Associations among age, playing position, dribbling, agility, and passing skills in elite youth football players graduating from a high-performance academy: an observational study

Abstract

Developing technical and motor skills during adolescence is critical for long-term football performance. This study aimed to examine the relationships between passing skill, dribbling skill, and agility performance in elite young football players and to determine whether these relationships differ according to playing position and age group. A total of 242 male football players aged between U14 and U19, competing in the academy teams of four professional clubs, participated in the study. Dribbling performance and agility were assessed using the Illinois Dribbling Test (IDT) and Illinois Agility Test (IAT), respectively, while passing skill was evaluated using the Loughborough Passing Test (LPT). Analysis of variance, Pearson correlation, and regression analyses were used to analyze the data. Pearson correlation analysis revealed a moderate and significant relationship between dribbling performance and agility (r = .482, p < .001, 95% CI [0.38, 0.57]). In addition, dribbling performance was significantly and positively correlated with passing skill (r = .234, p < .001, 95% CI [0.11, 0.35]), and agility performance showed a low but significant positive correlation with passing skill (r = .154, p = .016, 95% CI [0.03, 0.27]). Multiple regression analysis demonstrated that dribbling and agility performances together explained a low but significant proportion of the variance in passing skill [F(2,239) = 7.207, p < .001, R² = 0.057].

In conclusion, monitoring the development of elite youth football players using objective and quantitative performance indicators is of great importance for coaches and sports scientists. The findings indicate that passing, dribbling, and agility performances are significantly interrelated across all age categories, highlighting the integrated nature of technical and motor skill development in youth football. Based on these findings, incorporating passing, dribbling, and agility skills together within youth football training programs appears to be important for supporting holistic technical and motor skill development.

Introduction

Football is a complex sport where players’ performance and match results are influenced by numerous technical, tactical, and physical factors [1]. Technical proficiency encompasses the ability to dribble, pass, shoot, and control the ball, all of which are essential for success at the highest levels of competition [2]. Technical proficiency encompasses not only motor skills but also cognitive processes, such as visual perception and decision-making in dynamic environments, which are essential for success in football.

Passing and dribbling are the two technical moves that happen the most during a game, so they are often seen as the best signs of a player’s technical ability and overall skill [3]. Short-passing is a crucial component of technical development, reflecting a player’s capacity to maintain possession and cooperate with teammates [4]. The application of these skills is intrinsically linked to continuous decision-making processes, requiring players to make optimal choices under significant temporal and spatial constraints [5, 6].

Another important part of performance is agility, which means being able to change direction quickly. According to Bloomfield et al. (2007), players change direction more than 700 times during a single match [7]. Agility in football has traditionally been seen as a physical trait, but recent studies show that it also includes mental skills like being able to react to opponents and the ball, which are important for getting a tactical edge on the field [8, 9]. Testing a player’s ability to change direction with and without the ball, like the modified Illinois change of direction test, gives you useful information about their functional and game-related skills and can tell the difference between different levels of youth players [10].

To evaluate these complex skills, various football-specific tests have been developed, including the Loughborough Passing Test (LPT), which assesses a combination of striking technique, ball control, and the ability to execute accurate passes in response to external stimuli [11]. While the validity and reliability of such tests have been established across various populations [12], it is acknowledged that a comprehensive assessment of football skill must be multifaceted, evaluating a combination of passing, dribbling, and ball control to effectively differentiate between players of differing skill levels [13].

Talent identification and development in youth football represents a critical challenge for professional clubs and academies worldwide. According to Vaeyens et al. (2008), traditional cross-sectional talent identification models often exclude promising young players, particularly late-maturing adolescents, due to the dynamic and multidimensional nature of sport talent [14]. This limitation highlights the necessity for more comprehensive, longitudinal assessment approaches that consider both biological maturation and technical skill development across different developmental stages.

The assessment of technical skills in football is inherently complex and multifactorial. Reilly et al. (2000) demonstrated that distinguishing elite from sub-elite youth players requires a multidisciplinary approach incorporating various performance measures [15]. While previous research has established that skills such as passing, dribbling, and agility are fundamental to match performance [1, 4], few studies have systematically examined how these technical competencies associate with each other across consecutive age groups and playing positions within a high-performance academy setting [16–18].

