Assessing The Feasibility Of Nfl Sports Betting As An Alternative Investment Strategy Through Predictive Analytics

Sports gambling is a rapidly expanding industry throughout the United States, providing many individuals with the opportunity to achieve profits through several platforms and applications -- much like the financial markets. While these markets are structured differently at their core, many qualitative similarities exist that suggest that profitable investment can be accomplished through either market. Investors typically analyze a number of accounting and financial metrics to assess which financial assets should be added to their portfolios; with an abundance of information easily accessible to the public, a similar analytical approach -- using performance statistics and predictive modeling -- can be used in the sports betting market to determine on which outcomes to wager. This thesis explores the feasibility of predictive modeling strategies in the sports betting market of being an alternative to traditional investment strategies. Building on techniques explored in previous literature, a probit regression model using extensive, historical NFL data is developed and refined, and its returns are compared to those observed in the financial markets during the 2022 NFL season. The results of this study suggest that exploitable arbitrage opportunities exist that allow individuals to achieve returns at a rate greater than the financial markets, providing promising evidence than sports betting can be an alternative financial asset class.