nhl-bigquery 0.1.0__tar.gz

nhl_bigquery-0.1.0/.github/workflows/release.yml

@@ -0,0 +1,41 @@
+name: release
+
+on:
+  push:
+    tags:
+      - "v*"
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    outputs:
+      dist-path: ${{ steps.build.outputs.dist-path }}
+    steps:
+      - uses: actions/checkout@v4
+      - uses: astral-sh/setup-uv@v3
+      - run: uv python install 3.13
+      - name: Create venv
+        run: uv venv
+      - name: Install build tools
+        run: uv pip install build
+      - id: build
+        run: |
+          uv run python -m build
+          echo "dist-path=dist" >> $GITHUB_OUTPUT
+      - uses: actions/upload-artifact@v4
+        with:
+          name: dist
+          path: dist/
+
+  publish:
+    needs: build
+    runs-on: ubuntu-latest
+    environment: pypi
+    permissions:
+      id-token: write
+    steps:
+      - uses: actions/download-artifact@v4
+        with:
+          name: dist
+          path: dist/
+      - uses: pypa/gh-action-pypi-publish@release/v1

nhl_bigquery-0.1.0/.github/workflows/test.yml

@@ -0,0 +1,30 @@
+name: test
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.11", "3.12", "3.13"]
+    steps:
+      - uses: actions/checkout@v4
+      - uses: astral-sh/setup-uv@v3
+      - name: Set up Python ${{ matrix.python-version }}
+        run: uv python install ${{ matrix.python-version }}
+      - name: Create venv
+        run: uv venv
+      - name: Install
+        run: uv pip install -e ".[dev]"
+      - name: Lint
+        run: uv run ruff check .
+      - name: Type check
+        run: uv run pyright
+      - name: Test
+        run: uv run pytest -v --cov=nhl_bigquery --cov-report=xml
+      - name: Build
+        run: uv run python -m build

nhl_bigquery-0.1.0/.gitignore

@@ -0,0 +1,16 @@
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+dist/
+*.egg-info/
+.pytest_cache/
+.ruff_cache/
+.coverage
+htmlcov/
+.venv/
+venv/
+.env
+*.duckdb

nhl_bigquery-0.1.0/.python-version

@@ -0,0 +1 @@
+3.13

nhl_bigquery-0.1.0/CHANGELOG.md

@@ -0,0 +1,12 @@
+# Changelog
+
+## 0.1.0
+
+Initial release.
+
+- `sync` command writes six tables in lockstep chunks: nhl_plays (with on-ice arrays merged from shift-charts), games, game_officials, boxscore_stats, shifts, standings
+- `--resume` support via `_nhl_ingest_runs` log
+- 5 doc renderers (bq-apply / llm / dictionary / markdown / dbt) backed by ColumnSpec SSoT
+- `verify --source internal` runs 8 zero-tolerance consistency checks
+- `verify --source nhl-api` reconstructs and compares team-season / player-season / game-boxscore against authoritative endpoints
+- Backfill from 2010-11 onward (coordinate era)

nhl_bigquery-0.1.0/CONTRIBUTING.md

@@ -0,0 +1,12 @@
+# Contributing
+
+Bug reports and PRs welcome at https://github.com/blahovec-labs/nhl-bigquery.
+
+## Dev setup
+
+    pip install -e ".[dev]"
+    pytest -q
+
+## Capturing test fixtures
+
+    python scripts/capture_fixture.py --game-id 2024020001 --out tests/fixtures/games/

nhl_bigquery-0.1.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Jason Blahovec
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

