This prospective, observational, cross-sectional study was conducted in cooperation with two German elite youth football academies affiliated with professional men's teams competing in the first national and second national division. Due to the mandated gender segregation in youth football (DFB & DFL, 2022), the sample exclusively comprised male participants. Data collection took place during the annual Pre-Competition Medical Assessments (PCMAs) conducted in the pre-season period of June to September 2023. All procedures were carried out on-site at the training facilities of the respective clubs and at the Department of Sports and Rehabilitation Medicine of Ulm University Hospital. The assessments were embedded within the clubs' regular diagnostic routines and followed standardized protocols. The study was conducted in accordance with the latest version of the World Medical Association's Declaration of Helsinki—Ethical Principles for Medical Research Involving Human Subjects 2008 and approved by the ethics committee of Ulm University (No. 371/23).

The study included 221 male elite youth football players aged 11–18 years. All participants were recruited from two professional academies affiliated with first and second division clubs in the German national football league. Eligibility required enrollment in the respective academy for at least one full season, ensuring standardized exposure to elite-level training and competition. Players with chronic conditions preventing competitive football participation or whose consent was withdrawn (either by themselves or their legal guardians) were excluded. Verbal assent was obtained from all players and written informed consent from at least one parent or legal guardian was provided prior to participation.

All players competed within a structured academy framework, aligned with the national seasonal calendar (August–December and February–May). Players were assigned to age groups from Under-12 (U12) to Under-19 (U19), with training load progressively increasing with age. Training sessions followed standardized long-term development curricula implemented across both academies. U12 players trained approximately three times per week (60–75 min per session), focusing on coordination and basic technical skills. U13–U15 players trained four times per week (75–90 min), incorporating structured endurance, strength, and tactical elements. U16–U19 players trained five times per week (90–105 min), including high-intensity aerobic conditioning, strength and power sessions (twice weekly), and position-specific technical-tactical integration.

Competitive level also increased with age and development stage. Younger players (U12) typically competed in the 6th national division, U13–U15 teams in the 2nd to 3rd divisions, and U16–U19 players regularly participated in matches at the 1st or 2nd national division level. Match frequency ranged from one official league match per week (U12–U15) to one or two matches per week, including league, friendly, and international fixtures, in the older age groups. While individual training histories in total years were not assessed, the structured and age-graded progression within the academies ensured comparable developmental conditions and a consistently high level of competition across all participants.

The final sample of 221 participants had a mean age of 14.7 ± 2.3 years, a mean height of 168.7 ± 12.8 cm, a mean body mass of 57.4 ± 14.1 kg, and a mean body fat percentage of 11.1 ± 3.1%. Biological maturation was estimated using maturity offset (MO), with an average MO of 0.85 ± 2.03 years. Players reported an average training frequency of 3.6 ± 0.5 sessions per week, consistent with their respective academy schedules. Table 1 presents detailed participant characteristics and performance data.

Once eligibility for study was confirmed, participants underwent additional assessments of anthropometry and performance tests.

Body mass [kg] and body fat percentage [kg/m²] were measured using bioelectrical impedance analysis (InBody 770, Biospace Korea, Seoul, Korea) following the standard protocol described by Kyle and colleagues (40). Standing height and sitting height were recorded using a portable stadiometer with 0.1 cm precision (Seca 213, seca GmbH & Co. KG, Hamburg, Germany). With the same precision the leg length was measured as the distance between the anterior superior iliac spine and the medial malleolus at each ankle joint (27).

To assess biological maturity, a previously validated regression equation incorporating age, body mass, leg length, standing height, and sitting height was applied. The MO was calculated as the difference between the adolescent's chronological age and their estimated age at peak height velocity (PHV), providing an indicator of biological maturity relative to peers. A positive MO value indicates that the participants have passed their peak growth phase, whereas a negative value indicated that they are pre-PHV (38).

Handgrip strength (HGS) was measured using a hand dynamometer (SH1003, Saehan Corp., Donghae, Korea). Participants were instructed to apply maximal force to the dynamometer while maintaining a standardised posture: shoulder adducted to the body, elbow flexed at 90° (unsupported), and wrist in a neutral position (41). Participants were instructed to avoid compensatory movements that could compromise the accuracy of the test, and to ensure that the force was generated solely by the hand and forearm muscles. Each participant performed the test with both the dominant and non-dominant hand. Three measurements were recorded for each hand, and the mean value derived from the best measurements of each hand [kg] was used for analysis (42).

