The multicomponent, nonlinear, and dynamic nature of team invasion sports often generates challenges for practitioners when designing representative learning environments (10,15). A popular strategy to account for this dynamic interaction between technical, physical, and perceptual requirements has been to utilize small-sided games (SSGs) (3,6,18). Among other means, coaches typically construct SSGs by manipulating the field dimensions and playing numbers (5,18), with this manipulation believed to shape the activity profiles of players. For example, Klusemann et al. (9) demonstrated that increasing the number of players within a small-sided basketball game led to an increased number of successful passes and a decreased number of dribbles. Timmerman et al. (18) manipulated both pitch dimensions and player numbers within hockey games, noting the constrained environments that forced greater player density saw the emergence of fewer successful passes, skilled actions, and a change in the physical activity profiles of players. Accordingly, manipulating features of a performance environment could afford coaches with the capability to increase or decrease the complexities associated with SSGs, information which would be of value for the periodization of skill development in high-performance sport (7).
A theoretical underpinning to assist coaches with the effective manipulation of SSGs may be the constraints-led approach (14). This theoretical approach considers organismic, environmental and task constraints, and the role each play within skill acquisition (13,14). As each constraint class can be viewed independently of one another, this approach offers coaches the relative flexibility in the manipulation and control of certain constraints. Briefly, organismic constraints are defined by anthropometric (e.g., limb length), physiological (e.g., V̇o2max), or psychological (e.g., resilience) capabilities, and are categorized as either structural or functional (12). Environmental constraints are those viewed external to the human movement system, with examples including ambient temperature, light, and altitude (12). Task constraints are those which are specific to the task needing to be performed and relate to its outcome and rules (12). Central to this study, task constraint manipulation is likely to be of most use to coaches given the relative control that they are likely to possess over this constraint class during SSGs (11).
The informed and strategic manipulation of the aforementioned constraint classes during SSGs is likely to alter the perceptual information available to participants, potentially impacting on the emergence of their actions (4). For example, manipulating task constraints during SSGs, such as the field dimensions, could influence the spatial and temporal pressures imposed on participants through an increased or decreased player density. The relative spatial and temporal complexity generated by a reduction in playing space may result in the emergence of adaptive behavior (16), increase the difficulty of performing technical skills such as passing (18), and alter the physical loads placed on the human movement system (5). It is likely that field dimension manipulation would also implicate the physical activity profiles of players during SSGs. When afforded with greater field dimensions, it is likely that players will be able to reach higher running velocities relative to SSGs played within smaller field dimensions (8). Thus, this task manipulation is likely to be important for coupling skill acquisition and physical conditioning in team sports, as it could attune participants to pertinent environmental information to guide movement responses, while affording them with opportunities to engage in differing physical activities of use for conditioning.
Given multicomponent nature of game-play, SSGs are a commonly used training modality within Australian football (AF). Davies et al. (5) examined the influence of manipulating field dimension and player numbers on the agility demands in elite senior AF. It was noted that a reduction in space through a smaller field (increased player density) resulted in a small increase in agility maneuvers, and a large increase in 2-dimensional (mediolateral and anterolateral) load (5). Although this information would be of value for the physical conditioning of AF players, it remains unclear how the manipulation of field dimension (task constraint) impacts on the technical skill demands of SSGs in AF. The aim of this work was to determine the effect of this task manipulation on the technical and physical profiles of AF players. Given work conducted both in AF and other team invasion sports (5,18), it was hypothesized that this manipulation would alter the players' technical and physical activity profiles given the changing spatial and temporal complexity imposed on players. Specifically, it was expected that smaller dimensioned SSGs would incur greater erroneous technical activity and reduced total and relative distances in contrast to larger dimensions SSGs.
Experimental Approach to the Problem
To test the study hypothesis, an observational cross-sectional research design was used. All participants competed within 3 different sized SSGs, defined as “small,” “medium,” and “large” (Table 1). Each SSG was completed on an outdoor regulation AF oval, with the dimensions being strategically manipulated based on previous work in AF (5). Each SSG was contested between 2 teams of 5 players (5-a-side), with both teams being quasirandomized to ensure that no team had a bias of a certain playing position. In addition, the players in each team were randomized each week, whereas a “small,” “large,” and “medium” design was followed to limit a potential SSG learning effect. To accommodate 4 simultaneous SSG, field dimensions were set out so that the longest sides (i.e., 30 m for the small SSG) were adjacent to the longest side of the oval boundary.
Forty male participants (none under the age of eighteen) competing within a state-based AF competition were recruited (± SD 23.9 ± 3.5 years [age range: 20-27 years old]; 185.2 ± 4.1 cm; 85 ± 8.4 kg; career games 52 ± 31.7). Participants provided full informed written consent and were free of injury at the time of all data collection. Ethics approval was granted by the University of Notre Dame Fremantle Human Ethics Committee.