nhl_bigquery-0.1.0/PKG-INFO

@@ -0,0 +1,92 @@
+Metadata-Version: 2.4
+Name: nhl-bigquery
+Version: 0.1.0
+Summary: NHL play-by-play → BigQuery: idempotent ingestion + LLM-friendly docs + verification
+Project-URL: Homepage, https://github.com/blahovec-labs/nhl-bigquery
+Project-URL: Issues, https://github.com/blahovec-labs/nhl-bigquery/issues
+Project-URL: Changelog, https://github.com/blahovec-labs/nhl-bigquery/blob/main/CHANGELOG.md
+Author: Jason Blahovec
+License: MIT
+License-File: LICENSE
+Keywords: bigquery,data-engineering,hockey,nhl,play-by-play
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: MIT License
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Database
+Classifier: Topic :: Scientific/Engineering
+Requires-Python: >=3.11
+Requires-Dist: db-dtypes<2.0,>=1.0
+Requires-Dist: google-cloud-bigquery<4.0,>=3.20
+Requires-Dist: pandas<3.0,>=2.0
+Requires-Dist: pyarrow<19.0,>=15.0
+Requires-Dist: requests<3.0,>=2.31
+Provides-Extra: dev
+Requires-Dist: build>=1.2.0; extra == 'dev'
+Requires-Dist: pyright>=1.1.380; extra == 'dev'
+Requires-Dist: pytest-cov>=5.0; extra == 'dev'
+Requires-Dist: pytest>=8.0; extra == 'dev'
+Requires-Dist: responses>=0.25; extra == 'dev'
+Requires-Dist: ruff>=0.6.0; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# nhl-bigquery
+
+Idempotent NHL play-by-play (with on-ice arrays merged from shift-charts) → BigQuery ingestion, with first-class documentation for SQL/LLM agents and verification against the NHL public API.
+
+## Install
+
+    pip install nhl-bigquery
+
+## Quickstart
+
+    gcloud auth application-default login
+    nhl-bigquery sync \
+        --start 2024-10-01 --end 2025-06-30 \
+        --plays-table myproject.mydataset.nhl_plays
+
+This writes six tables to `myproject.mydataset.*`:
+
+- `nhl_plays` — one row per event, with `home_on_ice_ids` / `away_on_ice_ids` arrays
+- `games` — schedule dimension
+- `game_officials` — referees + linesmen per game
+- `boxscore_stats` — per-player per-game stats
+- `shifts` — per-shift per-player intervals
+- `standings` — daily team-standings snapshots
+
+## Backfill
+
+Backfill 15 seasons in resumable monthly chunks:
+
+    nhl-bigquery sync \
+        --start 2010-10-01 --end 2026-05-11 \
+        --chunk-by month --resume \
+        --plays-table myproject.mydataset.nhl_plays
+
+`--resume` skips chunks already recorded as `success` or `empty` in
+`<dataset>._nhl_ingest_runs`. Re-running with the same `--chunk-by` is
+safe; switching between runs will re-process (chunks must match exactly).
+
+## Documentation
+
+    nhl-bigquery docs --format llm > NHL_FOR_LLMS.md
+    nhl-bigquery docs --format bq-apply --table myproject.mydataset.nhl_plays
+
+Five formats: `bq-apply` (push descriptions to BigQuery), `llm` (one
+Markdown file packing every column for LLM context), `dictionary`
+(JSON rows for a data dictionary table), `markdown` (human reference),
+and `dbt` (dbt YAML schema stub).
+
+## Verification
+
+    nhl-bigquery verify --source internal \
+        --aggregation internal-consistency \
+        --table myproject.mydataset.nhl_plays
+
+    nhl-bigquery verify --source nhl-api \
+        --aggregation team-season --metric all --season 2024 \
+        --table myproject.mydataset.nhl_plays
+
+MIT licensed. This software does not include or distribute NHL data.

