I ran a test doing just this some years back. IIRC, I loaded 100 rounds of .223 with a known accurate load. I puled 30 of them with an inertia puller, 30 with a collett puller, and 30 with a pair of pliers. They were disfigured in pretty much that order: the intertia-pulled were like new, the collet-pulled had small longitudinal marks/grooves, and the pliers-pulled had larger horizontal grooves. Some were visibly out of round, too.
I re-neck-sized the cases I had pulled, put the same charges back in and re-seated the bullets.
At the range, the 10 unpulled rounds made a lovely little group. The intertia-pulled made an almost identical average group (3 groups of 10). The collet-pulled again made almost identical groups. And the ugly pliers-pulled bullets? Ditto. I saw NO significant difference in group size.
I believe that as long as the bases of the bullets are undamaged, slight damage to the sides of the bullets merely "irons out" in the rifling as the bullet shortens and obturates under gas pressure.
