...

Scientific Programme

Applied Sports Sciences

CP-AP13 - Statistics and Analyses

Date: 03.07.2024, Time: 16:30 - 17:30, Lecture room: Lomond Auditorium

Description

Chair TBA

Chair

TBA
TBA
TBA

ECSS Paris 2023: CP-AP13

Speaker A Jan Venzke

Speaker A

Jan Venzke
Ruhr-University Bochum, Department of Sports Medicine and Sports Nutrition
Germany
"Cracking the Code: Predicting Player Positions in Handball with Machine Learning"

INTRODUCTION: In team sports, an increasing amount of data is being collected to assess tactical behavior and monitor physical performance. Strong physical performance is a cornerstone of success in professional team sports. An individual or position-specific performance profile is highly desired by coaches for player monitoring. However, many analyses focus on selected parameters to describe position specific exercise load. A comprehensive understanding of player performance requires the consideration of multiple parameters. However, coaches are faced with an overwhelming number of parameters and require a more detailed view of the most meaningful parameters. Our aim is to identify the most influential predictor of locomotion data across different positions in handball. METHODS: A supervised machine learning model was used to classify positions with 10-fold cross validation. Our analysis included several conventional parameters (distance, mean velocity, speed/acceleration zones) and parameters derived from the metabolic power approach (energy cost, metabolic power). Local positioning system data (Kinexon Precision Technologies) from all 65 EHF EURO 2020 matches were used. 414 elite male handball players were included, resulting in 1596 datasets. We analyzed net playing time with durations longer than 1 min. Goalkeepers were excluded. RESULTS: 1437 datasets were used for training and validation and 159 for testing. The test accuracy of our model was determined to be 78.6%. Wings were the most accurate position to predict, with 80 out of 81 players correctly predicted. Center backs were often confused with outer backs. Pivots were also often predicted as outer backs. Average speed had the greatest relative influence on predicting position (47.4%) and weight the second greatest (31.2%), followed by average metabolic power (5.2%) and height (4.9%). CONCLUSION: Our study highlights the importance of using multiple parameters to accurately classify player positions in elite handball. Our model demonstrated robust performance, as evidenced by high accuracy on the test dataset. Wings consistently stood out as the most accurately predicted positions, reflecting their typically lower body weight. Average speed being the best predictor is consistent with recent research showing positional differences in average speed (Manchado et al., 2022). For example, wing players cover the most distance and spend the most time in the high-speed categories. These findings highlight the potential of machine learning techniques to improve our understanding of player performance dynamics in team sports such as handball.

Read CV Jan Venzke

ECSS Paris 2023: CP-AP13

Speaker B Hua Luo

Speaker B

Hua Luo
Universiti Putra Malaysia, Department of Sport Studies
China
"Emerging Trends and Knowledge Structures on Soccer Research: A Scientometric Analysis"

