Kaggle Datasets

Baseball-themed datasets published on Kaggle, generated with pybaseball and Baseball Savant.

Published Datasets (6)

1. Japanese MLB Players Statcast (2015-2025)

Comprehensive Statcast data for 34 Japanese MLB players across 10+ seasons.

• Pitchers: 25 players, 118,226 pitches (2015-2025)
• Batters: 10 players, 56,362 batted balls (2015-2025)
• Columns: 238 metrics (pitch velocities, spin rates, exit velocities, launch angles, xwOBA, etc.)
• Size: 75.5 MB
• DOI: 10.34740/kaggle/dsv/10697439
• Article: Zenn

2. MLB Bat Tracking (2024-2025) 🥈

MLB bat tracking leaderboard data from Baseball Savant (2024-2025).

• Data: 452 batters (226 per season)
• Columns: 19 swing metrics (bat speed, squared-up rate, blasts, swords, etc.)
• Size: 46 KB
• DOI: 10.34740/kaggle/dsv/10699103
• Article: Zenn

3. MLB Pitcher Arsenal Evolution (2020-2025)

Seasonal arsenal composition and performance metrics for MLB pitchers (2020-2025).

• Data: 4,253 pitcher-seasons (2020-2025)
• Columns: 111 metrics (pitch type usage %, velocity, movement, whiff rate, xwOBA, etc.)
• Size: 1.14 MB
• DOI: 10.34740/kaggle/dsv/10704532
• Article: Zenn

4. MLB Statcast + Bat Tracking (2024-2025)

Pitch-by-pitch Statcast data merged with Bat Tracking metrics.

• Data: ~1.4M rows, ~2.4 GB
• Metrics: bat speed, swing length, swing path tilt, Statcast pitch-by-pitch data
• Notebook:

5. Baseball Savant Leaderboards (2024-2025)

20 Baseball Savant & FanGraphs leaderboards as clean CSV files.

• Files: 20 CSVs — batting, pitching, fielding, catching, baserunning, park factors
• Sources: savant-extras v0.4.3 (17 leaderboards) + pybaseball (OAA, Outfield Jump, Pitcher Quality)
• Notebooks: Showcase · Defense & Pitching Quality

6. WBC 2026 Scouting - Statcast Data 🥈

Pitch-by-pitch Statcast data for WBC 2026 roster players, 20 countries.

• Batters: 338,811 pitches, 18 countries
• Pitchers: 220,385 pitches, 14 countries
• Summary: 109 batters / 90 pitchers (per-player stats)
• Rosters: 308 MLB-affiliated players across 20 countries
• Auto-update: GitHub Actions (update-wbc-dataset.yml) — triggers on workflow_dispatch

Workflow

WBC 2026 Scouting (automated)

Actions → Update WBC 2026 Kaggle Dataset → Run workflow

Internally: checkout wbc-scouting → run generate.py → clean rosters.csv → kaggle datasets version

Required secrets: KAGGLE_USERNAME, KAGGLE_KEY

Other datasets (manual)

1. Run generate.ipynb in Google Colab to generate CSV
2. Download CSV to local dataset folder
3. kaggle datasets create -p <folder> (first time) or kaggle datasets version -p <folder> -m "message" (update)

Related

• Kaggle Notebooks
• MLB Statcast Visualization