Position-specific demands in football are well-documented, with defenders, midfielders, and strikers exhibiting distinct physical and technical performance profiles during match-play [19, 20]. However, the extent to which standardized technical tests—such as the Loughborough Passing Test (LPT), Illinois Dribbling Test (IDT), and Illinois Agility Test (IAT)—can discriminate between positions and age groups in elite youth populations remains unclear. Furthermore, the adolescent period (ages 14–19) represents a critical window for talent development, where maturity-associated variation in sport-specific skills can substantially influence performance outcomes and selection decisions [21].

Despite these recognized challenges, there is a notable research gap regarding the holistic development of players. Previous literature has largely assessed technical skills in isolation or focused on limited age brackets. What distinguishes this study from prior research is its integrated and simultaneous evaluation of multiple technical and cognitive competencies (passing, dribbling, and agility) across a wide developmental spectrum (U14 to U19) and various playing positions within a single high-performance elite academy. By moving beyond isolated skill assessments, this research provides a more ecologically valid understanding of player development trajectories.

Therefore, the objective of this study is to establish empirical evidence regarding the interrelation of these technical skills. Specifically, this study aims to: (1) examine the associations between passing (LPT), dribbling (IDT), and agility (IAT) skills in elite youth football players; (2) determine whether these test scores differ significantly across playing positions; and (3) evaluate how these technical competencies and their associations evolve across consecutive age groups during the critical adolescent development period. It is hypothesized that significant relationships exist between the LPT, IDT, and IAT, and that performance in these assessments will exhibit distinct variations based on both playing position and age group.

Material and method

Study design

This study adopted a cross-sectional observational design and was conducted during the mid-season competitive period. The primary aim was to assess agility, dribbling, and passing performance in youth football players using standardized field-based performance tests.

Data collection took place over a 17-day period (1–17 April 2024) on weekdays, corresponding to team rest days, as participants were involved in official competitions during weekends.

Participants

A total of 242 male youth football players aged U14 to U19 participated in the study. All participants were registered players competing in the Youth Development Teams of four professional football clubs located in Izmir, Turkey. Participation was voluntary. An a priori power analysis using G*Power software (version 3.1) indicated that a minimum sample size of 128 was required to detect a moderate effect size (f = 0.25) with 80% power at an alpha level of 0.05 for a two-way ANOVA with five age groups and three positions. Thus, our sample of 242 participants exceeded the minimum requirement and was deemed adequate for the planned statistical analyses.

The players were informed about the purpose and procedures of the study prior to the tests. This study was deemed ethically appropriate by the Ege University Social and Human Sciences Scientific Research and Publication Ethics Board on 27.03.2024 (Protocol No: 2361). All procedures were conducted in accordance with the ethical principles of the Declaration of Helsinki (2013) [22]. Written informed consent was obtained from all participants prior to participation. For participants under the age of 18, written informed consent was also obtained from their parents or legal guardians and properly recorded. Only the data of participants who provided the required consent were included in the analysis.

Testing procedures

Players completed three performance assessments:

• Illinois Agility Test (IAT)

• Illinois Dribbling Test with the ball (IDT)

• Loughborough Passing Test (LPT)

The IAT and IDT were administered on the same day, each consisting of two trials, with standardized rest intervals between trials. The LPT was performed on the following day, also consisting of two trials.

Prior to testing, participants completed a 15-minute standardized warm-up and were introduced to the test course. They received detailed instructions emphasizing that they were required to navigate around the markers rather than over them, and the same test administrators conducted all trials to minimize inter-rater variability. Participants were also allowed to perform three to four submaximal familiarization trials to become accustomed to the test protocol before formal testing began, and they were instructed to perform each test as quickly and accurately as possible.

Environmental conditions and equipment

All tests were performed on artificial turf surfaces to ensure consistency and eliminate variability due to ground conditions. Environmental conditions remained stable throughout the testing period, with ambient temperatures ranging from 22 °C to 25 °C, and no precipitation was recorded.

