How yesterday’s picks performed
Four of the five published 2 November research selections resolved as forecast. These are verified event results, not a certified betting track record or proof of calibrated forecast accuracy.
| 2 November research market | Final score | Outcome |
|---|---|---|
| Sassuolo–Fiorentina · Over 1.5 match goals | 2–4 | WIN |
| Atalanta–Parma · Over 1.5 match goals | 0–3 | WIN |
| Benfica–Alverca · Benfica win | 1–0 | WIN |
| Atromitos–Volos · Over 1.5 match goals | 3–0 | WIN |
| Newcastle–Everton · Over 1.5 match goals | 0–1 | LOSS |
Outcome evidence: match results · Liga Portugal · Atromitos–Volos.
The Learning Loop
Every outcome is evidence. Every insight informs what we test next.
What Held Up
Four predictions settled correctly. All five remain in the transparent audit, including the loss. Successful outcomes do not erase negative expected values from their original prices.
Where the Model Fell Short
Newcastle–Everton finished 0–1, invalidating the over-1.5-goals forecast. One day is too small a sample to establish skill. Last-minute market and team news were not fully checked.
What We Refine
No model-weight change based on a single result. We stress-test probabilities, use lower estimates when models disagree and keep uncertified selections separate from BET approvals.
Which sports did we check?
Football
Tuesday's nine UEFA Champions League fixtures checked, including Barcelona–Aston Villa and LASK–Slovan Bratislava. Screening is not proof of exhaustive global coverage.
Basketball
Five EuroCup matchups examined, including Tofaş–Neptūnas. Rotation and player-availability validation remain incomplete.
Tennis
ATP Paris Masters order of play checked, including Rublev–Etcheverry, Davidovich Fokina–Kouamé and Djokovic–Vacherot. Start times may change.
Today’s CORE selection board
Rublev–Etcheverry · Over 19.5 total games
YELLOW · RESEARCH
77.8% outside model estimate. Indicative best-quoted 1.32; break-even 75.76%; nominal EV +2.70% but −1.26% if the model overestimates by 3 percentage points.
What could go wrong? Fast straight-set victory; 1.32 not available; an average price of 1.22 would not offer value.
Barcelona–Aston Villa · Barcelona under 3.5 team goals
YELLOW · RESEARCH
66.1% simple Poisson estimate derived from Barcelona expected goals λ=2.94; indicative odds 1.60, break-even 62.50%; nominal EV +5.71%, −3pp stress EV +0.91%.
What could go wrong? Strong Barcelona attack, high-score tail and disagreement between independent xG sources. Below the 70% high-probability threshold; not a certified model estimate.
Rublev over 19.5 total games × Barcelona under 3.5 team goals
Two different matches across two sports. Quoted individually at 1.32 and 1.60, resulting in 2.112 indicative combined odds. They were not confirmed together at one bookmaker.
*Assumes independence and uncalibrated external probabilities. A small sensitivity haircut removes the positive EV. This is neither a bookable bet slip nor a certified CORE Doubler.
Today’s Top 5 — estimated chances
Ranked by research priority and potential value, not just favourite status. “Forced” means publishing a transparent selection board, not recommending five bets.
#1 Rublev–Etcheverry — Over 19.5 games
Model 77.8%; indicative odds 1.32; break-even 75.76%. The edge disappears under a 3pp probability haircut; fast match risk. YELLOW
Numbers and model comparison
Pythia 77.8%; EV +2.70%; EV after −3pp −1.26%. Average listed quote 1.22 would mean negative EV.
#2 Barcelona–Aston Villa — Barcelona under 3.5 team goals
Simple uncalibrated Poisson 66.1%, example odds 1.60; break-even 62.5%. Barcelona's form and high-score tail challenge the model. YELLOW
Numbers and model comparison
LOAD90 Barcelona xG 2.94; P(goals ≤3) 66.07%; EV +5.71%; −3pp EV +0.91%; alternative xG estimates diverge.
#3 LASK–Slovan Bratislava — LASK match winner
Outside model 68%; example odds 1.47; break-even 68.03%. Strong favourite, but negligible or negative value. SAYS NO
Numbers and model comparison
Nominal EV −0.04%; −3pp stress EV −4.45%. Full-squads scenario drops the model to roughly 60%.
#4 Davidovich Fokina–Kouamé — Davidovich Fokina wins
Conservative external estimate 68.3% versus optimistic 76.2%; at price 1.37, the lower estimate is not value. SAYS NO
Numbers and model comparison
Pythia 68.3%, Dimers 76.2%; break-even 72.99%; nominal EV −6.43%; −3pp EV −10.54%.
#5 Tofaş–Neptūnas — Tofaş match winner
Outside model 78%, example price 1.19 requires 84.03%. Popular favourite with negative implied value. SAYS NO
Numbers and model comparison
Nominal EV −7.18%; −3pp EV −10.75%. Player availability and rotations not fully independently checked.
