Fielding teams at the professional level in soccer that include homegrown players, developed through a club's youth academy system has been described as cost-effective (25). Despite long-term financial benefits apparent in the development of homegrown players, a considerable outlay is required to ensure each player has access to adequate coaching and training facilities throughout their soccer education (25). Because of the scale of investment, it is important that clubs make informed decisions, with appropriate foresight, when recruiting, selecting, and releasing young players.
The relative age effect (RAE) is well documented within youth soccer and relates to the uneven distribution of players' birth date relative to the general population (13). Youth soccer is typically organized into 1-year age bands with a bias toward recruitment of players born in the first quarter of the selection year (9); a finding that has been reported in many countries (14). The existent research has documented the presence of the RAE in sport, yet it has failed to explain why the phenomenon exists (8). Of the proposed theories, the most commonly cited is the maturation-selection hypothesis (27). It is posited that relatively older players are more physically mature than their younger counterparts, which may be advantageous in sports, which involve physical contact, for example, soccer (21). Indeed, it is well understood that during the transition from childhood to adulthood, physical maturity influences many characteristics relevant to sporting performance including stature, mass, aerobic power, strength, and running speed (1,18). However, it is less clear if advanced physical maturity results in superior physical performance within the context of a 1-year age band.
It is unclear whether any relationships between physical maturity and measures of physical capacity are consistent throughout childhood and adolescence. Buchheit and Mendez-Villanueva (5) observed differences—varying in magnitude—in anthropometric and performance characteristics in relatively older and more physically mature under-15 players. In contrast, Carling et al. (7) reported few differences between relatively older and younger under-14 players. These conflicting studies illustrate that the relationships between relative age, maturity, and physical capacity in youth soccer players remain unclear. Furthermore, studies focusing on one age category reveal only a partial view of the influence of maturity on physical qualities and the RAE, especially because many players are registered to the same club for successive seasons. Furthermore, Figueiredo et al. (11) observed that within a wide range of age categories (under-11s to -14s), the influence of physical maturity on measures of physical capacity differed depending on the category analyzed. Similarly, Skorski et al. (23) and Lovell et al. (17) reported varying influence of relative age on physical performance markers across a wide range of age categories. These 2 studies, in addition to Buchheit and Mendez-Villanueva (5) are, to our knowledge, the only instances where magnitude-based inferences have been used to quantify the degree of influence relative age has on physical performance markers. The present study sought to contribute to this limited evidence base and report not only if physical maturity status had an influence on sprinting speed, within 1-year age bands, but also the degree of the relationship. Understanding these relationships has important implications for coaches and practitioners concerned with identifying players for selection, retention, and release at the end of each season.
The present study aimed to investigate the influence of relative age on physical maturity and sprinting speed within 6 consecutive age categories (U11-U17). Data were collected over 8 seasons within a professional soccer academy. The first hypothesis was that relatively older players would be more physically mature than their younger counterparts within all age categories. The second hypothesis was that physical maturity would influence anthropometric measurements (stature and mass) and sprinting speed but that the strength of these relationships would not be consistent between all age categories.
Experimental Approach to the Problem
An observational design was adopted for the present study. Anthropometric measures along with physical performance test results from youth players belonging to a professional soccer club academy were collected as part of routine fitness testing and analyzed retrospectively. Players were assessed over an 8-year period (season 2007/2008 to 2014/2015).
A total of 306 male elite youth players (age: 12.5 ± 1.7 years [range: 9.7–16.6 years]; stature: 156.9 ± 12.9 cm; mass: 46.5 ± 12.5 kg) who attended the same Scottish Premiership club academy participated. These players were drawn from 6 age categories including under-11, under-12, under-13, under-14, under-15, and under-17. During the observation period, some players were retained year after year and progressed through the age categories resulting in multiple observations in some instances (570 data points in total). All individuals joined the academy via a selection process administered by scouts affiliated with the club (subjective assessment) and were considered to be among the very best young players in Scotland. The benefits and risks associated with the current investigation were explained to the participants before signing an institutionally approved informed consent form. Written parental consent was also obtained before all physiological testing. The study was approved by The University of Glasgow, College of Medical and Life Sciences research ethics board, and conformed to the recommendations of the Declaration of Helsinki.