A FIFA-approved Molten F5A5000 size 5 football was used for all ball-related assessments. Timing was recorded using a Casio HS-80TW-1D stopwatch.

Loughborough Passing Test (LPT)

The LPT was conducted within a 12 × 9.5 m test area and consisted of 16 passes directed toward four color-coded target benches positioned at the midpoint of each boundary line. The test area included a central control zone and designated passing zones.

Each trial comprised:

• Eight long passes (green and blue targets)

• Eight short passes (white and red targets)

To prevent anticipation effects, the researcher randomized the inter-pass sequence into four predefined sequences. Target colors were announced immediately before the completion of the current pass.

Participants began the test from the central zone with the ball. Timing started upon the verbal start command and stopped when the ball made contact with the final target. One tester was responsible for timing, one for announcing target colors, and one for recording penalties. No feedback was provided during testing. To minimize inter-observer variability and ensure consistency in measurement, all LPT trials were manually timed by the same experienced assessor using a calibrated Casio HS-80TW-1D stopwatch.

Participants were instructed that passes:

• Must be executed within the designated passing lanes

• Required the ball to return to the central zone and cross the marked line opposite the next target before the subsequent pass attempt (Fig. 1).

Fig. 1.

Loughborough Football Pass Test [23]

Outcome measures and scoring

LPT performance was quantified using total completion time, adjusted by penalty additions or deductions according to established criteria [23]:

• + 5 s: Completely missing the target bench or striking an incorrect target

• + 3 s: Missing the central target area (0.6 × 0.3 m)

• + 3 s: Hand contact with the ball

• + 2 s: Passing from outside the designated area

• + 2 s: Ball contacting any cone

• + 1 s: For each second exceeding the 43-second reference time

• −1 s: Ball contacting the 10-cm central strip of the target

Penalty times were recorded in real time and added to the raw completion time to obtain the final LPT score.

Illinois Agility Test (IAT) and Illinois Dribbling Test (IDT)

Previous studies applying the IAT with and without the ball in team sport athletes and elite youth football playershave demonstrated high test–retest reliability. Additionally, strong correlations have been reported between the IAT and sprint performance, and the test has shown a strong ability to discriminate between youth football players of different competitive levels [10]. The Illinois Agility Test (IAT) and the Illinois Dribbling Test (IDT) were performed on an identical test course measuring 10 m in length and 5 m in width, with four markers placed in a straight line at 3.3 m intervals in the central section of the course. As illustrated in Fig. 2, the test consisted of 40 m of linear sprinting and 20 m of slalom running, incorporating 180° directional changes every 10 m.

Fig. 2.

Illinois Agility Test (IAT) and Illinoist Dribling Test (IDT) without and with ball [10]

After the course was prepared, a dual-beam photocell timing system with a measurement accuracy of 0.01 s was positioned at both the start and finish lines. The photocell gates were placed 1 m above ground level, with a distance of 3 m between the Gates [24, 25].

Test execution

Following familiarization, participants completed two maximal trials at test speed. Each trial began from a standing start at the starting line, initiated voluntarily by the participant.

For both tests, participants sprinted forward to the 10 m marker, performed a 180° turn, navigated the four central markers in a slalom pattern over a 10 m distance, returned through the same slalom section, performed another 180° turn at the 10 m marker, and completed the test at the finish line.

If an error occurred during the test (e.g., missing a marker), the trial was terminated and repeated after a 5-minute passive recovery period, ensuring that two successful trials were recorded. A minimum of 5 min of rest was provided between trials [26, 27].

Test order and rest ıntervals

Participants first completed the IAT without the ball, followed by the IDT with the ball, or vice versa. To control for order effects, half of the participants within each age group began with the IAT, while the remaining half started with the IDT.

The IDT with the ball was performed after 30 min of active recovery following completion of the IAT (Makhlouf et al., 2022). Both tests were conducted twice under fully rested conditions, and completion time was recorded in seconds. The mean value of the two trials was used for statistical analysis.

