Comment by Retric

Stars hitting stars isn’t the metric a baseball sized object clearly counts as a meaningful collision at these speeds. Add up all objects in terms of 2d cross sections and it’s the small stuff we’re going to care about, but getting near stars matters because from what we can tell there’s so much more small stuff near stars. The Oort Cloud sits 0.03 to 0.08 light years from the sun etc. https://en.wikipedia.org/wiki/Oort_cloud ~5 earth masses in vastly smaller objects is a lot of targets. (I recalled an earlier and far larger estimate but it’s still a lot of material.)

Ignoring gravity may be fine at 1.6 light years but a closer approach to even just say 2 light months means spending thousands of years much closer to the star which would matter here. So simple extrapolation based on random distribution of nearest approaches like you just proposed with that 100,000 star calculation is heavily biased in the wrong direction.