statcast

Pull current and historical baseball statistics using Python (Statcast, Baseball Reference, FanGraphs)

MCP server for advanced baseball analytics (statcast, fangraphs, baseball reference, mlb stats API)

A Python application and library that generates comprehensive advanced stat summary sheets for MLB players, customizable by year, providing in-depth analysis and visualizations. It can also be used as a library module, enabling users to develop their own features and extend functionality for custom applications and data processing needs.

Small client used to retrieve statcast data from Baseball Savant.

A data visualization tool that is an extension of pybaseball

Query MLB Statcast data using plain English through an MCP server that connects AI assistants to Baseball Savant, FanGraphs, and Baseball Reference.

Python package: matplotlib baseball field, Statcast spray charts and strike-zone plots

Python tools for baseball data from Statcast, the MLB Stats API, FanGraphs, Baseball-Reference, Retrosheet, and the Lahman Database.

⚾ Herramienta profesional para analizar probabilidades de bateo de jugadores de MLB con predicciones basadas en datos de Statcast

Python package: Baseball Savant leaderboards with date ranges and season checks (bat tracking, ABS challenges, minors)

Predicting MLB hit probability from 2026 Statcast data with random forests and neural networks.

XGBoost win probability model for MLB games, with WPA, Leverage Index, and a benchmark against classical sabermetrics. Live dashboard with pitch-by-pitch game replay.

Chase+ measures the run value of a hitter's chase decisions by comparing swing outcomes with the estimated value of taking. It adjusts for count, distance outside the zone, and the pitches each hitter was shown, and is scaled so 100 is league average each season.

MLB plate appearance outcome models (gradiant boosting, random forest, neural net) with point-in-time features, walk-forward backtests, and Monte Carlo game simulator. Projections from 2026 Statcast data, powered by machine learning and 10,000 game situations.

mlbproj — lock at T-10. Daily MLB Log5 × Statcast vs Polymarket. One pick per market. Append-only ledger. No orders.

Reverse-engineered MLB The Show 26 rating formulas, with a Flask app that builds cards from real MLB stats. Any player, any season, 1897–2025.

A proprietary Python-based analytics platform that audits player performance on a pitch-by-pitch basis, integrating Win Probability (WPA), Quality of Contact (xBA), and Stadium/Weather factors to determine true situational worth.

Simple pitch visualization tool that uses StatCast to generate charts for pitcher movement, velocity distribution, and pitch usage.

Predicting MLB team runs from lineups, pitchers and parks: a leakage-safe Statcast pipeline and a neural model benchmarked against a linear GLM.

Free-data, honestly-calibrated MLB prediction model (ML / RL / O-U). Walk-forward backtested 2014+. FastAPI desktop app, automated morning sync, weekly retrain, end-of-season self-evaluation.