The Anomaly Retired; the Defect Moved House: Muon g−2 After the Final Number

The Fermilab Muon g−2 experiment's final result (June 2025) fixes the experimental world average at a_μ^exp = 116592071.5 ± 14.5 (×10⁻¹¹). Against the 2020 Theory Initiative white paper (data-driven hadronic vacuum polarization), the defect was Δ = +261.5 ± 45.4 (×10⁻¹¹) — 5.8σ. Against the 2025 white paper, which replaced the data-driven HVP with the lattice-QCD evaluation, the defect is Δ = +38 ± 63 (×10⁻¹¹) — 0.6σ, nothing. The experiment did not move: both white papers face the same measurement. What moved is one theory input, the leading hadronic vacuum polarization, whose two independent reductions (dispersive (from e⁺e⁻ annihilation data) at 6931 ± 40 and lattice at 7132 ± 61 (×10⁻¹¹)) disagree with each other by 201 ± 73, a 2.76σ theory-versus-theory defect that is now the largest open wound in the quantity. The anomaly's death certificate therefore reads: killed not by a better measurement, and not by new physics, but by a change of theory method that is not itself yet settled — the e⁺e⁻ cross-section datasets behind the dispersive value contradict each other (the stated reason the 2025 paper abandoned them), and the second hadronic contribution, light-by-light, carries a pheno-versus-lattice split of 1.5σ with an error-inflation factor applied in print. Verdict: no case against the Standard Model — the anomaly is retired, and the defect has moved from experiment-versus-theory to theory-versus-theory, at 2.76σ, priced in §4. A discrepancy whose verdict flips with the choice of one input's reduction method was never a measurement of new physics. It was, all along, a measurement of the HVP problem, and the HVP problem is still open.