The annual edition of Ultraman Hawaii is limited to 40 athletes on an invitation-only basis and attracts participants from around the world. Participants must complete a 10-km open ocean swim, a 421-km cross-country non-drafting cycling, and an 84-km ultra-marathon. Each athlete has to be accompanied over the entire course by a support team of at least two persons. The events start around sunrise on each day. The swimming and first cycling events are held on day 1. Average water temperature for the open ocean swim is 21°C, and wetsuits are allowed. Following the swimming event, the participants have to cycle 145 km, with an elevation gain of approximately 2,500 m. On day 2, the participants cycle 276 km, with an elevation gain of approximately 2,600 m. On day 3, the athletes need to complete an 84-km ultra-marathon with an elevation gain of approximately 168.5 m. Each stage must be completed within 12 h.

Annual female and male winners (i.e., athletes with the fastest overall race time) and the annual overall top three women and men were analyzed to determine changes in the fastest overall race times and in the age of the fastest finishers during the 1983–2012 period. Participation and performance trends in the three split disciplines were analyzed only between 2001 and 2012 due to insufficient data in earlier years. The sex-related difference in performance was calculated using the equation ([race time in women] – [race time in men]) / [race time in men] × 100. The sex-related differences were transformed into absolute values prior to analysis. To calculate the mean sex-related difference in performance between the annual top three women and women, the differences for each pair (e.g., first man and first woman, second man and second woman, third man and third woman) were calculated separately and averaged. The age of the athletes was calculated by year of race – year of birth.

To optimize the reliability of data analysis, each set of data was tested for homogeneity of variances prior to statistical analysis, using Levene’s test in cases of two groups and Bartlett’s test in cases of more than two groups. Linear regression was used to determine the significance of changes in a variable (speed, age) across years. The variables between two groups were compared using unpaired t test. Statistical analyses were performed using IBM SPSS Statistics (Version 19, IBM SPSS, Chicago, IL, USA) and GraphPad Prism (Version 5, GraphPad Software, La Jolla, CA, USA). Statistical significance was accepted at p < 0.05 (two-sided for t tests). Data in the text are given as mean ± standard deviation (SD).

Between 1983 and 2012, the annual number of male finishers in Ultraman Hawaii remained unchanged (r ² = 0.03, p > 0.05), while the number of female finishers increased over time (r ² = 0.26, p < 0.01) (Figure 1). The number of female finishers increased from three women (14% of the total field) in 1984 to seven women (26% of the total field) in 2012. The largest number of female finishers (26.5%) was in the age group of 35–39 years, while the largest number of male finishers (24%) was in the age group of 40–44 years. Athletes from 15 countries accounted for 93% of all finishes. The largest number of male finishers was achieved by participants from USA (48%), followed by participants from Japan, Germany, Canada, and Brazil (Figure 2). Most of the female finishers (54.7%) originated from the USA, followed by athletes from Puerto Rico, Japan, Canada, and Brazil.

For the annual winners, the overall race time decreased across years for both women (r ² = 0.28, p < 0.01) and men (r ² = 0.14, p < 0.05) (Figure 3). The sex-related difference in overall race time decreased from 24.3% in 1983 to 11.5% in 2012 (r ² = 0.16, p = 0.04). When the top three finishers of each sex were considered, performance improved for both women (r ² = 0.36, p < 0.01) and men (r ² = 0.14, p = 0.02) (Figure 4). The sex-related difference in race time decreased over time (r ² = 0.39, p < 0.01). Figure 5 presents the change in performance for swimming on day 1 (Figure 5A), cycling on day 1 (Figure 5B), cycling on day 2 (Figure 5C), and running on day 3 (Figure 5D). During the 2001–2012 period, the swim time on day 1 (Figure 5A) remained unchanged for the annual top three women (r ² = 0.11, p > 0.05) and men (r ² = 0.02, p > 0.05). However, the sex difference in swimming performance decreased (r ² = 0.13, p = 0.04). The cycling time on day 1 (Figure 5B) showed no change for women (r ² = 0.02, p > 0.05) but decreased in men (r ² = 0.20, p < 0.01). The sex difference in cycling performance decreased significantly by 1.4% per annum (r ² = 0.28, p < 0.01). In contrast, the cycling time on day 2 (Figure 5C) decreased over time from 632 ± 50 min to 525 ± 35 min for women (r ² = 0.45, p < 0.01) and from 516 ± 20 min to 471 ± 14 min for men (r ² = 0.41, p < 0.01), while the sex differences remained stable (r ² = 0.08, p > 0.05). On day 2, women improved cycling performance by 11 min and men by 6 min per annum. For the marathon on day 3 (Figure 5D), sex difference decreased (r ² = 0.18, p = 0.01), but running time showed no change for both women (r ² = 0.01, p > 0.05) and men (r ² = 0.07, p > 0.05).

