Sporting governing bodies and professional clubs presently invest considerable resources to identify and develop youthful prodigies into hopefully becoming tomorrow's exceptional athletes. To identify and differentiate “adolescent potential,” once per year trials, subjective coaching/scout assessment, and a range of anthropometric and fitness testing measures are often applied. However, the benefit of these approaches has been recently questioned because they often fail to (a) consider maturation differences between youth athletes, (b) capture the all-round nature of sport contexts (e.g., psychological and physical), and (c) have shown a low ability to forecast who will become an adult elite athlete (4,18,27).
Maturation, defined as the timing and tempo of progress toward the adult state, can vary considerably between individuals (11). Many studies show that unsteady anthropometric and fitness development typically occurs around 12–15 years in boys and 11–14 years for girls (2,10,11,22,23). Higher chronological age (years) and relative age (months within a year) increase the possibility of entering and progressing through maturation early compared with age-grouped peers. This also adds to the potential differences between individuals on anthropometric and fitness measures (3,15). These changes can result in greater physical capacities such as aerobic power, muscular strength, endurance, and speed (16,19,24,28,29). So in sports where such qualities are required (e.g., soccer (12,13) and rugby league (5,6), they provide immediate physical performance advantages.
Because all youth will enter and progress through maturation, it follows that later maturing (also likely to be relatively younger) individuals could “catch up” on anthropometric and fitness measures in later adolescence (e.g., see Ref. (9)). To illustrate, recent research (25) presented data on 3 rugby league players, selected to a talent development program, were tracked over a 2-year period (i.e., aged 13–15) and then compared with a broader sample of age and skill-matched players (N = 1,172) using standardized z-scores. Z scores identify whether a particular score on a measurement is similar, below, or above the mean of all measured scores. In research, they can help identify how far away a score is from the mean. Some studies may examine particular scores within a sample; those scores at close or extreme ranges from the mean; or may examine how scores change when the means of broader sample change over time. Using standardized z scores, cases demonstrated differing initial profiles at the Under 13 stage, but then also showed different developmental trajectories over a two year period. For instance, a later maturing (and relatively younger) player improved their anthropometric (e.g., height = +9.2%) and fitness (e.g., 60-m sprint = −14.9%) characteristics more than the earlier maturing (relatively older) player (e.g., height = +2.0%, 60-m sprint = −0.7%) in the same period.
Their analysis (25) compared case values for a given measurement (e.g., height, body mass, 30-m sprint) against a mean value calculated from all players measured in a broader cohort (i.e., under 13s–15s included). However, this raises the question of whether this is the most accurate and sensitive approach to detect and compare player development. Therefore, in part 1 of this study, we examined the hypothesis that comparing cases against mean values from “across annual-age categories” may falsely increase the observed deviations in z-scores. Because of a given case player being potentially more or less different from the broader sample at particular time points, a “per year” calculation was tested. In other words, cases were compared against mean values of their peers at each age category (i.e., under 13s, 14s, and 15s).
In part 2, the aim was to verify if anthropometric and fitness characteristics previously presented (25) remained accurate when using the “per year” calculation. Compared against the age- and skill-matched broader sample, changes in anthropometric and fitness measures of 3 new youth rugby league players were examined. We hypothesized that our adjusted analysis would show that even within a relatively similar sample of players, (a) developmental variability would be apparent, (b) developmental changes were still feasible within and over the 2-year period, and (c) relatively later maturing players could show a reduction or negation of anthropometric and fitness differences in later adolescence. Confirmation of these predictions would strengthen the argument for long-term monitoring of “adolescent potential” beyond maturation, compared with one-off “pre-mature” assessments and (de)selection, if long-term athlete development is the overarching goal of sport systems.
