decisive_experiment

Scope: Decisive experiment design.
Audience: Reviewers and contributors.
Guarantees: Experiment is interpretable.
Non-Goals: Full protocol.
Why: This doc exists to record its single responsibility for review.

Overview

This doc defines one decisive experiment.
Architecture context lives in Architecture.
Claim context lives in Falsifiable Claim.

Contracts

Experiment compares recovery rate with and without regulator proposals.
Metrics follow Metrics.
The experiment uses deterministic replay from Execution Model.
Evidence uses tests/regression/test_negative_results.py.

Invariants

Experiment inputs remain fixed.
Experiment aligns with Core.
Evidence aligns with tests/regression/test_negative_results.py.

Failure Modes

Missing controls breaks interpretation.
Drift in setup breaks Falsifiable Claim.
Missing evidence breaks Docs Style.

Extension Points

Experiment updates follow Docs Style.
Extensions align with Experimental.
Evidence updates align with tests/unit/test_docs_contract.py.

Exit Criteria

This doc is obsolete when an experiment suite exists.
The replacement is Metrics.
Obsolete docs are removed.

Code refs: tests/regression/test_negative_results.py.