To assess dynamic balance, the Star Excursion Balance Test (SEBT) was performed according to the protocols of Dominguez-Navarro and colleagues (43) and Mohammadi Nia Samakosh and colleagues (44). Participants performed the test barefoot on a marked mat (Orthelligent Screening Mat, OPED Medical Inc., Braselton, USA). They were instructed to place both hands on their hips, balance one leg, lift the other leg off the floor, extend it as far as possible to lightly touch the floor, and return to the starting position without losing balance. The SEBT was performed in three directions: anterior, posteromedial and posterolateral. Each direction was tested alternately three times per leg and the maximum reach was recorded for each trial.

To calculate a normalized composite reach score for each leg, the sum of the reach distances in the three directions (anterior, posteromedial, and posterolateral) was divided by three times the leg length and the result was multiplied by 100. This procedure was performed separately for the dominant and non-dominant leg. Finally, the Limb Symmetry Index (LSI) was used to assess the symmetry between the compound scores of the less dominant leg and the dominant leg. An LSI of 100% indicates equal performance between the legs, while values below 100% suggest a balance asymmetry (45).

Vertical jump performance was assessed using three tests: the Counter Movement Jump (CMJ), the Abalakov Jump (AJ), and the football-specific Heading Jump (HJ). All tests were performed in accordance with current methodological standards for the assessment of athletic performance in adolescents (46) on a piezoelectric one-dimensional force plate (Quattro Jump, type 9290DD, Kistler, Winterthur, Switzerland), recording jump height [cm] and relative power [W/kg].

For the CMJ and AJ, participants stood upright on the platform with body weight evenly distributed between both legs. Following a verbal reference, they performed a rapid downward movement immediately followed by a maximal vertical jump. The CMJ was performed without arm movement, while the AJ included an active arm swing to enhance takeoff (47). The HJ was designed to replicate football-specific movement patterns. Players initiated the jump with a preparatory diagonal step and aimed to contact a suspended ball, simulating heading behavior under realistic game conditions. Arm movement was allowed to maintain ecological validity (48).

Each jump type was performed three times. The highest score for each trial was retained for further analysis. To minimize fatigue effects, a passive rest period of at least 2 min was observed between trials. All test procedures were performed by experienced personnel under standardized laboratory conditions to ensure high reliability and feasibility.

Horizontal jump performance was evaluated using the double-leg Broad Jump (BJ) and the Single-Leg Hop for Distance (SLHD) (49, 50).

For the BJ, participants stood with both feet parallel behind the 0.00 m mark on a floor-mounted measuring tape. For the SLHD, they positioned one foot directly behind the starting line. In both tests, participants were instructed to jump forward as far as possible. In the SLHD, they were required to land on the same leg used for take-off and maintain balance.

A trial was considered valid if the participant landed without falling, using their hands, or lifting the foot (SLHD) or feet (BJ) and maintained a stable stance for at least 3 s. The jump distance [cm] was measured from the take-off line to the nearest point of contact on landing, typically the heel. Each test was performed three times per leg (SLHD) or overall (BJ), with at least 2 min of rest between attempts. For analysis, the furthest valid distance was used for the BJ, while for the SLHD, the average of the best attempts from both legs was used.

Sprint performance was assessed using a 30-m sprint test on dry artificial turf, a standard and validated method for assessing linear speed in football (3, 51). Timing was recorded using a high-precision photocell system (Witty GATE System, type WIT002, MICROGATE, Bolzano, Italy) with gates positioned at 0 m, 10 m and 30 m. The system is characterized by excellent reliability (ICC = 0.96–0.99) and accuracy <.001 s (52) (WITTY Microgate, 2024).

Participants wore football boots and initiated each sprint from a standing, staggered position. They were instructed to run with maximal effort for 30-m and completed up to three trials. To minimize fatigue, rest intervals of at least 3 min were provided between trials. The fastest sprint time across trials was retained for analysis.

All statistical analyses were performed using JASP (version 0.19.2; JASP Team, 2024) and R (version 4.4.1, The R Foundation of Statistical Analysis) (R Core Team, 2024). Descriptive statistics (M, SD) were calculated for all study variables. Pearson correlation coefficients were calculated to assess associations between HGS and motor performance in sport. Due to the presence of outliers in some performance measures, additional percentage bend correlations were calculated (53). As both methods gave comparable results, the results of the Pearson correlations are presented in the manuscript. The strength of the correlations was interpreted according to Cohen's guidelines (54): small (r = 10–0.29), medium (r = 0.30–0.49) and large (r ≥ 0.50).

To examine the potential moderating effects of age and MO on the relationship between HGS and performance scores, a series of linear regression analyses were conducted. In these models, performance scores served as the dependent variable, HGS was included as a predictor, and either age or MO was entered as a moderator (HGS × Moderator).