Small-sided Game Rules
Regulation AF rules were imposed for each SSG (including full tackling), with accredited coaches (level 2 AF coaches) possessing more than 10 years' experience adjudicating each game. As is common practice within AF SSGs and in accordance with previous SSG work in AF (5), disposal mode was constrained, with participants only being eligible to handball. Participants “scored” by handballing the ball to a coach placed in the “goal zone,” with players being eligible to score at any stage throughout the course of the SSG.
Small-sided Game Procedures
Each SSG was competed using an Australian Football League (AFL) match-day adult-sized football, with all participants wearing their football boots. Testing took place over the course of a 3-week block at the end of the preseason phase of training. Each SSG (small, large, or medium) was blocked into weeks 1, 2, or 3. Accordingly, week 1 consisted of 2 “small” SSGs (performed at the beginning of 2 training sessions), where game-play was contested over three 60-second playing periods. Between each playing period, a passive recovery interval of 60 seconds was implemented, during which players did not receive any coach-driven augmented feedback. This protocol was repeated for weeks 2 (large) and 3 (medium).
To enable the coding of technical involvements, each SSG was recorded using a Casio Exilim EX-FH100 digital video camera (Casio, Sydney, Australia), recording at 25 Hz. The cameras were placed to enable a behind-the-goals aerial perspective, with pilot testing showing that this perspective offered comprehensive insight into the participant's movements. The 11 technical skill notations coded in this study, along with their subsequent description, are presented in Table 2. The notational analysis was performed retrospectively through the use of SportsCode (Sportstec Limited, Sydney, Australia). Given the subjective nature of this coding procedure, the interrater reliability of the notational analysis was measured. The lead investigator coded all SSGs, with a co-investigator coding 10% of the total samples to establish the interrater reliability. Coding was performed independently of one another to reduce a potential coding bias. Interrater reliability was assessed using intraclass correlation coefficient statistics using SPSS (version 21, SPSS, Inc., USA). Intraclass correlation coefficient range for the coded games was 0.92–0.99, showing excellent reliability.
Physical Activity Profiles
To ascertain the physical running demands of each SSG, players were fitted with an OptimEye S4 global positioning system unit (Catapult Innovations, Scoreby, Australia). Each unit sampled at 10 Hz and was positioned between each participant's scapulae in a custom-designed harness provided by the manufacturer. The criterion variables extracted from these devices included: total distance (m), relative distance (m·min−1), maximum velocity (m·s−1), PlayerLoad (AU; extracted from the triaxial accelerometers in each unit), and distances covered while light jogging (0–4.13 m·s−1), fast jogging (4.16–5.54 m·s−1), and sprinting (>5.55 m·s−1) (19).
Descriptive statistics (mean ± SD) were calculated for each technical and physical indicator relative to the SSG field dimensions. All analyses were conducted using SPSS (version 21, SPSS, Inc., USA), with the type-I error rate being set at p ≤ 0.05. A multivariate analysis of variance modeled the main effect of field dimension (3 levels: small, medium, and large) on the technical and physical criterion variables. Furthermore, Cohen's d effect size statistics were calculated according to the main effect, where d < 0.10 was considered trivial, d = 0.10–0.20 small, d = 0.21–0.50 moderate, d = 0.51–0.80 large, and d > 0.81 very large (2).
There was a significant effect of field dimension on the technical skill and physical performance criterions (V = 1.032; F38, 102 = 2.863; p ≤ 0.05). Thus, for brevity, the subsequent results have been partitioned into technical and physical criterion variables.
Technical Criterion Variables
As displayed in Table 3, the number of turnovers decreased as the field dimensions increased, with the “small” and “medium” SSGs generating a significantly greater number of turnovers relative to the “large” dimension (p ≤ 0.05; d = 0.99 and 0.82, respectively). Furthermore, the number of ineffective handballs decreased with the concurrent increase in field dimension (Table 3). The “large” SSG generated a significantly fewer number of tackles (p ≤ 0.05; d = −0.79) and greater number of bounces (p ≤ 0.05; d = 0.75) relative to the “small” SSG. The remaining technical criterion variables did not yield a significant effect.
Physical Criterion Variables
Each physical criterion variable yielded a significant “large” to “very large” effect (p ≤ 0.05; d > 0.51; Table 3). Total distance, relative distance, maximum velocity, PlayerLoad, fast jogging, and sprinting distances all significantly increased from “small” to “large” SSGs (Table 3). Conversely, the light jogging distances significantly decreased, as the SSG dimensions increased (Table 3).
