Free fold audit · Evidence before metrics

Walk-forward fold & temporal leakage auditor

Reconstruct what was knowable at every test boundary. One future feature, late outcome or repeated test match can make a clean-looking backtest answer a question that could never have existed in production.

Canonical publication record

Abstract

A browser-only audit of declared walk-forward fold membership, prediction cutoffs, feature availability, training-outcome boundaries, duplicate matches and repeated test evidence.

Author and publisher
Football Proof AI
Technical report
football-model-leakage-audit/1.0.0
Published
Last modified
Release status
Current release
Review status
Editorial technical note; not externally peer reviewed

Interactive · football-model-leakage-audit/1.0.0

Make time move in one direction.

Paste or choose a fold manifest. The deterministic audit traces each finding back to CSV line numbers without sending the manifest to a server.

Browser-only processing · no upload

Required: match_id, fold, split, prediction_at, kickoff_at, outcome_known_at, feature_available_at. Times must include Z or a numeric offset. Limit: 2 MB, 20,000 rows and 1,000 folds. The built-in rows are synthetic and intentionally contain audit findings.

Temporal integrity report
Ready for a fold manifest

Run the audit to reconstruct test boundaries and reveal evidence rows.

Input contract

Seven fields reconstruct the evidence boundary

Each row describes one match in one walk-forward fold. Timestamps must include a timezone because a naive local time cannot establish which fact existed first.

match_id
Stable match identity used to expose duplicate and cross-fold reuse
fold
Walk-forward evaluation fold or deployment replay window
split
Exactly train or test; same-fold membership in both is critical
prediction_at
Zoned time at which the forecast would have been produced
kickoff_at
Zoned match start; prediction_at must be strictly earlier
outcome_known_at
First time the settled training target was genuinely available
feature_available_at
Latest availability time among features consumed by that row

Download the synthetic leaky-fold example. Its findings demonstrate the contract and are not model-performance evidence.

Audit order

Fail the timeline before scoring the model

A Brier score or hit rate is uninterpretable until the fold manifest passes point-in-time integrity checks.

  1. 01

    Parse time

    Reject impossible dates and timestamps without Z or a numeric offset.

  2. 02

    Check inference

    Flag predictions at or after kickoff and features arriving after prediction.

  3. 03

    Lock outcomes

    Require every training outcome to be known strictly before the first test prediction.

  4. 04

    Separate windows

    Compare the latest train prediction with the earliest test prediction in each fold.

  5. 05

    Trace identity

    Expose exact duplicates, same-fold split crossings and repeated test matches.

  6. 06

    Bind receipt

    Store evidence rows, severity, formula version and the input SHA-256 together.

Severity semantics

Not every reused match is the same failure

Critical
Future feature, post-kickoff prediction, same-fold train/test crossing or unavailable training outcome.
High
Malformed timestamps, duplicate weighting or the same test match counted in multiple folds.
Review
Cross-fold reuse that may be valid for an expanding training window but needs a declared policy.

The verdict fails on any critical or high finding. A warning-only receipt remains “review” because the tool cannot infer the intended expanding- versus rolling-window protocol from timestamps alone.

Evidence boundary

A manifest audit is necessary, not sufficient

A CSV can declare a clean timeline while an undocumented feature store, manual join or training job used different data. The SHA-256 proves byte identity only; it does not prove historical existence, pipeline execution or independent certification. This tool has not received external peer review.

  1. Archive the exact feature snapshot.Keep immutable query parameters, extract time and source fingerprint.
  2. Store orchestration evidence.Bind code, model artifact, fold manifest and output probabilities in one run receipt.
  3. Repeat independently.Have a separate reviewer reconstruct at least one fold from primary data.

Continue the method

After time integrity, evaluate probability quality.

Pass the fold boundary first. Then report every out-of-sample row, uncertainty, calibration and proper scoring rule without deleting misses.

Read walk-forward protocolAudit prediction accuracyReview data provenance

Primary and authoritative sources

Methods behind the boundary checks

Cite this tool as: Football Proof AI. “Walk-Forward Football Model Leakage Auditor.” Version football-model-leakage-audit/1.0.0, 13 July 2026, https://footballproofai.com/tools/football-model-leakage-auditor. Editorial technical note; not externally peer reviewed.