Online Program Home
My Program

Abstract Details

Activity Number: 43 - SPEED: Statistics in Sports; Physical Activity/Sleep Studies, and Nonparametrics Part 1
Type: Contributed
Date/Time: Sunday, July 28, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistics in Sports
Abstract #306967 Presentation
Title: A SHINY Markov Machine for Decision-Making in Major League Baseball
Author(s): Jason Osborne*
Companies: North Carolina State University
Keywords: baseball; markov; simulation; shiny; shrinkage

We present a shiny app enabling the user to carry out Markov Chain calculations to assist with decision-making in baseball. These decisions, made both during and before the game, use summary statistics to estimate transition probabilities. The Markovian assumption leads to the calculation of the entire probability distribution of runs scored in the remainder of a game. An example of such a decision is whether or not to attempt to steal a base. While conventional analysis has phrased this problem in terms only of the base stealer's chance of success, a more informed decision would take account of the sequence of hitters who follow the batter in the lineup (and all other available information.) Other illustrations of this machinery include selection of batting order at the beginning of a game and deciding whether or not to attempt a sacrifice bunt or pinch-hit to exploit a hitter-pitcher handedness advantage. The app allows users to select the following inputs: MLB team and year, an ordering of nine players from each team, inning, outs, runners on base and pitcher handedness.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program