[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125995-en":3,"doc-seo-125995-105":30,"detail-sidebar-cat-0-en-105":92},{"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":11,"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},125995,687207024643,"Oliver","https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2",8,"Research & Report","Match predictions in soccer - Machine learning vs. Poisson approaches","Predicting soccer match outcomes draws strong interest from fans and from a large international sports betting market. This work compares machine learning models (e.g., neural networks and random forests) with Poisson-based approaches for single-match prediction across five European top leagues. Using season match results as features—excluding the target match—and applying equal weighting, the study finds no systematic team performance shifts within a season and shows that feature selection and model choice have only minor effects on prediction quality, with potential improvements discussed.","Match predictions in soccer: Machine learning vs. Poisson approaches  \nMirko Fischera, Andreas Heuera  \na Institute of Physical Chemistry, University of Münster,  \nCorrensstrasse 28/30, 48149 Münster  \nCorrespondence: mirko.fischer@uni-muenster.de, [andheuer@uni-muenster.de](andheuer@uni-muenster.de)  \nSummary  \nPredicting the results of soccer matches is of great interest. This is not only due to the popularity of the sport and the joy of private \"betting rounds\", but also due to the large sports betting market. Where previously expert knowledge and intuition were used, today there are models that analyze large amounts of data and make predictions based on them. In addition to Poisson models, approaches that belong to the machine learning (ML) category are increasingly being used. These include, for example, neural network or random forest models, which are compared in this article with each other as well as with Poisson models with regard to single-match prediction. In each case, the match results of a season are used as the data basis. The analysis is carried out for 5 European top leagues. A statistical analysis shows that the performance levels of the teams do not change systematically during a season. In order to characterize performance levels as accurately as possible, all match results, except from the match to be predicted, can be used as features with equal weighting. It can be seen that both, the exact choice of features and the choice of model, have only a minor influence on the prediction quality. Possible improvements in match prediction are discussed.  \n1 Introduction and motivation  \n1.1 Relevance of the match predictions  \nSince the 1960s, attempts have been made to systematically predict the results of soccer matches using statistical models (Dubitzky 2019), while previous predictions were essentially based on \"expert knowledge\". One of the main reasons for the high level of interest in such models and predictions is the growing international sports betting market, which is worth several billion US dollars per year (Etuk 2022) .  \nA match prediction essentially consists of two steps. First, the most meaningful information possible must be collected about the two teams. Matches are now analyzed and quantified in detail, especially using automated techniques (Goes 2020) . Teams can use this data to make important tactical decisions or identify suitable players for potential future transfers. In the context of match prediction, the performance strengths of the teams in particular can be estimated. Due to the limited information available before the match, the assessment of the performance of the two teams will always be subject to a certain degree of inaccuracy, but this can at least be minimized by optimizing the data used.  \nIn the second step, a match prediction model is required, which is \"fed\" with the available data. This shows that the result is not completely predictable. Reep and Benjamin (Reep 1968) already recognized that chance plays a significant role in the outcome of a game. For example, the teams with the higher market value do not always win-this is a major part of the appeal of soccer. Therefore, the actual prediction consists of determining probabilities for all possible outcomes. There are various fundamental reasons why exact predictions are not possible (Heuer 2012, Heuer 2014) .  \n(1) Relatively few goals are scored (on average between 2 and 3 goals in European professional leagues), so that the match result is generally a consequence of relatively few relevant actions and referee decisions, for example, have a major influence. This effect would disappear if (hypothetically) matches lasted much longer with the same level of performance.  \n(2) There are irreversible match-immanent effects (red cards, in-game injuries) that are unpredictable but can have a significant impact on the final result.  \n(3) There are match day-specific effects such as the prior absence of players due to injury or","cbCaipLi6Hsj5aUc","https://ap.wps.com/l/cbCaipLi6Hsj5aUc","pdf",563868,1,10,"English","en",105,"# Introduction and motivation\n## Relevance of match predictions\n## Machine learning in match prediction","[{\"question\":\"What are the two main steps in soccer match prediction described in the document?\",\"answer\":\"First, the most meaningful information about the two teams is collected and quantified. Second, a prediction model is trained with the available data to generate probabilities for possible outcomes.\"},{\"question\":\"Why are exact predictions of soccer match results fundamentally difficult?\",\"answer\":\"The document explains that only relatively few goals are scored, chance and irreversible match events (e.g., red cards and injuries) matter, and match-day-specific effects are hard to quantify, summarized as random effects.\"},{\"question\":\"How do the machine learning and Poisson approaches differ in this study?\",\"answer\":\"Poisson models assume goal-related random effects follow Poisson statistics, while the data-driven machine learning approaches used here do not make those assumptions and are instead trained from data.\"}]","Match predictions in soccer - Machine learning vs. Poisson approaches | PDF",1785902465,25,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"match-predictions-in-soccer-machine-learning-vs-poisson-approaches","",{"@graph":36,"@context":86},[37,54,69],{"@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/match-predictions-in-soccer-machine-learning-vs-poisson-approaches/125995/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-23","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":11},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What are the two main steps in soccer match prediction described in the document?","Question",{"text":76,"@type":77},"First, the most meaningful information about the two teams is collected and quantified. Second, a prediction model is trained with the available data to generate probabilities for possible outcomes.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Why are exact predictions of soccer match results fundamentally difficult?",{"text":81,"@type":77},"The document explains that only relatively few goals are scored, chance and irreversible match events (e.g., red cards and injuries) matter, and match-day-specific effects are hard to quantify, summarized as random effects.",{"name":83,"@type":74,"acceptedAnswer":84},"How do the machine learning and Poisson approaches differ in this study?",{"text":85,"@type":77},"Poisson models assume goal-related random effects follow Poisson statistics, while the data-driven machine learning approaches used here do not make those assumptions and are instead trained from data.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":21,"slug":134},"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]