Relative Age Effect
To investigate the birth date distribution of the players, data were obtained from the General Registrars Office for Scotland concerning the number of births within the general population for the relevant years (1993–2004). This allowed a comparison between the expected and observed birth date distribution in the sample population. Youth soccer in Scotland is structured such that the selection year follows the calendar year (1 January–31 December). Hence, players born in quartile 1 possessed a birth date in January, February, or March and players born in quartile 4 possessed a birth date in October, November, or December.
During the first week of September each season, players completed a series of physical assessment protocols. Club support staff conducted all tests; all possessed a postgraduate degree in sport science in addition to nationally recognized strength and conditioning certifications. Mass along with standing and seated stretch stature was recorded to the nearest 0.1 kg and 0.1 cm, respectively, using calibrated scales (Avery Weigh-Tronix, United Kingdom) and a wall-mounted stadiometer (Holtain Ltd., United Kingdom). For the anthropometric assessments players removed their footwear and wore a training t-shirt and shorts. Maturity offset was calculated using the equation developed by Mirwald et al. (20) and has been used in previous research as an indicator of somatic maturity among youth soccer players (4,6). Maturity offset represents the amount of time (in years) until or since an individual's predicted peak height velocity (PHV) and is calculated using an individual's stature, seated stature, mass, date of birth, and the date of measurement (19). Maturity offset offers a logistically feasible way to estimate physical maturity among large groups such as in the present study. Over the course of the 8-year observation period, a number of different tests were used to characterize the players' physical capabilities. As such, the results from season to season were not always directly comparable. For example, a variety of different yoyo tests were used during the observation period. The only physical performance test included in the analysis was the 0–15 m sprint because this test was used with all squads every season. After the players had completed the anthropometric assessments, they performed a standardized 15-minute warm-up comprising light aerobic exercise and dynamic stretches. The sprint test was always the first task to be performed in the test battery after the warm-up each year. The 0–15 m sprint test protocol allowed 3 attempts per player from a standing start 0.5 m behind the first timing gate; the fastest time was recorded for analysis. Players had approximately 3 minutes rest between efforts. The sprints were measured using electronic timing gates (Smartspeed; Fusion Sport, Brisbane, Australia) and conducted on the same indoor synthetic pitch each year. All participants wore soccer boots with molded studs. The technical error of measurement for the 0–15 m sprinting assessment according to the club's own quality control testing was 0.21 seconds.
Data are presented as the mean ± SD. Before all analyses, plots of the residuals versus the predicted values revealed no evidence of nonuniformity of error. In athletic research, it is not whether there is an effect but how big the effect is that is important; use of the p value alone provides no information about the direction or size of the effect or the range of feasible values (2). The odds ratio, with uncertainty expressed as 90% confidence intervals (CI), was used to examine birth date distribution of our players against an expected equal distribution (e.g., the general population). Here, all comparisons were made between quartile 1 and quartile 4, and the magnitude of the odds ratio was assessed against thresholds of trivial, >1.5; small, >3.4; and moderate, >9.0 (15). The effects of birth quartile (quartile 1 vs. quartile 4) on player maturity, stature, and mass were analyzed using a mixed linear model (SPSS v.22; IBM Corp., Armonk, NY, USA) with random intercepts. Standardized thresholds for small, moderate, and large changes (0.2, 0.6, and 1.2, respectively) calculated from between-player SD of all players in each respective squad, were used to assess the magnitude of all effects (15). Inference was subsequently based on the disposition of the CI for the mean difference to these standardized thresholds and calculated as per the magnitude-based inference approach using the following scale: 25–75%, possibly; 75–95%, likely; 95–99.5%, very likely; >99.5%, most likely (15). Inference was categorized as unclear if the 90% confidence limits overlapped the thresholds for the smallest worthwhile positive and negative effects (15). To interpret the magnitude of the variability in maturity offset within each squad, we doubled the SD for each respective squad and compared against a scale of 0.2 (small), 0.6 (moderate), and 1.2 (large) of the between-player SD across all squads (24). Finally, Pearson correlations were used to determine the relationship between player maturity and sprinting speed and the following scale of magnitudes was used to interpret the magnitude of the correlation coefficients: <0.1, trivial; 0.1–0.3, small; 0.3–0.5, moderate; 0.5–0.7, large; 0.7–0.9, very large; >0.9, nearly perfect (15).
