Invasion Team Sports Research Articles

Tackling Similarity Search for Soccer Match Analysis: Multimodal Distance Measure and Interactive Query Definition.

Analysts and coaches in soccer sports need to investigate large sets of past matches of opposing teams in a short time to prepare their teams for upcoming matches. Thus, they need appropriate methods and systems supporting them in searching for soccer moves for comparison and explanation. For the search of similar soccer moves, established distance and similarity measures typically only take spatiotemporal features like shape and speed of movement into account. However, movement in invasive team sports such as soccer includes much more than just a sequence of spatial locations. We propose an enhanced similarity measure integrating spatial, player, event as well as high level context such as pressure into the process of similarity search. We present a visual search system supporting analysts in interactively identifying similar contextual enhanced soccer moves in a dataset containing more than 60 soccer matches. Our approach is evaluated by several expert studies. The results of the evaluation reveal the large potential of enhanced similarity measures in the future.

Effects of pitch spatial references on players’ positioning and physical performances during football small-sided games

ABSTRACTThe aim of this study was to identify the effects of adding spatial references during football small-sided games in youth players’ tactical and physical performance. Twelve under-15 players performed a Gk+ 6v6+ Gk game under two playing conditions: (i) without spatial references (CONTROL condition); (ii) with spatial references, by dividing equally the pitch into three corridors and three sectors (experimental situation, LINES). Players’ positional data was used to compute time-motion and tactical-related variables. The results revealed that performance under LINES situation increased the regularity in the zones occupied (~14%, Cohen’s d: 0.5; ±0.3; p = 0.003) and in the distance between teammates’ dyads (~19%, 0.9; ±0.2; p < 0.001). Oppositely, LINES condition decreased the longitudinal synchronization of players’ displacements (0.4; ±0.2; p = 0.002), players’ average speed (0.5; ±0.3; p = 0.002) and distance covered at lower (0.9; ±0.3; p < 0.001) and moderate speed (0.5; ±0.3; p < 0.001). Adding spatial references seems to promote a more structured pattern of play and increase positional regularity. However, coaches should be aware that this constraint may decrease the synchronization between players. Overall, these findings may be generalized to most invasion team sports.

Team and seasonal performance indicator evolution in the ANZ Championship netball league

ABSTRACTAnalysis of netball has received scant attention in the literature and there is little understanding of the dynamics of netball game-play. This study aimed to analyse team and seasonal performance indicator (dis)similarity in the ANZ Championship (netball). Total season values for nine performance indicators were analysed for the ten ANZ Championship teams from 2009 to 2016. The data were analysed using a multivariate, distance-based, approach. Specifically, non-metric multidimensional scaling was used to examine seasonal and team (dis)similarity. After declining from 2009, shooting percentage, goal assists, centre pass receives, penalties and turnovers began to rise from 2011. Both penalties and turnovers declined from 2015, in addition to attempts at goal. The two-dimensional multivariate ordination plot showed relative similarity between each team and season over the observational period, suggesting stagnant game-play dynamics. Further, despite idiosyncratic seasonal profiles, teams generally followed similar directional progression across the ordination surface. Despite being observed in other team invasion sports, league-wide synchronous evolutionary tendencies were not observed within the ANZ Championships between the 2006 to 2016 seasons. However, certain teams did display seasonal fluctuation in their observed multivariate profile, with these seasonal idiosyncrasies being discussed relative to coaching and playing roster changes specific to the analysed team.

What's Next in Complex Networks? Capturing the Concept of Attacking Play in Invasive Team Sports.

The evolution of performance analysis within sports sciences is tied to technology development and practitioner demands. However, how individual and collective patterns self-organize and interact in invasive team sports remains elusive. Social network analysis has been recently proposed to resolve some aspects of this problem, and has proven successful in capturing collective features resulting from the interactions between team members as well as a powerful communication tool. Despite these advances, some fundamental team sports concepts such as an attacking play have not been properly captured by the more common applications of social network analysis to team sports performance. In this article, we propose a novel approach to team sports performance centered on sport concepts, namely that of an attacking play. Network theory and tools including temporal and bipartite or multilayered networks were used to capture this concept. We put forward eight questions directly related to team performance to discuss how common pitfalls in the use of network tools for capturing sports concepts can be avoided. Some answers are advanced in an attempt to be more precise in the description of team dynamics and to uncover other metrics directly applied to sport concepts, such as the structure and dynamics of attacking plays. Finally, we propose that, at this stage of knowledge, it may be advantageous to build up from fundamental sport concepts toward complex network theory and tools, and not the other way around.

