A key difference is that spinning the full satellite with the arm extended and measuring the center of rotation measures the mass of the samples collected regar

A key difference is that spinning the full satellite with the arm extended and measuring the center of rotation measures the mass of the samples collected regardless of the thrusters/reaction wheels. Measuring the trajectory of the reentry vehicle gives a parameter that can be used to estimate the sample mass assuming nominal thrust values, but you can't distinguish between an underperforming or overperforming rocket engine and heavier or lighter samples. This ambiguity doesn't really matter for the purposes of burn correction - you need to get to the correct reentry trajectory no matter what caused the deviation - but your sample mass estimate may be off. I don't think there's a whole lot of utility in knowing the sample mass, as long as you've got enough (and they've definitely got enough!), it would just help satiate the curiosity of the scientists and engineers who otherwise have to wait a few more years to see their results.