...I can't see why you couldn't simply measure off of the end of the stud for stretch (after the thread clearance had been taken up between the block female threads and the male threads on the stud). However, ARPs reply back to me this morning indicates using the torque value is what they suggest.
Hmm, my thinking was a bit different. Given the amount of stretch you'd expect to see I think it would be hard to consistently locate the stud well enough in the block to always take up the extra "wiggle room", and therefore give you a stable baseline measuring and starting point. After all, they normally say "put the studs in finger tight", and when the total stretch may only be 0.025" (guess only, but I'm comparing to a rod bolt that's typically in the 0.006-0.007 range) there is just too much room for an error to creep in when simply installing the studs.
So instead of the above I was thinking of using the OAL of the stud as an indicator to see if the stud was over-stretched, along the lines of the rod bolt check you'd normally do. For that to work you'd need some kind of a range from ARP that says: if the bolt OAL increases beyond this point that means it's lost it's ability to retain torque.
If that info was available you'd be able to torque it down to the advertised torque (110 ft-lbs in my case), and periodically (during refresh perhaps) pull the studs to re-measure the OAL to see if they were stretched past their initial length and therefore are suspect and should be replaced.
I use a stretch gauge on the bolts in the rods in a rod vice before and after tightening and sneak up on the torque 5 lbs. at a time to get the torque needed to be in the middle of the stretch specs 
