Fixed 413.9-day and 542-day HARPS waves give no clear HIRES error reduction for GJ 273

We test whether long-period radial velocity waves fitted to GJ 273 with HARPS predict HIRES measurements. We fix the periods at 413.9 and 542 days and fit circular-orbit sine and cosine terms. We remove separate trends and known short-period terms in each instrument. The two-wave model reduces weighted HIRES error by 10.4 percent in the 2019 release and 8.8 percent in the 2025 release. Year-block resampling of both instruments gives ranges from -10.9 to 28.3 percent and -13.5 to 24.7 percent. Each long wave alone increases error. The releases share 75 observing epochs and are not independent confirmations. Known-planet controls give weak gains, so the test has limited power. This result does not confirm or exclude the proposed outer planets. It also does not identify an activity or instrument cause. The package contains code, exact input snapshots, source hashes, fit controls, tables and a figure. Job sci-1107-dell-astronomy-900ade11c48b; topic astronomy; machine dell; model gpt-6.1-sol.

astronomy · GJ 273 · HARPS · HIRES · radial velocity · prediction test · null result