The aim of this study was to determine the effect that field dimension manipulation during SSGs had on the technical and physical profiles of Australian footballers. It was hypothesized that this manipulation would result in observable changes in the participants' technical and physical actions given the expected differing spatial and temporal complexities associated with each SSG. Supportive of this hypothesis, and consistent with others (5,18), results demonstrated that a reduction in playing space (increased player density) led to a greater count of turnovers, ineffective handballs, tackles, a reduction in running bounces, and a considerable change in physical activity profile relative to larger field dimensions (decreased player density). Accordingly, a reduction in field dimension may have limited the participant's capability to develop information-movement responses relative to their functional capabilities (i.e., identify who to pass the ball to and then execute that response) (17). These results provide AF coaches with insights into how a relatively simple task constraint manipulation within SSGs can alter the technical and physical involvements of players, with this potentially being of use for both skill and conditioning periodization in AF (7).
It was apparent that the “small” SSG generated the highest technical difficulty for participants relative to the “medium” and “large” dimensions. The greater erroneous activity recorded in the “small” SSG could have arisen from temporal and spatial constraints generated between the attackers and defenders. For example, the relatively greater count of turnovers and ineffective handballs could have stemmed from the spatial pressure imposed on the ball carrier from their opponent given the increased player density. This spatial pressure is likely to have temporally constrained the attacker's capability to visually search for, and process, the necessary environmental information needed to inform their motor response (9). Concurrently, it is possible that the smaller field dimensions enabled greater dyadic interactions between defenders and attackers (16). Notably, the greater density in the “small” SSG may have afforded a defender with a greater opportunity to impact the attacker. This is likely to place a greater emphasis on the execution of the disposal, as the defender may have been afforded with a greater capability to impact on the pass, resulting in the higher count of turnovers and tackles recorded in the “small” SSG. Despite a regulation adult AF game being contested on playing fields with dimensions between 130 and 150 m by 150–190 m (20), it is common for players to compete for ball possession in high congestion, as they strive to gain a “clearance” (the act of clearing the ball into space or to a teammate in space). Pertinently, elite senior AF teams who accrue a greater count of clearances are more likely to finish a season higher on the ladder (21). Accordingly, despite the “small” SSG incurring greater erroneous passes, the temporal and spatial constraints imposed in this condition may assist AF players to refine the perceptual and technical skills needed to manage congested scenarios experienced during match-play.
As expected, the “large” SSG resulted in greater total, relative, and high-speed running distances as well as PlayerLoad and maximum velocities when compared with the “small” and “medium” SSGs. It is likely that the greater field dimension and reduced player density in the “large” SSG afforded participants with the capability to use the space in an attempt to uncouple the attacker-defender dyad to maintain possession. These observations are in general agreement with work conducted in youth hockey (18) and soccer (1). Casamichana and Castellano (1) found that a relatively larger pitch area resulted in a higher physical and physiological workload in soccer, concluding that coaches could use this information when designing practice conditions at different phases of a competitive season. Furthermore, it was of note that the “large” SSG afforded participants with a greater capability to engage in running bounces relative to the “small” SSG. In AF, players must bounce the ball if they maintain possession while running greater than 15 m. Thus, the greater space afforded to the players in the “large” SSG seemed to result in the emergence of an action (the bounce) that was not used by participants within the “small” or “medium” SSGs. This observation is important for AF coaches, as it demonstrates that SSG design is likely to shape the types of actions players perform.
In summary, the “large” SSG generated fewer technical errors, tackles, more bounces, and a greater physical activity profile relative to the “small” and “medium” dimensions. These observations were likely driven by the increased spatial and temporal complexities associated with an increased player density incurred by reducing the field dimensions. Despite these promising results, the study is not without limitations that require acknowledgment. First, given data availability, we were unable to quantify the player's acceleration and deceleration profiles, information which is likely to be of use when planning the conditioning of players throughout a football season. Second, it is possible that the 60-second duration of the SSGs limited a player's capability to fully engage in game-play, particularly in the large dimension SSGs. Last, given the observational design of this work, it is difficult to ascertain the exact causation of the increased erroneous technical activity in the “small” SSGs (i.e., motor or perceptual). We hypothesize that errors were the resultant of both execution (motor) and perceptual (passing to the “wrong” player) misjudgments, but it would be an interesting avenue for future research to address given the implications that the subsequent findings could yield for skill periodization. Nonetheless, these limitations provide an enticing platform for future work, along with the examination into other task constraint manipulation, such as understanding how playing number (in)equalities impact the technical and physical profiles of AF players during SSGs.
