To Swing or Not To Swing: a Shiny App Plate Discipline Simulator

2026-03-21 · Source: Data Science on Medium · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Software Development & Engineering · Depth: Intermediate, medium

Summary

An interactive R Shiny application was developed using MLB Statcast data to simulate and optimize baseball plate discipline by evaluating the mathematical quality of swing decisions. The simulator calculates the Expected Value (EV) of swinging versus taking a pitch, factoring in pitch type, location (categorized into heart, shadow, chase, and waste zones), current count, and batter/pitcher handedness. It utilizes functions like `avg_woba_lookup`, `pitch_odds`, and `loc_odds` to assign probabilities and wOBA values, guiding users to make "GOOD DECISION" or "BAD DECISION" choices. The open-source tool provides a practical framework for training, aiming to simplify the complex art of plate discipline rather than creating "robots." Future enhancements include hitter archetypes and animated pitches.

Key takeaway

An R Shiny application uses MLB Statcast data to simulate and optimize baseball plate discipline by calculating the Expected Value (EV) of swinging versus taking. It probabilistically models pitch type, location (heart, shadow, chase, waste), count, and handedness via `avg_woba_lookup`, `pitch_odds`, and `loc_odds` functions to provide real-time decision feedback. This framework offers AI/ML professionals a practical, interactive tool for applying data-driven decision-making to complex, high-speed scenarios, shifting focus from outcomes to underlying decision quality.

Topics

• Plate Discipline Simulation
• R Shiny Application
• MLB Statcast Data
• Expected Value
• Baseball Analytics

Code references

• isaac-ciravolo/swingdecisionsimulator

Best for: Data Scientist, Software Engineer, AI Student

Related on AIssential

• NHL Simulator v1.0 — A Markov chain model can effectively simulate complex sports events using play-by-play data and transition probabilities.
• Scouting By Reward: VLM-TO-IRL-Driven Player Selection For Esports — Esports player evaluation can be reframed as an Inverse Reinforcement Learning problem to identify stylistic alignment.
• When Data Lies: Finding Optimal Strategies for Penalty Kicks with Game Theory — Optimal decisions in strategic interactions cannot be inferred from raw historical averages alone.
• A Constraint Programming Approach for $n$-Day Lookahead Playoff Clinching — A novel algorithm efficiently determines NHL playoff clinching scenarios by combining tree search with constraint programming.
• PAWN: Piece Value Analysis with Neural Networks — Full board state context, via CNN autoencoders, significantly enhances chess piece value prediction accuracy.
• Valet: A Standardized Testbed of Traditional Imperfect-Information Card Games — Valet standardizes imperfect-information card games for robust AI algorithm benchmarking.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Data Science on Medium.