Odds ratio revealed a clear bias in frequency, when compared with our reference population, of players born in quartile 1 vs. quartile 4 within each playing squad. The magnitude of this bias was small for under-11s (odds ratio, 2.7; 90% CI, 1.7–4.3), under-12s (2.1; 1.4–3.2) and under-13s (3.1; 2.0–4.9), and moderate for under-14s (3.7; 2.3–6.0), under-15s (4.7; 2.6–8.7), and under-17s (4.3; 1.7–10.6).
Effect of Birth Quartile on Player Maturity, Stature, and Mass
Descriptive anthropometry for each age category is presented in Table 1. The overall effect (all squads combined) of birth quartile (quartile 1 vs. quartile 4) was very likely small for player maturity (0.85 years; 90% CI, 0.44–1.26 years) and player stature (6.2 cm; 90% CI, 2.8–9.6 cm), and likely small for player weight (5.1 kg; 90% CI, 1.7–8.4 kg). Within-squad analyses for player maturity, stature, and mass are presented in Tables 2–4, respectively; differences ranged from unclear to large for player maturity and stature, and unclear to moderate for player mass. After doubling the SD of maturity offset within each playing squad, the magnitude of variability was small for under-11s and under-12s, and moderate for all remaining squads.
Relationship Between Player Maturity and Sprinting Speed
The magnitude of the relationship between maturity offset and 15-m sprinting speed was trivial for under-11s (r = 0.01; 90% CI −0.14 to 0.16) and under-12s (r = −0.04; −0.20 to 0.13), very likely small for under-13s (r = −0.26; −0.39 to −0.11), possibly large for under-14s (r = −0.53; −0.62 to −0.41), very likely large for under-15s (r = −0.62; −0.71 to −0.51), and likely small for under-17s (r = −0.26; −0.50 to 0.02).
The uneven birth date distribution observed was commensurate with that reported by many others (13,16). A widely reported explanation for the RAE phenomenon is the maturation-selection hypothesis, which proposes that relatively older players are more advanced in physical maturity than their younger counterparts and that this confers a performance advantage (27). This theory makes intuitive sense because it is well established that attributes relevant to soccer performance such as sprinting speed, strength, and aerobic capacity improve during growth and maturation (18). However, the magnitude of the relationship between physical maturity and such performance attributes within the context of 1-year age categories has not been widely investigated. Specifically, to our knowledge, only 3 other studies have assessed the practical relevance of the relationships between relative age, physical maturity, and physical performance measures using magnitude-based inferences (5,17,23).
Overall, physical maturity was related to chronological age, with older players displaying greater maturity offset values, although the strength of the relationship differed depending on the specific category considered (Table 2). This superior maturity status manifested itself as both greater stature (Table 3) and mass (Table 4) up until the under-17 age category when the trend was reversed; however, again the magnitude of the relationships varied depending on age category. The stature and mass of the players in the present study were comparable with results reported previously (17,23). The strength of the relationships between stature, mass, and birth quartile increased from the under-11 (“likely small” for both stature and mass) through to the under-15 age categories (“possibly moderate” for stature; “likely moderate” for mass) and then reversed among the under-17 players. This reversal should be interpreted with caution since the number of under-17 players observed in the current study was small. This is an interesting finding as it demonstrates that the influence of physical maturity is not necessarily consistent throughout childhood and adolescence. Vaeyens et al. (26) also reported that the influence of physical maturity on numerous performance parameters varied depending on age category. Indeed, our analysis demonstrates that the magnitude of variability in relation to maturity offset status differed between younger (under-11s and under-12s) and older (under-13s to under-17s) players perhaps explaining some of the inconsistencies.
Similarly, the influence of physical maturity on 0–15 m sprinting speed varied depending on age category. The greatest magnitudes were observed in the under-14 and under-15 age categories where physical maturity had a possible and very likely large positive effect, respectively. Combined with the fact that the older players in these 2 age categories were generally more physically mature than their younger counterparts, the maturation-selection hypothesis appears valid. It seems very plausible that scouts could associate physical precocity—in the form of sprinting ability and physical dimensions—with “talent” especially when one considers how valuable a commodity speed is within the sport of soccer (10). The most common action before scoring a goal at the professional level is straight-line sprinting, highlighting the importance of this attribute (10). Adolescent boys typically pass through their PHV around 14 years of age and peak weight velocity follows soon after (18,22). The greatest interindividual discrepancies in stature and muscle mass are likely to occur around the chronological age of 14 years when some players will be prepubertal and others will be postpubertal. Beunen et al. (3) reported that differences in physical maturity between players influenced physical performance to the greatest degree around the chronological ages of 14–15 years in Belgian teenagers, reinforcing this theory. Maturity-associated differences between players at this developmental stage are temporary and likely to diminish as less-developed players mature. Indeed, the present results hint at this, with minimal differences in sprinting speed observed among players of differing physical maturity status within the under-17 age category. The potential for players to be released from their clubs based on transient maturational differences during early adolescence may result in a loss of available talent at the upper echelons of the game when age categories are no longer important.
