Adding extra balls elicits tactical variability and external load in youth women's football training

Abstract

This study compares the effects of introducing multiple or different balls in training games on the tactical and physical performance, enjoyment, and perceived competence of young female association football players. Fourteen trained/developmental under-14 players participated in five 3 × 6-min eight-a-side games: normal conditions, and changing the number (two and three simultaneously) and type (tennis and fitball) of balls. Players’ positional data were collected to assess tactical (spatial exploration index, distances to the nearest teammate and to the nearest opponent, and their entropy measures) and physical (total and low-to-moderate, high speed, very high speed, and sprinting distances covered) performances. Participants also rated their perceptions of enjoyment and competence. Tactical and physical variables (p < .001) and enjoyment (p < .05) significantly differed across conditions, but not perceived competence. With multiple balls, players explored less (Cohen's d_unbiased = −0.96 to −1.04) and played further away from each other (Cohen's d_unbiased = 0.48 to 1.10), while playing closer (Cohen's d_unbiased = −0.53 to −2.38) with different balls. Using multiple balls also increased the entropy-based metrics of spatial exploration (Cohen's d_unbiased = 1.14 to 1.61) and proximity (Cohen's d_unbiased = 0.38 to 0.97), the total distance covered (Cohen's d_unbiased = 0.54 to 0.80), and the distances covered at the highest speeds (Cohen's d_unbiased = 0.39 to 0.74). Enjoyment was higher in normal conditions than in manipulated conditions (Cohen's d_unbiased = 0.49 to 0.99). Academy coaches can incorporate additional balls to increase tactical variability and external load in female footballers to support their development, while the outcomes of using different balls depend on their properties.

Introduction

From a socio-didactic perspective, ¹ training in association football can be seen as a communicative process involving three elements: the coach, the player and the task. The dyadic interaction between these elements gives rise to the three dimensions of any intervention process ¹ : the strategic dimension (coach–task), which lies in the design and periodisation of the training content; the stylistic dimension (coach–player), which refers to the instructional situations managed by the coaches; and the praxic dimension (task–player), which concerns the players’ performance of the tasks set by the coaches.

Considering the strategic dimension as the basis of the training process, task design consists of manipulating and combining structural and relevant features of football (i.e., task conditions)^2,3 according to the response that coaches seek from players. Coaches’ task design preferences tend to be strongly influenced by their previous experiences both on the field as athletes and on the bench as coaches, or by their academic background.^4,5 Research is needed to bridge the gap between coaches’ intentions or beliefs (i.e., desired effects) and the results for player behaviour of manipulating task conditions (i.e., achieved effects) to improve training effectiveness. ⁶ The manipulated features can be classified according to the four axes of internal logic and expressed in terms of relationships ³ : the relationship to space (e.g., area per player), ⁷ to time (e.g., bout duration), ⁸ to other players (e.g., number of participants) ⁹ and to equipment (mainly the ball, such as limiting touches). ¹⁰ The acute effects of manipulating task conditions have been extensively studied in men's football,^11,12 but there is still a paucity of research in women's football^13,14 to help coaches optimise player development. ¹⁵

Among the different task conditions implemented by team coaches in relation to the ball, limiting the number of touches has been the most used and studied modification in men's football,^11,12 but the acute effects of introducing multiple^16,17 or different types of balls ¹⁸ have also been evaluated. Santos et al. ¹⁹ and Coutinho et al. ²⁰ proposed these manipulations, among others, to increase the effectiveness of training programmes on the adaptability of academy footballers. Adaptability, understood as the capacity to adjust readily to different and changing game situations, is crucial in team sports such as football, ²¹ where players must cope with and respond to the inherent uncertainty of the game. Therefore, providing players with highly uncertain training scenarios can promote their tactical variability and help them to adapt better to different training and competitive situations.^19,20

