Markov Chain Theory with Applications to Baseball

The applications of Markov chains span a wide range of fields to which models have been designed and implemented to simulate random processes. Markov chains are stochastic processes that are characterized by their memoryless property, where the probability of the process being in the next state of the system depends only on the current state and not on any of the previous states. This property is known as the Markov property. This thesis paper will first introduce the theory of Markov chains, along with explaining two types of Markov chains that will be beneficial in creating a model for analyzing baseball as a Markov chain. The final chapter describes this Markov chain model for baseball, which we will use to calculate the expected number of runs scored for the 2013 College of Wooster baseball team. This paper finishes by displaying an analysis of sacrifice bunt and stolen base strategies through using the Markov chain model.