Effect of two alternated teaching units of invasion team sports on the tactical learning in primary schoolchildren

The “alternated teaching units” (ATU) is an innovative proposal that consists of delivering two related contents alternatively during the Physical Education (PE) classes, making students aware that both contents are based on the same learning’s principles, increasing the students’ global learning time and making the students perceive one unique learning. This study aimed to compare the use of ATU with traditional teaching units (TTU) on the achieved tactical learning for invasion team sports in primary school students. The sample consisted of 104 students (49 girls and 55 boys; average age = 10.62 ± 0.62 years). Ten PE lessons of soccer and basketball were consecutively (TTU) and alternatively (ATU) delivered. Results showed that the ATU participants had a statistically significant increase in the scores of adjust, guard/mark and general performance indices in basketball, soccer and overall invasion sports compared with the TTU participants (r = 0.24 to 0.64; p < 0.05). Additionally, the ATU participants had a statistically significant higher change of scores in decision-making index in overall invasion sports than the TTU participants (r = 0.25; p < 0.05). ATU applied to related sports should be considered instead of delivering consecutively and unconnectedly blocks of teaching in PE.

Spatio-temporal metrics that distinguish play outcomes: A pilot study

In team invasion sports, tactical behaviour can be examined using spatio-temporal data, i.e. the position of the players at a given time. A review of the spatio-temporal metrics used in team invasion sports performance analysis indicated that thousands of variations of metrics being used. Information about the distribution of metrics' individual effects can inform us of the best variable-selection method. The aim of this pilot study was to estimate the distribution of strong marginal effects of spatio-temporal metrics of field hockey plays. With institutional ethical approval, the Womens’ and Mens’ gold medal games from the EuroHockey 2015 field hockey tournament were recorded. Best, acceptable and worst-case outcomes for plays were described by 1,837 spatio-temporal metrics. Each metric's marginal effects were estimated using Cramer's V, Mutual Information and the I-score. Values for Cramer's V of 0.2 and 0.4 to mark the boundaries of small, moderate and large effects. Less than 1% of metrics show large effects with > 87% of all metrics showing small effects as per the Cramer's V thresholds. These large effect metrics where all within the 98th percentile of Mutual Information values and within the 96th percentile of the I-score values, which supports the Cramer's V distribution of marginal effects. Therefore, according to the recommendations of Tibshirani (1996), univariate variable-selection methods will be the most appropriate for selecting important metrics.

Psychometric Properties of the Spanish Adaptation of the Tactical Skills Inventory for Sports in the School Context.

Tactical skills in sport are the basis for current models of invasion team sports instruction, and relate to other psychological variables that exert a mediating influence on people's normal physical activity. This study aimed to validate the Spanish adaptation of the Tactical Skills Inventory for Sports by Elferink-Gemser, Visscher, Richart, and Lemmink (2004) and verify its factorial invariance. The sample included 540 participants 12.89 ± 1.73 years of age (10 to 17 years old; 333 males and 207 females). The adaptation and translation of the original questionnaire followed the methodological steps established by the International Test Commission. Then two measurement models with the original questionnaire's four-factor structure were compared. Confirmatory analyses were conducted using the software AMOS 21 and maximum likelihood estimation. Results indicated a four-factor structure (positioning and deciding, knowing about ball actions, knowing about others, acting in changing situations) was viable and showed goodness of fit to the data, with adequate indices of reliability (α between .72 and .86) and validity (GFI = .955; RMSEA = .044; CFI = .974), and strong evidence of stability in the factor structure. Moreover, factorial invariance was observed between participants who do and do not practice team sports, with practitioners scoring higher than non-practitioners on all factors (p < .001; effect sizes between .97 and 1.08).

Does training with 3D videos improve decision-making in team invasion sports?

We examined the effectiveness of video-based decision training in national youth handball teams. Extending previous research, we tested in Study 1 whether a three-dimensional (3D) video training group would outperform a two-dimensional (2D) group. In Study 2, a 3D training group was compared to a control group and a group trained with a traditional tactic board. In both studies, training duration was 6 weeks. Performance was measured in a pre- to post-retention design. The tests consisted of a decision-making task measuring quality of decisions (first and best option) and decision time (time for first and best option). The results of Study 1 showed learning effects and revealed that the 3D video group made faster first-option choices than the 2D group, but differences in the quality of options were not pronounced. The results of Study 2 revealed learning effects for both training groups compared to the control group, and faster choices in the 3D group compared to both other groups. Together, the results show that 3D video training is the most useful tool for improving choices in handball, but only in reference to decision time and not decision quality. We discuss the usefulness of a 3D video tool for training of decision-making skills outside the laboratory or gym.

