Sustainable Sport Development: The Inﬂuence of Competitive-Grouping and Relative Age on the Performance of Young Triathletes

: Competitive-grouping by chronological age is a common organizational strategy in competition which may unintentionally promote relative age e ﬀ ects, for the beneﬁt of older individuals within the same competitive-group, especially in young athletes. This work presents the aim of analyzing di ﬀ erences in young triathletes on their performance within each competitive group. A total of 1243 entries of both sexes, both children—13–14 years old—and cadets—15–17 years old—participated. Firstly, we identiﬁed the year in the competitive group and relative-age semester for all the triathletes who competed in a total of six seasons from 2013 to 2018. Secondly, the performance indicator was calculated in all the segments in a triathlon competition for all triathletes. The Kruskal-Wallis Test and U Mann Whitney Test was applied. It was observed that all cadet triathletes born in the ﬁrst semester of the year (S1, born in January–June) were faster; for both boys and girls. Likewise, it was observed that older triathletes who competed within the same category were faster, but only in males and for both competitive groups: children ( p < 0.0083), and cadet ( p < 0.0033). In conclusion, families, coaches and sports political agencies need a greater knowledge and understanding of the e ﬀ ects of relative age and competitive grouping to understand the important role of age in the development of sports talent demonstrated in this study.


Contextualization of Triathlon as Sustainable Sport for Personal Development
For a sport to be considered sustainable, it must contain four dimensions: its impact on the social dimension, its impact on the environment dimension, its impact on the economy dimension of the place and its impact on the institutions [1]. The triathlon could be considered a sustainable sport since: (a) there are events for both men and women, respecting gender equality; (b) it is an accessible sport for athletes with disabilities with a wide range of categories; (c) it is an outdoor sport that is practiced in a natural environment like the sea, and in the city, pausing traffic pollution during the event; (d) bicycles are used as an environment-friendly means of transport; (e) it encourages active and passive tourism, and (f) it is an Olympic sport represented by an international federation -International Triathlon Union-and over 120 national federations. However, a number of aspects that are appointed in

