The influence of speed abilities and technical skills in early adolescence on adult success in soccer: A long-term prospective analysis using ANOVA and SEM approaches

Abstract

Several talent development programs in youth soccer have implemented motor diagnostics measuring performance factors. However, the predictive value of such tests for adult success is a controversial topic in talent research. This prospective cohort study evaluated the long-term predictive value of 1) motor tests and 2) players’ speed abilities (SA) and technical skills (TS) in early adolescence. The sample consisted of 14,178 U12 players from the German talent development program. Five tests (sprint, agility, dribbling, ball control, shooting) were conducted and players’ height, weight as well as relative age were assessed at nationwide diagnostics between 2004 and 2006. In the 2014/15 season, the players were then categorized as professional (n = 89), semi-professional (n = 913), or non-professional players (n = 13,176), indicating their adult performance level (APL). The motor tests’ prognostic relevance was determined using ANOVAs. Players’ future success was predicted by a logistic regression threshold model. This structural equation model comprised a measurement model with the motor tests and two correlated latent factors, SA and TS, with simultaneous consideration for the manifest covariates height, weight and relative age. Each motor predictor and anthropometric characteristic discriminated significantly between the APL (p < .001; η2 ≤ .02). The threshold model significantly predicted the APL (R2 = 24.8%), and in early adolescence the factor TS (p < .001) seems to have a stronger effect on adult performance than SA (p < .05). Both approaches (ANOVA, SEM) verified the diagnostics’ predictive validity over a long-term period (≈ 9 years). However, because of the limited effect sizes, the motor tests’ prognostic relevance remains ambiguous. A challenge for future research lies in the integration of different (e.g., person-oriented or multilevel) multivariate approaches that expand beyond the “traditional” topic of single tests’ predictive validity and toward more theoretically founded issues.

Talent selection in youth football: Technical skills rather than general motor performance predict future player status of football talents

“…Within U12 data, dribbling and ball control show highest effect sizes (η 2 = .02, n = 22,843) for discrimination between the performance levels of players at the middle and late adolescent stages (U16-U19; Höner & Votteler, 2016). A second study, also within U12 data, reports a higher prognostic validity for SMP (η 2 mean = .010) compared to speed abilities (η 2 mean = .006) in predicting adult performance levels (n = 14,187; Höner et al, 2017). In a third study, the working group was able to show that the 20m sprint was the only variable that did not significantly separate adult performance levels within U12 to U15 data (n = 1,134; Leyhr, Kelava, Raabe, & Höner, 2018).…”

“…If current talent selection models can only handle a limited number of variables, this raises the question of which areas provide the most powerful way of discriminating between future performance levels of elite youth players, and which variables are worthy of inclusion in these less extensive models (e.g., a person-oriented model with four to six operating factors). In general, motor performance has been one of the most considered predictors within talent research in youth football and is also frequently used by practitioners in the field (Höner, Leyhr, & Kelava, 2017;Sarmento, Anguera, Pereira, & Araújo, 2018). Whilst the overall value of motor performance for talent selection is not doubted, it is unclear whether physiologically driven general motor performance (GMP) capacities (e.g., speed, endurance, vertical jump) or domain-specific motor performance (SMP) capacities (i.e., technical skills) best predict future performance levels of young football players (Dodd & Newans, 2018;Forsman et al, 2016;Gonaus & Müller, 2012;Höner et al, 2017;Murr, Raabe, & Höner, 2017).…”

“…In general, motor performance has been one of the most considered predictors within talent research in youth football and is also frequently used by practitioners in the field (Höner, Leyhr, & Kelava, 2017;Sarmento, Anguera, Pereira, & Araújo, 2018). Whilst the overall value of motor performance for talent selection is not doubted, it is unclear whether physiologically driven general motor performance (GMP) capacities (e.g., speed, endurance, vertical jump) or domain-specific motor performance (SMP) capacities (i.e., technical skills) best predict future performance levels of young football players (Dodd & Newans, 2018;Forsman et al, 2016;Gonaus & Müller, 2012;Höner et al, 2017;Murr, Raabe, & Höner, 2017). In particular, the prognostic validity of GMP for long-term predictions from childhood or early adolescence has been vigorously questioned due to lower specificity of the task and development-related influences such as biological maturation and relative age, which may at least influence strength and speed abilities (Lidor, Côté, & Hackfort, 2009;Malina, Cumming, Coelho-e-Silva, & Figueiredo, 2017;Müller, Gonaus, Perner, Müller, & Raschner, 2017;Romann, Rössler, Javet, & Faude, 2018).…”

Back to the roots in football science. Why it might be smart to invest in the youngest players

“…There is evidence that talent identification is not reliably possible until puberty is finished (Güllich 2014;Höner et al 2017) and that early specialization might be harmful for talent development (Sugimoto et al 2017). Many football academies start their talent development programs at very young ages, often in pre-pubertal children, and researchers jump on the bandwagon.…”

Interactive Improvements of Visual and Auditory Function for Enhancing Performance in Youth Soccer Players

“…Among physical parameters, skill-related PF differs between elite/semi-elite players and "normal" people. Among the physical parameters, passing and dribbling skills are the best elements on which to determine a player's performance level [16,17]. In particular, the dribbling skills of young players (13-16 years old) are clearly distinguished by their playing level [18].…”