Gonzalez-Artetxe et al. ¹⁷ computed the approximate entropy (ApEn) measures of spatial exploration during 11-a-side games played with a single ball or with two balls simultaneously, while Santos et al. ¹⁸ assessed the ApEn of tactical variables to compare behavioural (un)predictability in small-sided games with footballs, handballs and rugby balls. Both manipulations slightly increased the unpredictability of young male footballers’ spatial exploration indices (SEIs) ¹⁷ and distances between teammates ¹⁸ in 11-a-side and six-a-side games. These findings can be used by men's academy coaches to optimise the design of their training according to the desired level of (un)predictability. However, the acute effects on the tactical behaviour of female players of altering the number or type of balls remain to be seen. The lack of specific knowledge makes it difficult to select training tasks that increase tactical variability for players in women's football academies.

In addition to tactical behaviour, external load was assessed, as monitoring and appropriately periodising the training load is key to ensuring optimal post-match recovery, adjusting the dose-response relationship, and avoiding pre-match fatigue.^22–24 Knowing the physical demands of different training scenarios can help practitioners to plan the weekly training content appropriately. Young male footballers ran much more when playing 11-a-side with two balls than with one ¹⁶ ; meanwhile, compared to football, they showed small to moderate decreases in total distance and jogging, running, and sprinting distances, but moderate increases in walking distances, when playing six-a-side with handball and rugby balls. ¹⁸ Therefore, coaches should modulate task duration when introducing more or different balls to ensure similar physical demands to football. However, this practical knowledge cannot be used by women's football coaches until these manipulations have been tested in this population.

Players’ perceptions of enjoyment and competence after playing with different task conditions were also collected, as these are common reasons for children to be physically active.^25,26 Assessing these components of motivation ²⁷ may help to address the high dropout rate of over 25 per cent in youth women's football. ²⁸ Recent studies reported that placing obstacles on the pitch ¹³ or modifying the opponent interaction with an extra rule (if a player touches an opponent just before receiving the ball, her team wins the ball) ¹⁴ did not significantly reduce the enjoyment or perceived competence of female academy footballers compared to unmodified small-sided games. However, the impact on the subjective training experience of female players of manipulating the relationship with the ball remains unknown.

Assessing training with a broad, multidimensional^29,30 view that considers the players’ tactical and physical performance and their subjective experience would enrich the study and make it more comprehensive. Given the limited knowledge on manipulating task conditions in women's football, this study aimed to compare the effects on the tactical and physical performance, enjoyment, and perceived competence of female academy footballers of introducing multiple or different balls in training games. Using more and/or different balls would increase tactical variability and external load without affecting players’ perceptions of enjoyment and competence.

Material and methods

Participants

The under-14 (U14) team of a Spanish women's football first division club (Liga F) was invited to participate in the study in the middle of the season (February–March 2024). All available players (n = 16, 14 outfield players plus two goalkeepers), who had no health issues or injuries, took part in the study. Goalkeepers participated but were excluded from the analysis due to the behavioural differences associated with their socio-motor role. A sensitivity power analysis using G*Power (version 3.1.9.6 for macOS) ³¹ for repeated measures with five conditions, an α level of 0.05, and a desired power of 0.80 reported a minimum detectable effect size (Cohen's f) of 0.31, indicating a moderate effect. ³² Participants, who identified football as their principal sport and had been playing for more than five years, trained three times a week (approximately 60–75 min per session) at 18:30 on an outdoor artificial turf pitch. Training sessions included passing exercises, build-up and finishing routines, and small- and large-sided games with and without floaters and with mini or regular goals. They competed in the top eight-a-side U14 league at weekends and ended the season by winning the regional championship. They were therefore considered trained/developmental players. ³³

Players, their parents/guardians, coaches and the club's academy management were fully informed about the purpose and procedures of the intervention, as well as its potential risks and benefits, and were made aware that they could leave the study at any time. Parents/guardians gave their informed written consent for the children to participate. The investigation followed the Committee on Publication Ethics’ International Standards for Authors and ethical principles of the Declaration of Helsinki (2013) and was previously approved by the Ethics Committee for Research involving Human Beings (GIEB in Basque) of the University of the Basque Country (protocol code: M10_2021_328; approval date: 25 November 2021).

