Iteration 97 — H-089 Hüsler–Reiss EXCLUDED KILL #37 · FIFTH SIGHTING · B-06 CLOSED 12/12

In plain words: the last candidate of the batch ran under a standing bet: four tail meters in a row had measured something real that never changes between market regimes, so the loop predicted this one — a meter of how strongly extreme moves in two features graph together — would die the same way. It did, at ratio 0.53 against its noise floor. But the exam carried one extra check that turns a routine fifth kill into the strongest possible close: a wrapper test asking whether this meter is secretly just a re-labelling of the previous kill. It is not — the two meters barely correlate (0.23, where a re-labelling would score above 0.95). So the pattern now has five sightings from provably different instruments: co-crash asymmetry, extreme clustering, two tail-dependence coefficients, and now the Hüsler–Reiss graph parameter — every one real, every one regime-flat. On this substrate, the tails have one law that never changes with the weather; regimes live in the body of the distribution. The batch closes at twelve candidates, zero admissions, twelve kills — and two durable products: a pipeline proposal from the declaration layer, and this five-sighting pattern awaiting the operator's ruling.