Experimental Approach to the Problem
This study (re)investigated variation in the development of anthropometric and fitness characteristics of (a) 3 original (see Ref. (25)) and (b) 3 new youth rugby league players. A case study approach was applied with measures compared against a broader cohort of age and skill-matched players over a longitudinal period (i.e., 2 years, aged 13–15). The UK's Rugby Football League (RFL) used a talent identification and development model, named the Player Performance Pathway, from 2001 to 2008 (see Ref. (26)). Each year, regional representative selection of players occurred at the under 13s, 14s, and 15s annual-age categories. Anthropometric and fitness testing were undertaken on all players selected to regional representative squads. Between 2005 and 2008, 1,172 anthropometric and fitness assessments were conducted. Eighty-one players were selected to the pathway on 3 consecutive occasions (i.e., under 13s in 2005, under 14s in 2006, and under 15s in 2007). This data set thus contained both longitudinal and cross-sectional data. This permitted consecutively selected case players (and their characteristics) to be examined and compared against the broader sample to assess differing developmental trajectories.
Whether relating to part 1 (re-analysis) or part 2 (i.e., new case players), case players were deliberately identified with different maturational statuses, relative ages, and playing positions. Maturation was determined by years from peak height velocity (YPHV) in accordance with published procedures (14). For relative age, player's birth dates were categorized to reflect their birth quartile (i.e., Q1—Q4), with 1st September acting as the date for creating annual-age groups. That is, Q1 = birth dates between September and November; Q2 = December and February; Q3 = March and May; and Q4 = June and August. Playing position was categorized into 4 subgroups (i.e., “Outside-Backs,” “Pivots,” “Props,” and “Backrow”), as used in previous rugby league research (e.g., Ref. (21)).
For part 1 of the study, anthropometric and fitness data on the 3 players (i.e., players 1, 2, and 3; Ref. (25)) were extracted and taken forward for reanalysis (see data analysis section). For part 2, 3 new case players were identified. Player 4 was a relatively older (age = 13.87 years; Q1), “earlier maturing” (YPHV = 0.67 years) “Prop.” Similarly, player 5 was a Q2 (age = 13.64 years) “average maturing” (YPHV = 0.04 years) “Backrow.” Player 6 was a relatively younger (age = 13.11 years; Q4), “late maturing” (YPHV = −1.69 years) “Outside-Back.” Although data in part 1 represent a secondary data analysis, all original procedures (described below) were approved by a University Ethics Committee. All players and parents provided written informed consent before participating in any testing.
Anthropometric and fitness testing on all case players and the broader sample of players were conducted once per year at the same time of the day (i.e., early evening) and year (i.e., July) on each occasion. Assessments were conducted on 3 consecutive years (i.e., under 13s, 14s, and 15s). Before testing, all participants were instructed to refrain from strenuous activity 48 hours before testing and to consume their normal pretraining diet.
Height and sitting height were measured to the nearest 0.1 cm using a Seca Alpha stand. Body mass, wearing shorts only, was measured to the nearest 0.1 kg using calibrated Seca alpha (model 770; Seca, Birmingham, United Kingdom) scales. Sum of skinfold thickness was determined by measuring 4 skinfold sites (i.e., biceps, triceps, subscapular, suprailiac) using calibrated Harpenden skinfold callipers (British Indicators, West Sussex, United Kingdom). Skinfold procedures were in accordance with the recommendations by Hawes and Martin (7). Intraclass correlation coefficients (ICCs) and typical error measurements (TEMs) for reliability of skinfolds were r = 0.954 (p < 0.001) and 3.2%, respectively, indicating acceptable reliability based on established criteria (i.e., >0.80; Ref. (8)).
Maturation (Age at Peak Height Velocity)
To ascertain maturational status, an age at peak height velocity (PHV) prediction equation was used (14). This prediction method used a gender-specific multiple regression equation including height, sitting height, leg length, body mass, chronological age, and their interactions. The YPHV was calculated by subtracting age at PHV from chronological age.
