Fixed 414- and 542-day GJ 273 templates have HIRES gain intervals that include zero in three block tests

We test two proposed outer planet signals of GJ 273 with public radial velocity tables. We fit four fixed circular terms to 280 HARPS measurements and freeze each long-period template in 81 HIRES measurements. A common baseline gives weighted squared error gains of 0.5 percent at 413.9 days, 16.9 percent at 541.5 days, and 13.8 percent for the pair. Joint 100-day block resampling gives 95 percent intervals of -51.8 to 20.7, -9.6 to 33.2, and -47.7 to 32.8 percent. All nine intervals from 30-, 100-, and 365-day blocks include zero. Instrument corrections retain the point gain near 542 days. Published period limits can change its sign. The short-period controls are weak. The 2025 HIRES catalog adds only six epochs, of which three follow the 2019 candidate paper. Historical HIRES data were used to select the candidate periods. Thus, this is a test across instruments, not a fully independent period test. It confirms no planet and excludes no general planet model. A source overlap check resolves a velocity unit label error. Raw tables, hashes, code, controls, figures, and offline steps are included.

astronomy · GJ 273 · radial velocity · planet candidates · HARPS · HIRES · time-block resampling · stellar activity · null confirmation result