nhl_bigquery-0.1.0/README.md

@@ -0,0 +1,58 @@
+# nhl-bigquery
+
+Idempotent NHL play-by-play (with on-ice arrays merged from shift-charts) → BigQuery ingestion, with first-class documentation for SQL/LLM agents and verification against the NHL public API.
+
+## Install
+
+    pip install nhl-bigquery
+
+## Quickstart
+
+    gcloud auth application-default login
+    nhl-bigquery sync \
+        --start 2024-10-01 --end 2025-06-30 \
+        --plays-table myproject.mydataset.nhl_plays
+
+This writes six tables to `myproject.mydataset.*`:
+
+- `nhl_plays` — one row per event, with `home_on_ice_ids` / `away_on_ice_ids` arrays
+- `games` — schedule dimension
+- `game_officials` — referees + linesmen per game
+- `boxscore_stats` — per-player per-game stats
+- `shifts` — per-shift per-player intervals
+- `standings` — daily team-standings snapshots
+
+## Backfill
+
+Backfill 15 seasons in resumable monthly chunks:
+
+    nhl-bigquery sync \
+        --start 2010-10-01 --end 2026-05-11 \
+        --chunk-by month --resume \
+        --plays-table myproject.mydataset.nhl_plays
+
+`--resume` skips chunks already recorded as `success` or `empty` in
+`<dataset>._nhl_ingest_runs`. Re-running with the same `--chunk-by` is
+safe; switching between runs will re-process (chunks must match exactly).
+
+## Documentation
+
+    nhl-bigquery docs --format llm > NHL_FOR_LLMS.md
+    nhl-bigquery docs --format bq-apply --table myproject.mydataset.nhl_plays
+
+Five formats: `bq-apply` (push descriptions to BigQuery), `llm` (one
+Markdown file packing every column for LLM context), `dictionary`
+(JSON rows for a data dictionary table), `markdown` (human reference),
+and `dbt` (dbt YAML schema stub).
+
+## Verification
+
+    nhl-bigquery verify --source internal \
+        --aggregation internal-consistency \
+        --table myproject.mydataset.nhl_plays
+
+    nhl-bigquery verify --source nhl-api \
+        --aggregation team-season --metric all --season 2024 \
+        --table myproject.mydataset.nhl_plays
+
+MIT licensed. This software does not include or distribute NHL data.

nhl_bigquery-0.1.0/nhl_bigquery/__init__.py

@@ -0,0 +1,5 @@
+"""nhl-bigquery: NHL play-by-play → BigQuery ingestion."""
+
+from nhl_bigquery._version import __version__
+
+__all__ = ["__version__"]

nhl_bigquery-0.1.0/nhl_bigquery/_version.py

@@ -0,0 +1 @@
+__version__ = "0.1.0"

nhl_bigquery-0.1.0/nhl_bigquery/boxscore/__init__.py

@@ -0,0 +1 @@
+"""boxscore_stats table: per-player per-game stats."""