Study design

The investigation comprised five training sessions for five experimental conditions, conducted on the participants’ habitual training pitch during the first session of the week (Tuesdays), with at least 48 h rest between the weekend match and the following session (Thursdays). Figure 1 outlines the study design. Environmental conditions recorded by the nearest weather station were similar across sessions (temperature 11.6 ± 2.6°C, humidity 74 ± 12%, wind speed 7.3 ± 2.9 km/h).

Figure 1.

Study design.

Coaches divided the team into two balanced groups according to players’ levels and positions ³⁴ in a 1-3-1-3 formation with one goalkeeper; one centre back, two fullbacks; one midfielder; two wingers, and one striker. Each footballer maintained their habitual playing position for all sessions. After an eight-minute customary warm up, both groups faced each other in a 3 × 6-min (with a two-minute rest between bouts) eight-a-side game (seven each, plus goalkeepers) played on a 50 m long × 30 m wide pitch (94 m² × player). This format was chosen because the team formation, player positions and pitch size were the same as in the competition. Several balls were scattered around the pitch to ensure fast ball replacement. Coaches were silent and seated on the bench during the games to let their pupils play as they pleased, avoiding any comment or gesture that might influence players’ conduct.^19,34

The relationship with the ball was manipulated in a random order in four eight-a-side games, changing the number (two and three) and the type (tennis ball and fitball) of ball. Participants also played a regular, non-altered eight-a-side game (normal) to compare task conditions with a control scenario. Apart from these manipulations, the rules were the same as in the competition. All games were refereed by a Federation referee with more than six years’ experience. In the two-ball game, each team started with one ball; in the three-ball game, the third ball was introduced by a kick-off in the air by one of the researchers. In the case of multiple balls, offside in the penalty area was relative to each ball and stoppages were restarted by regulation while the other ball(s) remained in play. ¹⁷ Two of the researchers, who hold UEFA B and UEFA A licences and are experienced in youth football training, were present on the pitch during all sessions to assist the coaches and referee, record all events in a field diary, and ensure the fidelity and accuracy of the study. Three to six third-year sports science students with football experience also helped prepare the game scenarios, assisted the referee, picked up balls and collected data.

Data collection

Outfield players’ positional data during eight-a-side games were gathered using a global positioning system (WIMU PRO, version 2020, RealTrack Systems, Almeria, Spain) with a 10 Hz sampling frequency validated for time-motion analyses in football. ³⁵ Each player wore the same individual tracking device for all sessions to avoid inter-unit variability.

Tactical response was assessed by the central tendency and ApEn measures of the SEI (the distance [m] of each player to the mean position) and the distances (m) to the nearest teammate (NearTM) and the nearest opponent (NearOPP). The ApEn algorithm quantifies regularity in a time series by measuring the logarithmic likelihood that runs from patterns that are close to m contiguous observations remain close (within r tolerance) on subsequent incremental comparisons. ³⁶ The input value of the vector length (m) was 2.0; the tolerance factor (r) was 0.2 standard deviations. ³⁷ The obtained values are unitless real numbers ranging from 0 to 2, low values corresponding to more regular and predictable sequences of data points. ³⁶ In the present study, the ApEn values allowed quantification of the regularity of the adjustments in positioning that each player performed in relation to their teammates and opponents.

