More giant companions, or heavier ones?

Two different population questions

Finding more giant companions around one class of stars does not automatically mean those companions are heavier. Judah Van Zandt and Brendan Bowler examine both questions in a reanalysis of the California Legacy Survey.[1]

The study spans 719 F-, G-, K- and M-type stars, combining radial velocities with Hipparcos–Gaia astrometry to constrain true companion masses. It accounts for survey completeness and parameter uncertainty.[1]

The 2021 survey catalog drew on Keck-HIRES, APF-Levy and Lick-Hamilton radial velocities.[3]

The 2026 analysis covers companion masses of 0.4–50 Jupiter masses and orbital semi-major axes of 0.1–10 AU, including brown dwarfs as well as giant planets. The findings apply to this defined survey sample.[2]

A rate is not a host percentage

Above the Sun’s mass, the estimated rate is 0.39 companions per star (+0.06/−0.05), versus 0.11 ± 0.02 at or below it. These are rates of companions, not percentages of stars with companions.[2]

The higher-mass group is also more metal-rich. Because both properties vary, splitting on mass alone cannot isolate mass’s influence. The authors also compare subsamples split on both properties.[2]

What changes when the star sets the scale

The authors report only about 2–3 sigma evidence that the companion mass distribution shifts toward heavier companions. The stronger occurrence result should not be confused with this more tentative shape result.[1]

Using companion mass divided by stellar mass, the higher- and lower-mass host groups have distribution shapes consistent at 1.6 sigma. Doubling both masses, for example, leaves their ratio unchanged.[2]

A common relative-mass shape is a possibility, not proof that the stellar groups are identical. The analysis does not establish causation.[2]

This explanation uses arXiv v1, submitted 2 November 2026. Its author comment says submitted to AJ; journal acceptance is not established here.[1]