Latest research in football - week 34 - 2021

#1 Understanding gender differences in professional European football through machine learning interpretability and match actions data

Reference: Sci Rep. 2021 May 24;11(1):10805.  doi: 10.1038/s41598-021-90264-w.

Authors: Marc Garnica-Caparrós, Daniel Memmert

Download link: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8144211/pdf/41598_2021_Article_90264.pdf

Summary: After the great success of the Women's World Cup in 2019, several platforms have started identifying the reasons for gender inequality in European football. Even though these inequalities emerge from a variety of key aspects in the modern sport, we focused on the game and evaluated the main differential features of European male and female football players in match actions data under the assumption of finding significant differences and established patterns between genders. A methodology for unbiased feature extraction and objective analysis is presented based on data integration and machine learning explainability algorithms. Female ([Formula: see text]) and male ([Formula: see text]) data points were collected from event data and categorized by game period and player position. We set up a supervised classification pipeline to predict the gender of each player by looking at their actions in the game. The comparison methodology did not include any qualitative enrichment or subjective analysis to prevent biased data enhancement or gender-related processing. The pipeline included three representative binary classification models; A logic-based Decision Trees, a probabilistic Logistic Regression and a multilevel perceptron Neural Network. Each model tried to draw the differences between male and female data points, and we extracted the results using machine learning explainability methods to understand the underlying mechanics of the models implemented. The study was able to determine pivotal factors that differentiate each gender performance as well as disseminate unique patterns by gender involving more than one indicator. Data enhancement and critical variables analysis are essential next steps to support this framework and serve as a baseline for further studies and training developments.

 

 

 

#2 The effect of approach velocity on pelvis and kick leg angular momentum conversion strategies during football instep kicking

Reference: J Sports Sci. 2021 May 25;1-10.  doi: 10.1080/02640414.2021.1929008. Online ahead of print.

Authors: Simon Augustus, Penny E Hudson, Neal Smith

Summary: During football instep kicking, whole-body deceleration during the final stride has been associated with greater kick leg angular momentum and enhanced foot and ball velocities, but the influence of approach velocity on these mechanisms is unknown. This study assessed how approach velocity affects momentum conversion strategies of experienced players performing fast and accurate kicks. Eleven semi-professional footballers performed instep kicks from self-selected (3.34 ± 0.43 m/s), fast (3.71 ± 0.33 m/s) and slow (2.77 ± 0.32 m/s) approaches. Kicking motions and ground reaction forces under the support leg were captured using 3D motion analysis (1000 Hz). The players responded to perturbations in approach velocity by using the support leg to regulate whole-body deceleration and create ideal conditions for co-ordinated pelvic and kick leg momentums during the downswing. Further, the pelvis was key for generating transverse momentum at the kick leg, but the participants displayed distinctly different pelvis transverse rotation strategies. Identification of these inter-individual strategies may provide a basis for technical and strength training practices to be tailored for individual players. Future research might investigate if training practices that expose footballers to varying approach velocities of between 2.5 and 4.0 m/s promotes development of movement strategies that are robust to perturbations in approach conditions.

 

 

#3 Competitive anxiety or Coronavirus anxiety? The psychophysiological responses of professional football players after returning to competition during the COVID-19 pandemic

Reference: Psychoneuroendocrinology . 2021 May 14;129:105269.  doi: 10.1016/j.psyneuen.2021.105269. Online ahead of print. 

