Thesis

Bridging the gap between soccer and science: using technology and data to optimize training and performance

Authors:
  • FC Twente/Heracles Academie
To read the full-text of this research, you can request a copy directly from the author.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the author.

ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
This study aimed to: (1) develop an integrated approach to quantifying match physical-tactical performance and (2) comprehensively examine the validity and reliability of this novel approach. Both UEFA qualified coaches and performance analysts (n = 30) participated to verify the scientific robustness of this new method. The percentage of correct responses were used to verify the validity of the integrated approach and the minimum acceptable agreement was set at 80%. Two well-trained groups of observers analysed a randomly selected English Premier League match for inter- and intra-observer reliability using the kappa statistic. A high degree of validity was demonstrated as the mean percentage of correct responses by all participants, accounting for 91.8 ± 4.3% for all, 92.2 ± 4.7% for out-of-possession, and 91.6 ± 5.7% for in-possession physical-tactical variables. Inter- and intra-observer reliability were found to be strong (κ = 0.81) to almost perfect (κ = 0.94), respectively. Additional analyses demonstrated that there was a nearly perfect correlation between data derived from the novel filter used for the present study to capture high-intensity running and those obtained from the filter of the commercial data provider (r = 0.99; P < 0.01). The data demonstrates that the integrated approach is valid and reliable regarding the quantification of physical-tactical performances. Therefore, it is now possible to unveil unique high-intensity profiles of elite players related to key tactical actions. This may help coaches and practitioners better understand the physical-tactical performances of players, as well as effectively translate physical metrics into training.
Article
Full-text available
Seeding in tournaments is a process of creating a schedule based on performance in the recent past. It is used in many athletic disciplines to ensure that particularly attractive match ups do not occur until the later stages of the tournament. We exploit the discontinuous nature of the seeding system in the UEFA Champions League and the UEFA Europa League as a natural experiment to estimate the causal effect of being seeded. We find no evidence that seeding itself contributes positively to the team’s success in the tournament. This finding is surprising given the substantial drop in average strength of the opponents for seeded teams and in striking opposition to the findings of previous studies.
Article
Full-text available
The current technological age has created exponential growth in the availability of technology and data in every industry, including sport. It is tempting to get caught up in the excitement of purchasing and implementing technology, but technology has a potential dark side that warrants consideration. Before investing in technology, it is imperative to consider the potential roadblocks, including its limitations and the contextual challenges that compromise implementation in a specific environment. A thoughtful approach is therefore necessary when deciding whether to implement any given technology into practice. In this article, we review the vision and pitfalls behind technology's potential in sport science and medicine applications and then present a critical decision-making framework of 4 simple questions to help practitioners decide whether to purchase and implement a given technology.
Article
Full-text available
The aim of this study was to investigate position-specific endurance performance of soccer players. 136 professional players competing in the 1st and 2nd division in Germany were divided into the positional groups goalkeepers (GK), central defenders (CD), wingers (WI), central midfielders (CM), and forwards (FW). All players performed an incremental treadmill test with blood lactate sampling until exhaustion with the following endurance parameters being obtained: Fixed aerobic threshold (v2mmol/l), fixed anaerobic threshold (v4mmol/l), individual aerobic threshold (vLT), individual anaerobic threshold (vIAT), and maximum velocity (vmax). Results revealed significant differences between GK and all outfield playing positions for all endurance parameters (p ≤ 0.03; ES 0.87–2.19). No significant differences among outfield playing positions were evident for any of the parameters. However, trends were found in favor of the CM compared to the WI (p = 0.11; ES = 0.68) and the FW (p = 0.06; ES = 0.47) relating to vLT as well as in favor of the CM compared to the WI (p = 0.10; ES = 0.56) relating to vIAT. Findings suggest that goalkeepers possess the lowest endurance capacity compared to other playing positions. While outfield players in general showed similar endurance performance, CM seem to possess the highest aerobic capacity of all positions as indicated by all lactate-based thresholds, however, with only small to moderate ES. These findings could lead one to question the appropriateness of current endurance training regimes to prepare all players adequately for their positional match-running demands. Indeed, endurance training of players should be specific to their match-running demands. However, it remains unknown to what extent these demands are position or player specific.
Article
Full-text available
The increase in the volume of structured and unstructured data related to more than just sport events leads to the development and increased use of techniques that extract information and employ machine‐learning algorithms in predicting process outcomes based on input but not necessarily output data. Taking sports into consideration, predicting outcomes, and extracting valuable information has become appealing not only to sports workers but also to the wider audience, particularly in the areas of team management and sports betting. The aim of this article is to review the existing machine learning (ML) algorithms in predicting sport outcomes. Over 100 papers were analyzed and only some of these papers were taken into consideration. Almost all of the analyzed papers use some sort of feature selection and feature extraction, most often prior to using the machine‐learning algorithm. As an evaluation method of ML algorithms, researchers, in most cases, use data segmentation with data being chronologically distributed. In addition to data segmentation, researchers also use the k ‐cross‐evaluation method. Sport predictions are usually treated as a classification problem with one class being predicted and rare cases being predicted as numerical values. Mostly used ML models are neural networks using data segmentation. This article is categorized under: • Technologies > Machine Learning • Technologies > Prediction
Article
Full-text available
In the last few decades, a number of technological developments have advanced the spread of wearable sensors for the assessment of human motion. These sensors have been also developed to assess athletes’ performance, providing useful guidelines for coaching, as well as for injury prevention. The data from these sensors provides key performance outcomes as well as more detailed kinematic, kinetic, and electromyographic data that provides insight into how the performance was obtained. From this perspective, inertial sensors, force sensors, and electromyography appear to be the most appropriate wearable sensors to use. Several studies were conducted to verify the feasibility of using wearable sensors for sport applications by using both commercially available and customized sensors. The present study seeks to provide an overview of sport biomechanics applications found from recent literature using wearable sensors, highlighting some information related to the used sensors and analysis methods. From the literature review results, it appears that inertial sensors are the most widespread sensors for assessing athletes’ performance; however, there still exist applications for force sensors and electromyography in this context. The main sport assessed in the studies was running, even though the range of sports examined was quite high. The provided overview can be useful for researchers, athletes, and coaches to understand the technologies currently available for sport performance assessment.
