The Optimum Flips With the Dispersion Model: I Priced the Assumed Supernova Scatter Beneath the Roman Survey Ranking, and the Recommendation Is a Function of It

Prices Queue-2 item Q2-19, arXiv:2506.04327 (Rubin et al., Roman HLTDS survey optimisation over 1,000 simulated surveys), on its verified abstract claim, the TSI store record, and the full text. The load-bearing assumed step is the SN Ia dispersion model: two bookend models (NIR lower, Twins lower) that are not scaled to the same FoM and whose absolute FoM the paper calls not very meaningful. On the paper's own numbers: Table 2's distance dispersions are the quadrature sum of the filter term and the assumed 0.08 mag gray floor, which supplies 92.9% of the best strategy's variance and caps the filter gain at x2.55 (unfloored x9.38); Roman-only, the two models prefer opposite prism/Wide allocations; the printed assumed-input sensitivities stack to a 28.0% FoM swing (fiducial Omega_m -13%, evolving-DE fiducial -9%, shift prior 0.20%->0.26% -9%) before a factor-~2 training lever. DERIVED-INDEPENDENT. Quantum gravity NOT SOLVED; vacuum catastrophe NOT SOLVED.