Yeah that's not good. I have had dies do the same thing on bullets in a few calibers.
My solution was to drill out the stem and bed it to a seated bullet.
1) Drill out the stem so the hole is nice and straight and just slightly smaller than the bullet diameter. A mill is best, but a drill press will work too because the bedding will create perfect alignment. Stems are easy to drill with good bits.
2) Choose a case with a seated bullet that has virtually no runout (preferably .001" or less) and apply release agent to the entire bullet.
3) Apply release agent inside the seating die bullet seating chamber with long Q-tips
4) Clean then apply bedding in the drilled hole of the stem. Make sure you dont overdo the bedding to avoid having too much bedding overflow in the die.
5) Carefully re-seat the stem in the die a quarter turn shy of the seating depth used on the case you chose for the bedding job. This is to make sure no metal part of the stem contacts the bullet.
6) Run the case up in the die all the way into the stem and let it cure for a day.
7) When you break the bullet out the next day by stroking the press ram down, there is a very good chance you will pull the bullet out of the case neck. No big deal.
8) Remove the stem and lightly tap the bullet until it breaks free from the bedding.
9) Now drill out the center of the bedding to avoid the stem contacting the point of the bullet. The drill bit size used depends on caliber. You want to leave a good portion of the bedding in place to have more surface contact, but you dont want the bedding contacting the bullet too high on the ogive.
10) Trim overflow and chamfer bedding at base of stem. Now ready for very consistent seating.
Heres a list of seating depth and runout measurements I took on one of my standard Redding seating dies after bedding the stem. It is now extremely consistent. No bullets I measured from a small sample had more than .001" runout and the seating depths were all within .001" of each other save for one bullet that must have had a very slightly different ogive shape. But even that bullet measured only .0005" depth above the .001" window the others fell in.
