Graduation Semester and Year
2017
Language
English
Document Type
Thesis
Degree Name
Master of Science in Industrial Engineering
Department
Industrial and Manufacturing Systems Engineering
First Advisor
Kim Aera LeBoulluec
Abstract
In the new age of daily fantasy sports (DFS), fantasy football has become an enormous revenue generator for DFS sites, such as DraftKings and FanDuel. Both companies are valued over \$1 billion. However, previous analysis done by popular DFS site Rotogrinders has shown that only the top players are consistently winning, the top 10 players much more frequently than the remaining 20,000 players. Using complex statistical models they're able to identify top athletes and value picks (based on an athlete's draft `salary') that the average player might not be aware of. There is a need to evaluate which methods and algorithms are best at predicting fantasy football point output. These methods could then be applied to future DFS contests outside of football to see if they predict other fantasy sports point output well. There are few resources and little literature available on this subject. Several factors contribute. Daily Fantasy Sports are still relatively new, and many people are still just starting to get involved in them. Also, very few people have published their work on their custom models or significant variables, since they are generally developing these models for personal use in an attempt to gain an edge in DFS contests and win money. Thus, there is little to no motivation to make their research or methods publicly available. This research will attempt to predict the weekly point output of a quarterback based on a variety of attributes and metrics. Finding the important variables and statistical models and learning how to address the volatility in week-to-week performances for a quarterback will allow us to expand this to other player positions in the future. In addition to understanding the best algorithms to apply to weekly point prediction and the best variables to use to predict a quarterback's output, this research also seeks to answer the question that is a currently being debated in courtrooms across the country - should DFS be considered a legal game of skill, or a game of luck, and therefore online gambling?
Keywords
Data mining, Predictive analytics, Data science, Daily fantasy football, DraftKings, Regression
Disciplines
Engineering | Operations Research, Systems Engineering and Industrial Engineering
License
This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License.
Recommended Citation
King, Nicholas Aaron, "Predicting a Quarterback's Fantasy Football Point Output For Daily Fantasy Sports Using Statistical Models" (2017). Industrial, Manufacturing, and Systems Theses. 13.
https://mavmatrix.uta.edu/industrialmanusys_theses/13
Comments
Degree granted by The University of Texas at Arlington