Using Voronoi diagrams to describe tactical behaviour in invasive team sports: an application in basketball

Team sports are recognized as dynamic systems of interaction, where individual and collective patterns of behaviour emerge from a confluence of multiple constraints on the players. Players’ distribution on the field is often associated with tactical options, strategic decisions, and principles. Hence, the analysis of positional data has gained relevance and is now a hot topic in sport science research, feeding the interests of researchers, analysts and coaches. Recent studies tried to characterize teams’ and players’ tactical behaviour using variables derived from their spatial distribution (e.g., the area of the convex hull). The Voronoi diagrams (VD) have been introduced for the analysis of the spatial organization in team sports as it allows, unlike other models, defining the dominant region of both players and teams. This study considered the application of VD to describe o;ensive tactical behaviour in basketball. For this purpose, a formal game between two teams of Austrian basketball talents was observed, from where 19 ofensive sequences were selected for analysis. The results indicate that the percentage of area occupied by each team, at each instant, can be considered to describe teams’ patterns of spatial organization during a positional attack, allowing to classify and distinguish transition phase and organized phase. This approach may also help to reduce the time spent in video analysis by coaches, giving them more time to prepare training plans and sessions in order to improve the tactical and strategic performance of their teams.

Emergence of Contact Injuries in Invasion Team Sports: An Ecological Dynamics Rationale

The incidence of contact injuries in team sports is considerable, and injury mechanisms need to be comprehensively understood to facilitate the adoption of preventive measures. In Association Football, evidence shows that the highest prevalence of contact injuries emerges in one-on-one interactions. However, previous studies have tended to operationally report injury mechanisms in isolation, failing to provide a theoretical rationale to explain how injuries might emerge from interactions between opposing players. In this position paper, we propose an ecological dynamics framework to enhance current understanding of behavioural processes leading to contact injuries in team sports. Based on previous research highlighting the dynamics of performer-environment interactions, contact injuries are proposed to emerge from symmetry-breaking processes during on-field interpersonal interactions among competing players and the ball. Central to this approach is consideration of candidate control parameters that may provide insights on the information sources used by players to reduce risk of contact injuries during performance. Clinically, an ecological dynamics analysis could allow sport practitioners to design training sessions based on selected parameter threshold values as primary and/or secondary preventing measures during training and rehabilitation sessions.

Numerical relations and skill level constrain co-adaptive behaviors of agents in sports teams.

Similar to other complex systems in nature (e.g., a hunting pack, flocks of birds), sports teams have been modeled as social neurobiological systems in which interpersonal coordination tendencies of agents underpin team swarming behaviors. Swarming is seen as the result of agent co-adaptation to ecological constraints of performance environments by collectively perceiving specific possibilities for action (affordances for self and shared affordances). A major principle of invasion team sports assumed to promote effective performance is to outnumber the opposition (creation of numerical overloads) during different performance phases (attack and defense) in spatial regions adjacent to the ball. Such performance principles are assimilated by system agents through manipulation of numerical relations between teams during training in order to create artificially asymmetrical performance contexts to simulate overloaded and underloaded situations. Here we evaluated effects of different numerical relations differentiated by agent skill level, examining emergent inter-individual, intra- and inter-team coordination. Groups of association football players (national – NLP and regional-level – RLP) participated in small-sided and conditioned games in which numerical relations between system agents were manipulated (5v5, 5v4 and 5v3). Typical grouping tendencies in sports teams (major ranges, stretch indices, distances of team centers to goals and distances between the teams' opposing line-forces in specific team sectors) were recorded by plotting positional coordinates of individual agents through continuous GPS tracking. Results showed that creation of numerical asymmetries during training constrained agents' individual dominant regions, the underloaded teams' compactness and each team's relative position on-field, as well as distances between specific team sectors. We also observed how skill level impacted individual and team coordination tendencies. Data revealed emergence of co-adaptive behaviors between interacting neurobiological social system agents in the context of sport performance. Such observations have broader implications for training design involving manipulations of numerical relations between interacting members of social collectives.