Competitive Grouping and Relative Age in Sports Development
In a separate issue, in order to establish a competitive grouping by political organizations, one of the basic criteria used is the year of birth [9,10]. This is a simple approach, used by entrenched systems such as education, but with some disadvantages for the optimal individual development [11]. However, for a competitive grouping to be effective, we should compare the performance of the triathletes on an equal baseline and consider the maturity differences of the triathletes according to their year of birth or even their month of birth [12]. An individual born in January is almost twelve months older than any other individual born in December of the same year; therefore, both physical and psychological advantages and disadvantages that could appear could be attributable to that [9,12]. The consequences attributed to those differences have been studied as the Relative Age Effects-RAEs-which are more prominent in younger age groups [13][14][15][16]. In triathlon, like in other sports, it is considered that groups are brought together within competitive categories with a range of two years-children (13-14 years old)-or three years-cadets (15)(16)(17) years old [2,8,17]. The adequacy of this competitive organization has not been investigated and, therefore, it is necessary to analyze the effects of the RAE on young triathletes in order to the abovementioned differences could increase significantly within competitive grouping.
These differences can be interpreted as a systematic discrimination and/or unequal opportunities for those individuals born shortly before the cut-off date of the competitive selection year [18]. On the one hand, the comparison between triathletes with maturity differences at a given time could lead to some individuals having the sensation of not having a sufficient sporting competence or ability and to feel that not all the levels of success are reachable to those individuals [19], which could decrease the self-determined motivation towards sporting practice [20] and could increase the possibilities of dropout [21][22][23]. As an example of this, we could mention a scientific study with footballers, which shows that those athletes born in the last month of the year, who were not selected for a specific competition, dropout in their career at a young age [24]; this consequence has been studied by some authors as the Galatea effect in sport [25]. Otherwise, opposite results have been evidenced in shooting sports, where dropout was similar in boys and girls without influence from RAE [26]. Although the current research is not intended to assess the dropout rate, it is considered a necessary aspect in the state of art of the issue. On the other hand, decisions made by the coach or scout and the criteria set out in the technification programs, i.e., plans offered by federations to athletes who have shown their giftedness in order to reach high performance, may be conditioned by short-term performance, giving even more importance to maturity aspects than the own potential of the athlete [2,27]. This can undermine the chances for the development of the long-term sports talent [28], which could prevent the vulnerable individuals from the deliberate practices that are needed to reach the expert performance [16,29].
In this sense, regarding the RAEs' influence on the decisions of the coaches, it has been seen in handball matches that those sub-13 male players with a higher RAE had more chances of being selected to participate in the competition than those with a lower RAE of the same category [27]. This data emphasizes the idea that older individuals have greater chances of learning within a competitive environment. In addition, this higher expectation of coaches, or even family, would benefit the most advantageous RAEs individuals, a fact that is known as the Pygmalion effect in sport [25,30]. Thus, the consequences of the RAEs come to form a loop that feeds itself back and which is always in favor of the individuals who most benefit RAEs. This fact is also studied in sport, e.g., the Matthew Effect [25], which needs to be neutralized from a professional point of view.
RAEs appear to be complex phenomena, with sociocultural antecedents combining with interindividual age, physical differences and competitive grouping to affect sport attainment [9]. Some proposals have been recommended to resolve it. Among these proposals, a change in the age-group cut-off date (e.g., from January to June), rotating cut-off dates from year to year [31], or altering age-grouping bandwidths [9,32] have been recommended.
Many studies have investigated the RAEs in young people and adults in sports (see reviews [9,12,14]). The results are mixed, depending on the age of participants, the sport analyzed, and the sex, among other factors.
RAEs have been widely investigated in team sports. Some research reports the existence of RAE in soccer [15,23,24,[33][34][35][36][37][38], basketball [39], volleyball [33,40,41], baseball [33], hockey [42][43][44] and handball [27]. Conversely, some research reports no RAE in basketball, handball, American football and rugby [33]. However, literature is relatively scarcer in individual sports. Hancock et al. [16] found RAE in female gymnasts under 15, and inverse RAE in gymnasts over 15. Nakata et al. [33] investigated many Japanese top-level sports. These authors found RAE in badminton and Ekiden (like track and field but long distance). They found no RAE in golf, horse racing and sumo. Medic et al. [45] investigated the RAE in both swimming and track and field. These authors found that RAE exists in both sports and sex, highlighting a greater effect on males and in swimming. Some research evidences that the RAE exists and is consistent in athletics. Pizzutto et al. [46] claimed that the possibility of reaching the IAAF Top ten was dependent on the relative age. Saavedra et al. [47] found that RAE exist in all categories of male. In females, RAE exist in youth and absolute categories only. These authors showed a clear influence of the month of birth in terms of performance observed, especially in lower categories [47]. Similarly, Costa et al. [48] found RAE in young elite swimmers, with stress having a greater effect on boys. Finally, RAE have been observed in international triathletes who participated in the Olympic Games [49]. These authors assessed the quartile of birth of the total sample (male and female, n = 111) and found a greater frequency in triathletes born in the first months of the year (Q1 = January-March > Q2 = April-June > Q3 = July-September > Q4 = October-December; p = 0.07). However, the authors also analyzed the semester of birth including Q1 and Q2 in Semester 1 and Q3 and Q4 in semester 2. The authors concluded that only in males, there was a higher presence of Olympic triathletes born in the first semester. Ortigosa-Márquez et al. [17] analyzed the presence of RAE in the national selection processes of Spanish triathletes in both cadet (15-17 years old) and youth (18-19 years old). They also concluded that RAE was present only in males and more significantly in cadet category. In line with these results, Ferriz et al. [8] found a greater medal achievement in cadet Sustainability 2020, 12, 6792 4 of 13 triathletes born in the first months of the year, especially gold medal winners. Moreover, RAE was also present in technification programs.

