Most folks say that fired brass is better since it is "fire-formed" for your specific chamber. My thought on this is 100% true. A lot less movement during firing. Assuming you size only the necks when you reload. Maybe 11 firings is the end, though. Don't know - I try to keep several weighed lots of over 1000 cartridges each, so I ran over 10k through one particularly happy barrel, and almost 4k on the factory barrel before that and I just started cycle #5 on the brass.
But....there's a point where it needs some restoration and TLC.
My best 308 brass gets a bit of a "donut" around the inside the base of the neck after 4 firings. Inside neck reaming (before sizing) or uniforming with a mandrel and then neck turning the outside makes it all uniform again. Then trim it up to length.
Get a piece of your fired brass - before you resize. drop a bullet into it. If it drops right down inside - no donut. If it drops smoothly in and then hangs up - you have a donut. This is from brass flowing around during firing. If you're loading out long, this probably won't affect you - but if you're seating deep, back into this "donut" then the donut will force outward and make a bulge in the outside of the neck. That can cause chambering problems, or worse.
After LOTS of firing, the brass near the head of the casing is probably getting thin (that's where the excess brass forming the donut is coming from). So it's worthwhile to hacksaw a casing down the middle every once in a while to see how your "pile" is progressing.
The act of sizing brass makes it more brittle and moves brass around as well. (bend a coathanger enough and it eventually breaks -- same principle. The forming is probably more harmful than the heat of firing...) I think annealling the case mouths reduces these tendencies - but it seems like a lot of work so I can't be bothered.
Ideal situation (IMO) would be to use once fired brass from your own chamber -- size necks only with collett die. Use shoulder bump die to knock them back a bit if they start to get tough to chamber after 5 firings or so. Neck turn at the start ONLY if neck thickness variation is garbage (a lot of work) And then uniform from teh inside with a mandrel and neck turn right down to the shoulder after 4 firings or so ONLY ENOUGH to clear that "donut" out of the way. Or run a reamer through before resizing every once in a while to do the same thing.
Do you have a ball micrometer? - what's your neck wall thickness - more importantly, how consistent is the thickness. Possible that you have minimal runout becasue a straight die is keeping things straight -- btu there could be some heinous neck tension things going on. Also - seriously check (by cutting and visual inspection) the case wall down near the head of a few samples. Accuracy aside, I'm curious if you're on the road to a head separation...