How Databricks Helps Baseball Teams Gain an Edge with Data & AI

This post describes how baseball teams use Databricks to convert high-fidelity pitch data into game decisions through AI agents and a governed data lakehouse.

• •Analysts use Genie on Unity Catalog to query Statcast tables with natural language, generating hitter meeting one-pagers with pitch mix and location tendencies by hand and base-runner state
• •A Multi-Agent Supervisor built with Agent Bricks and deployed on Model Serving simulates bullpen matchup scenarios, calling UC SQL functions to compare each reliever's arsenal against specific hitter pockets
• •Pinch-hit decisions are pre-planned using the same agent framework, ranking bench bats by expected outcome against likely late-inning relievers before the game starts
• •Vector Search indexes pitcher embeddings in Unity Catalog to find comparable pitchers when direct Statcast head-to-head history is thin, enabling advance scouting reports