Aim and Hypothesis
Once a state of art picture is provided, more research is needed to establish how different variables affect talent development in young triathletes-age, sex, performance level, experience, etc. Therefore, this work presents the main aim of analyzing the existence of differences in young triathletes on their performance in competition within each competitive group-children and cadets-considering year and relative-age semester. Based on the results of the previous literature mentioned above, it was initially hypothesized that: Hypothesis (H1). Older male triathletes will have faster times than younger male triathletes in overall performance within the same competitive group.
Hypothesis (H2). Older female triathletes will not have faster times than younger female triathletes in overall performance within the same competitive group.

Sample
A total of 1243 entries of both sexes, both children-13-14 years old-and cadets-15-17 years old-participated in this investigation (Table 1). Valencian Federation of Triathlon in Spain was informed of the aim of the study and gave its consent for the publication of the data, anonymously, after signing a confidentiality agreement, in line with the principles of the Helsinki declaration [50]. This investigation was previously approved by the Ethics Committee of the University of Alicante (UA-2016-06-0). Table 1. Frequency of participation according to sex, competitive and relative-age semester group.  IF2  CD1  CD2  CD3  IF1  IF2  CD1  CD2  CD3   S1  69  69  52  43  26  81  100  85  69  37  S2  56  50  44  33  16  88  102  86  85  52  Total  125  119  96  76  42  169  202  171 154 89 IF1 = First season in Children; IF2 = Second season in Children; CD1 = First season in Cadet; CD2 = Second season in Cadet; CD3 = Third season in Cadet; S1 = Born in first semester (January-June); S2 = Born in second semester (July-December).

Procedure
We identified both year in the competitive group and relative age semester for all the triathletes that participated in Valencian Federation League of triathlon in Spain, one of the official Spanish leagues in which a larger number of young triathletes participate, during the previous six seasons 2013-2018-1Y2S = First season in competitive group and second relative-age semester; 1Y1S = First season in competitive group and first relative-age semester; 2Y2S = Second season and semester; 2Y1S = Second season and first relative-age semester; 3Y2S = Third season and second relative-age semester; 3Y1S = Third season and first relative-age semester.
Then, the performance indicator was calculated in all the segments that enclose a triathlon [8] except for the transitions for which data was unavailable. This performance indicator is calculated from the results in the competition, which are obtained through a chip-timing system that allows us to check a partial for each individual with a high level of precision.
In order to avoid bias in the investigation, we calculated the average value of the performance indicator of the competition for each triathlete and season, i.e., the same individual could register different entries for different seasons. Thus, this registry of the data contributes to a greater understanding of the case, due to the same triathlete being able to provide data related not only with its first year of the season-1Y2S-but also with its second year-2Y2S-of the next season.
A performance indicator is expressed from 0 to 10,000; with 10,000 being the score for the best time of the segment and, therefore, the best performance [3,8]. The rest of the triathletes' performances were calculated as a proportional part of the mentioned best time (Equation (1)). It was calculated for all the segments of a same race: swimming performance indicator (SPI) cycling performance indicator (CPI) running performance indicator (RPI) and overall performance indicator (OPI).

Statistical Analysis
All continuous variables in the data set were subjected to a normality test (Shapiro-Wilk for n < 50 and Kolmogorov-Smirnov for n > 50). The data were further subjected to a chi-square analysis and a Kruskal-Wallis Test for non-parametric samples. Firstly, chi-square analysis was used to analyze the representation of comparative groups (equal groups). Secondly, Kruskal-Wallis Test was applied to analyze differences in performance between groups. Moreover, post-hoc test through U Mann Whitney Test was applied to assess the significant differences between the groups. The level of significance was established at p < 0.

Distribution of the Sample According to Sex and Season
The distribution did not overrepresent any semester of birth in any competitive group evaluated (p > 0.05).

