Performance and pacing of professional IRONMAN triathletes: the fastest IRONMAN World Championship ever—IRONMAN Hawaii 2022

Pacing during cycling and running in an IRONMAN triathlon has been investigated in only one study with elite IRONMAN triathletes. We have, however, no knowledge of how professional triathletes pace during an IRONMAN World Championship. To investigate the split-by-split speed, pacing strategies and pacing variability in professional female and male IRONMAN World Championship participants in the fastest IRONMAN World Championship ever in IRONMAN Hawaii 2022. For both cycling and running, 25 specific split times were recorded in each discipline. The best 30 men and 30 women overall were chosen from the official IRONMAN website database for further analysis. They were divided into three performance groups: Top 10, 11–20th place, and 21st–30th place. Mean speed, individual linear regressions with the corresponding correlation coefficients, and coefficient of variation were calculated to assess split-by-split speed, pacing strategies, and pacing variability, respectively. In both men’s and women’s cycling and running segments, the top ten participants exhibited faster split times compared to the slower performance groups. Notably, no discernible differences existed between the 11–20th and 21st–30th place in men’s cycling and women’s running times. Conversely, in men’s running and women’s cycling segments, those in the 11–20th place displayed quicker times than those in the 21st–30th place. In the cycling segment across all groups, men demonstrated a more negative pacing pattern (indicating an increase in speed), whereas women exhibited more consistent pacing. In the running segment, the top 10 men and all women’s groups showcased relatively similar slightly positive pacing profiles. However, men ranking 11–20th and 21st–30th displayed more pronounced positive pacing strategies, implying a more significant decline in speed over time. In terms of cycling, the variability in pacing remained relatively consistent across the three performance groups. Conversely, during the running segment, the top ten male triathletes and those in the 11–20th place displayed lower pacing variability than their counterparts in the 21st–30th position place and all women’s groups. In summary, performance and pacing were examined in professional male and female IRONMAN World Championship participants during IRONMAN Hawaii 2022. Top performers showed faster cycling and running split times, with differences in pacing strategies between sexes. The pacing was more consistent in cycling, while running pacing varied more, particularly among male triathletes in different performance groups.


Method Data set and data preparation
The race data were manually downloaded from the official IRONMAN website (www.ironm an.com) where we selected the race data from the 2022 edition of 'IRONMAN Hawaii' .The website recorded the name, sex, performance level (professional or age group athlete), country of origin, times of the split disciplines (swimming, cycling, and running), including the transition times (swimming to cycling and cycling to running).For both cycling and running, 25 specific split times were recorded.From the database, only the best 30 men and women were chosen for further analysis.They were consequently divided into three groups: Top 10, 11-20th place, and 21st-30th place.

Statistical analysis
Prior to all analyses, descriptive statistics were calculated as mean, standard deviation, minimum and maximum.The Kolmogorov-Smirnov test and visual inspection of histograms and QQ plots confirmed the data distribution normality.To assess the split-by-split speed for cycling and running, two-way between-within ANOVAs were calculated separately for men and women.The main effects of split speed (25 splits), ranking (Top 10, 11-20th place, and 21st-30th place), and their interaction split speed × ranking was calculated.
Individual linear regressions were applied to the mean split speed for each of the 25 cycling and running laps.Individual linear regressions were calculated with the corresponding correlation coefficients (r) to assess pacing strategies (i.e., r > − 0.1 < 0.1 = Even pacing profile; r < − 0.1 = Positive pacing profile; r > 0.1 = Negative pacing profile).Furthermore, two-way between-between ANOVAs were performed separately on the Fisher Z-transformed correlation coefficients 24 for cycling and running.The main effects of sex (men and women), ranking (Top 10, 11-20th place, and 21st-30th place), and their interaction sex × ranking were calculated, as well as Bonferroni post hoc test.The same two-way between-between ANOVAs were performed on the coefficient of variation (CV) to assess pacing variability.Since the CV data were expressed as percentages, data were logtransformed for the ANOVAs, and then back-transformed according to existing methods 25 .Eta squared (ŋ 2 ) was calculated for the ANOVAs where the effect size 0.01, 0.06, and above 0.14 were considered small, medium, and

Ethics approval
This study was approved by the Institutional Review Board of Kanton St. Gallen, Switzerland, with a waiver of the requirement for informed consent of the participants as the study involved the analysis of publicly available data (EKSG 01/06/2010).The study was conducted in accordance with recognized ethical standards according to the Declaration of Helsinki adopted in 1964 and revised in 2013.

Split times and speed
Descriptive data (i.e., mean, standard deviation, minimum and maximum), specifically the time needed to complete each discipline, is shown in Table 1.
The data were presented separately for sex and ranking.The importance of this specific race was in the fact that the first ten men completed the IRONMAN World Championship triathlon in under 8 h.Furthermore, 6 out of 10 women achieved a time of under 9 h.That makes this race by far the fastest IRONMAN World Championship ever.A split-by-split analysis was performed separately for men and women to investigate this race further.

