Abstract

In this paper, we will discuss a method of building a predictive model for Major League Baseball Games. We detail the reasoning for pursuing the proposed predictive model in terms of social popularity and the complexity of analyzing individual variables. We apply a coarse-grain outlook inspired by Simon Dedeos' work on Human Social Systems, in particular the open source website Wikipedia [2] by attempting to quantify the influence of winning and losing streaks instead of analyzing individual performance variables. We will discuss initial findings of data collected from the LA Dodgers and Colorado Rockies and apply further statistical analysis to find optimal betting points using a coarse-grain approach. We will apply Bayes' Theorem to add predictive power to a naive model using winning and losing streaks. We will discuss possible shortcomings of the proposed using Bayes' approach and address the question as to whether or not baseball wins and losses can be produced using a random process.

Advisor

Kim In-Jae

First Committee Member

Deepak Sanjel

Second Committee Member

Han Wu

Date of Degree

2014

Language

english

Document Type

Thesis

Degree

Master of Science (MS)

Department

Mathematics and Statistics

College

Science, Engineering and Technology

Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial 4.0 License

Share

COinS