Article
Full-text available
Fairness has several interpretations in sports, one of them being that the rules should guarantee incentive compatibility, namely, a team cannot be worse off due to better results in any feasible scenario. The current seeding regime of the most prestigious annual European club football tournament, the UEFA (Union of European Football Associations) Champions League, is shown to violate this requirement since the 2015/16 season. In particular, if the titleholder qualifies for the first pot by being a champion in a high-ranked league, its slot is given to a team from a lower-ranked association, which can harm a top club from the domestic championship of the titleholder. However, filling all vacancies through the national leagues excludes the presence of perverse incentives. UEFA is encouraged to introduce this policy from the 2021-24 cycle onwards.
Article
Full-text available
SUMMARY: Success in artistic gymnastics certainly depends on the morphological characteristics of the athletes, some of which are basic body height and weight. The aim of this study is to determine whether there is a statistically significant difference in body height and weight between the male and female participants on the Olympic Games from 1996 to 2016. The following variables were chosen: body weight (kg), body height (cm). The results did show significant differences between women's artistic gymnastics in body weight and height over a 20-year period. In men's artistic gymnastics we didn't find the difference. This research is important in setting comparative parameters in the selection process and sport specialization regarding morphological characteristics according to different events of gymnasts at the elite level.
Article
Full-text available
Predicting the results of soccer competitions and the contributions of match attributes, in particular, has gained popularity in recent years. Big data processing obtained from different sensors, cameras and analysis systems needs modern tools that can provide a deep understanding of the relationship between this huge amount of data produced by sensors and cameras, both linear and non-linear data. Using data mining tools does not appear sufficient to provide a deep understanding of the relationship between the match attributes and results and how to predict or optimize the results based upon performance variables. This study aimed to suggest a different approach to predict wins, losses and attributes' sensitivities which enables the prediction of match results based on the most sensitive attributes that affect it as a second step. A radial basis function neural network model has successfully weighted the effectiveness of all match attributes and classified the team results into the target groups as a win or loss. The neural network model's output demonstrated a correct percentage of win and loss of 83.3% and 72.7% respectively, with a low Root Mean Square training error of 2.9% and testing error of 0.37%. Out of 75 match attributes, 19 were identified as powerful predictors of success. The most powerful respectively were: the Total Team Medium Pass Attempted (MBA) 100%; the Distance Covered Team Average in zone 3 (15-20 km/h; Zone3_TA) 99%; the Team Average ball delivery into the attacking third of the field (TA_DAT) 80.9%; the Total Team Covered Distance without Ball Possession (Not in_Poss_TT) 76.8%; and the Average Distance Covered by Team (Game TA) 75.1%. Therefore, the novel radial based function neural network model can be employed by sports scientists to adapt training, tactics and opposition analysis to improve performance.
Article
Full-text available
Body height is among the most important attributes of basketball players. Whether it differs among the basketball players who compete at the world basketball championship (FIBA-WC) is unknown. The aim of this study was to investigate the differences between the basketball players from the teams ranked 1-16 and those ranked below 16th place. The body heights of all players from the last three FIBA-WCs were collected and allocated according to the ranking at the FIBA-WC and analyzed by position in team. An independent sample t-test was conducted to analyze the difference in body height of players ranked 1-16 and players who ranked below 16th place. The players from the first 16 teams were significantly taller at three positions: point guards (Difference = 2.92 cm, p < 0.05), shooting guards (Difference = 2.16 cm, p < 0.05), and small forwards (Difference = 2.49 cm, p < 0.01). Body height seems to be an important factor for the performance of the basketball team at the FIBA-WC. Considering that all players at the FIBA-WC went through rigorous selection process to be in their national teams, body height of the higher-ranked players could be used as a reference value.
Article
Full-text available
(This paper has OPEN ACCESS) In 2020 the world faces an unprecedented challenge consequence of the coronavirus pandemic. Soccer, arguably, one of the most globalized industries, has reacted to the event by closing down much of its operations. This paper reviews the speed and breadth of such closures and compares it to another major event in history: The Second World War. During the latter, soccer continued operations despite the apparent difficulties. Total closures were, for the most part, temporary, and the sport was practiced whenever possible. The way that soccer operated during the war contrasts with how it has responded to the COVID-19 crisis. In record time, virtually all leagues in the world stopped operations, and for the first time since its foundation, with but a few exceptions, there were no games anywhere in the world.
Article
Full-text available
The present study aimed to validate and compare the football-specific measurement accuracy of two optical tracking systems engineered by TRACAB. The "Gen4" system consists of two multi-camera units (a stereo pair) in two locations either side of the halfway line, whereas the distributed "Gen5" system combines two stereo pairs on each side of the field as well as two monocular systems behind the goal areas. Data were collected from 20 male football players in two different exercises (a football sport-specific running course and small-sided games) in a professional football stadium. For evaluating the accuracy of the systems, measures were compared against simultaneously recorded measures of a reference system (VICON motion capture system). Statistical analysis uses RMSE for kinematic variables (position, speed and acceleration) and the difference in percentages for performance indicators (e.g. distance covered, peak speed) per run compared to the reference system. Frames in which players were obviously not tracked were excluded. Gen5 had marginally better accuracy (0.08 m RMSE) for position measurements than Gen4 (0.09 m RMSE) compared to the reference. Accuracy difference in instantaneous speed (Gen4: 0.09 m⋅s-1 RMSE; Gen5: 0.08 m⋅s-1 RMSE) and acceleration (Gen4: 0.26 m⋅s-2 RMSE; Gen5: 0.21 m⋅s-2 RMSE) measurements were significant, but also trivial in terms of the effect size. For total distance travelled, both Gen4 (0.42 ± 0.60%) and Gen5 (0.27 ± 0.35%) showed only trivial deviations compared to the reference. Gen4 showed moderate differences in the low-speed distance travelled category (-19.41 ± 13.24%) and small differences in the high-speed distance travelled category (8.94 ± 9.49%). Differences in peak speed, acceleration and deceleration were trivial (
Article
Full-text available
Purpose: To assess injury risk in elite-level youth football players based on anthropometric, motor coordination and physical performance measures with a machine learning approach. Methods: A total of 734 players in the U10 to U15 age categories (mean age: 11.7 +/- 1.7 years) from seven Belgian youth academies were prospectively followed during one season. Football exposure and occurring injuries were monitored continuously by the academies' coaching and medical staff, respectively. Preseason anthropometric measurements (height, weight, and sitting height) were taken and test batteries to assess motor coordination and physical fitness (strength, flexibility, speed, agility, and endurance) were performed. An extreme gradient boosting algorithms (XGBoost) was used to predict injury based on the preseason test results. Subsequently, the same approach was used to classify injuries as either overuse or acute. Results: During the season, half of the players (n = 368) sustained at least one injury. Of the first occurring injuries, 173 were identified as overuse and 195 as acute injuries. The machine learning algorithm was able to identify the injured players in the hold-out test sample with 85% precision, 85% recall (sensitivity) and 85% accuracy (f1-score). Furthermore, injuries could be classified as overuse or acute with 78% precision, 78% recall and 78% accuracy. Conclusion: Our machine learning algorithm was able to predict injury and to distinguish overuse from acute injuries with reasonably high accuracy based on preseason measures. Hence, it is a promising approach to assess injury risk among elite-level youth football players. This new knowledge could be applied in the development and improvement of injury risk management strategies, to identify youth players with the highest injury risk.