The mean age of overall finishers was lower for women with 32 ± 5 years compared to men with 35 ± 6 years (p < 0.001). The age of the annual winners increased from 28 to 47 years (r ² = 0.35, p < 0.01) for men, while it remained stable at 32 ± 6 years (r ² < 0.01, p > 0.05) for women (Figure 6). The age of the annual top three finishers increased from 33 ± 6 years to 48 ± 3 years for men (r ² = 0.32, p < 0.01) and from 29 ± 7 years to 49 ± 2 years for women (r ² = 0.11, p < 0.01) (Figure 7).

The results of the present study supported the hypothesis that the number of female finishers in Ultraman Hawaii increased during 1983–2012, while the number of male finishers remained relatively constant. The number of female finishers in ultra-endurance events generally increased over time, as is evident in Ultraman Hawaii with two female participants in 1983 and seven in 2012, as well as in Ironman Hawaii with 20 female participants in 1981 and more than 450 in 2007, respectively 6 . The reasons for the increase in female participation are not clear, but sociological phenomena such as an increased focus on physical fitness and on the positive influence of healthy living on longevity might be potential explanations 26 . For female ultra-marathoners, the strongest sources of motivation were health and personal achievement 27 . Men, on the other side, rather search for competition and try to reach high rankings. Lombardo suggested that sport began as a way for men to develop the skills needed in primitive hunting and warfare, then developed to act primarily as a competition where athletes display and male spectators evaluate the qualities of potential allies and rivals 28 . Successful ultra-endurance performance is characterized by the ability to sustain a higher absolute speed for a given distance than other competitors 29 . Hunter and Stevens 30 reported for marathoners that the greater sex difference in velocity occurring with age and lower performance was primarily explained by the lower number of female finishers compared to male finishers. These data provided evidence that a lower participation among female competitors could amplify the sex difference in running velocity above that due to physiological sex difference alone. In Ultraman Hawaii, the largest number of finishers originated from the USA, probably because long triathlons offered worldwide provide athletes from Europe and other parts of the world the opportunity to participate without having to travel such long distances 31 .

Overall race time for both sexes decreased significantly during 1983–2012, supporting our second hypothesis. The improvement in performance over time by both sexes in Ultraman Hawaii might be related to an increase in participation in ultra-triathlon events. The increasing number of athletes was probably due to improvements in training, nutrition, and experience-related pacing strategies 6 . During the last century, there has been a continuous decrease in performance times in different sports disciplines such as track and field and swimming 32 33 . Previous experience seemed to play an important role in ultra-endurance performance. Overall race time in a Triple Iron ultra-triathlon was more closely associated with personal best time in previous ultra-triathlons than with anthropometry or training 34 . Race time in a Deca Iron ultra-triathlon was also more closely related to the number of finished Triple Iron ultra-triathlons and personal best time than to anthropometric characteristics 21 . Considering these findings, the number of successfully finished Triple Iron ultra-triathlons and a fast personal best time in Triple Iron ultra-triathlon might reflect pre-race training volume and experience, where athletes with a higher training volume might compete faster in a longer ultra-triathlon than athletes with a lower training volume and less experience.

The sex-related difference in overall performance for both the top three athletes and the annual winners decreased over time. The annual top three women and men improved cycling time. The sex difference in performance decreased for day 1 annually by 1.4%. On day 2, both women and men became faster over time, but the sex difference in performance remained unchanged. Female athletes performed by 11 min faster per annum, while male by 6 min per annum. One third of the Ultraman Hawaii distance consists of cycling, so this mode of locomotion has a strong influence on overall race time. Cycling technology has improved consistently since 1981 35 , probably contributing to improved performance by both sexes in the cycling portion of the race. Based on the single top finishers and the top three finishers of each sex, men consistently performed better than women in Ultraman Hawaii.