nhl_bigquery-0.1.0/nhl_bigquery/boxscore/schema.py

@@ -0,0 +1,146 @@
+"""BOXSCORE_SCHEMA: per-player per-game stats (skaters + goalies)."""
+
+from __future__ import annotations
+
+from nhl_bigquery.games.schema import _col
+from nhl_bigquery.schema import ColumnSpec, PartitioningSpec
+
+
+def get_partitioning() -> PartitioningSpec:
+    return PartitioningSpec(
+        field="game_date", type="DAY", clustering=["player_id", "team_id"],
+    )
+
+
+BOXSCORE_SCHEMA: list[ColumnSpec] = [
+    # Common
+    _col("game_id", "INT64", "REQUIRED", "NHL game ID.",
+         "FK to games.game_id.", tags=["identifier", "join_key"], example=2024020001,
+         api_eq="id", api_src="id"),
+    _col("game_date", "DATE", "REQUIRED", "Game date.",
+         "Partition key for this table.", tags=["temporal"], example="2024-10-08",
+         api_eq="gameDate", api_src="gameDate"),
+    _col("player_id", "INT64", "REQUIRED", "NHL Player ID.",
+         "Canonical player identifier; FK to player landing API.",
+         tags=["identifier", "join_key"], example=8478402,
+         api_eq="playerId", api_src="playerId"),
+    _col("team_id", "INT64", "REQUIRED", "Team ID this game.",
+         "NHL team_id of the team the player suited up for. "
+         "Mid-season trades produce different team_ids across games.",
+         tags=["identifier", "join_key", "team"], example=10,
+         api_eq="(side teamId)", api_src="(side teamId)"),
+    _col("player_position_category", "STRING", "REQUIRED",
+         "Discriminator: 'skater' or 'goalie'.",
+         "Splits skater rows (with G/A/+-/PIM/...) from goalie rows (SA/SV/SV%/GAA/...).",
+         tags=["identifier"], values=["skater", "goalie"], example="skater",
+         api_eq="(derived: 'forwards'/'defense'/'goalies' bucket)",
+         api_src="(derived)"),
+    _col("position_code", "STRING", "NULLABLE", "Position code (C/L/R/D/G).",
+         "Single-letter position: C=center, L=left wing, R=right wing, D=defense, G=goalie.",
+         tags=["identifier"], values=["C", "L", "R", "D", "G"], example="C",
+         api_eq="position", api_src="position"),
+    _col("sweater_number", "INT64", "NULLABLE", "Jersey number worn this game.",
+         "Sweater number; not stable across trades.",
+         tags=["identifier"], range_=(1.0, 99.0), example=97,
+         api_eq="sweaterNumber", api_src="sweaterNumber"),
+    _col("toi", "STRING", "NULLABLE", "Time on ice (MM:SS).",
+         "Total time on ice. String MM:SS to preserve fidelity; use toi_seconds for arithmetic.",
+         tags=["measure"], example="20:34",
+         api_eq="toi", api_src="toi"),
+    # Skater-only
+    _col("goals", "INT64", "NULLABLE", "Goals scored.",
+         "Goals scored by this skater in this game.", tags=["measure"], example=1,
+         api_eq="goals", api_src="goals"),
+    _col("assists", "INT64", "NULLABLE", "Assists.",
+         "Assists in this game.", tags=["measure"], example=2,
+         api_eq="assists", api_src="assists"),
+    _col("points", "INT64", "NULLABLE", "Points (G + A).",
+         "Goals + assists. Stored directly to avoid query-time arithmetic.",
+         tags=["measure"], example=3, api_eq="points", api_src="points"),
+    _col("plus_minus", "INT64", "NULLABLE", "Plus/minus.",
+         "Plus/minus for this skater in this game.", tags=["measure"], example=2,
+         api_eq="plusMinus", api_src="plusMinus"),
+    _col("pim", "INT64", "NULLABLE", "Penalty minutes.",
+         "Penalty minutes assessed to this player.", tags=["measure"], example=0,
+         api_eq="pim", api_src="pim"),
+    _col("hits", "INT64", "NULLABLE", "Hits delivered.", "Body checks delivered.",
+         tags=["measure"], example=3, api_eq="hits", api_src="hits"),
+    _col("blocked_shots", "INT64", "NULLABLE", "Shots blocked.",
+         "Shots the player blocked.", tags=["measure"], example=1,
+         api_eq="blockedShots", api_src="blockedShots"),
+    _col("shots", "INT64", "NULLABLE", "Shots on goal.",
+         "Shots on goal (SOG); does not include missed or blocked shots.",
+         tags=["measure"], example=4, api_eq="shots", api_src="shots"),
+    _col("faceoffs", "INT64", "NULLABLE", "Faceoffs taken.",
+         "Total faceoffs taken (won + lost).", tags=["measure"], example=10,