Article
Full-text available
The benefits of differentiating between the physiological and biomechanical load-response pathways in football and other (team) sports have become increasingly recognised. In contrast to physiological loads however, the biomechanical demands of training and competition are still not well understood, primarily due to the difficulty of quantifying biomechanical loads in a field environment. Although musculoskeletal adaptation and injury are known to occur at a tissue level, several biomechanical load metrics are available that quantify loads experienced by the body as a whole, its different structures and the individual tissues that are part of these structures. This paper discusses the distinct aspects and challenges that are associated with measuring biomechanical loads at these different levels in laboratory and/or field contexts. Our hope is that through this paper, sport scientists and practitioners will be able to critically consider the value and limitations of biomechanical load metrics and will keep pursuing new methods to measure these loads within and outside the lab, as a detailed load quantification is essential to better understand the biomechanical load-response pathways that occur in the field.
Article
Full-text available
This study investigated the time-motion characteristics of football teams in the Spanish first division, in relation to their final competitive level as defined by league position (Champions League, Europa League, Upper mid-table, lower mid-table and relegation). Match observations (n = 9641) were collected using a multiple-camera computerized tracking system during the 2013-2014 competitive season. The following match parameters were analyzed: total distance, relative distance (m·min −1), distance < 14 km·h −1 , >14 km·h −1 , between 14-21 km·h −1 , >21 km·h −1 , and >24 km·h −1. Total distance and distance at different velocities (>14, 21, and 24 km·h −1) in and out of ball possession were also analyzed. A repeated analysis of variance and a comparison of effect sizes were carried out to compare the performance of the teams. The analysis of the data showed differences in physical performance characteristics between competitive levels. The volume of distance covered in the variables analyzed did not relate to success in soccer. Both successful and unsuccessful teams presented the same running requirements at higher velocities. These findings provide valuable information about the physical demands of the running requirements according to their final position in the league table.
Article
Full-text available
Research paradigms guide scientific discoveries through their assumptions and principles. Understanding paradigm-specific assumptions helps illuminate the quality of findings that support scientific studies and identify gaps in generating sound evidence. This article focuses on the research paradigm of positivism, examining its definition, history, and assumptions (ontology, epistemology, axiology, methodology, and rigor). Positivism is aligned with the hypothetico-deductive model of science that builds on verifying a priori hypotheses and experimentation by operationalizing variables and measures; results from hypothesis testing are used to inform and advance science. Studies aligned with positivism generally focus on identifying explanatory associations or causal relationships through quantitative approaches, where empirically based findings from large sample sizes are favored-in this regard, generalizable inferences, replication of findings, and controlled experimentation have been principles guiding positivist science. Criteria for evaluating the quality of positivist research are discussed. An example from health professions education is provided to guide positivist thinking in study design and implementation.
Article
Full-text available
Soccer analytics is attracting increasing interest in academia and industry, thanks to the availability of sensing technologies that provide high-fidelity data streams for every match. Unfortunately, these detailed data are owned by specialized companies and hence are rarely publicly available for scientific research. To fill this gap, this paper describes the largest open collection of soccer-logs ever released, containing all the spatio-temporal events (passes, shots, fouls, etc.) that occured during each match for an entire season of seven prominent soccer competitions. Each match event contains information about its position, time, outcome, player and characteristics. The nature of team sports like soccer, halfway between the abstraction of a game and the reality of complex social systems, combined with the unique size and composition of this dataset, provide an ideal ground for tackling a wide range of data science problems, including the measurement and evaluation of performance, both at individual and at collective level, and the determinants of success and failure.
Article
Full-text available
It is common practice amongst coaches and analysts to search for key performance indicators related to attacking play in football. Match analysis in professional football has predominately utilised notational analysis, a statistical summary of events based on video footage, to study the sport and prepare teams for competition. Recent increases in technology have facilitated the dynamic analysis of more complex process variables, giving practitioners the potential to quickly evaluate a match with consideration to contextual parameters. One field of research, known as machine learning, is a form of artificial intelligence that uses algorithms to detect meaningful patterns based on positional data. Machine learning is a relatively new concept in football, and little is known about its usefulness in identifying performance metrics that determine match outcome. Few studies and no reviews have focused on the use of machine learning to improve tactical knowledge and performance, instead focusing on the models used, or as a prediction method. Accordingly, this article provides a critical appraisal of the application of machine learning in football related to attacking play, discussing current challenges and future directions that may provide deeper insight to practitioners.
Conference Paper
Full-text available
Assessing the impact of the individual actions performed by soccer players during games is a crucial aspect of the player recruitment process. Unfortunately, most traditional metrics fall short in addressing this task as they either focus on rare actions like shots and goals alone or fail to account for the context in which the actions occurred. This paper introduces (1) a new language for describing individual player actions on the pitch and (2) a framework for valuing any type of player action based on its impact on the game outcome while accounting for the context in which the action happened. By aggregating soccer players' action values, their total offensive and defensive contributions to their team can be quantified. We show how our approach considers relevant contextual information that traditional player evaluation metrics ignore and present a number of use cases related to scouting and playing style characterization in the 2016/2017 and 2017/2018 seasons in Europe's top competitions.
Article
Full-text available
Evaluation of cardiorespiratory parameters is important in athletic population in order to monitor the training status and define training intensities. The aim of this study was to validate an easy-to-perform running test called RABIT® (Running Advisor Billat Training) for detecting aerobic (AerT) and anaerobic (AnT) threshold and maximum parameters. Fifteen trained-runners completed a graded (GRAD) and RABIT test (four self-selected pace steps: 1) ten minutes at free warm-up pace 2) five minutes at medium pace 3) three minutes at hard pace 4) ten minutes at easy pace). We compared the cardiorespiratory parameters and running speed of the RABIT with those corresponding to AerT, AnT or maximum parameters obtained by the GRAD. The ⩒O2max, HRmax, RERmax and running speed max measured during the three minutes hard pace of the RABIT were not statistically different from the maximum parameters measured during GRAD (p > 0.05). The ⩒O2, HR and RER measured during the medium and easy pace of the RABIT were not significantly different from the AnT and AerT parameters measured during GRAD. In conclusion, RABIT was validated for all the maximum parameters and for most of Ant- and AerT- related parameters and then it might be used for detecting training zones in athletes.
