The model's --use-xg path fits round(xG) as the per-match Poisson response in Dixon-Coles and Hierarchical Poisson (falling back to realised goals where xG is absent). It was structurally bounded by a tiny corpus: the JaseZiv FBref mirror that scripts/pull_intl_xg.py reads carries Opta xG for only 143 tournament matches — WC 2018 (64), Euro 2020 (51), Copa 2021 (28). WC 2022 is present-but-null there; Euro 2024 / Copa 2024 / AFCON 2023 are absent. With 143 xG-bearing matches out of ~49…
ملاحظة بحثية
Back-filling international xG from StatsBomb open data
الحالة: Production xG path WIRED (2026-05-29) — `auto-refit.yml` fits DC with `--use-xg` and refits the calibrator on the xG-enabled ensemble; HP excluded (fails the gate's ECE half). The xG-enabled artefacts (`dixon_coles.json` / `ensemble_calibrator.json` / `data.json`) regenerate on the first auto-refit after merge (not hand-committed — see "Production wiring" for why). Single-provider corpus (314 StatsBomb + 28 residual Opta Copa-2021 = 342 rows). Gate clears for DC + Ensemble on both evaluation slicesتاريخ الاختبار الرجعي: 29 May 2026الخلاصة + الملاحظة الكاملة · 1,755 كلمة
الخلاصة
الملاحظة الكاملة
Standard Pass
اقرأ الملاحظة البحثية الكاملة
Back-filling international xG from StatsBomb open data تحتوي على 1,755 كلمة. Standard Pass يفتح كل ملاحظة بحثية بالكامل، إضافة إلى التوقع الكامل وتقييمات كل منتخب وكل لاعب، صالح طوال البطولة.
Every forecast graded against the real result, scored on 987 matches since 2014. See the scorecard.
24h money-back, no questions asked·No subscription, no auto-renewal·Access through 31 Dec 2026. See refund policy.