PARTIAL / HOLD — NO CERTIFIED CORE RELEASE
This dated addendum corrects two material research gaps. It does not rewrite original pre-match observations, change the audited previous-day outcomes or issue a new recommendation.
01 · Arsenal vs OL Lyonnes — missing adverse models
The morning Top 5 cited external Lyon win probabilities of 65–70%. Additional models, checked on 19 November, gave only 43% (Oddbert) and 55% (Tippwise). This wider disagreement matters. These are external estimates, not calibrated AIONORA probabilities.
| Model assumption | Illustrative EV at comparison price 1.86 |
|---|---|
| p = 43% | −20.0% |
| p = 55% | +2.3% before uncertainty haircut |
| p = 52% (−3pp stress) | −3.3% |
There is no robust verified positive edge and no authenticated same-book executable Greek price. OL Lyonnes selection stays HOLD.
Sources: Oddbert and Tippwise.02 · Bayern basketball vs Baskonia — record correction
The initial Top 5 described Bayern's EuroLeague record as 3–8. Correct as of the pre-game 19 November report: 3–9, including a 0–4 EuroLeague home record. Bayern also lost 114–117 after overtime at Žalgiris on 17 November; the travel and shorter rest are adverse factors.
Sources: Bundesliga 18 Nov preview and FC Bayern official preview.03 · CORE Doubler and Bet Builder remain HOLD
The original illustrative Valencia × OL Lyonnes pair was 1.22 × 1.86 = 2.269. Under low external win probabilities (71% and 43%), simple mathematical independence would yield only 30.5% combined probability and about −30.7% nominal EV. This is not a calibrated joint model and cannot justify release. The separate same-game Bet Builder over 4.00 also lacks verified combined quote and correlation modelling.
04 · Evidence verdict
PARTIAL / HOLD, unchanged. The original public edition documented 19 event identities discovered and screened, five partly checked and none fully signed off. Full last-five/last-ten opponent data, confirmed lineups, Greek executable prices, certified calibration and all release gates remain incomplete. No validation score or positive release is assigned.
Later editorial source-quality correction; retain the immutable morning archive and the previous-day WIN/LOSS record. AIONORA Research · Independent sports research.