You have a point, but from my perspective, there don't seem to have been any failures related to actually overstressing the MMM. In fact, I don't think anyone's really come close to stressing it; Castle themselves have run a a 1521 1D on the MMM with 6S which is an insanely powerful and current-hungry motor, and not had a failure.
You can't really compare mechanical safety margins to electronic margins though, IMO. The MMM is designed for a maximum of 6S LiPo, and will run at that voltage perfectly every time, ignoring any other problems. Its components can handle that voltage just fine, with the FETs rated for 30V and caps for 35V, but not any higher because it they don't need to be to run 6S reliably. The input voltage will never stray above 25.2V, wheras in mechanical designs, stresses may temporarily move above its "design-strength" and you have to take material tolerance deviations into account. Castle certainly know about mechanical tolerances; when testing their motors, even the very worst samples were able to hit at least 85,000RPM before the rotors went (with others reaching over 100,000). As aresult, Castle are only rating them to 60,000.
Current-handling ability is where you need plenty of headroom, and the MMM has buckets of that. Only one person has managed to get it as high as 170F when it died, and that seems to be a serious setup error (it would appear so anyway); the MMM shouldn't even thermal until 220F.
I'm not trying to argue or anything, but that's just my opinion.