Please use this identifier to cite or link to this item: https://repository.iimb.ac.in/handle/2074/22035
DC FieldValueLanguage
dc.contributor.advisorRoy, Rishideep
dc.contributor.authorMandar, Akshay Kumar
dc.contributor.authorGupta, Shivansh
dc.date.accessioned2023-07-02T15:20:22Z-
dc.date.available2023-07-02T15:20:22Z-
dc.date.issued2022
dc.identifier.urihttps://repository.iimb.ac.in/handle/2074/22035-
dc.description.abstractPredicting the live winning probability of a sporting game is becoming increasingly important in the sports industry. It provides fans, bettor s, teams, and coaches with valuable insights into t he game and can help them make more informed decisions. By knowing the probability of winning at any given point in the game, fans and bettors can adjust their expectations and make better decisions about placing bets or making in-game predictions. Teams and coaches can use the predicted probabilities to inform their strategic decisions and increase the chances of winning. Additionally, broadcasters and sports analysts can use the predicted probabilities to provide real-time updates to viewers and provide more informed commentary and insights into the game. Compared to other sports, advanced statistics in Hockey are still in infancy. It has been suggested that the best models can only predict the winner 62% of the time due to variances in talent and "puck luck". The goal of this project was to quantify the uncertainty present in NHL hockey games by using the Kaggle NHL Game dataset to predict the initial and live [>r f bability of winning for both teams based on the game's current developments. We did statistical analysis of the game to assess what and how much individual actions contribute to t he outcome of the game. The project involved cleaning and pre-processing the data, developing a predictive model, and analysing the model's performance. We used a logistic regression model to predict the probability of each team winning a game. The model took into account various game features, such as game time, goals scored, and team performance statistics, to predict the probability of each team winning. We also developed a live prediction system that updated the probabilities in real-time as the game progressed. Overall, this project demonstrates the potential of using machine learning techniques to predict the outcome of sports games. The predictive model developed in this project can be used to inform betting strategies, as well as provide insights into team performance and game dynamics.
dc.publisherIndian Institute of Management Bangalore
dc.relation.ispartofseriesPGP_CCS_P22_179
dc.subjectSports
dc.subjectHockey
dc.subjectNational Hockey League
dc.subjectNHL
dc.titleEstimating the win probability in a National Hockey League (NHL) game
dc.typeCCS Project Report-PGP
dc.pages21p.
Appears in Collections:2022
Files in This Item:
File SizeFormat 
PGP_CCS_P22_179.pdf2.3 MBAdobe PDFView/Open    Request a copy
Show simple item record

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.