External load was assessed by total and low-to-moderate-speed running (zone [Z] 1: < 2.91 ms⁻¹), high-speed running (Z2: 2.91–4.73 ms⁻¹), very-high-speed running (Z3: 4.73–5.66 ms⁻¹) and sprinting (Z4: > 5.66 ms⁻¹) distances (m) covered. ³⁸ All computations were run with MATLAB (version 2018a for macOS, MathWorks, Natick, MA, USA) following existing procedures. ³⁹

Players rated their subjective experience 5 min after the end of each eight-a-side game using the enjoyment and perceived competence scale validated by Arias-Estero et al. ⁴⁰ and previously used with this population.^13,14 The players took less than 5 min to respond to the seven statements on this five-point Likert scale on paper, in silence, and sufficiently separated from each other to avoid peer influence. The average of the even and odd statements was used to determine enjoyment and perceived competence, respectively.

Statistical analysis

After preliminary inspections for distribution and assumptions, a repeated measures analysis was processed to identify the effect of the game condition (normal, tennis ball, fitball, two balls and three balls) on the considered variables. Pairwise differences were assessed with Bonferroni post hoc tests. The analysis was performed using SPSS Statistics version 29 for macOS (IBM Corp., Armonk, NY, USA); statistical significance was set at p < .05. An estimation techniques approach was employed to overcome the shortcomings associated with traditional N-P null hypothesis significance testing.^41,42 Cohen's d_unbiased (d_unb) with 95% confidence intervals (CI) as the effect size (an unbiased estimate has a sampling distribution whose mean equals the population parameter being estimated) was applied to identify pairwise differences. ⁴¹ The thresholds for effect size statistics were 0.2, 0.5 and 0.8 for small, moderate and large, respectively. ³²

Results

The descriptive and inferential results for the effect of game scenarios (normal, tennis ball, fitball, two balls and three balls) on the dependent variables are presented in Table 1.

Table 1.

Descriptive (mean ± standard deviation) and inferential analysis.

Variables	Game scenarios					F	p	η²p
	Normal	Tennis ball	Fitball	2balls	3balls
Tactical performance
Spatial exploration index (m)	8.1 ± 0.9	7.0 ± 0.7*	8.8 ± 1.6*, †	7.0 ± 1.5*, ‡	6.9 ± 1.4*, †, ‡	22.4	< .001	0.353
Distance to NearTM (m)	8.1 ± 1.2	6.6 ± 1.1*	5.7 ± 0.7*, †	7.9 ± 1.5†, ‡	8.6 ± 1.0*, †, ‡, #	68.5	< .001	0.626
Distance to NearOPP (m)	4.9 ± 1.1	4.3 ± 1.1*	4.0 ± 1.1*	5.7 ± 1.4*, †, ‡	6.4 ± 1.6*, †, ‡, #	58.1	< .001	0.586
Spatial exploration index (ApEn)	0.08 ± 0.02	0.10 ± 0.03*	0.08 ± 0.02†	0.11 ± 0.03*, ‡	0.11 ± 0.02*, †, ‡	23.9	< .001	0.368
Distance to NearTM (ApEn)	0.14 ± 0.03	0.16 ± 0.03*	0.15 ± 0.04*	0.17 ± 0.04*, †, ‡	0.17 ± 0.05*, ‡	12.0	< .001	0.226
Distance to NearOPP (ApEn)	0.18 ± 0.05	0.19 ± 0.04	0.18 ± 0.06	0.20 ± 0.06*, ‡	0.21 ± 0.05†, ‡	6.9	< .001	0.143
Physical performance
Total distance covered (m)	536.8 ± 68.6	580.0 ± 81.7*	515.9 ± 66.2†	599.9 ± 86.1*, ‡	576.4 ± 74.9*, ‡, #	27.7	< .001	0.403
Distance covered Z1 (<2.91 ms−1)	404.8 ± 31.9	427.2 ± 32.8*	396.7 ± 40.6†	411.3 ± 28.1†, ‡	412.7 ± 39.0†, ‡	10.8	< .001	0.209
Distance covered Z2 (2.91–4.73 ms−1)	115.9 ± 43.8	136.5 ± 59.9*	105.0 ± 41.4†	156.0 ± 58.7†, ‡	145.2 ± 46.5*, ‡	18.8	< .001	0.314
Distance covered Z3 (4.73–5.66 ms−1)	12.3 ± 9.5	13.4 ± 9.2	11.3 ± 8.8	20.3 ± 13.7*, †, ‡	16.0 ± 9.1*, ‡	6.6	< .001	0.139
Distance covered Z4 (>5.66 ms−1)	3.7 ± 5.5	2.9 ± 3.6	2.9 ± 5.3	12.2 ± 15.1*, †, ‡	2.5 ± 2.6‡, #	12.9	< .001	0.238
Perceptions
Enjoyment (a.u.)	4.1 ± 0.5	3.7 ± 1.0	3.3 ± 1.1*	3.8 ± 0.6	3.4 ± 0.8*	2.8	.033	0.158
Competence (a.u.)	3.5 ± 0.5	3.3 ± 0.4	3.1 ± 0.6	3.4 ± 0.5	3.4 ± 0.5	1.4	.244	0.086