+         api_eq="faceoffs", api_src="faceoffs"),
+    _col("faceoff_winning_pctg", "FLOAT64", "NULLABLE", "Faceoff win %.",
+         "Faceoff winning percentage as a decimal (0.0-1.0).",
+         tags=["measure"], range_=(0.0, 1.0), example=0.55,
+         api_eq="faceoffWinningPctg", api_src="faceoffWinningPctg"),
+    _col("power_play_goals", "INT64", "NULLABLE", "Power-play goals.",
+         "Goals scored while on the power play.", tags=["measure"], example=1,
+         api_eq="powerPlayGoals", api_src="powerPlayGoals"),
+    _col("power_play_points", "INT64", "NULLABLE", "Power-play points.",
+         "Goals + assists earned while on the power play.",
+         tags=["measure"], example=2, api_eq="powerPlayPoints", api_src="powerPlayPoints"),
+    _col("shorthanded_goals", "INT64", "NULLABLE", "Shorthanded goals.",
+         "Goals scored while shorthanded.", tags=["measure"], example=0,
+         api_eq="shorthandedGoals", api_src="shorthandedGoals"),
+    _col("sh_points", "INT64", "NULLABLE", "Shorthanded points.",
+         "Goals + assists earned while shorthanded.", tags=["measure"], example=0,
+         api_eq="shPoints", api_src="shPoints"),
+    _col("power_play_toi", "STRING", "NULLABLE", "Power-play TOI.",
+         "Time on ice while on the power play (MM:SS).",
+         tags=["measure"], example="2:14",
+         api_eq="powerPlayToi", api_src="powerPlayToi"),
+    _col("shorthanded_toi", "STRING", "NULLABLE", "Shorthanded TOI.",
+         "Time on ice while shorthanded (MM:SS).",
+         tags=["measure"], example="1:32",
+         api_eq="shorthandedToi", api_src="shorthandedToi"),
+    # Goalie-only
+    _col("shots_against", "INT64", "NULLABLE", "Total shots against.",
+         "Shots faced by this goalie in this game.", tags=["measure"], example=32,
+         api_eq="saveShotsAgainst (sum)", api_src="saveShotsAgainst"),
+    _col("saves", "INT64", "NULLABLE", "Total saves.",
+         "Saves made. Equals shots_against - goals_against.",
+         tags=["measure"], example=30,
+         api_eq="(saveShotsAgainst - goalsAgainst)", api_src="(derived)"),
+    _col("save_pctg", "FLOAT64", "NULLABLE", "Save percentage.",
+         "Save percentage as decimal (0.0-1.0).",
+         tags=["measure"], range_=(0.0, 1.0), example=0.938,
+         api_eq="savePctg", api_src="savePctg"),
+    _col("goals_against", "INT64", "NULLABLE", "Goals against.",
+         "Goals scored against this goalie in this game.",
+         tags=["measure"], example=2,
+         api_eq="goalsAgainst", api_src="goalsAgainst"),
+    _col("even_strength_shots_against", "INT64", "NULLABLE",
+         "Even-strength shots against.", "Shots faced at even strength.",
+         tags=["measure"], example=20,
+         api_eq="evenStrengthShotsAgainst", api_src="evenStrengthShotsAgainst"),
+    _col("power_play_shots_against", "INT64", "NULLABLE",
+         "PP shots against.", "Shots faced while shorthanded.",
+         tags=["measure"], example=8,
+         api_eq="powerPlayShotsAgainst", api_src="powerPlayShotsAgainst"),
+    _col("shorthanded_shots_against", "INT64", "NULLABLE",
+         "SH shots against.", "Shots faced while on the power play.",
+         tags=["measure"], example=4,
+         api_eq="shorthandedShotsAgainst", api_src="shorthandedShotsAgainst"),
+    _col("even_strength_goals_against", "INT64", "NULLABLE",
+         "EV goals against.", "Goals allowed at even strength.",
+         tags=["measure"], example=1,
+         api_eq="evenStrengthGoalsAgainst", api_src="evenStrengthGoalsAgainst"),
+    _col("power_play_goals_against", "INT64", "NULLABLE",
+         "PP goals against.", "Goals allowed while shorthanded.",
+         tags=["measure"], example=1,
+         api_eq="powerPlayGoalsAgainst", api_src="powerPlayGoalsAgainst"),
+    _col("shorthanded_goals_against", "INT64", "NULLABLE",
+         "SH goals against.", "Goals allowed while on the power play.",
+         tags=["measure"], example=0,
+         api_eq="shorthandedGoalsAgainst", api_src="shorthandedGoalsAgainst"),
+    _col("ingested_at", "TIMESTAMP", "REQUIRED",
+         "Ingestion timestamp.",
+         "UTC timestamp when this row was written.",
+         tags=["meta"], example="2026-05-22T17:00:00Z",
+         api_eq=None, api_src="(set by ingestion)"),
+]