Article
Full-text available
Ecological dynamics is a contemporary theory of skill acquisition, advocating the mutuality of the performer-environment system, with clear implications for the design of innovative training environments in elite sport. It contends that performance behaviours emerge, and are adapted, by athletes satisfying a confluence of constraints impacting on their structural and functional capacities, the physics of a performance environment and the intended task goals. This framework implicates contemporary models of coaching, training design and sport science support, to stimulate continuous interactions between an individual and performance environment, predicated on representative learning designs (RLD). While theoretical principles of RLD in ecological dynamics are tangible, their practical application in elite and high level (team) sports need verification. Here, we exemplify how data sampled from a high-performance team sport setting could underpin innovative methodologies to support practitioners in designing representative training activities. We highlight how the use of principles grounded within ecological dynamics, along with data from performance analytics, could suggest contemporary models of coaching and preparation for performance in elite sport.
Article
Full-text available
Due to the tremendous popularity of youth football, practitioners in this domain face the ongoing question of the most effective solutions in early talent selection. Although the scientific community has suggested multidimensional models for some time, coach assessments and motor performance tests remain common. Earlier research has determined the strengths and weaknesses within these different approaches. The current investigation directly compared the effectiveness of each approach in talent selection (coach assessment vs. motor performance tests vs. multidimensional data). A sample of 117 youth football players, their parents, and coaches participated in multidimensional measurements in the U14 age category (coach assessments, motor performance tests, psychological characteristics, familial support, training history, and biological maturation). The area under the curve (AUC [95% CI]) from receiver operating characteristic indicated the prognostic validity of each approach in predicting U19 player status five years after the assessments (professional vs. non-professional). Motor performance tests (0.71 [0.58; 0.84]) showed a lower AUC than the multidimensional data (0.85 [0.76; 0.94], p = 0.02), whilst coach assessments did not differ from the two others (.82 [.74; .90]). Further, combined talent selection approaches, especially the use of coach assessments and multidimensional data together, were significantly better at predicting U19 player status (0.93 [0.87; 0.98], p = 0.02 vs. multidimensional data only). Although certain limitations may impede further insights (summation of data, skipped use of non-linear statistics), scientific claims for using multidimensionality within talent selection were confirmed to be fruitful. In particular, the combination of the subjective coaches' eye with scientific data may buffer the mutual weaknesses of these different approaches. Future research should focus on optimizing the output of promising multidimensional models. Knowledge of detailed values relating to specific dimensions within these models and the implementation of enhanced non-linear statistics may enable further improvements in the field of talent selection.
Article
Full-text available
Martín-García, A, Gómez Díaz, A, Bradley, PS, Morera, F, and Casamichana, D. Quantification of a professional football team's external load using a microcycle structure. J Strength Cond Res 32(12): 3520-3527, 2018-The aims of this study were to (a) determine the external load of a football team across playing position and relative to competition for a structured microcycle and (b) examine the loading and variation the day after competition for players with or without game time. Training and match data were obtained from 24 professional football players who belonging to the reserve squad of a Spanish La Liga club during the 2015/16 season using global positioning technology (n = 37 matches and n = 42 training weeks). Training load data were analyzed with respect to the number of days before or after a match (match day [MD] minus or plus). Training load metrics declined as competition approached (MD-4 > MD-3 > MD-2 > MD-1; p < 0.05; effect sizes [ES]: 0.4-3.1). On the day after competition, players without game time demonstrated greater load in a compensatory session (MD + 1C) that replicated competition compared with a recovery session (MD + 1R) completed by players with game time (MD + 1C > MD + 1R; p < 0.05; ES: 1.4-1.6). Acceleration and deceleration metrics during training exceeded 50% of that performed in competition for MD + 1C (80-86%), MD-4 (71-72%), MD-3 (62-69%), and MD-2 (56-61%). Full backs performed more high-speed running and sprint distance than other positions at MD-3 and MD-4 (p < 0.05; ES: 0.8-1.7). The coefficient of variation for weekly training sessions ranged from ∼40% for MD-3 and MD-4 to ∼80% for MD + 1R. The data demonstrate that the external load of a structured microcycle varied substantially based on the players training day and position. This information could be useful for applied sports scientists when trying to systematically manage load, particularly compensatory conditioning for players without game time.
Conference Paper
Full-text available
Abstract: Many models have been constructed to quantify the quality of shots in soccer. In this paper, we evaluate the quality of off-ball positioning, preceding shots, that could lead to goals. For example, consider a tall unmarked center forward positioned at the far post during a corner kick. Sometimes the cross comes in and the center forward heads it in effortlessly, other times the cross flies over his head. Another example is of a winger, played onside, while making a run in past the defensive line. Sometimes the through-ball arrives; other times the winger must break off their run because a teammate has failed to deliver a timely pass. In both circumstances, the attacking player has created an opportunity even if they never received the ball. In this paper, we construct a probabilistic physics-based model that uses spatio-temporal player tracking data to quantify such off-ball scoring opportunities (OBSO). This model can be used to highlight which, if any, players are likely to score at any point during the match and where on the pitch their scoring is likely to come from. We show how this model can be used in three key ways: 1) to identify and analyze important opportunities during a match 2) to assist opposition analysis by highlighting the regions of the pitch where specific players or teams are more likely to create off-ball scoring opportunities 3) to automate talent identification by finding the players across an entire league that are most proficient at creating off-ball scoring opportunities.
Article
Full-text available
The aim of the present study was to evaluate the effect of match status, venue, and quality of opposition on the styles of play in soccer. Data were collected from 380 games of the English Premier League from the 2015–2016 season. Linear mixed models were applied to evaluate the influence of these contextual variables on membership scores for Direct Play, Counterattack, Maintenance, Build Up, Sustained Threat, Fast Tempo, Crossing, and High Pressure. The results showed that match status had a significant effect on the eight styles of play (all P < 0.001), venue had a significant effect on all styles of play (P < 0.01) except Counterattack and Maintenance, and quality of opposition had a significant effect on all styles of play (P < 0.05) except Counterattack. Moreover, the interaction between match status and quality of opposition, and venue and quality of opposition showed significant effects on some styles of play. The results of this study imply that contextual variables influence the use of styles of play in soccer match play. Consequently, this provides meaningful recommendations for practitioners in soccer.