Performance in Triathlon According to Competitive Grouping and Relative-Age Semester
First, based only on the relative-age semester in child category, it was observed that the best results were obtained by triathletes born in first semester of the year-S1, born in January-June-for both boys (87.5%) and girls (62.5%). In addition, all cadet triathletes born in S1, i.e., males and females performed better in all the dependent variables: SPI, CPI, RPI and OPI. This analysis is based only on the percentages observed in Table 2. Secondly, we observed a statistically higher performance of older triathletes who competed within the same category, but only in males, in both competitive groups: children (p < 0.0083), and cadet (p < 0.0033).
As a result, U Mann-Whitney test and effect size (Table 3) indicated that, on the one hand, in the children category the younger triathletes-1Y2S and 1Y1S-had statistically lower performance compared to their second season peers-2Y1S and 2Y2S-in all variables, except in RPI for 1Y1S (p = 0.009; Figure 1). On the other hand, in the cadet category, it was observed that younger male triathletes-1Y2S-had statistically lower performance than their older peers-3Y1S-in all the segments of a triathlon and lower than 2Y1S and 3Y2S in OPI (Figure 2). In addition, 1Y1S had a lower performance in OPI than older triathletes-3Y1S.

Discussion
The purpose of the current study was to analyze the existence of differences on young triathletes' performance in competition within each competitive group, children and cadets, considering year and relative-age semester.
The most relevant results of this investigation showed a significant performance difference in male in favor of older triathletes in both categories, especially in children. This allows us to accept the proposed research hypothesis (H1).
Although in this investigation we are not studying the RAEs themselves but the involvement of the competitive grouping by the age of birth criteria in the performance, these results are consistent with those observed by other authors, where the girls did not show RAEs in their performance in the national youth selection process in triathlon [17], swimming [51], athletics [13,18] or rugby [52]. Also, Werneck et al. [49] observed no RAEs in adult female triathletes during their participation in the * Figure 2. Bars graph representing the results of post-hoc Bonferroni test for the performance difference depending on relative-age semester and season of male cadet triathletes (* = p < 0.0033 compared to 1Y2S; † = p < 0.0033 compared to 1Y1S).

Discussion
The purpose of the current study was to analyze the existence of differences on young triathletes' performance in competition within each competitive group, children and cadets, considering year and relative-age semester.
The most relevant results of this investigation showed a significant performance difference in male in favor of older triathletes in both categories, especially in children. This allows us to accept the proposed research hypothesis (H1).
Although in this investigation we are not studying the RAEs themselves but the involvement of the competitive grouping by the age of birth criteria in the performance, these results are consistent with those observed by other authors, where the girls did not show RAEs in their performance in the national youth selection process in triathlon [17], swimming [51], athletics [13,18] or rugby [52]. Also, Werneck et al. [49] observed no RAEs in adult female triathletes during their participation in the Olympic Games of London in 2012. It is common to find studies which show the very disparate results between boys and girls [53]. Therefore, our results allow us to accept the proposed research hypothesis (H2). Apparently, RAEs and competitive grouping shown in the previous literature are both prominent factors which influence male triathletes in order to have the chance of reaching the highest level of development of sports talent [9].
Although statistical differences attributed to competitive grouping was not observed for girls in this investigation, it should be noted that 62.5% and 100% of the best performances were observed by the female triathletes born in the first half of the year in child and cadet category, respectively. While no statistical differences were found, this high percentage may partly make the difference needed to justify the predominance of females born in the first months of the year in the selection of talent in triathlon or in the fact of achieving a medal [8]. Nevertheless, in a revision and meta-analysis carried out by Smith et al. [54], they concluded that RAEs also affects the female sports context. Indeed, these authors claim RAE magnitudes reduced with age and performance level; possibly suggestive of a declining influence of growth and maturational processes on sporting involvement, that in women happens two years before men [55]. This early maturation by girls, or even a lower level of performance [54], could partially explain the results of this study. However, it should be mentioned that further studies are needed in relation to female triathletes to establish stronger conclusions for this sport. These studies should be based on the different categories and level of triathletes.
The performance shown in competition by triathletes, especially for boys, seems not only to depend on relative age but competitive grouping. As expected, the results of this work show lower performance for younger triathletes within each competitive group and, consequently, higher performance for older triathletes, according to other recent investigations [13]. These results appear to indicate that, on the one hand, relative age could influence the performance in triathlon, as noted that 87.5% and 100% of the best performances were observed by the male triathletes born in the first half of the year in child and cadet category, respectively. Those results may support the relative age as a factor that must be considered for the identification, selection and development of the talent in triathlon [8].
In line with the previous paragraph, there was a more significant effect on the difference in performance within competitive grouping for the lowest category of athletes, i.e., children between 13 and 14 years old, as other recent studies have been claimed [51,54,[56][57][58][59]. This point seems to reinforce the importance of considering both RAEs and maturation state [60] in the competitive grouping in order for the athlete to develop to the highest level of performance [14]. Being relatively older and, simultaneously, on a higher maturity status corresponds to a substantial advantage in the characteristics that are related to performance [61].
Finally, the identification of the factors for the development of sports talent is a very complex process [11]. There are influential factors of the first level such as genes or training [62]; and second level ones such as RAEs [63]. The results shown by this study prove that for competitive grouping of young triathletes, i.e., children (two-year range), and cadets (three-year range), by political organizations could also be an influential factor for the sustainable sport development of a triathlete. So far, the results of other studies in triathlon do not give any sound conclusion to be drawn from female triathletes. However, in male triathletes it seems to be evident that there exist differences caused by competitive grouping and RAEs [8] and that this needs to be neutralized from the early stages of the sports talent development pathway [17]. Families, coaches and sports political agencies need a greater knowledge and understanding of the effects of relative age and competitive grouping, especially in male triathletes, to understand the important role of age in the development of sports talent demonstrated in this study. Thus, it is necessary to make changes in the organization of competitive groups in order to neutralize those differences that benefit to older athletes, attributed to sport policies, especially for triathletes in the smaller and close categories to peak height velocity [55], i.e., approximately 13-14 years old in males; 11-12 years old in females.