Two main practical applications stem from this work. First, results demonstrate that strategic manipulation of task constraints (field dimension) can result in behavioral changes in Australian footballers. This could afford a coach with the capability to train players to attune their perceptions to pertinent environmental information of use to inform motor responses, particularly in periods of high player density. Second, coaches could use these results to assist with the periodization of the technical and physical loads generated in their practice environments (7). Specifically, in training phases that require higher perceptual and technical complexity (perhaps “in-season”), coaches may consider the use of “small” SSGs, while periods of the season requiring greater physical load (perhaps “pre-season”), coaches may consider the use of “large” SSGs. Given these results, it would be of interest for future work to ascertain how the manipulation of additional and interacting task constraints, such as player numbers, impacts on the technical and physical activity profiles of AF players.
The authors thank the relevant football club for their participation in this study.
1. Casamichana D, Castellano J. Time motion, heart rate, perceptual and motor behaviour demands in small-sided soccer games: Effects of pitch size. J Sports Sci 28: 1615–1623, 2010.
2. Cohen J. Statistical Power Analysis for the Behavioural Sciences (2nd ed.). Hillsdale, NJ: Erlbaum, 1988. pp. 1–14.
3. Corvino M, Tessitore A, Minganti C, Sibila M. Effect of court dimensions on players' external and internal load during small-sided handball games. J Sports Sci Med 13: 297–303, 2014.
4. Davids K. The constraints-based approach to motor learning. Implications for a non-linear pedagogy in sport and physical education. In: Motor Learning in Practice. A Constraints-led Approach. Renshaw I, Davids K, Savelsbergh GJP, eds. London, United Kingdom: Routledge, 2010. pp. 3–16.
5. Davies MJ, Young W, Farrow D, Bahnert A. Comparison of agility demands of small-sided games in elite Australian Football. Int J Sports Physiol Perform 8: 139–147, 2013.
6. Dellal A, Chamari K, Owen AL, Wong DP, Lago-Penas C, Hill-Haas S. Influence of technical instructions on the physiological and physical demands of small-sided soccer games. Eur J Sport Sci 44: 341–346, 2011.
7. Farrow D, Robertson S. Development of a skill acquisition periodisation framework for high performance sport. Sports Med 47: 1043–1054, 2017.
8. Gabbett TJ, Abernethy B, Jenkins DG. Influence of field size on the physiological and skill demands of small-sided games in junior and senior rugby league players. J Strength Cond Res 26: 487–491, 2012.
9. Klusemann MJ, Pyne DB, Foster C, Drinkwater EJ. Optimizing technical skills and physical loading in small-sided basketball games. J Sports Sci 30: 1463–1471, 2012.
10. Launder A. Player Practice: The Games Approach to Teaching and Coaching. Champaign, IL: Human Kinetics, 2001. pp. 19–31.
11. Magill RA, Anderson DI. Motor Learning and Control: Concepts and Applications (10th ed.). New York City, NY: McGraw-Hill Education, 2014. pp. 255–298.
12. McGinnis PM, Newell KM. Topological dynamics: A framework for describing movement and its constraints. Hum Mov Sci 1: 289–305, 1982.
13. Newell KM, Broderick MP, Deutsch KM, Slifkin AB. Task goals and change in dynamical degrees of freedom with motor learning. J Exp Psychol Hum Percept Perform 29: 379–387, 2003.
14. Newell KM. Constraints on the development of coordination. In: Motor Development in Children: Aspects of Coordination and Control. Wade MG, Whiting HTA, eds. Dordrecht, the Netherlands: Martinus Nijhoff, 1986. pp. 341–360.
15. Pinder R, Davids K, Renshaw I, Araújo D. Representative learning design and functionality of research and practice in sport. J Sport Exerc Psychol 33: 146–155, 2011.
16. Ric A, Torrents C, Goncalves B, Torres-Ronda L, Sampaio J, Hristovski R. Dynamics of tactical behaviour in association football when manipulating players' space of interaction. PLoS One 12: 1–16, 2017.
17. Savelsbergh GJP, van der Kamp J. Information in learning to co-ordinate and control movements: Is there a need for specificity of practice? Int J Sports Psychol 31: 467–484, 2000.
18. Timmerman EA, Farrow D, Savelsberg GJP. The effect of manipulating task constraints on game performance in youth field hockey. Int J Sport Sci Coach 12: 588–594, 2017.
19. Twist C, Highton J, Waldron M, Edwards E, Austin D, Gabbett TJ. Movement demands of elite rugby league players during Australian national rugby league and European super league matches. Int J Sports Phsiol Perform 9: 925–930, 2014.
20. Veale JP, Pearce AJ. Physiological responses of elite junior Australian rules footballers during match-play. J Sports Sci Med 8: 314–319, 2009.
21. Woods C. The use of team performance indicator characteristics to explain ladder position at the conclusion of the Australian Football League home and away season. Int J Perform Anal Sport 16: 837–847, 2016.