Skip to main content
← All predictions

J1 League · Regular Season - 13

Vissel Kobe crestVissel KobevsCerezo Osaka crestCerezo Osaka

Kickoff: Wednesday, April 29, 2026 at 1:00 AM ET · Noevir Stadium, Kobe

Outcome Probabilities

Confidence 5 out of 5 confidence

Home

Vissel Kobe

69%

Draw

Draw

19%

Away

Cerezo Osaka

12%

Model pick: Vissel Kobe

Expected goals: 2.8—1.0(total 3.8)
Upset risk: 6/10

Prediction Evolution

How the model's confidence changes as new info arrives

HomeDrawAway
25%50%75%?69%??T−48hT−24hPre-matchHalf-time

One snapshot logged so far · 3 more stages will appear as the kickoff approaches

Dixon-Coles dominates the ensemble (0.81 weight) and strongly favors Kobe at home (0.771 pH), but Elo's 140-point advantage and xG both pull toward a tighter match (Elo 0.397 pH, xG 0.626 pH). Recent form shows Kobe 2W–2L over last 4 J1 games with mixed away record, while Cerezo is 1W–3L–1D in recent outings, suggesting form uncertainty on both sides. No injuries to either squad. The sharp disagreement between Dixon-Coles and Elo (37pp gap on pH) signals low conviction; small downward adjustment to pH and modest upward to pA/pD respects the baseline while acknowledging Elo's structural skeptic

SPX AI Quant Prediction Model (DC 0.81 / Elo 0.19 / xG 0) + Claude Haiku 4.5 · 55/45

T−24h stage hash · locked 4/28/2026, 1:05:38 AM

2edf14c1e2aecf832e2d01ba23177a51c5c061fe83ffc355ebb6c7a5ae719ebe

SPX AI Quant Prediction Model

DC 0.81 · Elo 0.19 · xG 0

H 70%D 18%A 12%

Dixon-Coles (Poisson)

H 77%D 13%A 10%

xG: 2.89 — 0.97

Elo (Davidson 3-way)

H 40%D 43%A 18%

Elo gap: +140 pts

xG-Poisson

H 63%D 21%A 16%

Expected goals: 1.95 — 0.87

LLM contextual review

H 68%D 19%A 13%

Based on recent form, news, injuries

Data quality score: 62/100 · verdict publish

Integrity Lock

This prediction's latest stage was generated at 2026-04-28T01:05:38.315Z and its canonical JSON hashed with SHA256. Each prior stage carries its own hash inside the corresponding tab so you can audit every step independently.

2edf14c1e2aecf832e2d01ba23177a51c5c061fe83ffc355ebb6c7a5ae719ebe
✗ Hash mismatch — investigate

Verify yourself via the API.

How the model works →