Pre-match predictionLa LigaSun, 12 Jul

Real Sociedad vs Villarreal Prediction, Fair Odds & Result

Model answer: Home at 38%. Home is 38%, draw 31% and away 31%. These are clearly labelled sample figures until the production pipeline is connected.

My callAsk Kito
More proof tools
Audit history

One URL, complete history

Pre-match prediction

This is the original pre-match forecast. It will remain at this address during play and after settlement.

Kick-off18:30 WAT20:30 EAT

Calibrated 1X2 view

Home, draw and away probabilities

H · Real Sociedad
38%38%
D · Draw
31%31%
A · Villarreal
31%31%
Model probabilities and derived fair decimal odds. Fair odds are mathematical references, not bookmaker offers.
HReal SociedadD — DrawAVillarreal
38%probability31%probability31%probability
2.63fair odds3.23fair odds3.23fair odds

Calibration mirror

What does this confidence mean in reality?

38% is the largest probability, not certainty. The other two outcomes still carry 62% combined, and the top two outcomes are separated by 7%.

Demo cohort · All published models and leaguesBelow 40%
Settled forecasts
5
Mean forecast
36.8%
Observed frequency
20.0%
Calibration gap
16.8%

These labelled sample rows were already settled before this demo forecast was published. They are not evidence of production performance.

Open the current full accuracy ledger

Historical probability shapes

Probability Neighbours

Before publication, 10 settled forecasts with the closest full 1X2 probability shapes were selected from all published leagues and model versions. Actual outcomes were 4 home wins, 4 draws and 2 away wins. Their mean prediction was home 39.7%, draw 30.7% and away 29.6%; mean class-averaged Brier score was 0.2072. Mean Total Variation distance was 5.8% and the maximum was 10.0%. Every selected record is within the fixed 10% Total Variation window. This is a small sample below 30 records, so it is descriptive evidence rather than proof of model skill or a promise about this match.

Demo evidenceSmall sample

As of

Total Variation compares the entire H/D/A probability split.

A distance of 5% means five probability points must be redistributed to turn one 1X2 split into the other. Match results, scores, Hit/Miss labels and Brier scores never participate in neighbour selection. This historical view does not change the current prediction.

Settled neighbours
10
Selection scope
all published leagues and model versions
Mean distance
5.8%
Maximum distance
10.0%
Average Brier
0.2072

What actually happened

Observed H/D/A outcomes

Home win
440.0%
Draw
440.0%
Away win
220.0%

Swipe horizontally to inspect probabilities, result, distance and verdict.

All 10 selected historical forecasts, ordered by Total Variation distance
Historical recordPublished H / D / AActual resultTV distanceVerdict
Remo Stars vs Rivers UnitedNigeria Premier Football LeagueH 36%D 31%A 33%D · DrawFinal score 0–02.0%Demo Miss
Brentford vs FulhamPremier LeagueH 42%D 31%A 27%D · DrawFinal score 1–14.0%Demo Miss
AC Milan vs AtalantaSerie AH 37%D 28%A 35%A · Away winFinal score 1–24.0%Demo Miss
Sofapaka vs KCBFKF Premier LeagueH 43%D 31%A 26%H · Home winFinal score 1–05.0%Demo Hit
Athletic Club vs VillarrealLa LigaH 34%D 30%A 36%D · DrawFinal score 0–05.0%Demo Miss
Napoli vs LazioSerie AH 34%D 37%A 29%H · Home winFinal score 2–16.0%Demo Miss
Shooting Stars vs EnyimbaNigeria Premier Football LeagueH 32%D 30%A 38%A · Away winFinal score 0–17.0%Demo Hit
Bandari vs Kakamega HomeboyzFKF Premier LeagueH 45%D 30%A 25%H · Home winFinal score 1–07.0%Demo Hit
Gor Mahia vs TuskerFKF Premier LeagueH 46%D 30%A 24%H · Home winFinal score 2–08.0%Demo Hit
Roma vs FiorentinaSerie AH 48%D 29%A 23%D · DrawFinal score 1–110.0%Demo Miss

Publication proof

Forecast identity

Real Sociedad vs Villarreal pre-match probability proof card
Illustrative Demo proof card. The image is tied to this record's publication payload and can be opened at full size.
Published
Model version
FPAI-LGBM 0.9.3
Competition
LALIGA
Confidence
38%
Record SHA-256
Not issued for demo data

Demo notice: the timestamp, model version and figures on this preview are sample data, so no production record fingerprint is issued. Official records are generated automatically at least 24 hours before kick-off.

Kito calibration challenge

Make your call before the whistle

Set your own home, draw and away split. It stays only in this browser, saves the match to your private desk, and can be scored beside the locked model after the result.

Loading your device-local call…

Your selected probabilities never leave this device · anonymous aggregate use counters respect Global Privacy Control and Do Not Track · excluded from the public ledger, accuracy, APIs and structured data · not betting advice

Interactive counterfactual

What would have to change?

Move three transparent assumptions and watch the probability shape respond. This is an educational sensitivity test, not a second model and never an edit to the demo record.

Nearest illustrative flip

5-point grid

Find the smallest move from the neutral official baseline that puts another outcome at least 0.5 percentage points ahead.

This boundary is a sensitivity illustration—not a new prediction. Missing-information stress stays at 0 and the official record never changes.

Illustrative leaderReal Sociedad · 38%
Real Sociedad
38%+0 pts
Draw
31%+0 pts
Villarreal
31%+0 pts
Top-two gap
7 pts
Uncertainty
Very high
Official record
Unchanged

Illustrative Scenario v1. Method: apply visible log-probability shifts, then normalise with softmax; the stress control raises temperature toward a flatter split. Boundary search uses only team balance and draw pressure on a 5-point grid, holds stress at zero, and requires the target to lead by at least 0.5 percentage points.

Why the model leans this way

Leading match factors

  1. 01

    The home split creates a narrow Sociedad advantage.

  2. 02

    Both teams enter with seven points from their last five matches.

  3. 03

    Near-even Elo ratings make this the lowest-confidence call of the day.

Grounded record explainer

Ask Kito about this probability

Evidence mode

Kito can read only this labelled demo record. It cannot browse for hidden context, change the probabilities, suggest a stake, or promise a result.

Instant evidence mode is deterministic and makes no external model request. Optional AI response planning remains off until a server-side OpenAI key and budget switch are configured. When enabled, the model can only choose a full or compact view of server-authored evidence.

Kito

This is a labelled demo record. The top call is home win at 38%. Across a sufficiently large set of genuinely comparable 38% forecasts, that outcome should occur about 38 times in 100—not every time. The other two outcomes still hold 62% combined probability.

Evidence from the locked record
Kito Evidence Engine v1 · deterministically calculated from the locked record · probabilities unchanged

Locked feature context

Model-generated explanation

The highest published probability is home, but the split leaves meaningful room for the other outcomes. The top call is the largest of three uncertain probabilities, not a guaranteed result. Review the full probability table rather than reading the pick alone.

Generated only from the locked pre-match feature snapshot; maximum 120 words and no change to the official probabilities.