In addition, linear regression models were used to assess the predictive value of HGS and age on sport motor performance. Due to multicollinearity concerns (Variance Inflation Factor > 10) when both age and MO were included as predictors, age was selected as a proxy for biological development due to its greater practical relevance. Models were calculated using raw, unstandardized values to maintain applicability in practical contexts. The explanatory power of each model was assessed using the coefficient of determination (R²), with higher R² values indicating better model fit. Statistical significance was defined as p ≤ .05.

Table 2 displays the descriptive statistics and Pearson correlation coefficients between HGS and all performance tests. A small but statistically significant negative correlation was found between HGS and the SEBT. Significant negative correlations were also observed between HGS and sprint performance. In contrast, large positive correlations emerged between HGS and both vertical and horizontal jump performance.

This section presents the results of the moderation analyses, investigating whether age and MO independently moderate the relationship between HGS and sport-specific performance tests. For comparability, all variables were z-standardized. First, the moderating role of age is presented (Figures 1A–D), followed by the analyses with MO as the moderator (Figures 2A–D).

For dynamic balance performance (SEBT, Figure 1A), no significant main effect of HGS was found, but a negative main effect of age was found (β = −0.27, p = .008). Additionally, a positive interaction between HGS and age was observed (β = 0.14, p = .045), suggesting that the association between HGS and balance performance strengthened with increasing age.

Regarding vertical jump performance (Figure 1B), positive main effects of HGS were found for the countermovement jump (CMJ; β = 0.44, p < .001), the Abalakov jump (AJ; β = 0.50, p < .001), and the heading jump (HJ; β = 0.40, p < .001). Age also showed positive main effects on CMJ (β = 0.32, p < .001), AJ (β = 0.32, p < .001), and HJ performance (β = 0.41, p < .001). No significant interaction effects between HGS and age were observed for any of the vertical jumping tests.

For horizontal jump performance (Figure 1C), positive main effects of HGS were observed for the broad jump (BJ; β = 0.45, p < .001) and the single leg hop for distance (SLHD; β = 0.33, p < .001). Age also showed positive main effects on BJ (β = 0.38, p < .001) and SLHD (β = 0.51, p < .001) performance. A negative interaction between HGS and age was found for SLHD (β = −0.11, p = .01), suggesting a slight decrease in the strength of the association between HGS and SLHD with increasing age. No significant interaction effect was found for BJ.

Finally, for sprint performance (Figure 1D), negative main effects of HGS were found for both the 10 m sprint (β = −0.31, p < .001) and the 30 m sprint (β = −0.33, p < .001), indicating that higher HGS was associated with faster sprint times. Age also showed negative main effects on 10 m (β = −0.40, p < .001) and 30 m sprint performance (β = −0.52, p < .001). In addition, positive interaction effects between HGS and age were observed for both sprint distances (10m: β = 0.26, p < .001; 30 m: β = 0.24, p < .001), suggesting that the positive association between HGS and sprint performance became stronger as players aged.

For dynamic balance (SEBT, Figure 2A), no significant main effect of HGS was observed, but a negative main effect of MO was found (β = −0.33, p = .005). There was no significant interaction between HGS and MO.

For vertical jump performance (CMJ, AJ, HJ, Figure 2B), positive main effects of HGS were observed for the CMJ (β = 0.37, p < .001), AJ (β = 0.44, p < .001), and HJ (β = 0.31, p < .001). MO also showed positive main effects on CMJ (β = 0.38, p < .001), AJ (β = 0.37, p < .001), and HJ (β = 0.50, p < .001). No significant interaction effects between HGS and MO were found for any vertical jump test.

For horizontal jump performance (BJ, SLHD, Figure 2C), HGS showed positive main effects for both the BJ (β = 0.35, p < .001) and SLHD (β = 0.22, p = .003). MO also showed positive main effects for the BJ (β = 0.48, p < .001) and SLHD (β = 0.60, p < .001). No significant interaction effects between HGS and MO were found for the horizontal jumping tests.

In sprint performance (10 m and 30 m Sprint, Figure 2D), HGS showed negative main effects for the 10 m sprint (β = −0.23, p = .010) and the 30 m sprint (β = −0.24, p < .001). MO exhibited strong negative main effects for the 10 m sprint (β = −0.46, p < .001) and the 30 m Sprint (β = −0.59, p < .001). Additionally, positive interaction effects between HGS and MO were found for the 10 m sprint (β = 0.24, p < .001) and the 30 m sprint (β = 0.21, p < .001).