Statistical analysis

To analyze the data, descriptive statistics (mean and standard deviation) were calculated for all variables. Assumptions for parametric tests were examined: normality was assessed using the Shapiro-Wilk test, and homogeneity of variance was tested using Levene’s test. Two-way analysis of variance (ANOVA) was used to investigate whether test scores (IAT, IDT, and LPT) differed according to position and age group. Post-hoc pairwise comparisons were conducted using LSD tests following significant main effects. Although Bonferroni-adjusted analyses were also performed and showed consistent results, the LSD outcomes are reported, as they were intended for exploratory interpretation. This limitation regarding multiple comparison correction has been acknowledged. Pearson correlation analysis was conducted to examine the relationships between the three tests. Multiple linear regression analysis was performed to test the predictive power of IDT and IAT on LPT scores. The regression model assumptions (linearity, independence of residuals, homoscedasticity, and normality of residuals) were examined using residual plots and diagnostic statistics. Statistical significance was set at p < .05. All analyses were performed using SPSS version 25.0.

Results

Descriptive statistics (mean and standard deviation) for the Illinois Agility Test, Illinois Dribbling Test, and Loughborough Football Passing Test of the 242 young football players who participated in the study, according to their age groups and positions, are presented in Table 1.

Table 1.

Descriptive statistics of the Illinois Agility Test, Illinois Dribbling Test, and Loughborough Passing Test by age and position

Age	Position	Illinois Agility Test by age/position		Illinois Dribbling Test by age/position		Loughborough Passing Test by age/position		n (total)	n (D/M/S)
		M ± SD (sec)	M ± SD (sec)	M ± SD (sec)	M ± SD (sec)	M ± SD (sec)	M ± SD (sec)
14	D	16.11 ± 0.47		20.33 ± 0.68		55.50 ± 8.92		51	18
	M	15.88 ± 0.65	15.92 ± 0.59	20.03 ± 0.66	20.09 ± 0.67	51.84 ± 11.52	52.56 ± 10.36		26
	S	15.57 ± 0.53		19.71 ± 0.48		47.71 ± 7.65			7
15	D	16.36 ± 0.80		20.54 ± 1.43		51.63 ± 11.56		37	11
	M	16.00 ± 0.88	16.10 ± 0.84	20.10 ± 0.80	20.32 ± 1.10	48.10 ± 9.04	49.08 ± 9.95		19
	S	16.00 ± 0.81		20.58 ± 1.25		47.71 ± 10.43			7
16	D	15.73 ± 0.75		20.13 ± 0.62		47.95 ± 7.40		62	23
	M	15.68 ± 0.80	15.64 ± 0.72	19.44 ± 0.82	19.69 ± 0.80	45.68 ± 7.18	46.75 ± 7.23		25
	S	15.42 ± 0.51		19.22 ± 0.75		46.71 ± 7.24			14
17	D	15.64 ± 0.49		19.82 ± 0.80		49.41 ± 6.65		49	17
	M	15.17 ± 0.49	15.30 ± 0.58	19.43 ± 0.84	19.55 ± 0.89	44.95 ± 7.00	47.71 ± 7.32		23
	S	15.00 ± 0.70		19.33 ± 1.11		51.55 ± 7.33			9
19	D	15.27 ± 0.57		19.55 ± 0.85		49.00 ± 8.58		43	18
	M	15.05 ± 0.42	15.20 ± 0.51	19.11 ± 0.78	19.39 ± 0.79	45.29 ± 8.57	47.27 ± 8.25		17
	S	15.37 ± 0.51		19.62 ± 0.51		47.62 ± 6.63			8
T	D	15.78 ± 0.70		20.04 ± 0.90		50.48 ± 8.72
O
T	M	15.58 ± 0.75	15.62 ± 0.73	19.64 ± 0.85	19.79 ± 0.90	47.34 ± 9.13	48.62 ± 8.80	242 (D:87/M:110/S:45)
A
L	S	15.44 ± 0.65		19.66 ± 0.92		48.15 ± 7.64

D Defance, M Midfielder, S Striker, M Mean, SD Standard Deviation

A two-way analysis of variance (ANOVA) was conducted to investigate the effects of playing position and age group on agility, dribbling, and passing skills. The results of the analysis are summarized in Table 2.

Table 2.

