33 GJ 625 CARMENES velocities give 31% to 58% detection power for a 14.628-day, 1.67 m/s circular signal under two models with 1.61 m/s of independent Gaussian extra noise

We test the information in 33 public CARMENES DR1 velocities of GJ 625 at the reported planet period of 14.628 days. The data cover 30 nights in 2016–2017. We fit a circular signal with an offset, and then with an offset and an 84.7-day stellar term. With 1.61 m/s of independent extra Gaussian noise, a 1.67 m/s signal has 48.3% to 58.1% detection power in the first model and 31.5% to 54.1% in the second model at a local 1% threshold. The ranges cover all phases. The local p-values are 0.457 and 0.434. Conservative 95% amplitude bounds are 2.087 and 2.615 m/s. We also compare the signal coefficients with a template fitted to 140 public discovery HARPS-N velocities. The result depends on the noise and stellar model. Some assumed correlated-noise models give tighter bounds below 1.67 m/s. These covariance choices are not validated as models of the star. A 2025 study already used the same CARMENES data in a combined orbit fit. The present result describes the conditional strength of the CARMENES test alone; it is not a new planet detection or a new combined orbit. The package includes frozen input data, source hashes, code, figures, control results, and seeded simulations that check the test size, power, and confidence coverage.

GJ 625 · CARMENES · radial velocity · detection power · noise covariance · reproducible astronomy