Beating the Odds: - A Statistical Approach to Hockey Goals in the NHL
2024 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE credits
Student thesisAlternative title
Slå oddsen : - En statistisk undersökning av mål i NHL (Swedish)
Abstract [en]
This thesis explores the application of statistical models to predict goal outcomes in NHL hockey games and to identify advantageous betting opportunities. The study introduces two statistical models: a Poisson regression model and a Bayesian Poisson regression model, specifically designed to estimate the number of goals in NHL games based on historical team performance data. By examining discrepancies in bookmaker odds, the research aims to demonstrate that a data-driven approach can expose value bets that are typically overlooked. Both models are tested against historical betting odds, showing their potential to provide bettors with a statistical edge over traditional bookmaking methods. This study not only contributes to the sports analytics field by applying comprehensive statistical methods to betting strategies but also makes gamblers more aware and statistically informed by reducing the information asymmetry between bookmakers and bettors.
Abstract [sv]
Denna avhandling undersöker användningen av statistiska modeller för att förutsäga antalet mål i NHL-hockeymatcher och därigenom identifiera fördelaktiga bettingmöjligheter. Studien introducerar två statistiska modeller: en Poisson regressionsmodell och en Bayesiansk Poisson regressionsmodell, specifikt utformade för att uppskatta antalet mål i NHL-matcher baserat på historisk data. Genom att granska avvikelser i bettingbolagens odds, syftar forskningen till att demonstrera att en datadriven metod kan upptäcka "värdespel" som vanligtvis förbises. Båda modellerna testas mot historiska matchers odds, vilket visar deras potential att ge spelare en statistisk fördel över traditionella bettingmetoder. Denna studie bidrar inte bara till området för sportslig analys genom att tillämpa omfattande statistiska metoder på bettingstrategier, utan gör också spelare mer medvetna och statistiskt informerade genom att minska informationsasymmetrin mellan bettingbolagen och spelare.
Place, publisher, year, edition, pages
2024.
Series
TRITA-SCI-GRU ; 2024:103
Keywords [en]
Statistical Models, NHL (National Hockey League), Poisson Regression Model, Bayesian Poisson Regression Model, Goal Prediction, Betting Opportunities, Value Bets, Odds, Sports Analytics, Probability Theory, Information Asymmetry, Betting Strategy, Predictive Analysis, Sports Betting Industry
Keywords [sv]
Statistiska modeller, NHL (National Hockey League), Poisson-regressionsmodell, Bayesiansk Poisson-regressionsmodell, Målprognos, Bettingmöjligheter, Värdespel, Odds, Sportanalys, Sannolikhetsteori, Informationsasymmetri, Bettingstrategi, Prediktiv analys, Sportbetting
National Category
Probability Theory and Statistics
Identifiers
URN: urn:nbn:se:kth:diva-352557OAI: oai:DiVA.org:kth-352557DiVA, id: diva2:1894676
Subject / course
Applied Mathematics and Industrial Economics
Educational program
Master of Science in Engineering - Industrial Engineering and Management
Supervisors
Examiners
2024-09-032024-09-03