World Cup 2026 contextual model card
Kickoff Lens separates public match facts from model context. The model is designed to explain why a fixture may tilt one way, not to guarantee a result.
Current version
wc-analytics-v2.1-agentic-202606181200 · generated from public schedule, team, squad, result, and stable venue-profile data.
Current coverage
| Layer | Current coverage | Why it matters |
|---|---|---|
| Match model context | 72/104 matches | Shows whether a fixture has explainable model inputs beyond a raw percentage. |
| Factor explanations | 72/104 matches | Lists the concrete drivers behind the estimate: team quality, attack-defense fit, venue, travel, form, and pressure. |
| Venue and weather layer | 104/104 matches · 16 venue profiles | Uses climate, altitude, roof, travel, and kickoff context before more volatile live weather updates are available. |
| Squad data | 1248 listed players | Lets the model separate team depth and key-player influence from broad ranking strength. |
| Agent-readable summaries | 0/104 matches | Turns scoring inputs into reader-facing explanations and scenario notes. |
| Finished-match calibration | 24 match sample · direction 58.3% · Brier 0.2 | Tracks whether the model is drifting and where it misses, instead of hiding uncertainty. |
| Source-backed late signals | 0 signals | Reserved for reliable public updates such as squad news, result corrections, and late context. |
Inputs used
- team radar weighted quality
- FIFA ranking context
- key-player impact
- squad experience
- recent tournament form
- attack-versus-defense matchup
- venue climate and altitude adaptation
- travel and rest-day load
- style matchup volatility
- group-stage pressure
- pre-match-only table state
- source-backed late team news
- scenario-sensitive Agent explanation
- rolling post-match calibration
Two-system operating model
The data updater and the analysis Agent are separate layers. The updater handles schedule, score, squad, source freshness, and post-match refresh windows. The analysis layer reads that package and produces context indexes, factor drivers, confidence notes, and recap language. If the updater is rate-limited, the analysis layer keeps using the latest verified package instead of inventing facts.
How the Agent scores a fixture
- Start from team baseline: FIFA ranking context, Kickoff Lens radar, squad depth, core-player impact, and current tournament record.
- Adjust for matchup shape: attack versus defense, control versus transition risk, execution quality, style volatility, and group-stage pressure.
- Apply environment context: venue climate, altitude, roof profile, travel/rest load, kickoff timing, and likely adaptation risk.
- Apply source-backed late signals only when reliable public sources expose them. Missing injury, suspension, or lineup data is shown as uncertainty, not guessed.
- After full time, compare the estimate against the actual result and keep a rolling calibration log.
Accuracy operating loop
The Agent is tuned as a reading system, not a picks service. It should become more useful by explaining uncertainty better, adding verified late context faster, and learning where the model was overconfident after finished matches.
| Signal | Current behavior | Accuracy rule |
|---|---|---|
| Team strength | Ranking, radar, squad depth, key-player impact, and tournament record. | Strong teams can still be dampened when context, travel, or matchup style opens the balance lane. |
| Venue and weather | Stable climate, altitude, roof, heat, humidity, and travel/rest proxies. | Environment can move close matches, but it should not overpower a clear quality gap without verified late weather. |
| Lineup/news | Reserved for source-backed public signals. | If injury, suspension, or expected XI data is missing, confidence drops instead of inventing an adjustment. |
| Calibration | Finished results feed a rolling Brier and direction check. | When the model becomes overconfident, stronger lanes are dampened and balance lanes stay visible. |
| Post-match learning | Scores and event timelines are compared to pre-match factors. | Wrong calls should update the explanation layer so future previews show why uncertainty mattered. |
Confidence rules
- High confidence needs a clear composite edge, complete squad context, and low volatility.
- Medium confidence means the leading team-strength lane is visible but matchup or environment signals still matter.
- Cautious confidence means the estimate should be read as a scenario map, not a strong call.
- Missing late team news, missing event timelines, or stale source windows lower confidence and are shown in the reader note.
How the Worker keeps it current
The Cloudflare Worker wakes on a cron schedule, checks the next post-match windows, overlays verified public corrections when upstream caches lag, rebuilds contextual model reads, and keeps Pages functions reading the live package first with static data as fallback.
Known limits
- Weather is currently represented by stable venue climate profiles, not minute-by-minute weather update feeds.
- Injuries, suspensions, confirmed lineups, and late rotation are not guaranteed unless public sources expose them reliably.
- Model context is informational only and must not be treated as financial, legal, or medical advice.
What a restriction changes
API or Agent limits affect freshness, not the scoring framework itself. When live calls are limited, Kickoff Lens uses the most recent verified data package, marks stale or uncertain states, and waits for the next allowed refresh. That is safer than fabricating weather, injury, or event data. Accuracy improves when reliable late signals are available, but transparency improves by showing what is known and what is missing.
Next model upgrades
- Live weather values: temperature, humidity, wind, rain, roof status, and kickoff-time changes.
- Lineup and absence layer: injuries, suspensions, expected XI, and squad rotation risk.
- Calibration log: compare pre-match estimates against finished results and track model drift.