Ruck-Play as Performance Indicator during the 2010 Six Nations Championship: A Commentary

INTRODUCTION Analytics is the use of statistical analysis to support decision making in organisations. It is a growth area in many sectors including elite sport. Sports analytics is increasingly becoming part of the coaching toolkit as more and more teams analyse the numbers to go alongside their video analysis. Michael Lewis’s book, Moneyball: The Art of Winning an Unfair Game (2003), subsequently turned into an Oscar-nominated film starring Brad Pitt, tells the story of how the Oakland Athletics in Major League Baseball embraced the use of statistical analysis in player recruitment and has done much to publicise the possibilities of sports analytics as a David strategy for resource-constrained teams trying to become more competitive. Rugby union is one of the invasion team sports that has been more open than most to the possibilities for analytics as part of an evidence-based approach to coaching. As both a professor of business and sports analytics as well as a technical analyst with a leading English Premiership rugby union team, I welcome Kraak and Welman’s detailed study of ruck play using data from the 15 games in the 2010 Six Nations tournament. My comments on their study come mainly from the perspective of an active practitioner in elite sport. The main contribution of Kraak and Welman is the level of detail on ruck play, particularly the numbers of players involved in a ruck. The 15 games studied yielded 1,479 rucks all of which have been analysed. Although a mass of performance data on all aspects of the game is now available to teams from commercial providers, this data remains primarily tally data on the quantity of different activities and their outcomes, usually coded by player, time and field position. What we still lack is extensive data across teams and games on the micro technical detail of the activities. So, for example, teams are able to source data on the number of rucks contested and the outcomes but are less likely to have detailed data on the number of attackers and defenders involved in each ruck.

Invasion team sports: strategy and match modeling

Performance in invasion team sports (ITS) is highly dependent on the appropriate execution of an efficient team strategy by the players. Strategy enables individual and group actions to be organized in order to produce collective execution, increasing the diversity and unpredictabilityof the team’s actions. Nonetheless, the scientific literature has mainly analyzed match events without integrating information regarding the observed behavioral patterns and team strategies. As neither match dynamics nor its relation with strategy has been modeled, this study accomplishesthese goals through the following steps. First, the organization of collective play was defined by modeling the strategic control of players’ actions and the coordination of cooperative actions between two or more players. Second, a model was constructed that supports the design of strategies,including constraints that ensure effective usefulness of the strategy. Third, the integration between the strategy model and the dynamical structure of the confront was established through the match model. The presented model may provide insight for team training, match analysis, supportthe development of intelligent software to help coaches design team strategies and as prior information about the collective movements of the team players acquired from automatic player recognition using digital video images.

Providing sufficient opportunity to learn: a response to Grehaigne, Caty and Godbout

Background: Over the last 30 years, traditional skill-based game teaching models have gradually been supplemented by instruction under an inclusive banner of ‘Teaching Games for Understanding’ (TGfU). This approach focuses on developing tactical understanding through modified games and a philosophy that places the learner rather than the game at the centre of instruction. A recent paper by Grehaigne, Caty, and Godbout, ‘Modelling Ball Circulation in Invasion Team Sports: A Way to Promote Learning Games Through Understanding’, had a threefold focus.• to report the results of a qualitative study on various offensive configurations of football play observed in Physical Education (PE) lessons with young novice players• to propose a model of game play based on the analysis of such configurations of play• to promote a radical constructivist teaching approach based on ‘learning games through understanding’ that challenges the long-established TGfU methodologyPurpose: This paper critically examines the contention of Grehaigne, Caty, and Godbout that the presentation of tactical data collected through the observation of novices playing sport in ill-structured domains, e.g. team games such as football, represent a useful pedagogical model that promotes ‘learning games through understanding’. In rewording the familiar TGfU approach, and calling it ‘learning games through understanding’, Grehaigne, Caty, and Godbout challenge the evolving TGfU approach as too solutions based and not sufficiently student centred.Discussion: This paper challenges the use of radical constructivism as a construct for the development of a philosophy for instructing novices in team games, in this instance, football. It defines ill-structured and well-structured learning domains and suggests that effective pedagogy, the art of teaching, requires flexible attitudes towards the choice of pedagogy in games. By inference, it also challenges Grehaigne, Caty, and Godbout's assumption that the TGfU models previously published, e.g. TGfU, Game Sense, Play Practice, and interpreted in various texts, e.g. Transforming Play, Teaching Tactics and Game Sense, are not student-centred and teacher dominated.Conclusions: Grehaigne, Caty, and Godbout's work is important in the evolution of models of game instruction for use in PE contexts. The concept of ‘learning games through understanding’ is a timely reminder of the importance of pedagogies, for example, guided or discovery learning that induce effortful thinking. However, if we consider TGfU in its true philosophical light, that is, as a holistic and experiential approach to teaching, then it already encompasses ‘learning games through understanding’. Because no two students learn or conceive knowledge in exactly the same way, teaching contexts require a flexible approach to instruction, based on a methodological continuum of empirical to radical constructivism. In short, providing novice learners with sufficient opportunities to learn, requires flexibility and a holistic experiential approach to teaching that is appropriate for the learner, activity and context.