Before fitness testing, a standardized warm-up was conducted, and all players received full instructions of the tests. For each assessment, the highest value of 3 trials was used. Lower-body power was assessed using the vertical jump test (cm) and a Takei vertical jump meter (Takei Scientific Instruments Co. Ltd, Japan). A countermovement jump with hands positioned on the hips was used. The ICC and TEM for the vertical jump were r = 0.903 (p < 0.001) and 2.9%, respectively. A 2-kg medicine ball (Max Grip, China) chest throw measured upper-body power (20). Players attempted to throw the ball horizontally as far as possible (measured to the nearest 0.1 cm) while seated with their back against a wall. The ICC and TEM for the medicine ball chest throw were r = 0.965 (p < 0.001) and 0.6%, respectively. Running speed was assessed over 10, 20, 30, and 60 m using timing gates (Brower Timing Systems, IR Emit, Draper, UT, USA). From a standing start 0.5 m behind the initial timing gate, players started sprints in their own time. Times were recorded to the nearest 0.01 s. The ICC and TEM for the 10-, 20-, 30-, and 60-m sprints were r = 0.788 (p < 0.001), r = 0.852 (p < 0.001), r = 0.899 (p < 0.001), and r = 0.924 (p < 0.001), and 8.4, 4.5, 3.3, and 2.3% respectively. Change of direction speed was assessed using the agility 505 test. Players were positioned 15 m from a turning point with timing gates positioned 10 m from the start point. Players accelerated from the starting point, through the gates, turned on the 15-m line, and ran as quickly as possible back through the gates (5). Three alternate attempts on left and right turns were used, with times recorded to the nearest 0.01 s. The ICC and TEM for the agility 505 left and right were r = 0.823 and r = 0.844 (p < 0.001), and 3.5 and 3.1%, respectively. Estimated maximal oxygen uptake (V[Combining Dot Above]O2max) was assessed using the multistage fitness test (17). Players were required to run 20-m shuttles keeping in time with a series of beeps in which running speed progressively increased until they reached volitional exhaustion. Regression equations were used to estimate V[Combining Dot Above]O2max from the level reached during the test. The ICC and TEM for the multistage fitness test were 0.90 and 3.1% (6).
For part 1 and part 2 of the study, individual anthropometric and fitness profiles were first generated for each case player using z-scores. z-scores were calculated by the formula (x − μ/σ), where x is the raw score, μ is the mean of the broader sample of players, and σ is the SD of that same group of players. z-scores thus positioned a case individual, on any variable, against the broader sample means and their distribution at each time point in data collection. This allowed us to detect change in all anthropometric and fitness measures over time, relative to the broader sample of players. z-scores of −3, −2, −1, 0, 1, and 2 were calculated for each measurement at each annual-age group (i.e., under 13s, 14s, and 15s) to represent mean and SDs of the broader players at each respective year (Table 4). Note that this method is different to those previously used (25), which calculated comparison values on the basis of an average calculated across the year groups (i.e., under 13s–15s inclusive). For example, z-scores for body mass at the under 13s age category were −2 (41.7 kg), −1 (52.2 kg), 0 (62.7 kg), 1 (73.2 kg), and 2 (83.7 kg); whereas at under 14s, they were −2 (49.2 kg), −1 (59.7 kg), 0 (70.2 kg), 1 (80.7 kg), and 2 (91.2 kg), respectively. Estimates that lay between these z-scores were reflective of decimal place. With means of the broader sample acting as a comparison point (i.e., 0 in terms of a z-score), individual cases could then be positioned relatively for each year, and then be descriptively evaluated using tables or radar graphs.
Specifically for part 1 of the study, z-score values for the 3 case players—using an “across age category” calculation—were extracted from previous findings (25) and compared with z-score values from the newly deployed “per year group” calculation. Then, a degree of change (DOC; “across age category”—“per year” z-score) per variable was calculated. These were combined to provide an overall mean DOC across all variables (Tables 1–3). Paired t-tests were then conducted on the mean DOC to help determine whether the method of calculating z-score values affected the overall assessment of change and development in case players.
For part 2, longitudinal profiling on 3 additional new case players relative to the broader sample of players was conducted. Their raw anthropometric and fitness characteristics data (Table 4) were converted to z-scores applying the “per year” calculation method, and plotted onto radar graphs (Figures 1–3). This allowed descriptive comparison to the broader sample and case comparisons.