Authors: Amir Hossien Mehrsafar, Ali Moghadam Zadeh, José Carlos Jaenes Sánchez, Parisa Gazerani

Download link: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8118697/pdf/main.pdf

Summary: This study aimed to examine the relationship between competitive anxiety, fear/anxiety of COVID-19, and autonomic and endocrine stress responses in professional football players after returning to competition during the COVID-19 pandemic. Ninety male professional football players (age: 26.33 ± 2.48 yr) volunteered to participate in this study, which included an official competition. Psychophysiological responses, including the Fear of COVID-19 Scale, the Coronavirus Anxiety Scale, and the Competitive State Anxiety Inventory-2 Revised, were collected 30 min before the competition. In addition, salivary alpha-amylase (sAA) and salivary cortisol (sCort) were collected at 8 a.m. and 15 min before the competition. The main findings, based on the Pearson correlation, showed significant positive correlations between COVID-19 anxiety and somatic competitive anxiety (p = 0.01), cognitive competitive anxiety (p = 0.01), and competition response of sCort and sAA (p = 0.01). Moreover, fear of COVID-19 was positively correlated with COVID-19 anxiety (p = 0.01). On the contrary, the awakening response of sCort and sAA was not found to be correlated with psychological parameters (all p > 0.05). The analysis also indicated that there was no significant correlation between self-confidence with other psychological and physiological variables (all p > 0.05). The regression analysis showed that cognitive anxiety was a relevant predictor for the competition response of sCort and sAA (p < 0.05). Moreover, COVID-19 anxiety was the only predictor of somatic and cognitive anxiety (p < 0.05). The present study provides the first preliminary evidence that COVID-19 anxiety and competitive anxiety might pose a negative impact on the athletic performance of professional football players during COVID-19 pandemic competitions. Thus, research is needed to build a strategy to reduce the psychophysiological stress related to COVID-19 and competition response.

 

 

#4 Association of the British Athletic Muscle Injury Classification and anatomic location with return to full training and reinjury following hamstring injury in elite football

Reference: BMJ Open Sport Exerc Med. 2021 May 10;7(2):e001010. doi: 10.1136/bmjsem-2020-001010. eCollection 2021.

Authors: Ricky Shamji, Steven L J James, Rajesh Botchu, Kent A Khurniawan, Gurjit Bhogal, Alison Rushton

Download link: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8112435/pdf/bmjsem-2020-001010.pdf

Summary: The relationship between hamstring muscle injuries (HMIs) that involve the intramuscular tendon and prolonged recovery time and increased reinjury rate remains unclear in elite footballers. To determine the association of time to return to full training (TRFT) and reinjury of HMIs using the British Athletic Muscle Injury Classification (BAMIC) and specific anatomical injury location in elite-level football players. The electronic medical records of all players at an English Premier League club were reviewed over eight consecutive seasons. All players who sustained an acute HMI were included. Two experienced musculoskeletal radiologists independently graded each muscle using the BAMIC, categorised each injury location area (proximal vs middle vs distal third and proximal vs distal tendon) and reported second muscle involvement. TRFT and reinjury were recorded. Out of 61 HMIs, the intramuscular tendon (BAMIC 'c') was involved in 13 (21.3%). HMI involving the intramuscular tendon ('c') had a mean rank TRFT of 36 days compared with 24 days without involvement (p=0.013). There were 10 (16.4%) reinjuries with a significant difference of 38.5% reinjury rate in the group with intramuscular tendon injury ('c') and 12.5% in the group without (p=0.031). TRFT and reinjury involving a second muscle was statistically significantly higher than without. Most of the HMIs to the biceps femoris with reinjury (5 out of 9) were in the distal third section related to the distal tendon site involving both the long and short head. TRFT in HMI involving the intramuscular tendon ('c') of the Biceps femoris is significantly longer with significantly higher reinjury rate compared with injuries without, in elite football players. The finding that most reinjures of the biceps femoris occurring in the distal third muscle at the distal tendon site, involving both the long and short head, merits further investigation.

 

 

#5 The Effect of a Tailored Intervention on Female Soccer Players' Hydration Status

Reference: J Hum Kinet. 2021 Mar 31;78:131-140.  doi: 10.2478/hukin-2021-0047. eCollection 2021 Mar.

