Talent identification and selection have long been of great interest to coaches, researchers, communities, and governments. Talent identification (TID) is the process of finding the most talented individuals within a specific domain in a homogeneous talented population, a process, at least in theory, that should play a major role in sport (Pion et al., 2015). Similarly, talent selection (TS) is the process of making decisions on which athletes continue to progress in each system, and which ones are removed from the system. As international competition across many sports has improved and intensified, there has been a growing importance on identifying and selecting athletes at younger and younger ages who might be capable of top performance (Williams & Reilly, 2000).

Despite being an integral part of the selection process for elite-level athletes, TID programs remain a controversial topic in research. On the one hand, sports scientists advise against early TID due to its potential in leading to burnout and sport withdrawal (Güllich et al., 2023). On the other hand, some researchers argue that early investments are necessary to nurture athletes years before they reach peak performance (Hohmann, 2009; Pion et al., 2015; Woods et al., 2016; Lemoyne et al., 2022). Numerous studies (Matthys et al., 2011; Pion et al., 2015; Norjali et al., 2017; Robertson et al., 2022; Hohmann & Siener, 2021; Chapelle et al., 2023) have convincingly demonstrated that the identification of certain characteristics in young children can provide a strong foundation for identifying those most likely to excel at the international competition level. Many countries have adopted TID and talent development (the process of nurturing an athlete to help them flourish) programs in sport to increase their success on the global stage (Balyi et al., 2013; Gulbin et al., 2014). Some specific studies have generated valuable data that outlines the factors distinguishing athletes across various levels, providing a scientific basis for talent development programs (Faber et al., 2014), as well as the underlying performance characteristics that relate to international success (Robertson et al., 2018). While this area of study is important and helpful for athlete selection and development approaches, the evidence to date, is that there is much room for improvement.

One of the major concerns with the identification process for athlete selection, is the degree of bias that informs decision-making due to the athletes’ relative age and maturity timing. In this sense, the highest ranked players are often those who are born earlier in the year and/or are early maturers, at least in some sports like football and badminton (Sweeney et al, 2023). With earlier born children outperforming their later-born peers (Jakobsson et al, 2021), biological maturity status and timing have large implications for (de)selection decisions (Hill et al, 2023). In the context of badminton, for example, this system affects all de-selected players, since they will not have the chance to further develop at the highest level. Furthermore, relative age and maturity may remain a delayed risk of dropout for the selected players. In the end, the risk to de-select all the tested players is extremely large and costly.

In addition, despite the widespread use of TID programs by sports organizations, there is no clear set of variables that consistently predicts future success (Johnston et al., 2018). Many existing models overemphasize early identification rather than focusing on the long-term development of potentially talented athletes (Abbott et al., 2005). This imbalance may lead to suboptimal outcomes and missed opportunities to support athletes with latent potential. To enhance the development of athletes, well-founded decisions are critical, especially given the substantial financial investments in elite sports. Reliable benchmarks and a valid selection process are essential to optimize these investments and to identify individuals with the highest potential for long-term success.

A clear understanding of key priorities is needed to develop long-term plans for young athletes, rather than pushing them towards early success at the risk of burnout or attrition in sports. Importantly, many characteristics and skills can change through training, maturation, or good coaching. The criteria for elite performance appear to be idiosyncratic and may make up for deficiencies in a specific area and with strengths in another (i.e., compensation) (Simonton, 1999). Striking the right balance between broad-based development and targeted investments remains one of the core challenges in the field of TID. Ultimately, the effectiveness of TID programs depends on the implementation of reliable benchmarks that can accurately assess potential over time.

There are multiple strategies for determining reliable benchmarks, however on sucuh tool is the SportKompas. Characterised by a generic approach of anthropometric, physical, and motor competence measurements, the SporKompas has been recognized for being a helpful tool for determining benchmarks for a wide variety of sports (Pion, 2024). It offers possibilities to better understand the potential underlying performance characteristics of a given sport and allows for the tracking of young athletes across their developmental pathway. When comparing athlete data with benchmarks for heterogeneous populations, generic tests can be applied for the detection of the superior ‘movers’ in schools. When comparing an athlete’s testing results with sport-specific benchmarks, this approach can help with the identification of superiorly performing athletes in clubs and sports associations.