Tables 1–3 (players 1, 2, and 3, respectively) show the “across age category” and “per year” z-score anthropometric and fitness profiles (annually and longitudinally) for the 3 original case players. The DOC and overall mean DOC values are also reported.
At the under 13s age category, the “per year” method of z-score calculation, compared with the “across age category” (i.e., under 13s–15s) method, appeared to reduce the general degree of departure away from cohort mean values for all 3 players, respectively (mean DOC = 0.44; 0.41; 0.44; p < 0.001). This tendency was repeated at the under 15s age category with z-score values again reduced back toward broader player sample means (−0.40; −0.37; −0.0.43; p < 0.001, respectively). At under 14s, there was less indication of an influence on z-scores, with anthropometric and fitness variables showing minimal change (p > 0.05); with only player 1's mean DOC affected (p ≤ 0.05).
Table 4 shows the mean (±SD) anthropometric and fitness characteristics of the broader player sample at each annual-age category (i.e., under 13s, 14s, and 15s). It also shows individual characteristics for case players at each age category. Figures 1–3 (players 4, 5, and 6, respectively) show the anthropometric and fitness z-score profiles annually and longitudinally when plotted against the “per year” values of the broader sample of players.
Cases Compared With Broader Player Sample
In contrast to the broader sample at under 13s, player 4 was early maturing, relatively taller, and heavier. Between the under 13s and 15s age categories, the YPHV did not seemingly change, and minor z-score reductions in sitting height, height, and body mass (e.g., 1.8 to 1.5) were apparent. This suggests that the broader sample was growing relatively more in the same time period. Sum of 4 skinfolds (−2.4) also did not relatively change over this period, remaining very high throughout. Fitness characteristic z-scores did relatively vary though, and while MBT scores were above average, other characteristics were average or below at the under 13s. They remained that way (e.g., vertical jump) or deteriorated across the 2 years (e.g., 10- to 60-m sprint times; V[Combining Dot Above]O2max) (Figure 1).
Across all measures, a more “average” and rounded profile is apparent relative to the broader sample across under 13s–15s (Figure 2). The YPHV, body mass, and sum of 4 skinfolds remained stable at 0, −0.3, and 0.5, respectively. This suggests development in-line with the mean (or better) of the broader sample. From a slightly below average position at under 13s, height (−0.5 to 0.2) and sitting height (−0.3 to 0.1) showed relative improvements across the 2 years. In terms of fitness, most measurements fluctuated generally around the 0 z-score, or slightly below average (i.e., 0 to −0.5) at the under 13s time point. However, indications of relative but minor improvements in fitness performance over the 2 years are apparent (e.g., V[Combining Dot Above]O2max = −0.5 to 0) and are most evident for sprint speed (e.g., 30-m sprint: −0.3 to 0.5) (Figure 2).
Compared with the broader sample, player 6 can anthropometrically be described as a later maturing, smaller, and lighter player (YPHV = −2.9, height = −1.4, sitting height = −2.7, body mass = −1.1) at the under 13s time stage. These measures remain well below average from under 13s to 15s, intimating that he was a late maturer (e.g., age at PHV estimates = 14.6–14.8). For fitness characteristics, player 6 likewise performed below average on all measures (e.g., vertical jump = −1; agility 505 = −0.5) at the under 13s stage. However, across the under 14s and 15s age groups, incremental fitness improvements were made. Scores improved from −1.0 to 0.3 on 10 m, −0.9 to 0.6 on 30 m, and −1.2 to 0.5 on 60-m sprints; −1.0 to 0.1 on vertical jump; −0.5 to 0.0 on agility 505; and −0.4 to 0.1 on V[Combining Dot Above]O2max. Overall, an improving physical fitness trajectory is highlighted from under 13s to 15s when compared with the broader sample (Figure 3).