Authors: Kristien Van Campenhout, Laurent Chapelle, Tom Deliens, Eva D'Hondt, Peter Clarys

Download link: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8120958/pdf/hukin-78-131.pdf

Summary: Numerous studies have reported that athletes' pre-training/-game hydration status is of concern. Up to now, only two cohort studies have examined the effect of an intervention aimed at improving pretraining/-game hydration status; however, without including a control group. Therefore, the aim of this quasi-experimental study was to examine whether and to what extent an individually tailored intervention focused on the benefits of being optimally hydrated before training or game would alter the hydration status of female soccer players. Two teams of young adult female soccer players were allocated to an intervention (n = 22; 19.8 ± 3.0 years) or a control group (n = 15; 22.8 ± 4.0 years). Players in the intervention group received an individually tailored intervention, based on Urine Specific Gravity (USG) measurements, which took place in between two training weeks. Before each training and match play, a urine sample was collected to determine players' hydration status. Mixed modelling was applied to assess within and between differences in hydration status over time. The pre-training/-game hydration status of the intervention group improved significantly, with players' mean USG-value decreasing from 1.013 ± 0.001 g/mL to 1.008 ± 0.002 g/mL (p = 0.005). In contrast, the mean USG-value in the control group increased significantly from 1.012 ± 0.002 g/mL to 1.016 ± 0.002 g/mL (p = 0.023), indicating a deterioration of their pre-training/-game hydration status. The individually tailored intervention did show a positive effect even though female soccer players in the present study were already reasonably well hydrated before the start of the intervention.

 

 

#6 Does Youth Soccer Players' Group Cooperation Improve when Pitch, Goal, and Team Size are Reduced?

Reference: J Hum Kinet. 2021 Mar 31;78:151-160.  doi: 10.2478/hukin-2021-0035. eCollection 2021 Mar.

Authors: Antonio García-Angulo, Enrique Ortega-Toro, José María Giménez-Egido, Aurelio Olmedilla, Alexandre García-Mas 4, José Manuel Palao 5

Download link: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8120978/pdf/hukin-78-151.pdf

Summary: The aim of this study was to analyse the effects of the modification of the pitch size, goal size, and the number of players per team on the level of cooperation in youth soccer players (U-12). A quasi-experimental design was implemented to measure the effects of the changes in the number of players per team (8-a-side vs. 5-a-side), goal size (6 x 2 m vs. 3 x 2 m), and field size (58 x 38 m vs. 38 x 20 m) on the cooperation. Four soccer teams under-12 (n = 48 players) participated in three tournaments (standard rules; proposed rules; standard rules). The five variables of cooperation were measured using the "Sports Cooperation Questionnaire": conditional cooperation, cooperation with the coach, disposition of unconditional cooperation, situational cooperation with teammates, and situational cooperation outside the field of play. Players presented the same level of cooperation in conditional cooperation, situational cooperation with teammates, and situational cooperation outside the field of play with both types of rules. The modification of rules involved a reduction of the cooperation with the coach and a disposition of unconditional cooperation. The causes of the reduction may be related to players' adaptation to new tactical dispositions and game situations. Future studies are needed of player's cooperation and different competition rules in youth sports.

 

 

#7 Influence of Pitch Size on Short-Term High Intensity Actions and Body Impacts in Soccer Sided Games

Reference: J Hum Kinet. 2021 Mar 31;78:187-196.  doi: 10.2478/hukin-2021-0037. eCollection 2021 Mar.

Authors: Daniel Castillo, Javier Raya-González, Javier Yanci, Filipe Manuel Clemente

Download link: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8120965/pdf/hukin-78-187.pdf