Article
Full-text available
Sports periodization has traditionally focused on the exercise aspect of athletic preparation, while neglecting the integration of other elements that can impact an athlete's readiness for peak competition performances. Integrated periodization allows the coordinated inclusion of multiple training components best suited for a given training phase into an athlete's program. The aim of this article is to review the available evidence underpinning integrated periodization, focusing on exercise training, recovery, nutrition, psychological skills, and skill acquisition as key factors by which athletic preparation can be periodized. The periodization of heat and altitude adaptation, body composition, and physical therapy is also considered. Despite recent criticism, various methods of exercise training periodization can contribute to performance enhancement in a variety of elite individual and team sports, such as soccer. In the latter, both physical and strategic periodization are useful tools for managing the heavy travel schedule, fatigue, and injuries that occur throughout a competitive season. Recovery interventions should be periodized (ie, withheld or emphasized) to influence acute and chronic training adaptation and performance. Nutrient intake and timing in relation to exercise and as part of the periodization of an athlete's training and competition calendar can also promote physiological adaptations and performance capacity. Psychological skills are a central component of athletic performance, and their periodization should cater to each athlete's individual needs and the needs of the team. Skill acquisition can also be integrated into an athlete's periodized training program to make a significant contribution to competition performance.
Article
Full-text available
The aim of the present study was to assess prospectively the effect of pre-season fitness on injury rate during the competitive season among professional soccer players. Thirty-one players participated in the study during two consecutive competitive seasons (2015–16 and 2016–17; a squad of 22 players in each season). During the 6-week pre-season training period (8 training sessions and a friendly match every week, 14–18 training hours/week) there was a significant improvement in VO2max, a significant increase in ideal and total sprint time and no change in vertical jump, flexibility and repeated sprint test performance decrement. During the two consecutive seasons, 28 injuries were recorded. Ten injuries were classified as mild (missing 3–7 days of practice/match), 8 as moderate (missing 8–28 days) and 10 as severe (missing > 28 days). The rate of match injuries was higher (9.4 per 1000 match hours) compared to practice injuries (4.7 per 1000 training hours). Most injuries were overuse injuries (72%) of the lower limbs (71%). Most of match injuries occurred during the last 15 min of each half. There were no differences in fitness characteristics in the beginning of pre-season training between injured and non-injured players. However, improvements in VO2 max during the pre-season training period were significantly lower among injured players (0.9 ± 5.5 %) compared to non-injured players (10.4 ± 6.5%, p < 0.05). Our results emphasize the importance of pre-season training in professional soccer players not only for improvement in fitness but also for injury prevention during the following competitive season.
Article
Full-text available
Participation in soccer match-play leads to acute and transient subjective, biochemical, metabolic and physical disturbances in players over subsequent hours and days. Inadequate time for rest and regeneration between matches can expose players to the risk of training and competing whilst not entirely recovered. In professional soccer, contemporary competitive schedules can require teams to compete in excess of 60 matches over the course of the season with periods of fixture congestion occurring, prompting much attention from researchers and practitioners to the monitoring of fatigue and readiness to play. A comprehensive body of research has investigated post-match acute and residual fatigue responses. Yet the relevance of the research for professional soccer contexts is debatable, notably in relation to the study populations and designs employed. Monitoring can indeed be invasive, expensive, time inefficient, and difficult to perform routinely and simultaneously in a large squad of regularly competing players. Uncertainty also exists regarding the meaningfulness and interpretation of changes in fatigue response values and their functional relevance, and practical applicability in the field. The real-world need and cost–benefit of monitoring must be carefully weighed up. In relation to professional soccer contexts, this opinion paper intends to (1) debate the need for post-match fatigue monitoring; (2) critique the real-world relevance of the current research literature; (3) discuss the practical burden relating to measurement tools and protocols, and the collection, interpretation and application of data in the field; and (4) propose future research perspectives.
Article
Full-text available
The paper describes Dolores, a model designed to predict football match outcomes in one country by observing football matches in multiple other countries. The model is a mixture of two methods: (a) dynamic ratings and (b) Hybrid Bayesian Networks. It was developed as part of the international special issue competition Machine Learning for Soccer. Unlike past academic literature which tends to focus on a single league or tournament, Dolores is trained with a single dataset that incorporates match outcomes, with missing data (as part of the challenge), from 52 football leagues from all over the world. The challenge involved using a single model to predict 206 future match outcomes from 26 different leagues, played from March 31 to April 9 in 2017. Dolores ranked 2nd in the competition with a predictive error 0.94% higher than the top and 116.78% lower than the bottom participants. The paper extends the assessment of the model in terms of profitability against published market odds. Given that the training dataset incorporates a number of challenges as part of the competition, the results suggest that the model generalised well over multiple leagues, divisions, and seasons. Furthermore, while detailed historical performance for each team helps to maximise predictive accuracy, Dolores provides empirical proof that a model can make a good prediction for a match outcome between teams x and y even when the prediction is derived from historical match data that neither x nor y participated in. While this agrees with past studies in football and other sports, this paper extends the empirical evidence to historical training data that does not just include match results from a single competition but contains results spanning different leagues and divisions from 35 different countries. This implies that we can still predict, for example, the outcome of English Premier League matches, based on training data from Japan, New Zealand, Mexico, South Africa, Russia, and other countries in addition to data from the English Premier league.
Article
Full-text available
Sport science can contribute to the body of knowledge that influences practice and performance. Despite this, knowledge transfer from sport science to football coaches needs further improvement. The present study's purpose is to gain insight in current sport science needs and perceived barriers among professional football coaches. A 29-question digital survey was sent to a database of professional football coaches. Answering options were: check boxes, open fields, and 5-point Likert scales. 75 football coaches (mean ± SD age: 50.0±10.1 years) completed the survey. Coaches had 24.3±7.0 years of experience as a player and 25.5±13.7 years as a coach, and five had a university degree. The coaches evaluated their technical and tactical knowledge as good. Knowledge on physical skills was rated slightly lower and knowledge on mental skills lowest, but still fair. Top five of domains that the coaches wanted to know more about were Mental skills, Physical skills, Group dynamics, Monitoring load and capacity, and Talent development. The top five perceived barriers for using sport science in daily football practice was Conservatism in clubs, Lack of money, Lack of specific football knowledge of scientists, Poor applicability of sports science in practice, and Lack of time. Personal contact with sport scientists was most preferred to gain knowledge in sport science. Results suggest that there is an interest in applying sport scientific knowledge into football practice. An overview of coaches' sport science knowledge gaps and perceived implementation barriers could improve the integration of sport science and practical knowledge in work environments.