Age and Maturation
Table 4 illustrates that player 6 was relatively younger and later maturing (see age at PHV and YPHV) compared with player 5; whom was likewise chronologically and biologically younger than player 4. Although within the same annual-age group, maturational differences between players 6 and 4 during under 13s can be estimated as being approximately 2.36 years.
For height, player 4 was over 10 cm taller than player 5, and over 19 cm taller than player 6. Differences were apparent for the degree of change (DOC) in height and sitting height from under 13s to 15s, with player 6 increasing height the most in the period (i.e., player 4 = 3.3 cm; player 5 = 13.2 cm; player 6 = 15.7 cm). To add, the percentage of predicted height indicated that while player 4 was taller across under 13s–15s, he had almost attained his final adult height (i.e., % of predicted height at under 14s = 97.7%). In contrast, players 5 and 6 had lower values (i.e., player 5 = 96%; player 6 = 90.2%) at the same time point, indicating more expected growth in the future. Figures from predicted height also suggest that player 6 would go on to be a slightly taller individual, and that all 3 players would possibly—at adult height—be within 3 cm of each other (Figure 4).
For body mass, while all case players showed increases over the age groups, player 4 was over 22.6 and 29.1 kg heavier, respectively, than players 5 and 6 at the under 13s stage. This was partially reduced at under 15s (player 5 difference = 19.7 kg; player 6 = 19.8 kg). Similarly, player 4 exhibited consistently greater body fat across the same period (i.e., sum of 4 skinfolds—under 13s–15s = 84.6, 77.5, and 76.5 mm) compared with players 5 and 6 (e.g., player 5 under 13s–15s = 30.9, 34.7, and 31.1 mm), who over the same period maintained or reduced their sum of skinfold scores.
Vertical jump performance consistently improved across the 2 years for all 3 players (i.e., players 4 and 6 = 9 cm; player 5 = 7 cm). Similar jump heights were attained at under 15s. For medicine ball throw, player 4 threw almost 1 m further at the under 13s stage and similar differences were still apparent at the under 15s age category. In terms of sprint times, player 5 was generally quicker across the 10–60 m distances at under 13s. Over the 2 years, improvement was evident in all players; however, greater improvement was made in players 5 and 6 compared with player 4 (e.g., 30-m sprint—player 4 = −0.18 s, player 5 = −0.46 s, player 6 = −0.59 s). From similar starting points at under 13s (2.60–2.69 s) in the Agility 505, players 5 and 6 (−0.22 s) made better improvements while player 4's agility performance slightly deteriorated (i.e., 0.1 s). Finally in terms of V[Combining Dot Above]O2max, player 4 illustrated the lowest initial values and made the smallest incremental change from under 13s to 15s (i.e., 41.1–42.1 ml·kg−1·min−1). In the same period, both players 5 and 6 improved by 5.6% (45.2–50.8 ml·kg−1·min−1).
This study reexamined and presented evidence demonstrating differing developmental changes in anthropometric and fitness characteristics of youth rugby league players. Strengths of the study lay in the use of case profiling relative to a broader age and skill-matched sample on many anthropometric and fitness measures over a 2-year period (i.e., under 13s—15s). This approach revealed differing and unique case player trajectories that would not normally be identified; instead remaining hidden amongst the many “one-off” athlete assessments' or coach observations.
Specific findings illustrate that the method for comparing (i.e., “across age categories” vs. “per year” calculation) cases against broader player sample affected z-score values. An “across age category” method may falsely inflate (or deflate) z-score values. This suggests the possibility of greater perceived anthropometric and fitness changes over time, than actually would occur using the “per year” method. This latter method—at the under 13s and 15s categories—reduced the degree of departure away from extreme z-score values (i.e., −3 to 2) and back toward broader sample means (i.e., z-score of 0). Plotted on radar graphs, this generates less z-score dispersion from 0 for case players. When viewed over the long-term, the degree of likely (or potential) development change in anthropometric and fitness terms would appear less dramatic or severe than previously reported. When practically interpreted (e.g., by a strength and conditioner), this may affect perceptions of what appears to be different, normal or changing in athlete developmental terms. Still, the potential for developmental and trajectory change over a 2-year period should not be discounted. Part 2 of the study verified this assertion.
