[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124049-en":3,"doc-seo-124049-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},124049,1099514068035,"Ezra","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Improving the Evaluation of Defensive Player Values with Advanced Machine Learning Techniques","Quantifying defensive actions in football remains difficult because traditional analysis has historically emphasized offensive indicators. This study proposes a new machine-learning based evaluation method for defensive play, combining XGBoost and neural networks with established action-value metrics: On-Ball Value (OBV), Valuing Actions by Estimating Probabilities (VAEP), and eXpected Threat (xT). Experiments on Polish PKO BP Ekstraklasa seasons compare the model’s rankings to expert ratings and market values. Results demonstrate improved effectiveness for Defensive Player Value.","Improving the Evaluation of Defensive Player Values with Advanced  \nMachine Learning Techniques  \nMichał Zar˛eba  \nFaculty of Mathematics and Computer Science at Adam Mickiewicz University  \nPozna´n, Poland [michal.zareba@amu.edu.pl](michal.zareba@amu.edu.pl)  \nTomasz Piłka  \nFaculty of Mathematics and Computer Science at Adam Mickiewicz University  \nPozna´n, Poland [tomasz.pilka@amu.edu.pl](tomasz.pilka@amu.edu.pl)  \nTomasz Górecki  \nFaculty of Mathematics and Computer Science at Adam Mickiewicz University  \nPozna´n, Poland [tomasz.gorecki@amu.edu.pl](tomasz.gorecki@amu.edu.pl)  \nBartłomiej Grzelak  \nFaculty of Mathematics and Computer Science at Adam Mickiewicz University  \nPozna´n, Poland [bartlomiej.grzelak@lechpoznan.pl](bartlomiej.grzelak@lechpoznan.pl)  \nKrzyszof Dyczkowski  \nFaculty of Mathematics and Computer Science at Adam Mickiewicz University  \nPozna´n, Poland [krzysztof.dyczkowski@amu.edu.pl](krzysztof.dyczkowski@amu.edu.pl)  \nAbstract  \nQuantifying defensive actions, which offensive indicators have historically overshadowed, is challenging in football analysis. This study presents a novel approach using XGBoost and neural networks to evaluate defensive play using On-Ball Value (OBV), Valuing Actions by Estimating Probabilities (VAEP), and eXpected Threat (xT) indicators. The proposed evaluation of Defensive Player Value using machine learning techniques is presented. A comparative assessment of expert ratings and market values in a Polish PKO BP Ekstraklasa case study highlights the method’s effectiveness. The research contributes to the development of sports analytics by addressing the long-term challenge of evaluating the defensive play of football players.  \nKeywords: football, player evaluation, machine learning, deep learning, XGBoost  \n1. Introduction  \nIn football analytics, the main focus has been on player performance metrics, specifically expected goal (xG) and expected threat (xT) models. The xG metric aims to quantify shot quality, and numerous studies and data companies continue to refine methods for calculating shot quality [6] . In contrast, the xT model, introduced by Singh [14], uses a Markov model to assess the dynamics of ball possession, providing insight into how individual actions on the pitch contribute to creating goal-scoring opportunities. However, these models often focus on personal actions and overlook the interconnected nature of the events that lead to those actions. Some researchers advocate a more holistic approach, analyzing play sequences to understand the game’s dynamics better [13, 16] . This approach highlights the intricacies of football, where actions that directly lead to goals or assists are only a fraction of the total. It emphasizes the importance of player creativity and strategic decision-making. Recent methodologies have broadened this scope of analysis to include aspects such as player creativity [12], team performance evaluation, and pattern recognition in games [5] .  \nSignificant AI-driven developments have assessed player quality and on-field activity in recent years [17] . In particular, metrics such as OBV [15], VAEP [3], and xT have emerged to assess different facets of football play [11, 14] . However, existing metrics do not adequately evaluate defensive actions. Therefore, our work focuses on improving defensive assessment by integrating established methods and deep learning techniques. This research underscores the ongoing quest to refine football analytics to ensure a holistic evaluation of player performance, including defensive skills.  \n2. Dataset and methods  \n2.1. Dataset  \nThe dataset utilized for this research was sourced from StatsBomb 1 , covering the 2021/2022 and 2022/2023 PKO BP Ekstraklasa league seasons (612 games) . It includes detailed event data (Table 1), such as team and player information, possession chains, individual player actions, and event locations, providing a holistic view of the game’s dynamics. This event data logs every pa","cbCaibavRGhnevsT","https://ap.wps.com/l/cbCaibavRGhnevsT","pdf",190411,1,5,"English","en",105,"# Introduction\n# Dataset and methods\n## Dataset\n## Football metrics\n## Defensive Player Value (DPV)","[{\"question\":\"Why is evaluating defensive actions challenging in football analytics?\",\"answer\":\"Defensive actions are often overlooked because many existing analytics approaches historically prioritize offensive indicators. The study highlights the need for better defensive assessment that accounts for event interconnections.\"},{\"question\":\"Which machine learning methods and metrics does the study use?\",\"answer\":\"The approach uses XGBoost and neural networks together with OBV, VAEP, and xT indicators to quantify different facets of play and to construct a defensive value measure.\"},{\"question\":\"What data and league seasons support the experiments?\",\"answer\":\"Experiments use StatsBomb event data covering the 2021/2022 and 2022/2023 PKO BP Ekstraklasa seasons, totaling 612 games with detailed event logs used to derive model inputs.\"}]","Improving the Evaluation of Defensive Player Values with Advanced Machine Learning Techniques | PDF",1785820096,13,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"improving-the-evaluation-of-defensive-player-values-with-advanced-machine-learning-techniques","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/improving-the-evaluation-of-defensive-player-values-with-advanced-machine-learning-techniques/124049/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is evaluating defensive actions challenging in football analytics?","Question",{"text":75,"@type":76},"Defensive actions are often overlooked because many existing analytics approaches historically prioritize offensive indicators. The study highlights the need for better defensive assessment that accounts for event interconnections.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which machine learning methods and metrics does the study use?",{"text":80,"@type":76},"The approach uses XGBoost and neural networks together with OBV, VAEP, and xT indicators to quantify different facets of play and to construct a defensive value measure.",{"name":82,"@type":73,"acceptedAnswer":83},"What data and league seasons support the experiments?",{"text":84,"@type":76},"Experiments use StatsBomb event data covering the 2021/2022 and 2022/2023 PKO BP Ekstraklasa seasons, totaling 612 games with detailed event logs used to derive model inputs.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,109,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":21,"slug":137},19,"General","general"]