Article
Full-text available
This study examines the impact of chance in football using the case of goal scoring. For this purpose, chance in goal scoring is defined as goals involving uncontrollable or not planned aspects. “Chance variables” were defined according to the principle of downward estimation, i.e. when one of these variables is involved this is an uncontrollable or not planned aspect of scoring a goal, although chance may also have an impact in instances other than those collected. Six “chance variables” were identified: deflection, rebound, goalkeeper involvement, post/bar hits, long distance shots, and defence involvement. In the 2011/2012 season, the 875 goals in German Bundesliga and the 1056 goals in English Premier League were examined for the presence of these variables. Of the goals examined, 47.0% showed at least one chance variable (deflection: 4.8%; rebound: 9.4%, goalkeeper: 2.5%; post/bar: 5.2%; distance: 13.1%, and defence: 22.5%). The two leagues differed significantly only in the sense of more goalkeeper interventions in the Bundesliga and more distance goals in the Premier League. The rate of chance-goals drops significantly with the number of prior goals in a match (χ² = 17.59; p = 0.004). In both leagues, there is no significant correlation between final ranking and the percentage of chance-goals scored or conceded. If one accepts the selected chance variables as indicators of uncontrollable or not planned aspects of scoring, the results of this study can be taken as empirical proof for the existence and widespread prevalence of chance involvement in goal scoring in football with many interesting consequences.
Article
Full-text available
It is widely believed that an active cool-down is more effective for promoting post-exercise recovery than a passive cool-down involving no activity. However, research on this topic has never been synthesized and it therefore remains largely unknown whether this belief is correct. This review compares the effects of various types of active cool-downs with passive cool-downs on sports performance, injuries, long-term adaptive responses, and psychophysiological markers of post-exercise recovery. An active cool-down is largely ineffective with respect to enhancing same-day and next-day(s) sports performance, but some beneficial effects on next-day(s) performance have been reported. Active cool-downs do not appear to prevent injuries, and preliminary evidence suggests that performing an active cool-down on a regular basis does not attenuate the long-term adaptive response. Active cool-downs accelerate recovery of lactate in blood, but not necessarily in muscle tissue. Performing active cool-downs may partially prevent immune system depression and promote faster recovery of the cardiovascular and respiratory systems. However, it is unknown whether this reduces the likelihood of post-exercise illnesses, syncope, and cardiovascular complications. Most evidence indicates that active cool-downs do not significantly reduce muscle soreness, or improve the recovery of indirect markers of muscle damage, neuromuscular contractile properties, musculotendinous stiffness, range of motion, systemic hormonal concentrations, or measures of psychological recovery. It can also interfere with muscle glycogen resynthesis. In summary, based on the empirical evidence currently available, active cool-downs are largely ineffective for improving most psychophysiological markers of post-exercise recovery, but may nevertheless offer some benefits compared with a passive cool-down.
Article
Full-text available
The aim of this paper is to review the available literature on goal scoring in elite male football leagues. A systematic search of two electronic databases (SPORTDiscus with Full Text and ISI Web Knowledge All Databases) was conducted and of the 610 studies initially identified, 19 were fully analysed. Studies that fitted all the inclusion criteria were organised according to the research approach adopted (static or dynamic). The majority of these studies were conducted in accordance with the static approach (n=15), where the data were collected without considering dynamic of performance during matches and were analysed using standard statistical methods for data analysis. They focused predominantly on a description of key performance indicators (technical and tactical). Meanwhile, in a few studies the dynamic approach (n=4) was adopted, where performance variables were recorded taking into account the chronological and sequential order in which they occurred. Different advanced analysis techniques for assessing performance evolution over time during the match were used in this second group of studies. The strengths and limitations of both approaches in terms of providing the meaningful information for coaches are discussed in the present study.
Article
The session rating of perceived exertion (sRPE) method was developed 25 years ago as a modification of the Borg concept of rating of perceived exertion (RPE), designed to estimate the intensity of an entire training session. It appears to be well accepted as a marker of the internal training load. Early studies demonstrated that sRPE correlated well with objective measures of internal training load, such as the percentage of heart rate reserve and blood lactate concentration. It has been shown to be useful in a wide variety of exercise activities ranging from aerobic to resistance to games. It has also been shown to be useful in populations ranging from patients to elite athletes. The sRPE is a reasonable measure of the average RPE acquired across an exercise session. Originally designed to be acquired ∼30 minutes after a training bout to prevent the terminal elements of an exercise session from unduly influencing the rating, sRPE has been shown to be temporally robust across periods ranging from 1 minute to 14 days following an exercise session. Within the training impulse concept, sRPE, or other indices derived from sRPE, has been shown to be able to account for both positive and negative training outcomes and has contributed to our understanding of how training is periodized to optimize training outcomes and to understand maladaptations such as overtraining syndrome. The sRPE as a method of monitoring training has the advantage of extreme simplicity. While it is not ideal for the precise recording of the details of the external training load, it has large advantages relative to evaluating the internal training load.
Article
The ongoing advances and prevalence of technology in football is spurring a more prominent role for technology and data in driving decisions and practice. Sports science practitioners naturally play an integral part in the adoption of these technologies and are among the most enthusiastic proponents of a data-driven approach that promotes objective metrics and seeks to minimise or eliminate subjectivity. With the speed of these developments, there are a number of pitfalls that have been overlooked. In this commentary we describe the pervasive forces driving the adoption of technology solutions and critically examine the logic underpinning the present drive for metrics-based practice in football. In addition to highlighting some important gaps in what we are presently able to capture, we uncover some common flaws in how we implement new technology and use data. We propose that there is a fundamental need to reframe how we are seeking to employ data and more specifically make the necessary switch from being data-driven to data-informed. We propose some solutions to assist practitioners in being purposeful in their use of these tools and leverage the benefits of technology and data in a way that better supports decision-making and complements coaching practice.
Article
Aim To quantify the occurrence of near-to-maximal speed-running bouts in elite soccer players. Methods Tracking match data from 35 professional players (23 ± 3 yrs) were analysed over four seasons (2015–2019). The number of runs reaching >80, >85 and >90% of players’ individual maximal sprinting speed (MSS) was computed. Differences in peak speed occurrences between halves and positions were analysed. Results Occurrences were not different between halves, but slightly-to-moderately position-dependent, with wingers (3.5 ± 2.0 runs >80% per half) and central midfielders (1.6 ± 1.5 runs >80% per half) performing the most and the least number of near-to-maximal speed-running bouts. Players did not reach >90% of MSS at all in 35% (Attackers) to 65% (Midfielders) of their matches; they reached ≥3 times >90% MSS per match in 2% (Midfielders) to 11% (Attackers) of their matches only. The maximal number of cumulated match occurrences during congested periods was observed in acentral defender (2 runs >90% of MSS every 4 days for 31 days) and a winger (1 run every 4 days for 52 days). Conclusions The occurrence of near-to-maximal speed-running bouts is low in elite soccer. Match context, playing position and individual player profile are the main determinants of those occurrences.