Using the “per year” method for case-broader sample comparisons, descriptive findings and radar plots verified hypotheses that even within a relatively homogenous larger sample (a) developmental variation was apparent, (b) developmental changes occurred within and across a 2-year period, and (c) that it remained possible for a relatively later maturing player to more rapidly develop beneficial anthropometric and fitness characteristics over a 2 year period. In alignment with previous research (27), these findings help highlight the present limitations in applied practice. Notably, the reliance on one-off player assessments before or during maturation and early (de)selection in athlete development programs.
Supporting previous assertions (e.g., Ref. (1)) that human development does vary in inconsistent ways, case and broader sample comparisons showed relative age, maturational, anthropometric, and fitness differences between players at the under 13s stage, as well as unique change over the 2 years of tracking. For example, bar sum of skinfolds and estimated V[Combining Dot Above]O2max, player 4 reflected a more mature and better performing (in fitness terms) individual at the under 13s age category, suggesting “good (talented) potential” for the future. Yet across the next 2 years, the trajectory of player 5 is more accelerated in preferred anthropometric (e.g., height) and fitness (e.g., sprint speed) terms. Meanwhile, player 4 maintains or regresses on the same measures over the same time period. Player 6 also demonstrates improving fitness development. At under 15s, players 5 and 6 now possibly reflect greater “athlete potential” than player 4 despite being less mature. Further, based on these profiles, their future trajectory would appear more positive. Together, these findings propose that a broader pattern may be apparent. That adolescents who demonstrate advanced anthropometric and fitness characteristics at an earlier stage of adolescence may not (or to such an extent) improve on these attributes throughout adolescence and into adulthood, thus not maintaining their initial advantages (9,25).
Because developmental change was detectable, even among what may be considered a relatively similar group of players (i.e., age-matched “representative regional level”), and within a relatively short period of time (i.e., 2 years), these findings have significance. If present here, it first indicates that patterns of anthropometric and fitness change exist and are likely correlated with the stage of maturation. Second, as many youth athletes in team sport contexts are selected on the basis of their maturity-associated advantages, it implies that developmental variation may be wider still in different skill groups of rugby league players, as well as wider nonsporting adolescence. Compared with cases presented in this study, the potential rapidity of developmental change may be fast in some cases and generally slower in the broader population. But if cases of “later maturing nonselected athletes” were tracked over time and had exposure to appropriate training, then they too may also demonstrate “good potential” at later time points (e.g., under 16s).
The case study approach and selection bias can also be considered as limitations in this study. While acknowledged, case analysis should be seen as an appropriate research design to examine different and changing athlete development trajectories. To help address such concerns, we used a large age- and skill-matched reference sample to ascertain “normative” baseline values and guide case evaluations. In terms of selection bias, we deliberately identified variable cases. If athlete cases were examined randomly, then it is likely that a “more average” (e.g., within one SD, or z-scores within +1 to −1) player profile and trajectory would be illustrated. Although the majority of players may be less diverse or changeable in their development, this does not mean that variability and change do not occur. In fact, trajectory change may generally be more detectable over a longer time period (i.e., occurring at a slower pace). Unfortunately, however, our data only spanned the under 13s–15s. On-going research will need to assess the degree of potential developmental change at later stages of adolescence and beyond (i.e., 15–20 years of age). Determining if, and how, developmental trajectories generally converge or widen (e.g., reduction in sum of skinfolds associated with improvements in fitness parameters) and the number of athletes who follow these trajectories will prove valuable.