nhl_bigquery-0.1.0/nhl_bigquery/boxscore/transform.py

@@ -0,0 +1,93 @@
+"""Boxscore transform: /gamecenter/{id}/boxscore → per-player rows (skater + goalie)."""
+
+from __future__ import annotations
+
+from datetime import UTC, datetime
+from typing import Any
+
+import pandas as pd
+
+_SKATER_FIELDS = (
+    "goals", "assists", "points", "plusMinus", "pim", "hits", "blockedShots",
+    "shots", "faceoffs", "faceoffWinningPctg",
+    "powerPlayGoals", "powerPlayPoints", "shorthandedGoals", "shPoints",
+    "powerPlayToi", "shorthandedToi",
+)
+
+_GOALIE_FIELDS = (
+    "saveShotsAgainst", "goalsAgainst", "savePctg",
+    "evenStrengthShotsAgainst", "powerPlayShotsAgainst", "shorthandedShotsAgainst",
+    "evenStrengthGoalsAgainst", "powerPlayGoalsAgainst", "shorthandedGoalsAgainst",
+)
+
+
+def _camel_to_snake(s: str) -> str:
+    out = []
+    for c in s:
+        if c.isupper():
+            out.append("_" + c.lower())
+        else:
+            out.append(c)
+    return "".join(out).lstrip("_")
+
+
+def transform_boxscore_to_df(bs: dict[str, Any]) -> pd.DataFrame:
+    """Extract per-player per-game rows from a boxscore response."""
+    game_id = int(bs["id"])
+    game_date = bs.get("gameDate")
+    pgs = bs.get("playerByGameStats") or {}
+    rows: list[dict[str, Any]] = []
+    ingested_at = datetime.now(UTC)
+
+    for side in ("awayTeam", "homeTeam"):
+        team_id = int((bs.get(side) or {}).get("id"))
+        side_block = pgs.get(side) or {}
+        for cat, players in (
+            ("skater", side_block.get("forwards") or []),
+            ("skater", side_block.get("defense") or []),
+            ("goalie", side_block.get("goalies") or []),
+        ):
+            for p in players:
+                row: dict[str, Any] = {
+                    "game_id": game_id,
+                    "game_date": game_date,
+                    "player_id": int(p["playerId"]),
+                    "team_id": team_id,
+                    "player_position_category": cat,
+                    "position_code": p.get("position"),
+                    "sweater_number": p.get("sweaterNumber"),
+                    "toi": p.get("toi"),
+                    "ingested_at": ingested_at,
+                }
+                # Skater fields
+                for f in _SKATER_FIELDS:
+                    row[_camel_to_snake(f)] = p.get(f)
+                # Goalie fields (with renames)
+                # saveShotsAgainst may be "saves/shots" string (e.g. "19/22") or numeric
+                save_shots_raw = p.get("saveShotsAgainst")
+                goals_against = p.get("goalsAgainst")
+                shots_against_int = p.get("shotsAgainst")
+                saves_int = p.get("saves")
+                # Parse "saves/shots" string format if needed
+                if isinstance(save_shots_raw, str) and "/" in save_shots_raw:
+                    parts = save_shots_raw.split("/")
+                    try:
+                        saves_int = saves_int if saves_int is not None else int(parts[0])
+                        shots_against_int = (
+                            shots_against_int if shots_against_int is not None else int(parts[1])
+                        )
+                    except (ValueError, IndexError):
+                        pass
+                elif isinstance(save_shots_raw, (int, float)):
+                    shots_against_int = (
+                        shots_against_int if shots_against_int is not None else int(save_shots_raw)
+                    )
+                row["shots_against"] = shots_against_int
+                row["saves"] = saves_int
+                row["save_pctg"] = p.get("savePctg")
+                row["goals_against"] = goals_against
+                for f in _GOALIE_FIELDS[3:]:  # skip the three above
+                    row[_camel_to_snake(f)] = p.get(f)
+                rows.append(row)
+
+    return pd.DataFrame(rows)