Table 1.
Descriptive data of the time needed to complete each IRONMAN triathlon discipline.*Note that the transition times were outside our research focus, and consecutively, they are not included in this table.However, the total time consists of the transition times since the placement was based on that result.

Pacing strategy
To assess the triathletes` pacing strategy, the two-way between-between ANOVAs were performed on Fisher Z-transformed correlation coefficients (derived from individual linear regression) separately for cycling and running.In cycling (Fig. 3

Discussion
This study intended to investigate the split-by-split speed, pacing strategies, and pacing variability in professional female and male IRONMAN World Championship participants in the fastest IRONMAN World Championship ever in IRONMAN Hawaii 2022.By comparing pacing between the top 10, the 11-20th place, and the 21st-30th place for both sexes, we hypothesized for all performance groups that running would be more predictive of overall performance than cycling, and pacing would be positive (i.e., the speed would decrease across the split) in both cycling and running.
The main findings were as follows: The first important finding was that ten fastest women and men were faster than athletes in both cycling and running in the 11-20th and 21st-30th place.In contrast, the top 10 male IRONMAN triathletes had slower swimming times (Table 1) than other groups.However, no split times were provided for swimming, thus not allowing further investigation.For the two slower performance groups, the 11-20th were faster than the 21st-30th place in the men's running and women's cycling, implying that men would need to improve running and women cycling to enter the top twenty in IRONMAN Hawaii, and athletes who were in the 11-20th group would need to improve both running and cycling to enter to top ten in IRONMAN Hawaii.
The first important finding was that the top ten male and female athletes surpassed those in 11-30th positions in both cycling and running.In contrast, the leading male IRONMAN triathletes showed slower swimming times (Table 1) compared to other groups, with no available split times for further analysis.Among the slower performance groups, those ranked 11-20th showed better performance in men's running and women's cycling than those in 21st-30th positions.This highlights the necessity for men to enhance their running and women to improve their cycling to secure a top twenty position in IRONMAN Hawaii.Moreover, athletes in the 11-20th group would need to excel in both running and cycling to attain a top ten position in the competition.
Little was known regarding pacing in the IRONMAN triathlon 23 .More studies investigated pacing in shorter 27 or longer 28 triathlon distances.Especially, little to nothing was known regarding pacing in triathlons by performance groups.In the Sprint distance, Olympic distance, and the IRONMAN 70.3 distance triathlon races, an even swimming pacing strategy was adopted across all distances.During cycling, pacing varied for the three distances.In running, the pacing was negative in the Sprint distance triathlon and positive in the Olympic distance and the IRONMAN 70.3 distance triathlon 27 .In races covering x-times the full IRONMAN distance, the fastest triathletes spent relatively more time swimming and cycling and less time running, showing the importance of running in longer triathlon races 28 .Future studies would need to investigate differences between performance groups in professional IRONMAN triathletes regarding anthropometry, training, and equipment.This might provide further insights in the performance differences of the fastest professional IRONMAN triathletes.
2. Differences in pacing strategies during cycling and running between women and men A second important finding was that men overall paced more negatively in cycling while women paced rather evenly, whereas in running the ten fastest men and all three female groups paced slightly more positive.Men in the two slower performance groups showed more pronounced positive pacing implying that slower male triathletes might try to adopt more even pacing to improve results.
The differences between the sexes might be explained by differences in anthropometry and training in female and male triathletes [29][30][31] .Female triathletes had a lower body mass, a lower body height, a lower skeletal mass and a lower BMI compared to male triathletes 30 , but a higher percent body fat 30,31 .It is reported that female triathletes invested fewer weekly training hours on average and covered fewer training kilometers in all split disciplines than male triathletes 31 .Furthermore, swimming and cycling speed in training was higher in male triathletes 30 .However, running speed in training was higher in female triathletes 30 .Sex differences between the energetic metabolisms could also be associated with the fact that the women paced rather evenly.Women showed a greater proportional area of type I muscle fibers, economize more carbohydrates during prolonged exercise, and are more effective in the usage of fatty acids than the males, demonstrating a more uniform pacing strategy and less fatigue following long-distance aerobic exercise 32 .It has also been shown that variables of anthropometry and training were differently related to overall race times in female and male triathletes 31 .For example, it has been reported that the body fat percentage is related to the overall race times in male triathletes 31 , where average weekly training volume was not related 30 .In contrast, in female triathletes, the average weekly training volume was related to the overall race times 30,31 .Another study found that the somatotype was a strong predictor in male triathletes.For female triathletes, however, no correlation between somatotype, training, and the race could be found 29 .This highlights the lack of research on female triathletes, where anthropometrics, training intensity and volume, and racing performance seem to be important topics for future research.
Psychological differences between males and females have an effect on pacing as a result of competitiveness.Males are widely recognized for their stronger inclination toward direct competition in contrast to females 33 , a characteristic often attributed to their heightened sense of competitiveness 34 .This affects the pacing strategy in the form of a more aggressive start resulting in a steeper decline.

