Archived project / Final laboratory record
SETTLED / C4 MISSThe final score was outside the cloud.
Spain 0-0 Argentina after regulation. Turtle's locked C4 remains unchanged below. The current exact-score experiment failed and is closed pending rebuild.
Original regulation-time C4None landed
Actual: 0-0. It ranked sixth at 7.17%, only 0.67 percentage points below the fourth node, but the selector left it outside C4.
What Project Turtle tested
The prediction was written down. The match judged it.
Football analysis becomes effortless to rewrite after full time. Turtle preserved its evidence, published a small score cloud before kickoff, and kept the misses beside the hits.
The final test failed. That does not erase the artifacts, but it ends the claim that this protocol is ready for practical exact-score prediction. The archive remains public so a future method must start from the real record.
- Before kickoff
- Evidence, score cloud, timestamp
- After full time
- Settlement, failure class, correction
- Never rewritten
- The locked score nodes
Public track record
The hits are worth showing. The misses stay visible.
8 / 16labeled card or branch hitsEight misses remain in the same archive: a 50.0% descriptive hit rate, not a calibration claim.
World Cup Final / C4 missSpain 0-0 Argentina
The model ranked 0-0 sixth but omitted the surviving scoreless-draw branch from C4.
Semifinal / exact nodeEngland 1-2 ArgentinaNode two of the locked C4 landed in regulation.
Quarterfinal / C4 hitBelgium 1-2 SpainThe wall-break Spain-win branch survived the clean-sheet fork.
Round of 16 / underdog branchBrazil 1-2 NorwayThe fourth-node Norway branch captured the upset.
Round of 32 / leaderColombia 1-0 GhanaThe leading exact-score cell landed.
Round of 32 / draw branchAustralia 1-1 EgyptThe preserved draw node settled exactly.
First public cloudPortugal 2-1 CroatiaThe 2-1 branch landed inside Turtle's opening cloud.
Clean-sheet failurePortugal 0-1 Spain
Spain's verified non-concession run was not preserved.
Ruthless-tail failureBelgium 1-4 USA
The model tested 3-1 but stopped the scoring ladder too early.
Promotion failureFrance 2-0 Morocco
Inspect every settled row
The clean 2-0 scenario existed but was not promoted into C4.
The operating method
Six gates between a fixture and four scores.
Each gate leaves a downloadable artifact. A score cannot enter the public cloud merely because it sounds plausible.
- 01Verify
Fixture, venue, orientation, kickoff, and the no-leak cutoff.
- 02Admit data
Regulation-time rows with source proof, dates, scores, and opponent ratings.
- 03Compare realms
How each team performs against opponents near today's strength level.
- 04Build threads
Result, margin, total goals, BTTS, clean sheets, and exact-score mass.
- 05Attack nodes
Every candidate faces the strongest fact arguing against it.
- 06Freeze and audit
Four scores are timestamped; full time adds a separate postgame record.
Final evidence room
What was checked, and what the checks failed to establish.
Market prices were excluded from prediction inputs. The files expose both the reproducible model and its failed selection.
Fifty regulation-time matches for Spain and fifty for Argentina.
Major score families were challenged before selection.
All declared perturbations retained an eligible four-node set.
The frozen model copies returned the same card.
Current World Cup1.86 GF0.14 GA86% clean sheets
The tournament clean-sheet signal forces 1-0 to remain in the card.
Current World Cup2.14 GF0.86 GAscored in 7/7
Argentina's tournament scoring rate preserves 1-1, 0-1, and 1-2.
Inside C437.7%
Combined model mass of the four published nodes.
Outside C462.3%
Most modeled probability remains outside any four exact scores.
Actual exclusion0-0
Rank six at 7.17%, only 0.67 percentage points behind the fourth node.
Sparse evidenceArgentina realm
Effective sample 3.25; shrinkage was required and uncertainty remains.
Full final analysis
Open the rows, comparisons, score table, contradictions, and stress tests.
Public analysis archive
A hit has context. A miss has a name.
The archive links every settled result back to its match analysis. Personal bankroll activity is never part of this record.
DateMatchLocked cloudFinalSettlement
Spain vs Argentina1-1, 0-1, 1-0, 1-20-0C4 miss
England vs Argentina1-1, 1-2, 0-1, 2-11-2Exact hit
Belgium vs Spain0-1, 0-2, 1-2, 1-11-2C4 hit
France vs Morocco2-1, 3-1, 2-2, 3-22-0Miss
Argentina vs Egypt2-0, 3-0, 2-1, 1-13-2Miss
Belgium vs USA2-1, 1-2, 3-1, 1-11-4Miss
AI agents and researchers
Start from the same settled record humans see.
Structured discovery files now identify the `0-0` result, the C4 miss, and the closed protocol state without rewriting the original prediction.