Based on present findings, several applications can be considered for strength and conditioners, coaches, and development systems. It is essential for strength and conditioning coaches to undertake individual and longitudinal player assessments to monitor progression, and understand developmental processes. Strength and conditioning professionals can then consider such information so as to better inform their own decision making, as well as that of sport coaches and administrators. Athletes can be monitored over the long-term using z-score and radar graph techniques to assist interpretation. This will help in showcasing the diverse and changing trajectories of athletes in adolescence. For accuracy in monitoring, it is advisable to consider the method of z-score calculation by “age and maturational stage matching” against a broader sample of players.
Strength and conditioning professionals can also assist in the identification and development of players. For example, z-score and radar graph plotting may assist when making decisions about training programs for individual athletes. This may include (a) assessing strengths and weaknesses of individual players, and thus helping devise “player-specific” training and nutrition interventions, (b) helping identify (and select) youth athletes who are, or likely, to change their developmental trajectories, (c) generating expectations of anthropometric and fitness changes given the stage of maturation and training load exposure, and (d) being able to advise on decisions like restraining training load during peak growth phases; helping avoid soft tissue damage, joint instability, and injury.
More broadly speaking, athlete development systems which resemble the contexts of rugby league need to carefully factor in and better consider growth, maturation, and development to validate any form of (de)selection and differentiation in youth athletes. Unstable development in adolescence influences anthropometric and fitness characteristics as well as subsequent performance. This makes the ability to accurately assess athletic potential relative to peers and predict future performance challenging. For what may be deemed as “exceptional” at 1 age and stage, may not remain the same (and thus the same individual) at a later time points. As shown, later maturing athletes may close the “fitness and performance gap.” This recommends not only measuring anthropometric and fitness characteristics over the long-term but also promotes the need for a “mind-set” change in practitioners working within such systems. If the desired outcome is to develop adult athletes, and if there is variation and instability during adolescence (i.e., difficult to assess and predict), then (de)selection in this period needs to be avoided. Such a stance requires replacement of the present emphasis on immediate performance success in youth to one of longer-term inclusive player development.
This research was supported by the Rugby Football league (RFL), and the authors thank the RFL for providing the data to support this study. The results of this study do not constitute endorsement of a product by the authors or the National Strength and Conditioning Association.
1. Abbott A, Button C, Pepping GJ, Collins D. Unnatural selection: Talent identification
and development in sport. Nonlinear Dyn Psychol Life Sci 9: 61, 2005.
2. Baxter-Jones A, Helms P, Maffulli N, Baines-Preece J, Preece M. Growth and development of male gymnasts, swimmers, soccer and tennis players: A longitudinal study. Ann Hum Biol 22: 381–394, 1995.
3. Beunen G, Malina RM. Growth and biological maturation
: Relevance to athletic performance. In: The Encyclopedia of Sports Medicine: The Child and Adolescent Athlete. Bar-Or O., ed. Oxford, United Kingdom: Blackwell Scientific Publications, 1996. pp. 3–24.
4. Cobley S, Baker J, Schorer J. Identification and development of sport talent: A brief introduction to a growing field of research and practice. In: Talent Identification
and Development in Sport: International Perspectives. Baker J., Cobley S., Schorer J., eds. London, United Kingdom: Routledge, 2011. pp. 1–10.
5. Gabbett T, Herzig P. Physiological characteristics of junior elite and sub-elite rugby league players. Strength Cond Coach 12: 19–24, 2004.
6. Gabbett T, Kelly J, Ralph S, Driscoll D. Physiological and anthropometric characteristics of junior elite and sub-elite rugby league players, with special reference to starters and non-starters. J Sci Med Sport 12: 215–222, 2009.
7. Hawes MR, Martin AD. Human body composition. In: Kinanthropometry and Exercise Physiology Laboratory Manual: Tests, Procedures and Data. Eston R., Reilly T., eds. London, United Kingdom: Routledge, 2001. pp. 7–43.
8. Hopkins WG. Measures of reliability in sports medicine and science. Sports Med 30: 1–15, 2000.
9. Lefevre J, Beunen G, Steens G, Claessens A, Renson R. Motor performance during adolescence and age thirty as related to age at peak height velocity. Ann Hum Biol 17: 423–435, 1990.
