Other ways to work with Retrosheet data¶
There are several good options. This is a fair summary as of October 2026; check each project's own page for current details.
| Option | What it is | Needs | Good for |
|---|---|---|---|
| chadwickpy | Pure-Python port of the six Chadwick tools | Python 3.11+ only | Chadwick's exact tables with one pip install, anywhere |
| Chadwick (C) | The reference tools | C compiler to build, or a packaged build | Speed; the original |
| pychadwick | Python bindings to Chadwick's C library | The compiled C library | Calling Chadwick from Python; per its page it supports event data only |
| pyretrosheet | Python library that downloads and parses Retrosheet play-by-play into Python objects | Python | Working with games as Python objects |
| Retrosheet's CSV downloads | Pre-parsed yearly tables from Retrosheet | Nothing | A table with no tooling, using Retrosheet's columns |
R retrosheet |
R package that imports Retrosheet data | R | R users |
Where chadwickpy fits¶
- It reproduces Chadwick's output, verified on every season with game data, 1897 to 2025 (see how it was verified), without needing the C library.
- It covers all six tools, not just
cwevent. - It is slower than the C original.
If you only want a CSV and are happy with Retrosheet's columns, you may not need any tool. If you want Chadwick's definitions (the ones most published baseball research and tutorials use), chadwickpy is the easiest way to get them.
Corrections welcome
If something here is out of date or unfair to another project, open an issue.