In contrast, the influence of physical maturation on sprinting speed within the younger age categories (under-11s to under-13s) was minimal. This suggests that relatively older and more physically mature players in the earlier age categories were not selected because they were faster than their younger counterparts. Within the younger age categories (under-11s, 12s and 13s), the mean differences in stature and mass between those born in quarters 1 and 4 were small to nonexistent; ranging from 1 to 4 cm and 1–2 kg, respectively (Tables 3 and 4). It is questionable whether such small differences could have resulted in such a profound influence on selection. This raises the question, if differences in stature, mass, and sprinting ability are so small why were relatively older players disproportionally chosen? At the elite youth level, it may be that only the most biologically advanced late-born players are considered for selection, thus, creating homogenous groups. Gil et al. (12) reported superior sprinting ability, agility, and stature among relatively older compared with relatively younger nonelite youth soccer players. The RAE may simply appear to be unrelated to physical capacity at the elite youth level because of the formation of homogenous groups.
The present results demonstrate some likeness to previous findings; however, some discrepancies are apparent. Lovell et al. (17) found the greatest disparities in birth date distribution at the youngest age category observed (under-9) in addition to the age categories around expected PHV (under-13s to under-16s). The under-11 age category was the youngest observed in the present study and so a direct comparison cannot be made; however, like Lovell et al. (17), we observed the greatest RAE to be present among under-15 players. In contrast to Lovell et al. (17) and Skorski et al. (23), we investigated the relationship between physical maturity (rather than birth quartile directly) and sprinting ability. However, we also demonstrated that physical maturity and birth quartile were likely related (Table 2). Lovell et al. (17) reported superior anaerobic performance—including sprinting ability—among relatively older players in the under-10 to under-14 age categories. In contrast, the present results indicate minimal advantages in sprinting ability related to relative age within the under-11 to under-13 age categories. The explanation for this discrepancy is unclear; however, it may be attributable to differences in the sample populations. The data presented herewith originate from a single academy, whereas Lovell et al. (17) included data from 17 separate clubs. The present data may be indicative of a particular selection strategy at the club in question. However, because data were collected over the course of 8 seasons, any nuances related to the club's selection strategy at least highlight a consistent approach. In addition, the academy observed was attached to a Scottish top-division club, whereas the club academies observed by Lovell et al. (17) represented the third and fourth tier of English professional soccer.
The current results support the maturation-selection hypothesis but only at specific developmental stages (under-14s and under-15s). However, questions remain especially within the earlier age categories, which are synonymous with players' initial selection into performance programs. At the under-14 and under-15 age categories, relatively older players were generally more mature and this manifested as faster sprinting speed. However, at the younger age categories, although the older players were generally more mature, this did not translate to superior sprinting ability. Practitioners should be aware that the influence of physical maturity on sprinting speed varies throughout physical development. Crucially, it would appear that making decisions about which players to retain and release should not be based on sprinting ability around the under-14 and under-15 age categories because any interindividual differences may be confounded by transient inequalities in physical maturity status.
The authors wish to thank Mr. John Murray for his long-term support and encouragement during the data collection period.
1. Balyi I, Way R, Higgs C. Long-Term Athlete Development. Champaign, IL: Human Kinetics, 2013.
2. Batterham AM, Hopkins WG. Making meaningful inferences about magnitudes. Int J Sports Physiol Perform 1: 50–57, 2006.
3. Beunen GP, Malina RM, Lefevre J, Claessens AL, Renson R, Simons J. Prediction of adult stature and noninvasive assessment of biological maturation. Med Sci Sports Exerc 29: 225–230, 1997.
4. Buchheit M, Mendez-Villanueva A. Reliability and stability of anthropometric and performance measures in highly-trained young soccer players: Effect of age and maturation. J Sports Sci 31: 1332–1343, 2013.
