data/README.md

Data

raw/challenge_matches.json contains a curated historical challenge set. The displayed pre-match Elo values are approximate research inputs and are not represented as official ratings.

Run:

PYTHONPATH=src python scripts/prepare_matches.py

This derives lambda_home and lambda_away from the pre-match Elo difference. The match outcome is not used to derive the rates. The checked-in bootstrap coefficients are transparent defaults; a later research release should fit them on a separately licensed historical training set while excluding all challenge matches.

The challenge set is for product demonstration, not evidence that the model is state of the art.