The tilt is now a measurement

In the accepted preprint, six transits observed with NIRPS and HARPS separated the spectral lines hidden by the planet across the rotating stellar surface. The Rossiter–McLaughlin analysis gave a three-dimensional angle of 136 (+24/−18) degrees between the stellar spin and orbital plane. In this measurement, GJ 3090 b is genuinely retrograde.[1]

Explaining this architecture with a close encounter requires an unseen massive culprit. The paper’s imaging, astrometric, and radial-velocity limits exclude a wide stellar companion and sharply restrict massive outer planets, leaving only some highly inclined arrangements open. “No companion found” does not erase every small planet, but it reduces the parameter space available to a simple outer-body scattering explanation.[1]

The disk explanation is testable too

The authors favor a second gas disk accreted at an angle after the star formed, followed by migration within that disk. That route could turn the inner planet retrograde without dismantling the multiplanet system. The study does not observe the old disk, however; the same measurement can still accommodate a lower-mass undetected companion or another early torque history. The measurement narrows the model space but does not select a unique formation story.[1]

The discriminating measurement lies in the system’s other orbits. If the 13-day candidate and the confirmed 16-day planet share a similar tilt to the stellar spin, a common disk geometry gains weight; if only GJ 3090 b is tilted, a planet-specific dynamical event fits better. Longer radial-velocity monitoring and tighter imaging limits also reduce the remaining companion space. The 136-degree result is therefore more than an unusual angle: it is a blueprint connecting two families of formation models to different observations.[1]