5. Buchheit M, Mendez-Villanueva A. Effects of age, maturity and body dimensions on match running performance in highly trained under-15 soccer players. J Sports Sci 32: 1271–1278, 2014.
6. Buchheit M, Mendez-Villanueva A, Mayer N, Jullien H, Marles A, Bosquet L, Maille P, Morin JB, Cazorla G, Lambert P. Locomotor performance in highly-trained young soccer players: Does body size always matter? Int J Sports Med 35: 494–504, 2014.
7. Carling C, le Gall F, Reilly T, Williams AM. Do anthropometric and fitness characteristics vary according to birth date distribution in elite youth academy soccer players? Scand J Med Sci Spor 19: 3–9, 2009.
8. Cobley S, Baker J, Wattie N, McKenna J. Annual age-grouping and athlete development: A meta-analytical review of relative age effects in sport. Sports Med 39: 235–256, 2009.
9. Deprez D, Coutts AJ, Fransen J, Deconinck F, Lenoir M, Vaeyens R, Philippaerts R. Relative age, biological maturation and anaerobic characteristics in elite youth soccer players. Int J Sports Med 34: 897–903, 2013.
10. Faude O, Koch T, Meyer T. Straight sprinting is the most frequent action in goal situations in professional football. J Sports Sci 30: 625–631, 2012.
11. Figueiredo AJ, Gonçalves CE, Coelho Silva EMJ, Malina RM. Youth soccer players, 11-14 years: Maturity, size, function, skill and goal orientation. Ann Hum Biol 36: 60–73, 2009.
12. Gil SM, Badiola A, Bidaurrazaga-Letona I, Zabala-Lili J, Gravina L, Santos-Concejero J, Lekue JA, Granados C. Relationship between the relative age effect and anthropometry, maturity and performance in young soccer players. J Sports Sci 32: 479–486, 2014.
13. Helsen WF, Baker J, Michiels S, Schorer J, Van Winckel J, Williams AM. The relative age effect in European professional soccer: Did ten years of research make any difference? J Sports Sci 30: 1665–1671, 2012.
14. Helsen WF, van Winckel J, Williams AM. The relative age effect in youth soccer across Europe. J Sports Sci 23: 629–636, 2005.
15. Hopkins WG, Marshall SW, Batterham AM, Hanin J. Progressive statistics for studies in sports medicine and exercise science. Med Sci Sports Exerc 41: 3–13, 2009.
16. Jiménez IP, Pain MT. Relative age effect in Spanish association football: Its extent and implications for wasted potential. J Sports Sci 26: 995–1003, 2008.
17. Lovell R, Towlson C, Parkin G, Portas M, Vaeyens R, Cobley S. Soccer player characteristics in English lower-league development programs: The relationships between relative age, maturation, anthropometry and physical fitness. PLoS One 10: e0137238, 2015.
18. Malina RM, Bouchard C, Bar-Or O. Growth, Maturation and Physical Activity. Champaign, IL: Human Kinetics, 2004.
19. Malina RM, Rogol AD, Cumming SP, Coelho e Silva MJ, Figueiredo AJ. Biological maturation of youth athletes: Assessment and implications. Br J Sports Med 49: 852–859, 2015.
20. Mirwald RL, Baxter-Jones AD, Bailey DA, Beunen GP. An assessment of maturity from anthropometric measurements. Med Sci Sports Exerc 34: 689–694, 2002.
21. 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.
22. 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.
23. Skorski S, Skorski S, Faude O, Hammes D, Meyer T. The relative age effect in German elite youth soccer: Implications for a successful career. Int J Sports Physiol Perform 11: 370–376, 2015.
24. Smith TB, Hopkins WG. Variability and predictability of finals times of elite rowers. Med Sci Sports Exerc 43: 2155–2160, 2011.
25. Stratton G, Reilly T, Williams AM, Richardson D. Youth Soccer: From Science to Performance. Abingdon, United Kingdom: Routledge, 2004.
26. Vaeyens R, Malina RM, Janssens M, van Renterghem B, Bourgois J, Vrijens J, Philippaerts RM. A multidisciplinary selection model for youth soccer: The ghent youth soccer project. Br J Sports Med 40: 928–934, 2006.
27. Wattie N, Schorer J, Baker J. The relative age effect in sport: A developmental systems model. Sports Med 45: 83–94, 2015.