Analysis of Variance Tests of Between-Subjects Effects Dependent Variable: IAT IDT, LPT

TEST	Interaction	df	F	p	η2
IAT	Age	4	12.153	p < .001	0.176
	Position	2	5.624	p < .001	0.047
	Age * Position	8	0.819	0.586	0.028
IDT	Age	4	8.368	p < .001	0.128
	Position	2	7.104	p < .001	0.059
	Age * Position	8	0.820	0.585	0.028
LPT	Age	4	2.137	0,077	0.036
	Position	2	3.997	p < .05	0.034
	Age * Position	8	0.699	0.693	0.024

IAT Illinois Agility Test, IDT Illinois Dribbling Test, LPT Loughborough Passing Test, η2 Effect Size

*p≤0.05

There were no significant Age × Position interaction effects for any of the performance variables (p > .05). Therefore, the analysis primarily focused on the main effects of age and playing position. As shown in Table 2, a statistically significant main effect was found for both position [F (2, 227) = 5.624, p = .004, η² = 0.047] and age [F (4, 227) = 12.153, p = .000, η² = 0.176] on Illinois Agility Test (IAT) scores. However, the position-age group interaction effect was not significant [F (8, 227) = 0.819, p = .586, η² = 0.028]. Similarly, for the Illinois Dribbling Test (IDT) scores, a significant main effect was found for both position [F (2, 227) = 7.104, p = .001, η² = 0.059] and age group [F (4, 227) = 8.368, p = .000, η² = 0.128], but the interaction effect was not significant [F (8, 227) = 0.820, p = .585, η² = 0.028]. For the Loughborough Football Passing Test (LPT) scores, the main effect of position was significant [F (2, 227) = 3.997, p = .020, η² = 0.034], while the main effect of age group [F (4, 227) = 2.137, p = .077, η² = 0.036] and the interaction effect [F (8, 227) = 0.699, p = .693, η² = 0.024] were not statistically significant.

Post-hoc LSD tests were conducted to determine which groups differed from each other for the significant main effects found in the ANOVA results. The multiple comparison results by position are presented in Table 3.

Table 3.

IAT, IDT, LPT, LSD Multiple comparisons by Position

Test	Position		M.D (sec)	S.E	p	95% CI
						Lower	Upper
						Bound	Bound
IAT	Defense	Midfielder	0.19*	0.09	p < .05	0.016	0.38
		Striker	0.33*	0.11	p < .05	0.1	0.57
	Midfielder	Defense	− 0.19*	0.09	p < .05	− 0.38	0.01
		Striker	0.13	0.11	0.233	− 0.08	0.36
	Striker	Defense	− 0.33*	0.11	p < .05	− 0.57	− 0.1
		Midfielder	− 0.13	0.11	0.233	− 0.36	0.08
IDT	Defense	Midfielder	0.40*	0.11	p < .001	0.16	0.63
		Striker	0.37*	0.15	p < .05	0.07	0.67
	Midfielder	Defense	− 0.40*	0.11	p < .001	− 0.6	− 0.16
		Striker	− 0.02	0.14	0.873	− 0.31	0.26
	Striker	Defense	− 0.37*	0.15	p < .05	− 0.67	− 0.07
		Midfielder	0.02	0.14	0.873	− 0.26	0.31
LPT	Defense	Midfielder	3.13*	0.22	p < .05	0.72	5.55
		Striker	2.32	0.56	0.14	− 0.76	5.42
	Midfielder	Defense	-3.13*	0.22	p < .05	-5.55	− 0.72
		Striker	− 0.81	0.51	0.593	-3.79	2.17
	Striker	Defense	-2.32	0.56	0.14	-5.42	0.76
		Midfielder	0.81	0.51	0.593	-2.17	3.79

M.D. Mean Difference (seconds), S.E. Standard Error, IAT Illinois Agility Test, IDT Illinois Dribbling Test, LPT Loughborough Passing Test CI Confidence Interval

*p ≤ .05

When Table 3 is examined, it is seen that according to the IAT results, strikers and midfielders were statistically significantly faster than defenders. In the IDT results, it was found that midfielders were significantly faster than defenders. According to the LPT results, midfielders showed significantly better passing performance compared to defenders. The multiple comparison results by age group are summarized in Table 4.