Differences in pacing variability between performance groups and sexes
A third and last important finding was that pacing variability did not change across the three performance groups in cycling.It should be highlighted that the athletes adopted a negative pacing strategy (i.e., an increase in speed during the race) in cycling.An interpretation of this pacing strategy from a biophysical perspective might be the elevations of the course, especially considering the decrease of elevation towards the end of the cycling race.It has been proposed that an optimal pacing strategy is increasing power in uphill sections and decreasing power when traveling downhill 35,36 .
Drafting during running also has an effect on the running performance 37 and is permitted in an IRONMAN triathlon.However, drafting is less effective in running than in cycling due to the lower achieved speed 38 .As a result, we might expect that less prepared triathletes will vary more in marathon pacing due to a slowdown caused by an inability to stay in the pack.Furthermore, frequent acceleration and deceleration can increase the risk of injury due to greater impact forces on the musculoskeletal apparatus 39 .
Particularly novel findings of this study were that the two fastest male performance groups showed a lower running variability (i.e., a more even pacing) than the slowest performance group and all three female performance groups.However, previous studies showed that women paced more evenly in marathons than men 40,41 .In contrast, differences seemed to exist between the sexes regarding pacing in marathon running since another study reported that men seemed to slow more in a marathon than women 42 .Little is known, however, regarding the sex differences in pacing in triathlon.In ULTRAMAN Hawaii, a triathlon covering three stages during three days, women paced differently than men.Performance in the fastest women decreased on day 1 but was maintained on days 2 and 3. Performance in the fastest men decreased on day 1 and day 2 but improved on day 3 43 .Future studies need to investigate reasons for pacing differences between female and male triathletes.
Evidently, the pacing variability seems not quite as important for professional triathletes in cycling, while less variable groups achieve better results in the running.In particular, slower men and all women might try reducing pacing variability to achieve better results.To date, very little is known regarding the sex differences in triathlon performance 44 and in pacing during triathlon 45,46 .Stevenson et al. investigated the sex differences in performance in the top ten female and male age group triathletes for age groups 18-24 to 60-64 years for Sprint, Olympic, and IRONMAN 70.3 triathlons during the 2009-2011 World Championships.Sex differences in overall race times were highest in both the youngest and the oldest age groups for all triathlon distances, where the largest sex differences existed for swimming 44 .Le Meur et al. 45 investigated the differences in pacing between female and male elite triathletes a World Cup Olympic distance competition with drafting during cycling.Speeds in swimming and running decreased similarly for both women and men.Male athletes were faster than women in the transition from swimming to cycling, pushing the pace harder during the swim-to-cycle transition, contrary to the women.Women showed more changes in speed due to sloping changes during cycling and running 45 .However, drafting is allowed in Olympic distance triathlon but not in the full IRONMAN distance triathlon; therefore, comparisons are difficult.A study investigating pacing in multi-stage triathlons covering the daily distance of 5-times, 10-times, and 20-times the full IRONMAN distance showed that men achieved a stable cycling performance whether they competed in 5-times, 10-times, or 20-times the full IRONMAN distance.The cycling performance influenced the subsequent running split depending upon whether they competed in 5-times, 10-times, or 20-times the full IRONMAN distance 46 .Future studies need to investigate the pacing differences between the sexes and the split disciplines more deeply.We also need to consider the course profile www.nature.com/scientificreports/ in cycling (Fig. 5) and running (Fig. 6), showing that the racecourse is not flat.Up-hill and down-hill sections might differently influence female and male cycling and running performance 8,10 .

Limitations
Τhis study is not free of limitations since aspects of food 47 and fluid 48 intake, the influence of dehydration 49 and heat 50 , mental toughness 51 and equipment 51 for IRONMAN could not be included.On the other hand, its strength is its novelty and practical applications considering that this data analysis provides insights on the pacing strategy of the fastest professional IRONMAN triathletes.

Conclusion
In summary, this study investigated split-by-split speed, pacing strategies, and pacing variability in male and female IRONMAN World Championship participants during the IRONMAN Hawaii 2022 race.The analysis revealed that the top ten performers exhibited faster cycling and running split times compared to their slower counterparts, while differences in pacing strategies emerged between men and women, with men showing more negative pacing during cycling and women maintaining a steadier pace.Additionally, pacing variability remained consistent in cycling across performance groups, whereas in running, differences were observed with lower pacing variability seen in the top ten male triathletes and those ranking 11-20th compared to those in the 21st-30th position and all female groups.These findings highlight the significance of pacing strategies in IRONMAN triathlons and suggest areas for further research into factors influencing pacing differences and their impact on overall performance.

Figure 1 .
Figure 1.Men's split-by-split cycling and running speed.

Figure 5 .
Figure 5.The course profile for the cycling split.

Figure 6 .
Figure 6.The course profile for the running split.