You are leaving knowledge on the table.
If you have a laod that works well, try the same bullet in the lands square. (Rifling marks are as deep as they are wide). And also try laod just barely off the rifling.
Of the 3 loads, one might be better than the other.
The "into" the rifling load has issues. If you open the bolt, the bullet may pull out of the case and dump powder. No good for a hunting round, for sure.
I've been working on the assumption that a particular pressure will beget a certain velocity. This velocity with its attendant rate of acceleration in the bore may coincide with a node if the powder charge is correct for attaining such.
I have not fiddled with seating depths except for COL fitting magazine or between 10 & 15 thou off the rifling if it does.
I have been using QuickLoad for load development and it does not hint at any particular seating depth being good or bad, but as long as it is known then it can be entered into the calculation.
Is seating depth used to fine tune when other parameters are unknown?
I think seating depth is more of a function of the bullet than the rifle. I load Sierra match bullets at 20 thou off the rifling and have never found that other depths make much difference.
But I have used some VLDs where the popular wisdom was that they work best on the rifling. I tried that and found no difference, but the point is - I tried it.
If you have a load that works and is near max, seating a bullet into the rifling is probably going to make it worse. i would try it at a half and full grain less than usual, just to be sure you are not getting a potential bug hole group load.
I start load development with all bullets 20 thou back and find the best powder charge and then try the same load 80 thou back and 100. I find small increments don't change much.