Abbreviations: NearTM: nearest teammate; NearOPP: nearest opponent; ApEn: approximate entropy; Z: zone; a.u.: arbitrary units.

Superscripts indicate significant differences at p < .05.

*different from normal.

^†different from tennis ball.

^‡different from fitball.

^#different from 2balls.

Additionally, Figure 2 illustrates Cohen's d_unb differences for all considered variables, along with the 95% CI.

Figure 2.

Cohen's d _unbiased differences for the considered variables according to the game scenarios. Error bars indicate uncertainty in the true mean changes with 95% confidence intervals (CI). Abbreviations: NearTM: nearest teammate; NearOPP: nearest opponent; ApEn: approximate entropy; Z: zone; a.u.: arbitrary units.

Significant differences (p < .001) were found for spatial exploration across scenarios. SEI values were substantially greater in the normal scenario (8.1 ± 0.9 m) in comparison to the tennis ball scenario (d_unb = –1.50 [−1.04, −2.01]) and scenarios with multiple balls (two balls: d_unb = –0.96 [−0.54, −1.41]; three balls: d_unb = –1.04 [−0.57, −1.53]). SEI values were substantially greater in the fitball scenario in comparison to the normal scenario (d_unb = 0.50 [0.08, 0.94]).

Proximity measures showed significant differences (p < .001) across game scenarios. Distance to NearTM values were greater in the normal scenario (8.1 ± 1.2 m) in comparison to the fitball (d_unb = −2.38 [−1.79, −3.05]) and tennis ball (d_unb = −1.21 [−0.87, −1.59]) scenarios, but lower than in the three-balls scenario (d_unb = 0.48 [0.12, 0.84]). Similarly, distance to NearOPP values were greater in the normal (4.9 ± 1.1 m) than in the fitball (d_unb = –0.78 [−0.49, −1.10]) and tennis ball (d_unb = −0.53 [−0.25, −0.82]) scenarios, but lower than in the two-balls (d_unb = 0.66 [0.34, 1.00]) and three-balls scenarios (d_unb = 1.10 [0.73, 1.50]).

Entropy-based metrics also revealed significant differences (p < .001). Notably, the two-balls and three-balls scenarios exhibited substantial greater ApEn values for the SEI, distance to NearTM and distance to NearOPP variables in comparison to the normal scenario, indicating significantly greater irregularity and unpredictability in spatial positioning compared to the other scenarios. The differences were large for SEI (two balls: d_unb = 1.14 [0.72, 1.59]; three balls: d_unb = 1.61 [1.14, 2.12]) and NearTM (two balls: d_unb = 0.97 [0.62, 1.36]; three balls: d_unb = 0.79 [0.49, 1.11]), and ranged from small to moderate for NearOPP (two balls: d_unb = 0.38 [0.14, 0.63]; three balls: d_unb = 0.57 [0.30, 0.85]).

