I believe that variability would be the domain of the control system and launch/recovery parameters to handle no matter what the fin design. The shorthand way to think about it is that the fin should not add any variability that the control system does not have the authority to overcome (and by some margin above that) - else one would not choose that combination of design elements; apoligies if im retreading something you already know...
Yes, true - I just wasn't articulating that as clearly as you have.
I was thinking of "we don't want to rely on drag as a function of descent" rather than "drag doesn't matter, because the control system is fully advanced and capable enough to deal with any variance brought on by drag"