Table 4.

IAT, IDT LPT, LSD Multiple comparisons by AGE

TEST	Age		M.D (sec)	S.E	p	95% CI
						Lower Bound	Upper Bound
IAT	14	15	− 0.18	0.14	0.185	− 0.46	0.08
		16	0.27*	0.12	p < .05	0.03	0.51
		17	0.61**	0.12	p < .001	0.35	0.87
		19	0.71**	0.13	p < .001	0.44	0.97
	15	16	0.46**	0.13	p < .001	0.19	0.72
		17	0.80**	0.14	p < .001	0.52	1.08
		19	0.89**	0.14	p < .001	0.61	1.18
	16	17	0.33*	0.12	p < .001	0.09	0.58
		19	0.43**	0.12	p < .001	0.18	0.68
	17	19	0.09	0.13	0.476	− 0.17	0.36
IDT	14	15	− 0.22	0.17	0.200	− 0.58	0.12
		16	0.40*	0.15	p < .05	− 0.97	0.71
		17	0.54*	0.16	p < .001	0.22	0.87
		19	0.70*	0.17	p < .001	0.36	1.03
	15	16	0.63**	0.17	p < .001	0.29	0.97
		17	0.77**	0.17	p < .001	0.42	1.13
		19	0.93**	0.18	p < .001	0.56	1.29
	16	17	0.14	0.15	0.36	− 0.16	0.45
		19	0.29	0.16	0.07	− 0.02	0.62
	17	19	0.15	0.17	0.368	− 0.18	0.49
LPT	14	15	3.48	1.84	0.06	− 0.15	7.12
		16	5.81**	1.61	p < .001	2.62	8.99
		17	4.85**	1.71	p < .05	1.48	8.22
		19	5.28**	1.76	p < .05	1.80	8.77
	15	16	2.32	1.77	0.19	-1.17	5.82
		17	1.36	1.86	0.46	-2.30	5.03
		19	1.80	1.91	0.34	-1.97	5.57
	16	17	− 0.95	1.63	0.55	-4.17	2.26
		19	− 0.52	1.69	0.75	-3.86	2.82
	17	19	− 0.43	1.78	0.80	-3.08	3.95
IAT Illinois Agility Test, IDT Illinois Dribbling Test, LPT Loughborough Passing Test, CI Confidence Interval

*p ≤ .05

**p ≤ .001

According to Table 4, in the IAT results, it was found that players in the U19 and U17 age groups were significantly faster than in the other age groups (U14, U15, U16). In the IDT results, it was reported that there were significant differences between the U14, U15, U16, U17, and U19 age groups. In the LPT results, it was stated that there were significant differences between the U14 age group and the U16, U17, and U19 age groups, and that after the U16 level, the LPT pass test scores were similar to each other.

Pearson correlation analysis and multiple regression analysis were conducted to examine the relationships between the tests. The results of the Pearson correlation analysis conducted to determine the relationship between IAT, IDT, and LPT are shown in Table 5.

Table 5.

Pearson correlation table

	IAT	IDT	LPT
IAT	1
IDT	,482**	1
LPT	,154*	,234**	1

IAT Illinois Agility Test, IDT Illinois Dribbling Test, LPT Loughborough Passing Test

*p ≤ .05

**p ≤ .001

According to the correlation analysis results presented in Table 5, based on the pooled sample to reflect overall relationships across all age groups, a statistically significant and positive relationship was found between IDT and IAT (r = .482, p = .000), between IDT and LPT (r = .234, p = .000), and between IAT and LPT (r = .154, p = .016).Multiple regression analysis using the enter method to test whether IAT and IDT scores could explain a significant portion of the variance in LPT showed that IAT and IDT tests could explain a small but significant portion of the variance in LPT [F(2,239) = 7.207, p < .001, R² = 0.057]. The standardized beta coefficient for IDT was β = 0.208 (SE = 0.701, 95% CI [0.650, 3.412]), which was statistically significant (p = .004), indicating that dribbling performance is significantly associated with passing performance in the regression model. In contrast, the standardized beta coefficient for IAT was β = 0.054 (SE = 0.865, 95% CI [-1.051, 2.357]), which was not statistically significant (p = .451), suggesting that agility is not significantly associated with passing performance when controlling for dribbling ability. Although IDT showed a statistically significant association with LPT (p = .004), the modest explanatory power of the model (R² = 0.057) indicates that a substantial proportion of variance in passing performance is explained by other unmeasured variables.

