silly-kicks/xshot-occurrence-v1

xShotOccurrence v1 (sc_extended) — Shot-Occurrence Propensity from Tracking State

Read this first. This repo serves the sc_extended variant, which is NOT bundled with the silly-kicks wheel — it is trained on restricted owner-tier data and cannot be redistributed there. That is a licensing constraint, not a quality one: sc_extended cleared its pre-registered paired test. Pick the variant that matches your situation — see Which variant should I use?.

Model Description

XShotOccurrenceModel is a deterministic-XGBoost classifier estimating P(a shot is attempted by the in-possession team within ~1 s of a tracking frame) — the xS surface of the GKDV research arc (TF-16).

Paper-faithful 27-feature extractor in goal-relative coordinates via the shared _geometry helper: ball r/θ/z/speed, an openGoal goal-mouth obstruction term, GK distance/bearing, and the 5 nearest defenders + 5 nearest attackers.

• Domain filter: alive-ball, attacking third
• Trained on 2,024,373 rows / 366,834 positives (18.9% positive rate)
• prepare_xshot_training_data always returns the faithful class distribution; negative subsampling lives in a separate train-only helper.

Which variant should I use?

Variant	Corpus	Where it lives	Use it?
default (public)	17 matches — SkillCorner + IDSSE, redistributable	bundled in the wheel	Default choice — offline, fully reproducible, no restricted data
sc_extended	+ 98 owner-tier SkillCorner matches (179 total)	this repo (HF-only)	Yes, if you can accept a Hub download and the corpus caveats below — it beat public on held-out folds

sc_extended is HF-only because it is trained on restricted data and cannot be redistributed inside the PyPI wheel. Only learned parameters are published here — no raw provider tracking data, and no artifact from which raw rows can be reconstructed (every booster leaf aggregates ≥ 9 samples by the binding min_child_weight floor).

Why this variant is HF-only (it did NOT fail its paired test)

sc_extended CLEARED the pre-registered paired test. It is HF-only because it is trained on restricted owner-tier data and cannot be redistributed inside the PyPI wheel (ADR-038) — a licensing constraint, not a quality verdict.

The Stage-B run's own record:

"shipped": "sc_extended",
"why": "sc_extended clears; full does not dominate it -- ties go to less data"

Stage-B deltas vs public, held out on public folds: [-0.008, +0.009, +0.005, +0.029, +0.023] — positive in 4 of 5 folds, mean +0.0117. The registered rule (scripts/_paired.py::clears_rule: positive in ≥ K−1 of K folds and a positive mean) returns True.

Do not misread the BUNDLED artifact's paired block

The bundled default artifact carries a different run's record — Stage A, whose corpus contained no owner-tier SkillCorner rows at all. There, cand_masks["sc_extended"] = is_public | is_sc_private with is_sc_private empty, so the sc_extended candidate was the same mask as public: the same model scored against itself. Its deltas are [0.0, 0.0, 0.0, 0.0, 0.0] by construction — a degenerate self-comparison, not a measured null — and since the rule requires strictly positive values, zero fails and the fixed sequence stops. That is why public is the bundled default.

Reading those zeros as "sc_extended failed" is a trap; an earlier revision of this card fell into it. docs/research/tf19_pr2/decision_table.md has the Stage A / Stage B split correct.

The separate 4.9.0 verdict — that owner-tier Gradient Sports data degraded public held-out PR-AUC in all 5 folds — concerned the full arm, not sc_extended.

Held-out CV (5 folds, out-of-fold)

Metric	Value	Baseline
PR-AUC	0.5968 (± 0.0133)	base rate 0.1888
Brier	0.1110	base-rate Brier 0.1532
Log loss	0.3613	—

All four acceptance gates pass. Estimates are CV, not the shipped fit.

TF-19 status: the shot arm has never been measured

Unlike its xCross sibling, this model carries no GK-substitution probe, no GK-block ablation and no permutation importance in metrics.json. That is blocked, not missing:

• The registered xS probe rule and its locked constants shipped in silly-kicks 4.47.0 (tracking/_model_eval.py::evaluate_xs_probe, PROBE_WRAPPERS["xs"]).
• xs_substitution_probe consumes ghost-substituted targets produced by the silly_kicks/gkdv/ engine, which is ADR-037's PR-3 and does not exist yet.

A TF-19 spec must not assume the shot arm is healthy. It is unmeasured, not validated.

Usage

from silly_kicks.tracking import XShotOccurrenceModel

model = XShotOccurrenceModel.from_variant("default")       # recommended, bundled, offline
model = XShotOccurrenceModel.from_variant("sc_extended")   # this repo, downloads from the Hub

Requires pip install silly-kicks[xshot] and silly-kicks ≥ 4.51.0 (earlier versions have no sc_extended routing).

Integrity and load-time guards

load() is fail-closed on two independent checks:

1. SHA256SUMS verified before anything is parsed.
2. Chirality fingerprint (ADR-040) — the model re-runs its own outputs on a fixed y-asymmetric probe frame and compares to the recorded fingerprint. It raises on a mismatch and on a missing one, because every pre-PR-2 artifact belongs to the y-mirrored mis-served class. This enforcement is what caught the xgboost 3.x base_score serialization bug, which 2.x silently drops to 0.5.

Limitations

• Not the bundled model (restricted corpus, so it cannot ship in the wheel). This is a redistribution limit, not a performance one.
• No GK measurement exists for this arm at all — see the TF-19 section.
• Trained on a corpus that is 179 matches, heavily Real Madrid — the 98 owner-tier additions are one club, so club/style confounding is real and unquantified here.
• SkillCorner keepers are detected in only ~19.6% of frames (~80% interpolated), which is why GKDV measurement is registered to Gradient Sports frames only (ADR-038 §5).
• Estimates are cross-validated, not a held-out test of the shipped fit.

References

See the NOTICE file in the silly-kicks repository for full bibliographic citations.

• Attribution: arXiv:2512.00203.
• Decisions: ADR-011 (trained-model lifecycle), ADR-037 (TF-19 re-gate), ADR-038 (corpus + visibility), ADR-040 (chirality enforcement).

Model Files

File	Purpose
model.json	XGBoost booster (pickle-free)
metadata.json	features, hyperparameters, chirality fingerprint, provenance
metrics.json	CV metrics and acceptance record
SHA256SUMS	integrity manifest, verified by load()

More Information

https://github.com/karsten-s-nielsen/silly-kicks