You have to ask yourself too about how sharp a cut-off there is in this matter. That is, is there a grey area where some individual bullets are stressed, others are OK, but all still reach the target intact? If so, do the stressed examples begin to display odd behaviours on the target at long ranges? I always think of the rifling twist / bullet stability issue. For a long time most shooters said 'the holes are nice and round, so stability must be OK. Now we know that marginal stability can give reasonable groups and round holes, but moving to a higher level can give still tighter groups and/or higher effective BC values. Does marginally over-high rotation produce a similar range of invisible but still negative effects?
Like you, I simply stick to 7 twist and if I lose a little BC, so be it. (The only 224 bullets I ever blew up in many thousands of shots were some 52gn Hornady AMaxes from a well-worn Lilja 8 twist many years ago. I'd tested the load in cool conditions and it had performed very well. A day in the 80s with fouling building up in the barrel proved different. First few shots in a clean barrel fine, then the next half dozen vertically strung, then no more shots on the paper with everybody telling me when I got up that they'd disintegrated in clearly visible grey puffs about 80, 90 yards downrange.

)
Laurie - you raise some interesting points. In my hands, the line of demarkation between normal bullet function and jacket failure can be quite sharp. I have never lost a jacket in a 7.0-twist barrel. I have lost both 90 VLDs and 88 ELDMs in an otherwise identical (i.e. 5R, 30" finish length, same reamer specs) 6.8-twist barrel. Surprisingly, that same 6.8-twist barrel has not [yet] caused any jacket failures with 90 gr SMKs at comparable velocity with the 90 VLDs, or with 95 gr SMKs at ~2750 fps. With a load velocity of ~2850 fps, 7.0-twist and 6.8-twist barrels would be producing spin rates of of ~293K and 302K RPM, respectively.
When you talk with Berger about this issue, their apparent "red zone" with regard to bullet spin rates an jacket performance is 300K RPM. Clearly with a load using 90 VLDs over H4895 at ~2850 fps from a 30" barrel, the bullet RPM value from a 7.0-twist barrel is below 300K RPM, whereas that produced in a 6.8-twist barrel is not. I have bore-scoped that 6.8-twist barrel three ways from Sunday. If there is some innate problem with the bore itself, I cannot find it. Nonetheless, I am not able to state with certainty whether the problem lies in the barrel itself, or in the fast 6.8-twist rate. I have another apparently identical 6.8-twist barrel (both were purchased together, then chambered at the same time, by the same smith, with the same reamer). Eventually, I will use that barrel on the rifle. However, even over some period of time, any failure to observe jacket failure with the 2nd barrel is not 100% proof the problem lay within the 1st barrel/bore itself, although it is certainly suggestive of that. Further, I am not inclined to load bullets I suspect might fail, merely for test purposes, as reproducing the failures might take a while, and it's something you never want to show up in a match. My point is it may not always be easy to determine the cause of bullet jacket failures with absolute certainty. It is much simpler to make a change and hope the problem vanishes.
As I mentioned, I have never lost a bullet in a 7.0-twist barrel, so that is what I will continue to use when re-barreling my .223s in the future. If necessary, I may also consider adopting the use of 0.219" bore, although I haven't gone that far as yet. My opinion is that although Berger's Twist Rate Calculator suggests running the 90s through a 7.0-twist barrel versus a faster twist means giving up a small amount of intrinsic bullet BC, the difference is exactly that...small. So small in fact, that I don't believe anyone can actually shoot the difference. So in essence, the only thing a 7.0-twist barrel is probably giving up to a faster twist rate barrel is an increased propensity for jacket failure, IMO. I just had a couple more 7.0-twist barrels chambered for my original .223 Rem for shooting the 90s, and if anything, it is shooting even better now than it did with the original two 7-twist barrels.
I can easily envision that the dimensions (i.e. bearing surface length) of the 85.5 gr bullet might render it more resistant to jacket failure as compared to the 90 VLD. For those individuals that have effectively found the 90 VLDs to be unusable for F-TR competition due to jacket failures, I can also envision that switching to the 85.5s is a no-brainer. One fails, one doesn't - pretty easy choice, right? Nonetheless, at least in F-TR where the requirement is to use only the .223 Rem cartridge, in my mind there is no question that when properly loaded and tuned, the mighty 90 VLD is not going to allow its dominance to be questioned by any lighter weight newcomers, even its own siblings from Berger. In my hands, the 90 VLDs have always been quite easy to tune in, almost stupid-easy. I am aware that other shooters may not have always been as fortunate when using the 90 VLDs, but as long as I can load them with insanely good precision and shoot them in matches without fear of losing jackets (i.e. using a 7.0-twist barrel and/or a 0.219" bore), I can't imagine ever switching to a lighter bullet, even another Berger offering.