Summary: The aim of this study was to compare external training loads between small-sided games (SSGs) and largesided games (LSGs) in soccer players. Twenty outfield soccer players (14.8 ± 0.6 years old) who competed in the Spanish U16 Provincial Division and belonged to the same team participated in the study. The soccer sided games were played at different individual interaction space (IIS) per player (i.e., SSG = 100 m2 and LSG = 200 m2) and were disputed in the same format (five-a-side plus goalkeepers) on two different pitch sizes (i.e., 38 x 26 vs. 53 x 37 m) defending an official soccer-goal. The sided games' duration was 4 bouts of 6 min with 2 min rest intervals between bouts. The results of this study showed no meaningful differences in the total distance and intensity of accelerations and decelerations between SSGs and LSGs except for the lower distance covered at medium intensity (2.5 - 4 m·s-2) observed during LSGs (-10.2%; ES (effect size): -0.51). Players registered greater sprints, maximum velocity (Velmax) and body impacts at different intensities (i.e., I5-6g, I6-6.5g, I6.5-7g, I7-8g, I8-10g,) in LSGs in comparison to SSGs. These findings suggest that an increase in the pitch size (i.e., IIS per player) can induce higher external loads for soccer players.

 

 

#8 The influence of randomness on goals in football decreases over time. An empirical analysis of randomness involved in goal scoring in the English Premier League

Reference: J Sports Sci. 2021 May 24;1-16.  doi: 10.1080/02640414.2021.1930685. Online ahead of print.

Authors: Fabian Wunderlich, Alessandro Seck, Daniel Memmert

Summary: Performance analysis in football predominantly focuses on systematic contributions to success, thus neglecting the role of randomness. The present paper pursues a direct approach to quantify and analyse randomness in football by identifying random influences in the goal scoring process. The dataset includes all matches from the seasons 12/13 to 18/19 of the English Premier League, adding up to a total of 7,263 goals, that were checked for the occurrence of six variables of random influence. Additionally, the influence of nine situational variables was investigated. Results show that randomness was present for almost 50% of all goals. Moreover, it was demonstrated that the proportion of random goals decreased over the seven seasons (p < .001), is more pronounced for weaker teams (p < .05) as well as if the current scoreline is a draw (p < .05) and depends on the match situation (open play, freekick, corner, penalty). An improved understanding of randomness in football has important implications for both researchers and practitioners. Performance analysts should acknowledge randomness as a crucial factor to distinguish clearly between performance and success. Coaches could even consider the conscious creation of uncontrollable situations as a possible tactic to provoke random influences on goal scoring.

 

 

#9 Individual acceleration-speed profile in-situ: A proof of concept in professional football players

Reference: J Biomech. 2021 May 15;123:110524. doi: 10.1016/j.jbiomech.2021.110524. Online ahead of print.

Authors: Jean-Benoit Morin, Yann Le Mat, Cristian Osgnach, Andrea Barnabò, Alessandro Pilati, Pierre Samozino, Pietro E di Prampero

Summary: Assessing football players' sprint mechanical outputs is key to the performance management process (e.g. talent identification, training, monitoring, return-to-sport). This is possible using linear sprint testing to derive force-velocity-power outputs (in laboratory or field settings), but testing requires specific efforts and the movement assessed is not specific to the football playing tasks. This proof-of-concept short communication presents a method to derive the players' individual acceleration-speed (AS) profile in-situ, i.e. from global positioning system data collected over several football sessions (without running specific tests). Briefly, raw speed data collected in 16 professional male football players over several training sessions were plotted, and for each 0.2 m/s increment in speed from 3 m/s up to the individual top-speed reached, maximal acceleration output was retained to generate a linear AS profile. Results showed highly linear AS profiles for all players (all R2 > 0.984) which allowed to extrapolate the theoretical maximal speed and accelerations as the individual's sprint maximal capacities. Good reliability was observed between AS profiles determined 2 weeks apart for the players tested, and further research should focus on deepening our understanding of these methodological features. Despite the need for further explorations (e.g. comparison with conceptually close force-velocity assessments that require, isolated and not football-specific linear sprint tests), this in-situ approach is promising and allows direct assessment of football players within their specific acceleration-speed tasks. This opens several perspectives in the performance and injury prevention fields, in football and likely other sprint-based team sports, and the possibility to "test players without testing them".

