 26th mile (race finish). Male runners were split into three groups based on their final classification (1st–3rd position; 4th–10th position; higher than 10th position). The Kruskal–Wallis test was used to compare running speed among ranking position groups. Tracking was tested through autocorrelations. The confidence interval was fixed at 95%. Results: For males, 14%, 13%, and 5% of the performance in the last mile was explained by running pace in mile 4, for the three groups, respectively. For women, about 60% of the pace variance in mile 24 was explained by the pace in mile 4. Autocorrelation results showed strong stability in running pace along the race (0.90, 0.91, and 0.76 for 1st–3rd position, 4th–10th position, and >10th position, respectively). Conclusion: Among finishers of the ‘New York City Marathon’ competing from 2011 to 2019, the performance in the last mile was correlated with the performance in the first mile studied (4th mile), regardless of ranking position. Marathon organizers could employ data-driven strategies to monitor and support runners throughout the race.

Purpose: Ultramarathon running is increasingly popular, where the time-limited races offer formats from 6 hours to 10 days. To date, the origin of the best 6-day ultramarathoners and where the fastest races are held are yet to be determined. The aim of the present study was to investigate where these runners originate from, and where the fastest races are held.Methods: A total of 8,889 race records (6,737 from men and 2,162 from women) from 3,226 unique age group runners (2,413 men and 813 women) from 54 countries from age groups 18 to 75 years and participating in 141 races held in 25 different countries between 1874 and 2022 were analysed. A machine learning model based on the XGBoost Regression algorithm was built to predict running speed based on the athlete’s age, sex, country of origin, and where the race occurs. Model explainability tools were then used to investigate how each independent variable would influence the predicted running speed.Results: Most athletes (62.5%) were from the USA, France, South Africa, Australia, Germany, and the UK. Almost 60% of the 6-day races took place in the USA and France. Athletes from Lithuania, Slovenia, and Namibia were the fastest. Ukraine holds the fastest 6-day races, ahead of Austria and Australia. The model rated the country where the race takes place as the most important predictor. Men were ~0.4 km/h faster than women except for the 75 years age group. The fastest runners were in the 35–39 years age group.Conclusions: East-European runners from Lithuania and Slovenia were the fastest in 6-day ultramarathons, where most of the races took place in the USA and France. The fastest 6-day races were in Ukraine, where the races were held as track races.

The 100-km ultra-marathon is one of the most popular ultra-marathon distances. While we have a lot of scientific knowledge, no data exist about the influence of race course characteristics and other geographical aspects, on race performance. Therefore, the aims of this study were (i) to investigate where the fastest 100-km races are held and where the fastest runners originate from, (ii) to evaluate a potential influence of specific race characteristics (i.e., influence of elevation and race course characteristics) on performance, and (iii) to assess the influence of individual athlete performance against the other investigated factors. A total of 858,544 race records (732,748 from men and 125,796 from women) from 317,312 unique runners originating from 103 different countries and participating in 2,648 100-km races held in 80 different countries worldwide between 1892 and 2022 were analyzed using several descriptive, inferential and predictive methods, including a machine learning XG Boost Regression model. We evaluated the influence on the average running speed (in km/h) of factors such as gender of the athlete, age group, country of origin of the athlete, country where the race was held, course characteristics (i.e. mountain, trail, road, or track race) and elevation (i.e. flat or hilly course). The relative effect of the individual athlete performance was also investigated through a Mixed Effects Linear model. Discounting the fact that individual athlete performance is between 3 and 4 times ahead in race speed influence compared to the other factors, the model rated elevation (0.85) as the most important variable ahead of the country where the race was held (0.07), gender (0.02), age group (0.02), the country of origin of the runner (0.02) and the course characteristics (0.02). Running on a track (9.32 km/h) was the fastest ahead of road running (8.11 km/h), trail running (6.21 km/h) and mountain running (5.74 km/h). Flat running (8.85 km/h) was faster than running on a hilly course (6.57 km/h). The fastest athletes originated from African and Eastern European countries, with Swaziland (13.15 ± 0.88 km/h), Botswana (11.61 ± 2.22 km/h), Belarus (11.10 ± 2.29 km/h), Kazakhstan (10.74 ± 3.78 km/h), and Cape Verde (10.49 ± 2.26 km/h) in the top five. Africa, the Middle East, and Europe hold the fastest 100 km races, with Botswana (12.23 ± 1.35 km/h), Qatar (12.10 ± 1.63 km/h), Belarus (11.24 ± 1.27 km/h), Jordania (11.05 ± 1.58 km/h), and Montenegro (10.63 ± 1.90 km/h) in the top five. In summary, elevation was the most important variable in 100-km ultra-marathon running ahead of the country where the race was held, gender, age group, country of origin of the runner and course characteristics. Running on a track was the fastest ahead of road, trail and mountain running. Flat running was faster than running on a hilly course. Africa, the Middle East, and Europe hold the fastest 100 km races. Common for the fastest 100-km race courses was the fact that they were mainly indoor races and/or Championships. The fastest runners originated mainly from former republics of the dissolved Soviet Union. Future studies might select the fastest 100-km race courses.

Purpose: Ultra-marathon running is highly popular, with races including distance-limited, time-limited, and multistage events. The 12-hour run is the second shortest time-limited ultra-marathon, though little is known about the origin of athletes or where these races are preferably held. Therefore, the present study investigated where the fastest 12-hour runners originate from and where the fastest 12-hour race courses are located.Methods: A machine learning model based on the XG Boost algorithm was developed to predict running speed based on athlete age, sex, country of origin, and the country where the races were held. After the model was built and trained, explainability tools were used to investigate how each independent variable influenced the predicted running speed.Results: A total of 103,334 race records of 53,700 unique runners from 69 countries participating in races held in 55 countries were analysed. The United States of America (USA) accounted for about one-third of the 12-hour race records for country of origin and country of the event, followed by Taiwan, several European countries (e.g., Germany, France, Italy, and Norway), and others from the Anglosphere (e.g., Australia, United Kingdom, and South Africa). Athletes from Lithuania, Israel, Russia, Hungary, Croatia, and Namibia achieved the fastest average running speeds. The fastest running speeds were achieved in races held in Russia, the Netherlands, Israel, Slovakia, the Czech Republic, Croatia, and Hungary. There was a positive correlation between country of origin and country of event, indicating that athletes competed mainly in their home country. Men were about 0.5 km/h faster than women on average. Most athletes were in the 45–49 age group, while the fastest runners were in the 40–44 and 45–49 age groups.Conclusions: Most 12-hour ultra-marathon athletes originated from the USA and competed in the USA. However, athletes from Lithuania, Israel, Russia, Hungary, Croatia, and Namibia achieved the fastest running speeds.