The results revealed significant differences (p < .001) in all external load variables (total and zones distances) across the game scenarios. Players covered a substantially greater total distance (d_unb = 0.80 [0.49; 1.12]), and Z2 (d_unb = 0.76 [0.45; 1.109]), Z3 (d_unb = 0.66 [0.23; 1.11] and Z4 (d_unb = 0.74 [0.29; 1.22]) distances during the two-balls scenario in comparison to the normal scenario. Players covered a substantially greater total distance (d_unb = 0.54 [0.27; 0.83]) and Z2 (d_unb = 0.64 [0.33; 0.96]) and Z3 (d_unb = 0.39 [0.01; 0.77]) distances during the three-balls scenario in comparison to the normal scenario. Players covered a substantially greater total distance (d_unb = 0.56 [0.32; 0.83]) and Z1 (d_unb = 0.68 [0.41; 0.96]) and Z2 (d_unb = 0.39 [0.13; 0.65]) distances during the tennis ball scenario in comparison to the normal scenario.

Analysis of subjective experience revealed significant differences (p = .033) in enjoyment across game scenarios. Substantially greater enjoyment levels were declared after the normal scenario in comparison to the three-balls (d_unb = −0.99 [−1.75, −0.32]) and the fitball (d_unb = −0.92 [−1.70, −0.21]) scenarios. Perceived competence did not present significant variations (p = .244).

Discussion

This study set out to compare the effects of manipulating the relationship with the ball on the tactical behaviour and external load of female academy footballers and their perceptions of enjoyment and competence. The study hypotheses were partially met: 1) the inclusion of multiple balls decreased spatial exploration but increased the distance between the closest players; 2) the use of different types of ball meant that players played closer to each other; 3) additional balls introduced a layer of uncertainty that disrupted players’ tactical behaviour, but the effect of the different ball types was limited; 4) adding extra balls meant an increase in external load, especially playing with two balls; and 5) the regular game was more enjoyable than playing with manipulated conditions, while perceived competence did not differ across game scenarios.

Manipulating the relationship with the ball led to substantial changes in the tactical performance of young female players, but it varied with the number and type of balls. The inclusion of several balls meant substantial diminution of the space explored by players in comparison to the normal scenario. Similarly, SEI values of U14 male players were moderately lower during two-ball 11-a-side games in comparison to the standard one-ball games. ¹⁷ The reduction in spatial exploration may be because most players (i.e., defenders and forwards) stay near the goals to prevent or score goals rather than chasing the ball. ¹⁷ The use of different ball types had different effects on spatial exploration: female footballers explored much more when playing with a fitball, but less with a tennis ball, compared to the normal scenario. The specific properties of each ball, such as size and bounce, ¹⁸ may affect players’ exploratory behaviour, as ball handling may be compromised. Therefore, the acute effects of different balls ought to be assessed according to their characteristics.

Proximity measures also showed significant differences across game scenarios but differed according to the number and type of balls. Interestingly, adding more than one ball tended to keep players, both teammates and opponents, away from each other. Similarly, the distance between the back line and the midfielders was much greater when U14 male 11-a-side games were played with two balls instead of one. ¹⁶ Defenders’ and forwards’ positioning closer to the goals can split the team into two ¹⁷ and increase interpersonal distances. The use of different types of ball meant that players played closer to each other, with respect to both teammates and opponents, compared with the normal scenario. In contrast, Santos et al. ¹⁸ reported trivial differences in young male players’ interpersonal distances when using handballs and rugby balls in comparison to footballs. The specific properties of the balls may pose greater technical challenges to players than a standard football would. ¹⁸ Consequently, players can position themselves closer to their nearest teammates or opponents either to support them or to put pressure on them.

