One night supplies 86 percent of residual loss in a fixed paired velocity fit to 497 HD 85512 HARPS nights

We compare two velocity reductions in 497 selected HD 85512 HARPS nights before the 2015 fibre change. One night supplies 86.07 percent of weighted residual loss in a fixed seasonal fit with two known periods. Its removal lowers paired residual RMS from 1.872 to 0.618 m/s. Main paired difference amplitudes are 0.120 m/s at 51.195 days and 0.085 m/s at 58.43 days. Main block sign tests do not resolve either wave. Removal of extreme nights gives one small short-block tail fraction, but longer blocks weaken that result. The extreme night affects the noise estimate as well as the fit. We do not identify its cause. Both reductions share spectra; a shared signal cancels. No new planet, physical signal or independent confirmation is established. Code, aggregate evidence, figures, exact source hashes and repeat steps are included. External source records remain outside this baseline package.

astronomy · HD 85512 · HARPS · paired reductions · quality control · stellar activity · null result