NORMA eResearch @NCI Library

Using supervised learning techniques to predict kicking outcomes in the NFL

Gibney, Rory (2022) Using supervised learning techniques to predict kicking outcomes in the NFL. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (650kB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (451kB) | Preview


This purpose of this research project is to use supervised machine learning techniques in order to predict the outcome of kicks in the National Football League (NFL). There are 2 types of kicks that will be analysed in parallel; field goals and point after touchdowns (PATs). The motivation for this project came from a personal interest in the sport, and the knowledge that the sport is becoming every increasingly influenced by data driven decision making. Kickers are amongst the most important players on any NFL team. Understanding the conditions in which kickers perform better or worse can greatly help in the decision-making process for NFL team coaches. Models that predict kicking outcomes can also be retrospectively used to analyse whether a kicker, in the past, has made the kicks the model has predicted they should male. This paper aims to investigate various classification techniques to establish an optimal model in predicting whether a kicker will successfully make a kick or not, and potentially use this model to make real-time, in-game decisions. Feature selection will play a pivotal role in this process and will distinguish this research from other similar research undertaken to date. This analysis will help NFL team’s make decisions around what kickers are considered better than others, and ultimately help with difficult roster decisions. Various models were implemented, including naïve bayes, logistic regression and random forest. The overarching results summissed that predicting unsuccessful kick attempts is a much more difficult proposition than successful kick attempts; with a ranging accuracy between 25 and 90% respectively.

Item Type: Thesis (Masters)
Hasanuzzaman, Mohammed
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
G Geography. Anthropology. Recreation > GV Recreation Leisure > Sports
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 18 May 2023 14:43
Last Modified: 18 May 2023 14:43

Actions (login required)

View Item View Item