Limitations and Future Lines of Research
The study contains some limitations. Although the variable used, i.e., performance indicator in triathlon, is considered an adequate tool for testing the performance in triathlon, especially for tracking, this variable is limited to winner time. This is better than using average time of all triathletes like proposed Clotet and Perez [64] but winner time is up to that individual. As mentioned above in the materials and methods section, the performance indicator in a triathlon was calculated as the average value of the performance indicator of the competitions for each triathlete within a season. Sometimes, for some triathletes, there are two competitions, but for others there are three competitions. Future research should seek to test possible solutions to RAEs in sport, especially in triathlons. However, testing solutions to RAEs is difficult in reality because interventions could potentially be putting youths' future athlete development at risk [65]. This would also be interesting for future research by analyzing how sports development and active lifestyle (for avoiding dropout) are influenced by a coach's style affects (motivational climate, training, formation, autonomy support, etc.), and family or friends' expectations, among other things. Quantitative and qualitative research are necessary to analyze the dropout rate and its causes in triathletes at both sex and level of performance.

Conclusions and Practical Proposals
There is a subtle relative age effect in observed performance; i.e., triathletes born in the first half of the year showed a higher performance (although not statistically significant) than the same group born in the second half of the year. Moreover, older male triathletes have a significantly higher performance than younger triathletes within both children-13-14 years old-and cadets-15-17 years old. Contrastingly, female triathletes do not show the same behavior as male triathletes, that is, older female triathletes do not show a significantly higher performance. For this reason, the results shown in this work should be considered in order to rethink a fairer competition system and encourage the development of sports talent, especially when talking about triathlon. It is necessary to make a more specific competitive grouping criterion in which the athletes' ages are closer together. As a proposal, it could be appropriate to establish a separated ranking within the same race, i.e., a different classification for each group-1Y1S; 1Y2S; 2Y1S, etc. Another proposal that could be appropriate is to reserve several places for the youngest triathletes which may reduce dropout [23,53] and the loss of sports talent.