Detailed Top 5 value comparison
| Rank | Exact market | External / derived P | Odds | Break-even | EV | −3pp EV |
|---|---|---|---|---|---|---|
| 1 | Rublev–EtcheverryOver 19.5 games | 77.8% | 1.32 | 75.76% | +2.70% | −1.26% |
| 2 | Barcelona–VillaBarcelona under 3.5 goals | 66.1% | 1.60 | 62.50% | +5.71% | +0.91% |
| 3 | LASK–SlovanLASK wins | 68.0% | 1.47 | 68.03% | −0.04% | −4.45% |
| 4 | Fokina–KouaméFokina wins | 68.3% | 1.37 | 72.99% | −6.43% | −10.54% |
| 5 | Tofaş–NeptūnasTofaş wins | 78.0% | 1.19 | 84.03% | −7.18% | −10.75% |
Illustrative prices only. EV=(probability × decimal odds)−1. The −3pp check is a sensitivity illustration, not a measured confidence bound. These are external estimates, not calibrated AIONORA probabilities. Source transparency: LASK model · Fokina model · Fokina alternative · Tofaş model/market.
Two-game combination and Bet Builder
Rublev total games × Barcelona team goals
Indicative combined odds 2.112; point-estimate joint p 51.40%; break-even 47.35%. A mild −3pp stress reduces the EV to −0.37%. Prices come from different providers. Decision: Research only.
Barcelona–Aston Villa · three conditions
- Barcelona win and both teams score
- Lamine Yamal scores
- Nicolas Jackson 1+ shot on target
Third-party published combined price 6.00; break-even 16.67%. No defensible correlated joint model probability; not an approved bet.
AIONORA SAYS NO
Some opportunities look attractive. Our evidence says otherwise.
LASK – Slovan Bratislava · LASK to win
NO BETEstimated probability: 68.0% external model
Reference odds: 1.47 · Break-even: 68.03%
EV indicator: −0.04%Why we pass: The cautious estimate barely covers break-even; a more adverse squad scenario lowers the assessed chance. Risk: YELLOW — Price and model uncertainty.
Davidovich Fokina – Kouamé · Fokina to win
NO BETEstimated probability: 68.3% vs 76.2% external models
Reference odds: 1.37 · Break-even: 72.99%
EV indicator: −6.43% using lower modelWhy we pass: Two model estimates differ by 7.9 percentage points; the conservative estimate cannot justify the market price. Risk: YELLOW — Model disagreement.
Tofaş – Neptūnas · Tofaş to win
NO BETEstimated probability: 78.0% external model
Reference odds: 1.19 · Break-even: 84.03%
EV indicator: −7.18%Why we pass: A strong favourite can be poor value. Rotation and availability checks are incomplete. Risk: YELLOW — Value and team-news uncertainty.
Rublev totals × Barcelona team goals · CORE Doubler
NO BETEstimated probability: 51.40% independence estimate*
Reference odds: 2.112 combined (indicative) · Break-even: 47.35%
EV indicator: −0.37% after −3 pp per legWhy we pass: The price is unverified at one bookmaker and the advantage disappears with a small probability stress. Risk: YELLOW — Price and robustness.
Barcelona – Aston Villa · Three-leg CORE Bet Builder
NO BETEstimated probability: Calibrated joint probability unavailable
Reference odds: 6.00 (published example) · Break-even: 16.67%
EV indicator: Not establishedWhy we pass: Barcelona win plus BTTS, Yamal goal and Jackson shot on target lack a validated correlated joint estimate. Risk: YELLOW — Correlation and lineups.
*Independence-based research calculation, not a calibrated AIONORA model probability. Rejected markets can remain in the Top 5 for audit transparency. Prices are illustrative, not verified executable offers. No official BET release is claimed.
How confident can we be?
The five picks, Doubler and Bet Builder are published as auditable research candidates. None has a demonstrated official AIONORA 80/100 Validation Score, independently calibrated probability, verified positive conservative EV and executable operator price simultaneously.
| Quality gate | Status |
|---|---|
| Event universe | Three-sport featured fixture scan completed, not exhaustive worldwide |
| Odds & live execution | Indicative public sources only; bookmaker pairing unverified |
| Players, teams and lineups | Not fully confirmed; pre-event update necessary |
| Independent models | External values compared; important differences retained |
| Certified AIONORA model calibration | Unavailable |
| Validation Score ≥80 | N/A — cannot independently compute |
| Prospective learning history | No genuine history stored in currently connected Neon backend |
| Official BET gate | FAIL CLOSED |
Latest production research-sync status: Neon read-only health ↗. No certified track record, archive or model weights were changed for this edition. Publication is not the same as statistical validation.