Comparing baseball players across eras via novel full house modeling

Yan, Shen

Permalink

https://hdl.handle.net/2142/124575 Copy

Description

Title

Comparing baseball players across eras via novel full house modeling

Author(s)

Yan, Shen

Issue Date

2024-04-26

Director of Research (if dissertation) or Advisor (if thesis)

Eck, Daniel J

Doctoral Committee Chair(s)

Eck, Daniel J

Committee Member(s)

• Burgos, Adrian
• Zhao, Sihai
• Douglas, Jeffrey

Department of Study

Statistics

Discipline

Statistics

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Order Statistics
• Nonparametric Estimation
• Extrapolation Techniques
• Era-adjustment
• Sports Statistics.

Language

eng

Abstract

We propose a statistical model for era-adjusting the baseball statistics in Major League Baseball and Negro League. Also, we consider an intriguing paradox that the component with more talent underperforms the component with less talent under our model setting. In the first project, we propose a new methodological framework suitable for era-adjusting baseball statistics developed in this article. Within this methodological framework, specific models are motivated. We call these models Full House Models. Full House Models work by balancing the achievements of Major League Baseball (MLB) players within a given season and the size of the MLB talent pool from which a player came. We demonstrate the utility of Full House Models in an application of comparing baseball players' performance statistics across eras. Our results reveal a new ranking of baseball's greatest players which includes several modern players among the top all-time players. Modern players are elevated by Full House Modeling because they come from a larger talent pool. Sensitivity and multiverse analyses which investigate how results change with changes to modeling inputs including the estimate of the talent pool are presented. In the second project, we continue our work on the Full House Model for computing era-adjusted or system-adjusted statistics that balances the size of the eligible population in each system with the component's performance via "versus their peers" in a specific system. We assume that each component has a latent aptitude that can be calculated from this balancing act provided the system inclusion mechanism for components of the system is known. Without making any assumptions about the distribution of the system's components, the components' distribution is well-established from the nonparametric probability distribution. Then we consider an intriguing paradox that the component with more talent underperforms the component with less talent. We compute the probability of this paradox under different scenarios. In the third project, we continue our work on the Full House Model with the application of Negro League baseball statistics. With the estimation of the talent pool for the negro league, we compare the negro league players with all MLB players across eras. This third project is under development and in our further study, our estimation of the era-adjusted statistics for negro league players will be more accurate and convincing with the updated negro league baseball statistics and adjustments in our Full House Modeling in the context of negro league.

Graduation Semester

2024-05

Type of Resource

Text

Handle URL

https://hdl.handle.net/2142/124575

Copyright and License Information

Copyright 2024 Shen Yan

Owning Collections