Long-distance events were generally associated with large sex-related differences in performance 4 8 11 36 . Men were physically stronger and had greater aerobic capacity than women 37 . Men also tended to have stronger ego involvement in athletic competitions than women 38 . Female ultra-marathoners were more task-orientated than ego-orientated and set goals other than winning 27 . Deaner 39 and Deaner and Mitchell 40 argued that the ultimate benefit of studying sex differences in enduring competitiveness from an evolutionary perspective is not to support an essentialist position. The payoff, instead, should be improved in the understanding of the various factors that affect long-term achievement motivation. Deaner showed the importance of sex but also highlighted that motivation should be modulated by whether achievement in the domain yields status or resources and whether one perceives that they possess talent in the domain 39 . In particular, men should be relatively more motivated to achieve in context where their talent allows them to compete for status whereas women would be relatively more likely to compete for resources.

Previous studies investigating performance in ultra-triathlons showed that changes in male and female performance varied among the three locomotion modes. In Ironman Hawaii, for example, the sex-related difference remained stable in swimming (+0.1% per decade), increased slightly in cycling (+0.8% per decade), and decreased in running (−2.8% per decade) between 1981 and 2007 6 . Rüst et al. showed that for Ironman Hawaii from 1983 to 2012, the sex differences decreased over years for overall race time and running but not for swimming and cycling 17 . Sex-related differences in running were predicted to diminish as distance increases beyond the marathon 7 because women have greater fat stores and an improved fat metabolism compared to men and should be more fatigue-resistant than male athletes during ultra distances 41 . In Ultraman Hawaii, running performance during day 3 is the last part of this multi-stage event. During the day, the temperature may increase up to 30°C in Hawaii 42 . Presumably, the athletes might have had only little reserves left for the run after two races of racing. The training and previous experience might provide the athletes with the empirical knowledge to select the optimal pacing strategy with the ability to maintain race pace 21 . During multi-stage races such as Tour the France and Vuelta de Espana 22 23 , race intensity is rather high. Rodríguez-Marroyo et al. showed that experienced athletes modulated the intensity of each stage due to the total race duration 23 . Ironman triathletes have a special strategy during the cycling phase dependent upon the wind 43 . In a multi-stage triathlon such as Ultraman Hawaii, a specific pacing strategy is needed including also the aspect of recovery and nutrition between the stages 20 21 .

The results of the present study only partly supported the hypothesis that the average age of finishers increased over time. The age of the annual top male finishers increased, while the age of annual top female finishers showed no changes. It must be stated that the increase in the age of the male winner across years was not due to the same athlete. Several different athletes were able to win Ultraman Hawaii 24 . Considering the annual top three finishers, age increased in both women and men. Participation in master athletes in triathlons 44 and ultra-triathlons 5 44 45 increased over the last two decades, so it was not surprising that the mean age increased in finishers. Master athletes were generally defined by the age at which the world record in open elite sport peaks, typically older than 35–40 years 46 . The lack of a comparable increase in the age of female finishers over time might be attributable to the small number of female finishers over 40 years of age or to a low number of female master athletes in general. Overall, the decline in athletic performance with age was greater in women than in men 47 . The age-related decrease in maximal oxygen uptake seemed to be the most consistent contributor to the decline in endurance performance with advancing age, and age-related reduction in lactate threshold might also be important 48 . One reason for the increase in the age of peak performance could be the increasing importance in training and competition strategies 49 as the number of competitors was growing in endurance sport events 2 3 4 5 6 50 . Another reason for the increasing age of peak performance might be the popularity of Ultraman Hawaii, which attracted more athletes and especially more master athletes in recent years.

The major limitation of the present study is the small number of participants (maximum of 40 invited athletes per year) and finishers in the Ultraman Hawaii. This is especially true for female participants who have never made up more than 26% of the total group. The inclusion of individual participants who finished in more than 1 year probably did not strongly influence the results. Another limitation was the lack of information about performance-related factors such as training 34 37 , motivation 27 , nutrition 20 51 52 , fluid metabolism 53 54 , previous experience 21 34 , weather 55 , equipment 35 56 , and anthropometry 57 58 of individual participants, which could significantly affect individual performance and overall strength of the athlete. We did not check for disqualification of athletes due to doping. One athlete was disqualified in 1998 due to unsportsmanlike conduct. For practical applications, the fastest finishers improved performance, and the age of the fastest finishers increased in Ultraman Hawaii as has also been found in Ironman Hawaii 59 . However, the fastest finishers in Ironman Hawaii seemed younger than the fastest finishers in Ultraman Hawaii 59 . The upper limits in the age of peak performance in both Ironman Hawaii and Ultraman Hawaii seemed not reached for both women and men.