Article
The main aim of this work was to review the use of technological tracking methods to assess collective spatial-positioning variables in team sports. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and PICO design for systematic reviews, study identification was performed in four databases (PubMed, SPORTDiscus, ProQuest Central, and Web of Science). Articles were selected if they focused on player position and technological tracking methods. After duplicate removal, 2194 articles were identified based on the established search criteria, of which 72 articles were selected and analysed. Semi-automatic optic-based systems, Global Positioning System/Global Navigation Satellite Systems, and local positioning systems were used in 60%, 33% and 7% of the studies, respectively. All studies that measured tactical variables by local positioning system technology in team sports used local position measurement technology. Optic-based systems were used more often in the early years to analyse collective tactical behaviour during competition. Later, Global Positioning System/Global Navigation Satellite Systems became more frequent to measure behaviour in team sports during the training process. The possibility of using the same system during competition and training will facilitate the assessment of collective tactical behaviour in team sports.
Article
The number of studies examining associations between training load and injury has increased exponentially. As a result, many new measures of exposure and training-load-based prognostic factors have been created. The acute:chronic workload ratio (ACWR) is the most popular. However, when recommending the manipulation of a prognostic factor in order to alter the likelihood of an event, one assumes a causal effect. This introduces a series of additional conceptual and methodological considerations that are problematic and should be considered. Because no studies have even tried to estimate causal effects properly, manipulating ACWR in practical settings in order to change injury rates remains a conjecture and an overinterpretation of the available data. Furthermore, there are known issues with the use of ratio data and unrecognized assumptions that negatively affect the ACWR metric for use as a causal prognostic factor. ACWR use in practical settings can lead to inappropriate recommendations, because its causal relation to injury has not been established, it is an inaccurate metric (failing to normalize the numerator by the denominator even when uncoupled), it has a lack of background rationale to support its causal role, it is an ambiguous metric, and it is not consistently and unidirectionally related to injury risk. Conclusion : There is no evidence supporting the use of ACWR in training-load-management systems or for training recommendations aimed at reducing injury risk. The statistical properties of the ratio make the ACWR an inaccurate metric and complicate its interpretation for practical applications. In addition, it adds noise and creates statistical artifacts.
Article
Background Preseason training develops players’ physical capacities and prepares them for the demands of the competitive season. In rugby, Australian football, and American football, preseason training may protect elite players against in-season injury. However, no study has evaluated this relationship at the team level in elite soccer. Purpose/Hypothesis The aim of this study was to investigate whether the number of preseason training sessions completed by elite soccer teams was associated with team injury rates and player availability during the competitive season. It was hypothesized that elite soccer teams who participate in more preseason training will sustain fewer injuries during the competitive season. Study Design Descriptive epidemiology study. Methods We used the Union of European Football Associations (UEFA) injury dataset to analyze 44 teams for up to 15 seasons (total, 244 team-seasons). Separate linear regression models examined the association between the number of team preseason training sessions and 5 in-season injury measures. Injury-related problems per team were quantified by totals of the following: (1) injury burden, (2) severe injury incidence, (3) training attendance, (4) match availability, and (5) injury incidence. Results Teams averaged 30 preseason training sessions (range, 10-51). A greater number of preseason training sessions was associated with less injury load during the competitive season in 4 out of 5 injury-related measures. Our linear regression models revealed that for every 10 additional preseason training sessions that the team performed, the in-season injury burden was 22 layoff days lower per 1000 hours ( P = .002), the severe injury incidence was 0.18 severe injuries lower per 1000 hours ( P = .015), the training attendance was 1.4 percentage points greater ( P = .014), and the match availability was 1.0 percentage points greater ( P = .042). As model fits were relatively low (adjusted R ² = 1.3%-3.2%), several factors that contribute to in-season injury outcomes were unaccounted for. Conclusion Teams that performed a greater number of preseason training sessions had “healthier” in-season periods. Many other factors also contribute to in-season injury rates. Understanding the benefit of preseason training on in-season injury patterns may inform sport teams’ planning and preparation.
Conference Paper
The task of performance analysts and coaches in football (and other team sports) is manifold: they need to assess the performance of individual players of their team, they need to monitor the interaction between players of their team and their tactical compliance, and they need to analyze other teams. For this, they usually have to consider various sources of information: video footage, tracking data, event data, and aggregated statistics. On the basis of this information, analysts have to generate quantitative summaries of events including their spatial and temporal distribution, and the qualitative assessment of individual events by considering the associated video footage. In this paper, we present SportSense, a system for sports video retrieval, that seamlessly combines quantitative and qualitative analysis. For this, SportSense provides dedicated filters that help analysts in selecting the events they are interested in. Moreover, it supports the comparative analysis of stored queries with respect to specific parameters. Essentially, SportSense allows to easily switch between qualitative and quantitative analyses to support coaches and analysts in a best possible way in their task. Based on a user study, we show the effectiveness of the proposed approach.
Article
Across countries and continents, football (soccer) has drawn increasingly more attention over the last decades and developed into a huge commercial complex. Consequently, the market of bookmakers providing the possibility to bet on the result of football matches grew rapidly, especially with the appearance of the internet. With a high number of games every week in multiple countries, football league matches hold enormous potential for generating profits over time with the use of advanced betting strategies. In this paper, we use machine learning for predicting the outcome of football league matches by exploiting data about match characteristics. Based on insights from the field of statistical arbitrage stock market trading, we show that one could generate meaningful profits over time by betting accordingly. A simulation study analyzing the matches of the five top European football leagues from season 2013/14 to 2017/18 presented economically and statistically significant returns achieved by exploiting large data sets with modern machine learning algorithms. In contrast to these modern algorithms, the break-even point could not be reached with an ordinary linear regression approach or simple betting strategies, e.g. always betting on the home team.