An above real time training intervention for sport decision making

ObjectivesA speeded video-based decision-making training intervention was used to assess the impact of above real time training on decision-making skill in sport. Design and methodsThree groups completed pre tests and either five weeks of fast speed video training, normal speed video training or no training, followed by a post test and two retention tests in subsequent weeks. Decision accuracy was measured by awarding three, two, one, or no point(s) based on independent coach ratings of each situation. ResultsResults revealed that those trained in above real time improved performance earlier in the training intervention compared to those trained in normal speed. The above real time group also retained more of the performance improvements. The transfer test for decision accuracy showed improvement following the training intervention for all groups, trends in the data reflected a higher retention rate for the fast speed group choosing the bet option more frequently than normal and control groups. ConclusionsThe results lend support to the general use of video-based decision-making training for team invasion sports. A greater impact is that they provide a new paradigm by adapting above real time training to decision making, to create a more game-like training scenario.

Configurations of play in invasion team sports and learning by analogy

Configurations of play in invasion team sports and learning by analogy

MODELO DE JOGO E PROCESSO DE ENSINO NO FUTEBOL: PRINCÍPIOS GLOBAIS E ESPECÍFICOS

A dimensão tática deve ser entendida como centro do processo de ensino-aprendizagem no futebol, assumindo seu caráter organizativo, multidimensional, intencional e consciente. Este artigo teve como foco, apresentar conceitos e conteúdos para sustentação de um processo de ensino aprendizagem, baseado na dimensão tática e na interdependência de três aspectos fundamentais: metodologia para compreensão tática, princípios de jogo e modelos de jogo. Foram sugeridos e descritos em específico, princípios e modelos de jogo direcionados para diferentes faixas etárias, sendo estes compostos por peculiaridades comuns aos jogos esportivos coletivos de invasão.

Modelling ball circulation in invasion team sports: a way to promote learning games through understanding

Background: Sport Education and ‘Tactical decision learning model’ (TDLM) are two curriculum models used by physical education teachers in France to help students in the development of a tactical intelligence of game play in the didactics of team sports. Purpose: Identify prototypic configurations of play in the sense that they represent an original model, archetype of a model that reproduces itself. Participants and setting: We conducted this research with three age groups of secondary school students (grade 6, 11/12 years old; grade 9, 14/15 years old; and grade 12, 17/18 years old). The team sport selected was soccer with a 4 vs. 4 player format. Research design: Based on the video observation of a final sample of 188 sequences of play having led to a shot at goal or a goal, we considered three categories of criteria combined and came up with eight prototypic configurations of play often displayed by novice players. Data collection: The following observational categories were retained: (1) location of ball recovery with reference to the position on the field: high ball recovery (in the opponents’ side of the pitch); recovery in the middle area of the pitch; low recovery (in one's own side of the pitch); (2) location of recovery with reference to the position within the effective (occupied) play-space (EP-S): recovery at the front of the EP-S; recovery at the rear or in the middle area of the EP-S; and (3) selection of play after recovery of the ball: (a) ball exchange with a teammate (long pass, short pass, number of passes); and (b) dribble. Findings: Eight prototypic offensive configurations of play are presented and a related didactical model is discussed. Conclusions: In team sports with a constructivist approach, such modelling, models and efficient-action rules imply a construction by students. It requires an evolution of the various concepts involved, a progressive integration, step-wise or level-wise, to a thinking process. For this purpose, in the teaching/learning system in team sports, we propose to switch from Teaching Games for Understanding to Learning Games through Understanding.

How to become a better sport

Teaching the skills common to all invasion games and team sports. By Crispin Andrews, a primary teacher, qualified sports coach and freelance writer from Aylesbury, Buckinghamshire.