The entropy outcomes suggest that additional balls introduce a layer of uncertainty that disrupts players’ tactical behaviour, whereas the effect of the different ball types was limited. These results partially support the hypothesised increase in tactical variability by using more and different types of balls. ¹⁹ On the one hand, more balls involve more tactical unpredictability in both female and male ¹⁷ young footballers. Playing with more balls creates more opportunities for interaction and introduces more focal points, which makes the game more unpredictable and leads players to explore less and spread out more. Players may make strategic decisions in order to deal with the uncertainty of handling multiple balls. As seen in males,^16,17 young footballers may position themselves closer to the goals (defenders and forwards) or perform box-to-box transitions (midfielders), depending on their positional role, to adapt better to a scenario with greater variability. On the other hand, the effect of different ball types was less pronounced than with more balls. Similarly, a handball or rugby ball slightly increased the entropy measures of the distance between teammates in male players. ¹⁸ Thus, introducing several balls seems the most effective way to increase tactical variability in youth women's football training, whereas the consequences of changing the ball type need to be further investigated taking into account the properties of different balls. ¹⁸

Multiple-ball scenarios were the most demanding, with players running further and faster. Similarly, U14 male strikers and midfielders covered much greater distances and at higher speeds with two balls in 11-a-side games than with one. ¹⁶ Although players explored less with the addition of extra balls, it seems that greater uncertainty requires more physical effort, especially at high intensities, to respond to game scenarios with multiple balls. Alternatively, the technical difficulties involved in playing with different balls can slow down the pace of the game and result in players covering lower distances at the highest speeds. Similarly, the use of handball and rugby balls increased the walking distances of young male players, but not the overall running demands. ¹⁸ Adding more balls can enhance the external load of young footballers, whereas changing the ball type elicits lower intensity efforts. Academy strength and conditioning coaches could adjust training task duration according to the ball manipulation implemented and their physical fitness goals.

In addition to key aspects of coaching style to maintain player wellbeing, ⁴³ planning of training should consider the impact of tasks on players’ perceptions of enjoyment and competence.^8,44 The levels of enjoyment and perceived competence reported by female academy footballers in the five game scenarios were above 3 out of 5, in line with previous studies conducted with this population.^13,14 In this way, coaches can focus their attention on football-specific goals, knowing that both components of motivation ²⁷ will remain high regardless of the task conditions implemented. However, incorporating additional or different balls resulted in a small to large decrease in players’ enjoyment compared to normal conditions. Similarly, despite non-significance, placing obstacles on the pitch also largely impaired female academy footballers’ enjoyment, ¹³ while differences were trivial when manipulating the socio-motor relationship with opponents. ¹⁴ Otherwise, altering the relationship with the ball, space, ¹³ or opponents ¹⁴ did not affect the perceived competence of female academy footballers. Alongside previous findings relating to this population,^13,14 the results of this study indicate that coaches should anticipate reduced enjoyment when regular task conditions are altered, with no impact on perceived competence.

Although the findings of this study are promising, further investigation is necessary to expand the research body on the design of youth women's football training. Extending studies with a multifaceted approach about the acute effects of manipulating task conditions to more age groups, from grassroots to senior, and competitive levels, from recreational to elite, would strengthen the current knowledge on women's football training and assist coaches in optimising their practice. As ball properties influence playing responses, future studies will focus on the tactical adaptations, physical demands, and perceptual consequences of playing with different types of ball and explore the relationship between ball characteristics and their acute effects.

Conclusion

Introducing multiple balls (two or three) decreases spatial exploration, increases the distance to the nearest teammates and opponents, and boosts behavioural unpredictability, together with prompting higher external load. The consequences of using different ball types depend on the specific properties of the balls, such as their size or bounce. Both tennis balls and fitballs reduce distances to nearest players, but different trends were found for spatial exploration, entropy-based metrics and distances covered. Acute effects of incorporating different types of ball therefore need to be independently assessed taking into account the characteristics of each ball. In both cases, manipulating the number or the type of balls impaired female academy footballers’ enjoyment compared to normal conditions, with no differences in perceived competence across task conditions.