 

 

#10 Collective states and their transitions in football

Reference: PLoS One. 2021 May 24;16(5):e0251970.  doi: 10.1371/journal.pone.0251970. eCollection 2021.

Authors: Mitchell Welch, Timothy M Schaerf, Aron Murphy

Download link: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0251970

Summary: Movement, positioning and coordination of player formations is a key aspect for the performance of teams within field-based sports. The increased availability of player tracking data has given rise to numerous studies that focus on the relationship between simple descriptive statistics surrounding team formation and performance. While these existing approaches have provided a high-level a view of team-based spatial formations, there is limited research on the nature of collective movement across players within teams and the establishment of stable collective states within game play. This study draws inspiration from the analysis of collective movement in nature, such as that observed within schools of fish and flocking birds, to explore the existence of collective states within the phases of play in soccer. Order parameters and metrics describing group motion and shape are derived from player movement tracks to uncover the nature of the team's collective states and transitions. This represents a unique addition to the current body of work around the analysis of player movement in team sports. The results from this study demonstrate that sequences of ordered collective behaviours exist with relatively rapid transitions between highly aligned polar and un-ordered swarm behaviours (and vice-versa). Defensive phases of play have a higher proportion of ordered team movement than attacking phases, indicating that movements linked with attacking tactics, such as player dispersion to generate passing and shooting opportunities leads to lower overall collective order. Exploration within this study suggests that defensive tactics, such as reducing the depth or width to close passing opportunities, allows for higher team movement speeds and increased levels of collective order. This study provides a novel view of player movement by visualising the collective states present across the phases of play in football.

 

 

#11 Applying the Input-Process-Outcome Model to Team Learning in Sport Sciences: An Exploratory Qualitative Study on Twenty Soccer Players

Reference: J Hum Kinet. 2021 Mar 31;78:251-262. doi: 10.2478/hukin-2021-0040. eCollection 2021 Mar.

Authors: Thibault Kérivel, Cyril Bossard, Gilles Kermarrec

Download link: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8120967/pdf/hukin-78-251.pdf

Summary: In sport science literature, referring to the Input-Process-Outcome (IPO) model, few studies demonstrated links between team training and team learning despites several calls for empirical studies. Thus, this study aimed at exploring systemically the building process of the interpersonal coordination by focusing on (1) a specific antecedent (i.e., video feedback during practice), (2) the influence of this antecedent on the team learning process, and (3) outcomes from this process as shared cognitive contents. Thus, this study was original by examining empirically the IPO model in a sport-training context. Our study showed that Input influenced the specific learning Process during practice (five processes) and video feedback sessions (five processes) and produced Outcomes (six typical shared cognitive contents). Finally, results are discussed in relation to team learning processes theoretically identified in the literature and an IPO soccer model adapted to team learning in a soccer context is proposed.

 

 

#12 Structure of the training program during the Covid-19 confinement in Spanish professional football: A coach survey

Reference: Phys Sportsmed. 2021 May 24.  doi: 10.1080/00913847.2021.1932633. Online ahead of print.

Authors: Iván Asín-Izquierdo, Luis Gutiérrez-García, Juan Carlos Zapardiel, Marcos Chena

Summary: COVID-19 has produced an exceptional situation for sport due to confinement and restrictions. The usual training programs and competition have been interrupted in world football, requiring an adaptation of training to the new situation. The aim was to describe and analyze the training programs carried out during the COVID-19 forced confinement in men´s professional football in Spain. The sample consisted of thirty-six coaches representing thirty-six professional men's soccer teams in the Spanish first and second division. Training programs developed during confinement prioritized conditioning and functional emphasis, in addition to general and non-specific resistance training, due to contextual limitations. The structure of training during the COVID-19 confinement was limited by contextual circumstances. This study has made possible to record the training and strategies used in professional football during a confinement due to a worldwide state of alarm, with the aim of resuming competitive activity in the best possible conditions.