1
Match facts first
Every analysis starts with the fixture, teams, competition, kickoff time, score state, recent form, head-to-head context and available statistics.
Public methodology
How FootmeshAI explains a football match
FootmeshAI explains football matches through a three-lens methodology: structured match facts, prediction model probabilities, KG evidence, market signals, quality gates and clear uncertainty labels.
Methodology key questions
Before using the methodology, answer four questions
The methodology page turns every match read back into facts, model lens, change conditions and a reviewable workspace route.
Where do the facts come from?
Start with the three-lens method and separate match facts, model probability, KG context and market signals.
Is the model the conclusion?
No. The model is one lens and must be read with facts, KG context, market movement and quality gates.
What happens when data is missing?
Missing lineups, injuries, statistics or model output should stay missing instead of being filled with guesses.
Where do I go after reading this?
Return to the match center for a fixture, or open My FootmeshAI to continue a saved research route.
2
Model output as one lens
1X2 probabilities, goals outlook, BTTS, score ranges and risk flags are treated as model evidence, not as guaranteed outcomes or isolated recommendations.
3
KG and market context
KG Intelligence links the match to related entities while market movement, odds shifts and company split are displayed as changing context signals.
Quality gates
Explainability matters more than filling the page
Language-safe display: non-Chinese pages do not use Chinese entity names or Chinese body text as visible fallback.
Availability labels: unavailable, partial, stale, simulated and production-ready states must remain visible.
No invented facts: missing lineups, injuries, statistics or model output stay missing instead of being guessed.
Human-readable uncertainty: model probabilities are paired with change conditions and risk notes.
Reading policy
FootmeshAI does not turn probability into certainty
The product is designed to place scores, fixtures, team form, statistics, KG facts, model probabilities and market movement into one evidence matrix and research session before returning to the match workspace. Model output may change as coverage, time, lineups, markets and match state change.
Verification protocol
Verify the evidence boundary before saving a read
The methodology page is not a glossary of model terms. It gives every match read a reviewable route: where the facts came from, which lens the model represents, and what changes would make the read worth revisiting.
Step 1
Trace the facts
Start from the fixture, teams, league, time, score state, injuries, lineups and statistics that are actually available on the match page.
Step 2
Compare the lenses
Read model probabilities together with KG context, market movement and recent form instead of treating any single signal as the answer.
Step 3
Keep the uncertainty
Save the read only after coverage, freshness, missing-data notes and change conditions are clear enough to revisit later.
Next step
Bring the methodology back into your workspace
After reading the methodology, return to the match route first: choose a fixture, open the AI brief, ask, review changes, coverage, the evidence matrix and the research session, then return to the match workspace before keeping useful work in My FootmeshAI.
Match route
Choose a fixture, then continue through AI brief, Evidence review, change watchlist, coverage, evidence matrix, research session and match workspace.
Back to match centerMy FootmeshAI
Return to saved matches, reminders, recent reads and the personal AI brief.
Open My FootmeshAIAnalysis library
Continue saved model reads and evidence notes.
Open analysis librarySaved matches
Keep priority fixtures in the watchlist and return when data changes.
Open watchlist