MONEYBaRL: Exploiting pitcher decision-making using Reinforcement Learning

Abstract

This manuscript uses machine learning techniques to exploit baseball pitchers' decision making, so-called "Baseball IQ," by modeling the at-bat information, pitch selection and counts, as a Markov Decision Process (MDP). Each state of the MDP models the pitcher's current pitch selection in a Markovian fashion, conditional on the information immediately prior to making the current pitch. This includes the count prior to the previous pitch, his ensuing pitch selection, the batter's ensuing action and the result of the pitch.The necessary Markovian probabilities can be estimated by the relevant observed conditional proportions in MLB pitch-by-pitch game data. These probabilities could be pitcher-specific, using only the data from one pitcher, or general, using the data from a collection of pitchers.Optimal batting strategies against these estimated conditional distributions of pitch selection can be ascertained by Value Iteration. Optimal batting strategies against a pitcher-specific conditional distribution can be contrasted to those calculated from the general conditional distributions associated with a collection of pitchers.In this manuscript, a single season of MLB data is used to calculate the conditional distributions to find optimal pitcher-specific and general (against a collection of pitchers) batting strategies. These strategies are subsequently evaluated by conditional distributions calculated from a different season for the same pitchers. Thus, the batting strategies are conceptually tested via a collection of simulated games, a "mock season," governed by distributions not used to create the strategies. (Simulation is not needed, as exact calculations are available.)Instances where the pitcher-specific batting strategy outperforms the general batting strategy suggests that the pitcher is exploitableknowledge of the conditional distributions of their pitch-making decision process in a different season yielded a strategy that worked better in a new season than a general batting strategy built on a

A hierarchical Bayesian model of pitch framing

“…In 2006, the TV broadcast company Sportvision began offering the PITCHf/x service to track and digitally record the full trajectory of each pitch thrown using a system of cameras installed in major league ballparks. During the flight of each pitch, these cameras take 27 images of the baseball and the PITCHf/x software fits a quadratic polynomial to the 27 locations to estimate its trajectory (Sidhu and Caffo, 2014). This data is transmitted to the MLB Gameday application, which allows fans to follow the game online (Fast, 2010).…”

“…We begin this section with a brief overview, adapted primarily from Fast (2010) and Sidhu and Caffo (2014), of our pitch tracking dataset before introducing the hierarchical logistic regression model used to estimate each umpire's called strike probability.…”

Artificial Intelligence in Elite Sports—A Narrative Review of Success Stories and Challenges

“…To highlight another publication in football, FIFA ( 2014 ) use DL algorithms to create ghost teams to analyze reactions of different tactical approaches. Examples of publications in the big four American sports are Tian et al ( 2020 ) (basketball), Joash Fernandes et al ( 2020 ) (American football), Kononenko ( 2001 ) (hockey) and Sidhu and Caffo ( 2014 ) (baseball).…”

ML-Based Energy Consumption and Distribution Framework Analysis for EVs and Charging Stations in Smart Grid Environment

“…In our proposed energy consumption and distribution framework analysis for EVs and CSs in a Smart Grid Environment, we adopt a statistical approach to transform a conventional smart grid into an intelligent one. Unlike conventional machine learning techniques, our methodology is rooted in SLT [36] [37]. Let X be the vector space of all possible inputs and Y be the vector space of all possible outputs.…”