Determining Optimal Outfield Shifting Strategies for Major League Baseball Teams

"Infield shifts have become increasingly popular and effective in the modern era of Major League Baseball. Conversely, the outfield cannot be nearly as well covered as the infield given its size relative to the number of outfielders. We propose a method for determining analogous fielding shifts in the outfield. In this thesis we use existing statistics generated by years of Major League data such as win probability to develop an algorithm that, given a game state and distribution of outfield hits, will search over possible fielding ranges, calculate the probability of transitioning to the next game state, the resultant change in win probability, and ultimately output the optimal fielding position for that scenario. Additionally, in order to make this analysis practical, we develop an equivalence relation to partition the set of all batters into equivalent classes such that the number of distinct fielding positions can be effectively implemented in real game scenarios."

In collections

File details
ID Label Size Mimetype Created
OBJ OBJ datastream 1.84 MiB application/pdf 2021-05-28
TN TN 3.25 KiB image/jpeg 2021-05-28