To evaluate the predictive value of HGS and age on sport-specific performance outcomes in elite youth football players, a series of linear regression analyses was performed. For each performance test, HGS [kg] and age [years] were included as predictors. The predictive strength of each model was assessed using the coefficient of determination (R²), representing the proportion of variance in performance explained by the two predictors.

The models showed significant predictive validity for almost all tests, except for the SEBT, where no significant prediction was observed after adjusting for age. For all other sport-specific tests—including vertical jumps, horizontal jumps, and sprint performance—HGS and age together accounted for a meaningful proportion of performance variability. Full regression coefficients, equations, and R² values for each outcome are detailed in Table 3.

SEBT, as a parameter of dynamic balance, was significantly influenced by both age and MO, but not by HGS. Older and more biologically mature players had lower SEBT scores, likely reflecting growth-related changes in limb proportions and center of mass that temporarily impair postural stability during adolescence (60, 73). These findings are consistent with previous findings indicating superior balance control in less biologically mature athletes (74).

Interestingly, a modest positive interaction between HGS and age (β = 0.14, p = .045) suggests that greater strength may slightly attenuate the age-related decline in balance performance. As athletes progress beyond PHV, improvements in neuromuscular control may facilitate more effective integration of strength into postural stability tasks (75). In contrast, the lack of a significant interaction between HGS and MO suggests that biological maturity alone does not significantly alter this relationship. Overall, dynamic balance appears to depend more on lower limb strength, proprioceptive abilities, and flexibility than on general measures of strength such as HGS (55).

While dynamic balance performance showed only limited associations with HGS, jumping and sprinting abilities were strongly linked to strength capacities across all stages of maturation. Specifically, higher HGS was associated with better vertical (CMJ, AJ, HJ; β = 0.31–0.60, p < .001) and horizontal jump performance (BJ, SLHD; β = 0.33–0.60, p < .001), as well as faster sprint times over 10 m and 30 m (β = −0.23 to −0.33, p < .001), independent of chronological age or MO. These findings align with previous research by Pichardo and colleagues (39), who identified muscular strength as a key determinant of sprinting and jumping capacities in adolescent athletes.

Chronological age and MO independently influenced motor performance, with older and more biologically mature players performing better. This observation is consistent with previous research linking maturity-related increases in muscle mass, neuromuscular coordination, and biomechanical efficiency to improved physical performance (76, 77).

No significant interaction effects between HGS and either age or MO were observed for the vertical jumping tasks (CMJ, AJ, HJ), suggesting that the relationship between strength and jump height remains stable across different stages of maturation. Similar patterns have been reported in previous studies with children and adolescents (37).

For horizontal jumps, no interaction between HGS and age was found for the BJ. However, a small but significant negative interaction was found for the SLHD (β = −0.11, p = .01), suggesting that the contribution of HGS to unilateral jump performance may decrease slightly with increasing age. This trend likely reflects the increasing importance of technical skill and movement efficiency over pure strength (25, 26).

The strength of the association between HGS and sprint performance increased with advancing chronological age and biological maturity, as reflected by positive interaction effects for both the 10 m (β = 0.24, p < .001) and 30 m sprints (β = 0.21–0.26, p < .001). These findings suggest that HGS, as a marker of general upper-body and total-body strength, becomes increasingly important during the sprint acceleration and maximal velocity phases of adolescent maturation (67, 69, 78).

Among the various performance tests examined, HGS and age showed distinct predictive patterns. For dynamic balance performance, as assessed by the SEBT, HGS showed no significant predictive ability. This lack of association highlights the need for task-specific measures when assessing balance-related abilities, which are likely to be more dependent on flexibility, proprioception, and neuromuscular control (55). In addition, anthropometric variables such as height and body weight may have a stronger influence on SEBT performance than general strength measures (58).

In contrast, for explosive movements such as jumping and sprinting, HGS and age proved to be strong predictors. For vertical jumping, AJ (R² = 0.610) and HJ (R² = 0.601) had the highest predictive values, highlighting the importance of upper-body strength and coordination in jumping tasks involving arm swing dynamics. These findings are consistent with previous research suggesting that vertical jump performance is not solely dependent on lower body strength, but also on the effective interplay between upper and lower body mechanics (79). The slightly lower predictive ability observed for the CMJ (R² = 0.523) supports this notion, as the CMJ is performed without an arm swing and relies more heavily on isolated lower-body strength and stretch-shortening cycle efficiency (80). Similar results have been reported in young basketball players, where HGS was identified as a significant predictor of vertical jump height (81).

