I wanted to try to build the "most accurate AR-15 I could build." I won't lie, that upper cost me a mint to put together. And it mostly just sits on a shelf collecting dust!
The last experimenting I did on it was with seating depth of 77 grain SMK's. I've had a working theory for a while that bullets with the base seated deep in to the casing, with the bearing surface "floating in space", get perturbed and distorted prior to fully entering the throat. I first noticed the phenomenon on 300 win mag, 220 grain bullets, seated with the bearing surface deep in to the cartridge (3.300 OAL). I lost an entire MOA of accuracy if the bullet's bearing surface was seeted deeper than the neck, vs. an OAL matching my lands with the base NOT extending in to the cartridge!
The same thing happened when I tested it with 223. Seating a 77 grain bullet deep in to the case, so it fits in a magazine, shoots much less accurately than the same bullet seated over-long, with the start of the boattail matching the end of the neck. My group sizes at 1/4 mile (440 yards) were averaging about 3" with the rounds loaded long, and single loaded, versus an average of 6.5" groups from rounds seated deep and fed from the magazine. The groups opened up in each direction, but also grew oblong vertically, and the chronograph showed much a higher velocity spread on rounds seated "deep."
My theory for this is the that A] the bullet has enough bearing surface for the charge to cause it to "tilt" prior to fully entering the throat, and B] the base of the bullet actually
expands under the high pressure, and then re-forms to bore diameter as it enters the throat. (Copper & lead are pretty malleable, and deform much like a wad of jello when faced with 55k PSI of pressure... The rear of the bullet is getting shoved in to a front that is facing resistance, causing the back of the bullet to balloon out slightly, then re-form to bore diameter as it enters the throat.)
Granted, I don't have the ability to SEE this happening, and can only form hypothesis based on external data gathered, but the velocity spread difference was pretty telling!
I ALSO believe that the 1:9 twist issue on 77 grain that SOME people (including myself) have seen, with bullets not stabilizing, while OTHER people can shoot those just fine out of the same twist barrel, has a LOT to do with where the leads are located and how much throat erosion there is.
If the bullet has sufficient free-bore (from an eroded throat) it will "skip" ahead in the bore much easier, the base won't expand as much under pressure, and the bullet will be a little less likely to tilt under pressure as it engages the lands.
The only data I have to support that theory is from shooting 77gr loaded to magazine depth out of a 16" barrel 1:9 twist that keyholed at 440 yards. The barrel was brand new, the lands were sharp, and the measured velocity spread was MUCH greater than from my experimental 223 wylde chamber which has an additional .25" of freebore over the standard AR barrel (by design, for seating 80 gr bullets overlong).
The same load in one rifle was engaging the lands almost immediately, while the base was still very deep in the cartridge, while on the other rifle it skipped forward without resistance .25" allowing the base to almost clear the cartridge before engaging the lands. The velocity and accuracy was still not close to what I got with the bullet loaded long, the freebore pays a price...if you don't take advantage of it.
So my theory on that is the 1:9 twist barrel CAN stabilize a 77 gr bullet *IF* the lands are somewhat worn. (There might also be a chrome lined barrel vs. steel difference, too, but I haven't tested the differences as I don't have a non-lined steel barrel handy that's 1:9, only CM.)
Anyway, that's the whole story on the "experiment."
I have 12 different AR barrels sitting on a shelf downstairs, various gas length, various designs, various profiles, various muzzle breaks, it's fun to experiment if your willing to take the time with a barrel wrench.