Yeah it is a great explanation and gets you going quickly on the idea.  The Hornady Podcast is an absolute TROVE of knowledge though.  What they are basically presenting is that in reality to actually get a good feel for the accuracy of any rifle, you need about 30 data points... 30 shots.  a couple 3 or 5 shot groups simply won't cut it because you haven't shot enough rounds for the edges of the dispersion to fill out.  What you might think is a "flier" is actually another part of the dispersion that just hasn't been filled out yet.

Think of tossing a coin.  what is the probability of getting heads?  well 50% right?  But notice there's a chance you throw the coin three times and get three heads in a row.  That does not mean the coin is "heads all day long so long as you do your part" The reality is that the more you toss the coin, the closer your average is going to get to the real number of 50% and you need a certain number of trials to get the average to fill out to be truly representative of reality.  Same with rifles.  You just need more shots.  

There's a really good reason Molon does three 10-shot groups and compiles them into one 30 shot group for his testing.

Hmm.  This makes me lean even more heavily toward an AUG when I finally get around to getting back into a bullpup.

I really like that the new fad/hotness 77gr AAC is in your test. Doesn't looks that fun.

I didn't know it was the hotness really.  I chose it because it was cost effective.  My goal for testing cartridges is to select two go-to loadings.  I want one for range/training use, and one for shtf/self defense use.  So think of it like I want the best of the cheapest and the cheapest of the best.  

There are others that I am sure would group better like Federal GMM, but honestly for 1.50 a piece I'm just not really going to shoot that ammo ever so there's not much point in testing it. Especially since it's going to take two boxes to get a solid data set to make any kind of judgement on it's capability.

When I was bullpup shopping I seriously considered the AUG again, but current bbls are nitride instead of CHF/CL like the X95.

I know….   That was a major hang up for me.  But I am ever so slowly coming around to nitriding being an acceptable alternative.   Chrome lining is preferred, but some of the reports I’ve read about accuracy of the x95 are abysmal and as I said early in the thread it appears to be very random.  I don’t know, my mind has not been made up.