Article
The purpose of this study was to quantify the physical loads of programmed pre-season training in four different professional Dutch and Portuguese soccer teams. Eighty-nine professional players were monitored daily during a five-week period. We monitored the physical loading of training by measuring the external load measures of total distance covered, walking distance, jogging distance, running distance, sprinting distance, high-intensity sprint distance, player's load and number of sprints using a 10 Hz GPS technology. Weekly external load and intra-week external load variations were tested. Repeated measures did not show significant differences between weeks in terms of weekly loads based on total distance and sprinting distance. Significant differences were found between training days considering the duration (p = .011), walking distance (p = .017), running distance (p = .004), player's load (p = .040) and number of sprints (p = .006). Variations between weeks were small, however intra-week variations were observed namely considering the measures associated with great volume and lower intensity.
Article
Objective To compare injury rates among professional men’s football teams that have a winter break in their league season schedule with corresponding rates in teams that do not. Methods 56 football teams from 15 European countries were prospectively followed for seven seasons (2010/2011–2016/2017)—a total of 155 team-seasons. Individual training, match exposure and time-loss injuries were registered. Four different injury rates were analysed over four periods within the season, and linear regression was performed on team-level data to analyse the effect of winter break on each of the injury rates. Crude analyses and analyses adjusted for climatic region were performed. Results 9660 injuries were reported during 1 447 011 exposure hours. English teams had no winter break scheduled in the season calendar: the other European teams had a mean winter break scheduled for 10.0 days. Teams without a winter break lost on average 303 days more per season due to injuries than teams with a winter break during the whole season (p<0.001). The results were similar across the three periods August–December (p=0.013), January–March (p<0.001) and April–May (p=0.050). Teams without a winter break also had a higher incidence of severe injuries than teams with a winter break during the whole season (2.1 severe injuries more per season for teams without a winter break, p=0.002), as well as during the period January–March (p=0.003). A winter break was not associated with higher team training attendance or team match availability. Climatic region was also associated with injury rates. Conclusions The absence of a scheduled winter break was associated with a higher injury burden, both before and during the two periods following the time that many European teams take a winter break. Teams without a winter break (English clubs) had a higher incidence of severe injuries following the time of the year that other teams (other European clubs) had their scheduled break.
Article
To account for the individual intensity of locomotion tasks, individualised speed thresholds have been proposed as an alternative to global speed thresholds. Methodologies to determine individual speed thresholds have typically been laboratory based, time consuming and expensive, rendering them inappropriate for applied practitioners working with large squads. The current investigation utilised easy to administer field tests to individualise speed thresholds. The aim was to investigate differences between high-speed locomotion measured using global and individual speed thresholds. Nineteen, male, professional soccer players completed maximum sprint and maximum aerobic speed protocols and were divided into groups dependent upon maximum aerobic speed performance (high, medium and low). Locomotion data were collected using portable Global Positioning System units and analysed using global and individual analysis methods to determine distances travelled performing high-speed running, very high-speed running and sprinting. In low athletes, the individual analysis method produced significantly higher percentages of high-speed running, very high-speed running and sprinting compared to global (mean differences 7.8%, 6.1% and 1.7%, respectively, all p < 0.001). In medium athletes, no significant differences were found between analysis methods for high-speed running and very high-speed running. In high athletes, the individual analysis method produced significantly lower high-speed running and very high-speed running percentages compared to global (mean differences 11.0% and 6.8%, p < 0.001). Results concluded that global thresholds produced high-speed locomotion percentages significantly higher or lower than individual thresholds for 47% of athletes. The current investigation recommends the use of field tests to individualise speed thresholds, allowing applied practitioners to accurately quantify individual athlete intensity.
Article
IN SOCCER, GLOBAL POSITIONING SYSTEM (GPS) MONITORING OF PLAYER WORKLOADS IS NOW EXTENSIVELY USED ACROSS ALL LEVELS OF THE SPORT. TO MAKE BETTER USE OF THIS TECHNOLOGY IT IS IMPORTANT TO APPRECIATE HOW IT WORKS. FURTHER, WHEN THE LIMITATIONS OF GPS USE ARE APPRECIATED AND THE RATIONALE OF USE IS AGREED AND ARTICULATED, THEN THE POTENTIAL OF GPS MONITORING CAN BE EFFECTIVELY REALIZED TO BETTER MANAGE PLAYERS' PERFORMANCE, WORKLOAD AND WELFARE. (SEE VIDEO, SUPPLEMENTARY DIGITAL CONTENT, NUMBER 1, WHICH SUMMARIZES GPS USE, LIMITATIONS, AND POTENTIAL IN SOCCER, HTTP:// LINKS.LWW.COM/SCJ/A238).
Article
This study aimed to identify the game-related statistics that discriminated between Euroleague basketball players and European basketball players playing in the NBA, when competing in the same event (EuroBasket 2015). There was a total of 78 matches played by 24 teams in two groups of analysis: NBA, participants in the European Championship who played in the NBA season of 2014-2015 (n = 26); Euroleague, participants in the European Championship who played in the Euroleague season of 2014-2015 (n = 82). The players’ performance variables were normalized to the time they spent on the court. To identify which variables best discriminated between the NBA and the Euroleague performance profiles, a descriptive discriminant analysis was conducted. Structure coefficients (SC) from the matrix greater than |0.30| were interpreted as meaningful contributors to discriminating between the groups. The results revealed a significant function (p = 0.008, canonical correlation of 0.51, Λ = 0.74, reclassification = 84.2%) and substantial performance differences in game-related statistics much related to the influence of body size (body height and mass), such as two-point field goals made (SC = 0.42) and missed (SC = 0.40), free-throws made (SC = 0.55), defensive rebounds (SC = 0.62), blocks (SC = 0.48) and suffered fouls (SC = 0.34). No differences were found at the level of game-related statistics indirectly related to perception, such as assists, turnovers or steals. Also, the greater body size in NBA players was likely related to higher variability in performance, thus, being an important topic for coaches and recruiters to analyse. Key words: performance profile, analysis, game-related statistics, discriminant scores
Article
The aim of this study was to identify the key physical and technical performance variables related to team quality in the Chinese Super League (CSL). Teams’ performance variables were collected from 240 matches and analysed via analysis of variance between end-of-season-ranked groups and multinomial logistic regression. Significant physical performance differences between groups were identified for sprinting (top-ranked group vs. upper-middle-ranked group) and total distance covered without possession (upper and upper-middle-ranked groups and lower-ranked group). For technical performance, teams in the top-ranked group exhibited a significantly greater amount of possession in opponent’s half, number of entry passes in the final 1/3 of the field and the Penalty Area, and 50–50 challenges than lower-ranked teams. Finally, time of possession increased the probability of a win compared with a draw. The current study identified key performance indicators that differentiated end-season team quality within the CSL.