Discussion

Specifically, this study examined whether these relationships vary by playing position and age group, extending previous research in this area. The study was designed to first find the links between these tests and then see how performance changes based on age and playing position. The results demonstrated moderate to weak relationships between agility, dribbling, and passing performances. Specifically, a moderate positive correlation was found between dribbling and agility (r = .482, p < .001), while weak positive correlations were observed between dribbling and passing (r = .234, p < .001) and between agility and passing (r = .154, p = .016). Although IDT scores demonstrated statistically significant relationships with both IAT and LPT scores, the magnitude of these associations indicates that dribbling ability represents only one of several factors contributing to agility and passing performance. Similarly, while dribbling and agility emerged as statistically significant predictors in the regression model, the practical explanatory value of the model was limited, accounting for only 5.7% of the variance in LPT scores. These findings underscore the multifactorial nature of football-specific skills, suggesting that passing performance is influenced not only by motor abilities such as dribbling and agility but also, and perhaps more substantially, by perceptual-cognitive processes, decision-making, tactical understanding, and other contextual performance factors. Furthermore, the absence of a significant Age × Position interaction effect indicates that positional differences in passing, dribbling, and agility follow similar developmental trajectories across the examined age groups. This suggests that while absolute performance improves with age, the relative technical and physical profiles of different playing positions remain relatively consistent throughout the adolescent development period.

As expected, a general trend of improved performance with increasing age was observed across all three tests. In the IAT, completion times decreased progressively with age, indicating enhanced agility performance among older players. However, the difference between the 17- and 19-year-old groups was not statistically significant, suggesting that agility development may have approached a plateau during late adolescence. Furthermore, the 15-year-old participants performed significantly worse than all older age groups. Previous research has suggested that this phenomenon may be related to rapid growth-related changes in limb length and body proportions, which can temporarily disrupt established motor patterns [28, 29]. However, these age-related differences should be interpreted with caution. Since biological maturation indicators, such as peak height velocity, maturity status, or maturity offset, were not assessed in the present study, it remains unclear whether the observed performance variations are primarily attributable to chronological age or to individual differences in biological maturation. Therefore, alternative explanations, including variability in training experience, anthropometric characteristics, and developmental trajectories, may also have contributed to the observed differences in agility performance.

In the IDT, performance times also went down as people got older, but the 15-year-old group still took a lot longer to finish than other age groups. This supports the idea that coordination may be temporarily disrupted. The absence of a notable disparity in dribbling performance between the 16 and 17-year-old cohorts, as well as between the 17 and 19-year-old cohorts, suggests that dribbling proficiency may become stable during these later stages of adolescence. These results align with research indicating non-linear enhancements in motor coordination and skill execution during adolescence, significantly influenced by individual maturation status [30–32].

Position-specific differences were evident across the performance measures. Midfielders demonstrated significantly faster dribbling times compared to defenders (p = .001) and superior passing performance (p = .011). These findings align with the known technical demands of midfield play, which typically requires frequent ball manipulation in congested areas, necessitating enhanced close control, dribbling proficiency, and rapid decision-making to maintain possession and initiate attacks [33–35]. Conversely, defenders may prioritize positional awareness and tactical positioning over high-speed dribbling, which could explain the observed performance differences. Strikers exhibited intermediate performance levels, suggesting that their technical profile differs from both defenders and midfielders. These position-specific patterns underscore the importance of tailoring technical assessment and training interventions to the unique demands of each playing position.

