sportsdataverse-py NOTICE
=========================

sportsdataverse-py is distributed under the MIT License (see LICENSE).

It contains code and test-only fixtures derived from the following third-party
projects, both licensed under the Apache License, Version 2.0
(http://www.apache.org/licenses/LICENSE-2.0). Per Apache-2.0 Section 4, their
copyright and attribution notices are retained below. Neither upstream project
ships a NOTICE file of its own.

-------------------------------------------------------------------------------
cbb-on-off-analyzer  —  https://github.com/Alex-At-Home/cbb-on-off-analyzer
  Copyright (c) Alex-At-Home and contributors. Licensed under Apache-2.0.

  The TypeScript stats utilities LineupUtils.ts, RatingUtils.ts, LuckUtils.ts,
  RapmUtils.ts, PositionUtils.ts, and buildStrengthAdjustedStats.ts (with
  helpers from OverrideUtils.ts and PositionalManualFixes.ts) were ported into:
      sportsdataverse/mbb/mbb_lineup_stats.py
      sportsdataverse/mbb/mbb_ratings.py
      sportsdataverse/mbb/mbb_luck.py
      sportsdataverse/mbb/mbb_rapm.py
      sportsdataverse/mbb/mbb_positions.py
      sportsdataverse/mbb/mbb_ncaa_strength.py
  and re-exported by reference from the parallel sportsdataverse/wbb/ modules.

-------------------------------------------------------------------------------
cbb-explorer  —  https://github.com/Alex-At-Home/cbb-explorer
  Copyright (c) Alex-At-Home and contributors. Licensed under Apache-2.0.

  The Scala NCAA play-by-play ingest engine (EventUtils, PossessionUtils,
  StateUtils, ExtractorUtils, LineupUtils, LineupErrorAnalysisUtils,
  DataQualityIssues, the ncaa HTML parsers, and the associated models) was
  ported into sportsdataverse/mbb/mbb_ncaa_*.py and re-exported by reference
  from sportsdataverse/wbb/wbb_ncaa_*.py.

-------------------------------------------------------------------------------
fastRhockey  —  https://fastRhockey.sportsdataverse.org
  Copyright (c) SportsDataverse and contributors. Licensed under Apache-2.0.

  The NHL expected-goals (xG) feature-preparation + booster-scoring pipeline
  (``helper_nhl_prepare_xg_data`` / ``helper_nhl_calculate_xg``) was ported into
  sportsdataverse/nhl/nhl_xg.py, reusing the published ``nhl_xg_models``
  boosters verbatim (no retrain). The player-impact spine's RAPM / GSAx /
  GAR-WAR modules (nhl_rapm.py, nhl_gsax.py, nhl_unit_ratings.py,
  nhl_special_teams.py, nhl_war.py) follow published methodology
  (EvolvingHockey, MoneyPuck) rather than ported source code and carry no
  license obligation beyond the docstring citations.

-------------------------------------------------------------------------------
Each port is a faithful translation of the upstream logic; documented
divergences are noted in the respective module docstrings. Test-only oracle
fixtures vendored byte-exact from the upstream repositories live under
tests/fixtures/hoop_explorer/ and tests/fixtures/ncaa/ (see each directory's
README for per-file provenance); these fixtures are not shipped in the
distributed wheel or sdist.
