The Reactor Antineutrino 5 MeV Bump as Bookkeeping: Pricing the Branching Fix and the Forbidden-Transition Fix

Reactor experiments see ~10% more antineutrinos near 5 MeV prompt energy than the Huber-Mueller model predicts. From the published Daya Bay spectrum ratio (window-averaged ratio 1.09 over prompt 4-6 MeV), the excess is 0.040 antineutrinos per fission in the 4.78-6.78 MeV window (1.24e18/s per GW thermal). In the summation method the six largest contributing beta branches supply only 8.4% of the window flux, so manufacturing the bump from their branching ratios or cumulative fission yields requires a coherent +108% change against a ~6% evaluated uncertainty scale, with only ~5% branching headroom on the two largest branches (92Rb, 96Y, branching ratios ~95%): that route cannot carry the bump. First-forbidden decays carry an adopted 62% of the window flux, requiring a +15% mean shape-factor enhancement; published shell-model and ab initio calculations give +8% to +10% on that component (+5% to +6% total), carrying 56-67% of the bump. Only the forbidden-transition route is of the required size; it leaves a 3-4 point residual. All computations are in the deposited compute.py.