The utilization of the Loughborough Passing Test (LPT), which has been contested regarding its categorization as an open or closed skill, was a significant component of this study. While some researchers argue that structured assessments primarily measure closed skills, the LPT incorporates an important element of unpredictability that enhances its ecological validity [36]. In our application of the LPT, players were presented with a new task immediately following the completion of the previous one, necessitating rapid decision-making. This reactive component closely mimics the dynamics of a real match, where players must continuously perceive, process, and act based on the evolving game situation, thereby characterizing it as an open skill. The cognitive aspect, manifested through decision-making, constitutes a crucial element of skill that differentiates expert performers [37, 38].

To improve ecological validity, players were instructed that they could touch the ball an unlimited number of times, reflecting the demands of modern football, which requires players to adapt their skills to situational contexts. This approach contrasts with protocols that restrict touches, potentially diminishing the test’s utility for evaluating on-field performance. The primary focus was on the player’s ability to select and execute the appropriate technique effectively—the essence of skill—rather than solely on strict technical execution. Match analyses indicate that elite football is defined by enhanced sprint performance, the ability to execute repeated high-intensity actions, and the capability to traverse distance while dribbling at speed [34, 39].

Crucially, our findings demonstrate that these technical and physical attributes exhibit considerable variation depending on playing position. The observational data identified significant associations between Illinois Dribbling Test (IDT) performance and playing position, as well as between passing skill (LPT) and playing position. These results indicate that among elite youth football players, dribbling and passing skills display distinct patterns across different roles on the pitch. For instance, players in wide or attacking positions may demonstrate superior high-speed dribbling capabilities due to the tactical demands of their roles, whereas central midfielders might excel in passing accuracy and reactive decision-making under pressure. Although no significant position-age interaction was observed, the pronounced impact of position on both IDT and LPT outcomes underscores the necessity for position-specific profiling. By separating players based on agility and technical skill tests, talent identification and development programs can more effectively recognize and cultivate the unique skill sets required for specific tactical roles [40, 41].

For coaches and sports scientists, standardized assessments such as the Illinois Agility Test (IAT), IDT, and LPT provide objective, quantitative data to monitor player progression and identify talent. The strong correlations observed among these tests across all age groups suggest their utility as a comprehensive testing battery. Practitioners can leverage this information to design targeted training interventions, formulate position-specific development strategies, and track developmental trajectories, particularly for identifying players who may require specialized support to maintain their progression.

Limitations

This study has several limitations that should be acknowledged. First, the cross-sectional design does not permit causal inferences. Second, biological maturation was not assessed, despite its potential influence on physical and technical performance during adolescence. Third, although the order of the Illinois Agility Test (IAT) and Illinois Dribbling Test (IDT) was counterbalanced, potential order effects were not statistically examined. Additionally, different timing methods were used across tests, with the Loughborough Passing Test (LPT) measured using a handheld stopwatch and the agility and dribbling tests recorded via photocell systems, which may have introduced a small degree of measurement variability. Furthermore, although widely used in football research and practice, the closed-skill tests employed may not fully capture the perceptual-cognitive, decision-making, and reactive demands of real-game performance. Finally, the sample consisted exclusively of male academy players from Turkey, limiting the generalizability of the findings. Future research should address these limitations by incorporating measures of biological maturation, using more ecologically valid assessments, ensuring methodological consistency across performance measures, and examining more diverse populations.

Authors’ contributions

A.E.C. was responsible for data collection. M.Z.O. performed the data analysis. A.E.C., H.V., and M.Z.O. wrote and revised the manuscript. All authors read and approved the final manuscript.

Funding

The author(s) declare that no financial support was received for the research, authorship, and/or publication of this article.

Data availability

The datasets generated and analyzed during the current study will be deposited in an open repository (Open Science Framework) and will be publicly available upon publication.

Declarations

Ethics approval and consent to participate

This study was approved by the Ege University Social and Human Sciences Scientific Research and Publication Ethics Board on 27.03.2024 (Protocol No: 2361). All procedures were conducted in accordance with the ethical principles of the Declaration of Helsinki (2013) [22].

All participants were informed about the purpose and procedures of the study prior to participation. Written informed consent was obtained from all participants. For participants under the age of 18, written informed consent was also obtained from their parents or legal guardians. Only the data of participants who provided the required consent were included in the analysis.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.