INTRODUCTION: The sport of soccer has received increasing attention and sparked multidisciplinary exploration. There is a lack of comprehensive scientometric analyses of this change over time. Our objective is to understand emerging trends and knowledge structures in the field of study through scientometric analyses. METHODS: The scientometric scrutiny conducted in this investigation was underpinned by the Web of Science Core Collection as the designated database. The search strategy encompassed the employment of the following query: TS= ("football" OR "soccer"). The ambit of citation indexing was confined to SCI-E, SSCI, and A&HCI indices. Within the spectrum of document classifications, exclusivity was accorded to entries categorized as Article or Review Article. The temporal scope spanned from December 1, 2003, to December 1, 2023, with an absence of linguistic constraints. Redundancies were expunged utilizing the CiteSpace tool, culminating in a corpus of 9069 unique documents, subsequently imported into CiteSpace version 5.7.R5 for ensuing analysis. RESULTS: The five most cited references within this dataset were as follows: Gabbett TJ, 2016, titled "The training—injury prevention paradox: should athletes be training smarter and harder?". Ekstrand J, et al., 2011, titled "Injury incidence and injury patterns in professional football: the UEFA injury study". Ekstrand J, et al., 2016, titled "Hamstring injuries have increased by 4% annually in mens professional football, since 2001: a 13-year longitudinal analysis of the UEFA Elite Club injury study". Rampinini E, et al., 2009, titled "Technical performance during soccer matches of the Italian Serie A league: Effect of fatigue and competitive level". Hopkins W, et al., 2009, titled "Progressive statistics for studies in sports medicine and exercise science". The five most cited journals were the Journal of Sports Sciences, Sports Medicine, Medicine & Science in Sports & Exercise, Journal of Strength and Conditioning Research, and British Journal of Sports Medicine. The United States was the country with the highest number of publications and citations, followed by the United Kingdom, Australia, Spain, and Brazil. The top five most cited institutions were Liverpool John Moores University, Victoria University, Australian Catholic University, Edith Cowan University, and University Technology Sydney. The top 5 authors with the highest number of total citations in our dataset were Hopkins WG, Bangsbo J, Reilly T, Rampinini E, and Gabbett TJ. CONCLUSION: This study reveals three prominent and discrete general research trends that have occurred over the last five years. These trends were denoted as "training load and athlete monitoring," "sports injury risk and prevention," and "sports performance and analysis". This study identified emerging trends and knowledge structures in the field of soccer research. These findings can inform the future direction of funding agencies and research groups.

Read CV Hua Luo

ECSS Paris 2023: CP-AP13

Speaker C Clément Maviel

Speaker C

Clément Maviel
Université de Toulon, Laboratoire Jeunesse-Activité Physique et Sportive - Santé (J-AP2S)
France
"Quantifying Match Time Required for Reliable In-Situ Profiling in Rugby Union players."

INTRODUCTION: The force-velocity profile (P-FV) is commonly used to assess sprint-related physical qualities in sports, to orient training based on athletes performance. However, this method has limitations, particularly the inability to gather physical engagement’s information under actual match conditions in rugby. In this context, the use of Global Position System (GPS) technology has paved the way for collecting in-situ acceleration and speed data during rugby matches. Morin et al.[1] have introduced an in-situ method using GPS to generate an Acceleration-Speed (AS) profile conceptually close to the P-FV [2]. It is therefore crucial to understand players profiles in a match within a minimal time frame to avoid the confounding measures relative to the effects of training. This study aims to determine the saturation point of the AS profile in-match for rugby players, to provide a precise minimal time window for obtaining a meaningful AS profile in a match situation. METHODS: The playing time of 25 professional rugby players was recorded using GPS technology and segmented into four groups from 40 to 160 minutes of play. The analysis was conducted over 8 official matches during a period of 49 days. This segmentation enabled the gradual incorporation of data into the AS profile analysis. For each interval, the impact of integrating new match data on the AS profile outputs was assessed, aiming to identify the saturation point where additional data did not induce significant changes and thus, altered the profile. A repeated measures ANOVA was applied, and the significant differences were then explored using Bonferroni post-hoc tests, allowing for detailed comparisons between time windows. This methodology led to identifying the saturation point for theoretical maximum acceleration (A0) and maximum acceleration (S0), namely the threshold beyond which adding new gameplay data in the analysis does not result in statistically significant changes in these parameters. RESULTS: The ANOVA revealed a significant effect under all tested conditions with a loss of statistical significance from 120 minutes of play for acceleration (p = 0.12) and speed (p = 0.15), with an intra-subject variability of 3.29% for A0 and 1.99% for S0. Beyond 160 minutes, a lack of significant effect was observed for A0 (p = 1.00) and S0 (p = 0.99), with an intra-subject variability of 1.51% for A0 and 1.20% for S0. CONCLUSION: These findings indicate that the significant effect is not observed after 120 minutes of play, despite an important intra-subject variability. However, beyond 160 minutes, this variability is reduced, suggesting a clear saturation point. Therefore, it is recommended for practitioners to use 160 minutes of actual match play (equivalent to about 2 complete games), to derive a reliable AS profile. REFERENCES: 1 J.-B. Morin et al., (2021) 2 P. Clavel et al., (2022)

Read CV Clément Maviel

ECSS Paris 2023: CP-AP13