Despite the SportKompas’ potential benefits, there are limitations to establishing benchmarks. The common practice of placing children into age groups for sport will still acutely benefit those who are more developed physically, emotionally, and cognitively. Those born later in the year appear to be at a disadvantage because they are less developed than their peers (Malina et al., 2015; Cumming et al., 2017; Pion, 2024). Furthermore, it is difficult to copy a champion’s profile. Collecting data from champions and drawing causal conclusions comes with limitations.

One such strategy to minimize the impact of those limitations, is to apply a ‘sliding benchmark’ approach to selection. To minimize the impact of an athlete’s chronological and biological age-related, a sliding benchmark approach can be used to control for maturity and chronological age. For example, two 12-year-old boys can differ by nearly an entire year based on when their birthdays are relative to a cut-off date for selection. The difference of one year is a significant advantage requiring recalculation to the appropriate proportions. Moreover, the difference in biological age (maturity) also results in a physical advantage. The sliding benchmarks method can be compared to an equaliser to adjust music output; it is possible to slide the raw test scores along the benchmarks of both the age group and the maturity group to obtain adjusted scores for chronological and biological age. To compute these sliding benchmarks, the dataset must be recalculated for the younger and older players in the cohort as well as for the less mature players and the more mature players.

To date, however, there is no known study that empirically investigates a sliding benchmark approach within a high-performance sport context. To address this, the present study aims to critically examine the impact of a sliding benchmark approach in neutralising performance advantages and disadvantages due to relative age and maturity status. Specifically, the study explores the possibility of applying the sliding benchmark approach to a selected group of elite young badminton players. The research question is twofold, 1) is it possible to successfully draft 12-year-old late mature boys, born later in the year when applying sliding benchmarks? 2) what is the predictive value when looking years ahead?

Basic descriptive statistics (Table 1) were used to benchmark the 12-year-old boys (n=33)

The individual results based on age-group benchmarks for physical (x-axis) and motor performance (y-axis) were plotted in the theoretical selection model proposed by Baker et al. (2018). The figure visually represents which players can be selected (Figure 1).

Figure 1 Individual scores for physical performance and motor performance based on age group benchmarks for 12-year-old boys (badminton), plotted in the theoretical selection model according with Baker et al. (2018).

Players n° 1; 2; 3; 4 and 20 are categorized as obvious talented with above-average performance; players n° 5; 6 and 10 are high potentials that meet performance standards.

Unfortunately, RAE and maturity distort the results in Figure 1, since relatively younger players are disadvantaged by the age group benchmarks (Table 2).

Figure 2 Individual scores for physical performance and motor performance based on age group benchmarks for 12-year-old boys (badminton), plotted in the theoretical selection model of Baker et al. (2018). Red dots PAH >89; green dots PAH <85, yellow dots 85> PAH <89.

When sliding the sum scores according to the relative age and maturity status (%PAH), the advantages and disadvantages become more visible, and the selection looks different (Figure 3).

Figure 3 Sliding benchmark approach for 12-year-old boys (badminton).

Figure 4 Individual scores for physical performance and motor performance adjusted with the sliding benchmarks for 12-year-old boys (badminton), plotted in the theoretical selection model Baker et al. (2018). Red dots PAH >89; green dots PAH <85, yellow dots 85> PAH <89.

Six years after the selection process, it was checked which players are still part of the national team. 9 players out of 33 are still competing for BAM.

Figure 5 Active players after a 6-year development period. plotted in the theoretical selection model Baker et al. (2018). Red dots PAH >89; green dots PAH <85, yellow dots 85> PAH <89.