10. Malina RM. Physical growth and biological maturation
of young athletes. Exerc Sport Sci Rev 22: 280–284, 1994.
11. Malina RM, Bouchard C, Bar-Or O. Growth, Maturation
, and Physical Activity. Champaign, IL: Human Kinetics Publishers, 2004.
12. Malina RM, Eisenmann JC, Cumming SP, Ribeiro B, Aroso J. Maturity-associated variation in the growth and functional capacities of youth football (soccer) players 13–15 years. Eur J Appl Physiol 91: 555–562, 2004.
13. Malina RM, Ribeiro B, Aroso J, Cumming SP. Characteristics of youth soccer players aged 13–15 years classified by skill level. Br J Sports Med 41: 290–295, 2007.
14. Mirwald RL, Baxter-Jones GAD, Bailey DA, Beunen GP. An assessment of maturity from anthropometric measurements. Med Sci Sports Exerc 34: 689, 2002.
15. Musch J, Grondin S. Unequal competition as an impediment to personal development: A review of the relative age effect in sport. Dev Rev 21: 147–167, 2001.
16. Philippaerts RM, Vaeyens R, Janssens M, Van Renterghem B, Matthys D, Craen R, Bourgois J, Vrijens J, Beunen G, Malina RM. The relationship between peak height velocity and physical performance in youth soccer players. J Sports Sci 24: 221–230, 2006.
17. Ramsbottom R, Brewer J, Williams C. A progressive shuttle run test to estimate maximal oxygen uptake. Br J Sports Med 22: 141–144, 1988.
18. Reilly T, Williams AM, Nevill A, Franks A. A multidisciplinary approach to talent identification
in soccer. J Sports Sci 18: 695–702, 2000.
19. Sherar LB, Baxter-Jones ADG, Faulkner RA, Russell KW. Do physical maturity and birth date predict talent in male youth ice hockey players? J Sports Sci 25: 879–886, 2007.
20. Stockbrugger BA, Haennel RG. Contributing factors to performance of a medicine ball explosive power test: A comparison between jump and nonjump athletes. J Strength Cond Res 17: 2003.
21. Sykes D, Twist C, Hall S, Nicholas C, Lamb K. Semi-automated time-motion analysis of senior elite rugby league. Int J Perform Anal Sport 9: 47–59, 2009.
22. Tanner JM. Growth and maturation
during adolescence. Nutr Rev 39: 43–55, 1981.
23. Tanner JM. Foetus Into Man: Physical Growth From Conception to Maturity. Portland: Harvard University Press, 1990.
24. Till K, Cobley S, O'Hara J, Chapman C, Cooke C. Anthropometric, physiological and selection characteristics in high performance UK junior rugby league players. Tal Dev Excel 2: 193–207, 2010.
25. Till K, Cobley S, O'Hara J, Cooke C. An individualized longitudinal approach to monitoring the dynamics of growth and fitness development in adolescent athletes. J Strength Cond Res 27: 1313–1321, 2013.
26. Till K, Cobley S, Wattie N, O'Hara J, Cooke C, Chapman C. The prevalence, influential factors and mechanisms of relative age effects in UK Rugby League. Scand J Med Sci Sports 20: 320–329, 2010.
27. Vaeyens R, Lenoir M, Williams AM, Philippaerts RM. Talent identification
and development programmes in sport: Current models and future directions. Sports Med 38: 703–714, 2008.
28. Viru A, Loko J, Harro M, Volver A, Laaneots L, Viru M. Critical periods in the development of performance capacity during childhood and adolescence. Eur J Phys Educ 4: 75–119, 1999.
29. Viru A, Loko J, Volver A, Laaneots L, Karelson K, Viru M. Age periods of accelerated improvement of muscle strength, power, speed and endurance in the age interval 6-18 years. Biol Sport 15: 211–228, 1998.
Keywords:Copyright © 2014 by the National Strength & Conditioning Association.
talent identification; maturation; anthropometry; physical fitness; coaching