Regression models further confirmed the predictive value of HGS and age for horizontal jump performance, with results for BJ (R² = 0.621) and SLHD (R² = 0.636) exceeding those observed for vertical jumps. Even higher predictive values have been reported in previous studies, such as Vaidya and Nariya (82) in college students (R² = 0.677). Similarly, Nara et al. (64) validated the use of HGS as a predictor of horizontal jump performance in male collegiate athletes. The greater biomechanical complexity of horizontal jumps may explain their stronger association with HGS and age compared to vertical jumps. Unlike vertical jumps, which primarily involve upward forces, horizontal jumps require both vertical and forward forces, which more intensely engage the hip extensors and core stabilizers (83). Furthermore, unilateral tasks introduce additional balance and stability demands, increasing the need for neuromuscular control (25, 26). This may explain why the SLHD demonstrated slightly higher predictive power than the BJ, as greater postural stability demands place additional emphasis on strength-related skills. Additionally, the greater neuromuscular control required for single leg jumping is strongly influenced by maturational development (75), suggesting that age plays a more prominent role in predicting single leg jump performance.

Finally, HGS and age also significantly predicted sprint performance, with the 30 m sprint (R² = 0.672) demonstrating the strongest predictive ability of all motor tests. This finding highlights the broader relevance of HGS as an indicator of whole-body strength, particularly for longer sprint distances where sustained speed and upper-body stabilization become increasingly important (69, 67). The 10 m sprint (R² = 0.464), while still significantly predicted, had a lower predictive capacity, consistent with its greater emphasis on short-term explosiveness and technical skill rather than sustained strength over an extended sprint phase (66).

The findings of this study were interpreted with critical consideration of the study design and current evidence, aiming to provide a differentiated understanding of the role of HGS in elite youth football.

Despite the strengths of this study, several methodological considerations must be acknowledged to appropriately interpret the findings. The cross-sectional design, with data collection at a single pre-season time point, limits the ability to capture seasonal variations in performance related to training load, fatigue, and ongoing maturation. Future research should employ longitudinal designs to track individual performance trajectories over time. Moreover, all performance tests were conducted on the same day, which may have introduced fatigue-related effects despite standardized rest periods. Although rest intervals were carefully managed, the possibility of cumulative fatigue cannot be entirely excluded. Another methodological consideration involves the use of multiple examiners due to the large sample size (n = 221). Although standardized protocols were followed, minor variations in instruction, encouragement, or measurement techniques may have occurred. To ensure consistency, a maximum of three different examiners were assigned per test station. Nevertheless, manual measurement of horizontal jump distances, while practical, may have introduced slight inaccuracies; future studies should consider using laser-based or force plate systems to enhance measurement precision. In addition, although the moderating effects of chronological age and MO were accounted for, HGS alone does not fully represent the complex biomechanical and neuromuscular determinants of sport motor performance. Finally, as the study sample consisted exclusively of elite male youth football players, the generalizability of the findings is limited. Future studies should examine whether the observed relationships differ by sex, competition level, age group, or sport-specific demands, and also account for potential sources of inter-individual variability in HGS values, such as hand size, hand dominance, and minor fluctuations in fatigue or motivation.

Based on these findings, several practical applications for elite youth football develop. The results highlight the value of HGS testing as a simple, efficient, and accessible tool for assessing strength-related abilities, particularly in settings where more comprehensive performance diagnostics may not be feasible (84). Given its associations with sprinting and lower-body, HGS offers a practical alternative for talent identification and athletic monitoring in resource-limited environments. The regression models developed in this study suggest that coaches can estimate sprint and jump performance based on HGS and chronological age. While these estimates are not intended to replace direct performance tests, they may support initial screening and serve as supplementary indicators when full testing is impractical.

From an applied perspective, HGS testing is quick (under 2 min), requires minimal equipment (e.g., a handheld dynamometer), and can be administered by a single trained coach. These characteristics make regular implementation feasible, even in environments with limited resources (84). Integrating HGS into routine assessments could support tracking neuromuscular development across different stages of maturation. As a general indicator of neuromuscular readiness, HGS reflects strength adaptations during periods of growth (85). Defining normative values according to age and maturation status would further enhance its utility by enabling coaches to compare players to developmental benchmarks and support talent identification and long-term monitoring (38).

However, it is important to note that this study did not reveal any direct implications for injury prevention based on dynamic balance testing. Given that dynamic balance performance is multifactorial and best assessed through dedicated tools such as the Y-Balance Test or dynamic postural stability measures (24, 26).

Taken together, HGS assessment has potential as a scalable and complementary tool in athlete monitoring and talent development strategies in elite youth football.