This study sought to explore the value in using sliding benchmarking to help neutralise the advantages and disadvantages that certain athletes have due to relative age and maturity status (Pion, 2024). Performance benchmarks from the badminton test battery related to the SportKompas are an objective approach to selecting high potentials and to support coaches’ decisions, albeit that the age group benchmarks are biased for relative age and maturity (Table 1). The relative younger boys (< 12.20y) are overall 3% smaller, and the relative older boys (> 12.60y) are 2% taller compared to the measurements for the complete cohort. Late mature boys (PAH <85%) are 5% smaller, and early mature boys (PAH > 89%) are 8% taller. Using these benchmarks within this age group, is making the mistake of comparing apples to oranges. When comparing Figures 2 and 4, it appears that the late mature (red dots) shift to the left, and the early mature (green dots) to the right. Comparing within the age group without considering RAE and PAH clearly shows why 12-year-old late-maturing boys, born later in the year can hardly be selected for the Malaysian U13 Badminton Organisation (Figure 2). When applying the sliding benchmarks approach, the selection probability increases (Figure 4).

Our findings revealed that relative age and maturity distort the selection process. In this study the coaches made a preliminary selection of U13 players (n=33) and the sport scientists used the data of the badminton test battery to support their decision for a final selection of nine players each. Shifting the benchmarks reveals advantages and disadvantages. Figure 3 shows what the disadvantage for late-maturing boys (PAH<85) and the advantage for precocious boys (PAH>89), combined with an estimate of the disadvantage for relatively younger boys (>12.20y) and the advantage for relatively older boys (>12.60y) looks like.

Younger and less mature players are being overlooked during the selection procedure, since their morphology, physical fitness and motor capacities are less developed than their peers (Cumming et al., 2017; Mann et al., 2017). Talent identification takes place at a certain moment in time and so far, the objective measurements refer to age group benchmarks. It is important to highlight the distinction between identification and development of high potential athletes.

For decades, RAE and maturity have a huge impact on ‘here and now’ assessments of young athletes for identification and selection purposes (Barnsley, 2025). In any case, one must consider adjustments for RAE and maturity whether to selection contains objective or subjective criteria? This is one of the first studies to show that it is possible and feasible to adjust strategies by using sliding benchmarks. In this respect, the article is also a model for many other sports. Although it is important to acknowledge some limitations of this study. Firstly, the sample size was small. This is mainly because the larger population of elite badminton players is relatively small, and the authors feel lucky enough to take part in the selection process of a very successful cohort of players. To provide context on the size of the sample, in 2018, 33 players were screened. The dropout rate of 73% is high (24/33 players). On the other hand, the 9 remaining players who are still members of the national team achieve exceptional results and became national U21 champions, international match silver, 2 BWF junior World Champions and 2 semi-finalists at the junior World Championships (Figure 5). The decision to include two late mature players with average physical scores but superior motor competence during the badminton draft in 2018 was successful. Without the sliding benchmark intervention, they would never have been selected. These players benefited from a specific development in the elite sport school and became junior world champions six years later.

Second, the sum scores for the physical and motor performance test were not weighed. With the data of a small sample, it is difficult to build accurate prediction models, but in the future, it will be possible with the big data collected during the talent identification sessions. Nevertheless, the benchmarks for age groups and especially during the growth spurt are biased and therefore the results should be equalised. A coach can equalise the (dis)advantage due to relative age and maturity in the same way as equalising the sound of the radio with a bit more bass or treble. To be able to make good predictions in the future, it is important to collect data year after year and to keep adjusting the benchmarks.

Morphology, physical fitness, and motor competence are much more common to measure than social and psychological characteristics. In this badminton identification tool only morphology, physical fitness and motor competence are included. In a holistic approach, also the technical, social, and psychological development should be considered to assess the potential of youth athletes (Faber et al., 2016; Nijenhuis et al., 2024; Schoof et al., 2024; Teunissen et al., 2024).

Coaches’ decisions on talent selection in sport can be approached from two angles. On the one hand, a more objective approach is possible, in which coaches apply a multi-faceted formula for scoring as a means of predicting their future success. On the other hand, a subjective approach is possible, also known as the coach’s eye, in which these professionals select or de-select athletes based on their personal observations and impressions (Bar-Eli et al., 2024). Both approaches are often seen as complementary. Selecting talented players is often a matter of feeling and expertise. Although, in the future, accurate prediction models, based on a combination of observations and performance tests can be applied to support the selection of the next cohorts.