Boy, this thread looks to be taking off (pun intended).
However, derated thrust settings, if available, are allowed and may allow an increased performance limited TOW.
We know what you mean but better to cite "handling-limited takeoff". With derate, you may get a benefit from a lower Vmcg or Vmca which allows for a lower V1 or Vr which can provide for a better runway-limited set of figures. Perverse situation but there you are. I am guessing that, in most cases, the benefit will be from a reduction in Vmcg. All depends on the particular Type's performance numbers in the AFM.
Dare I ask, what's the logic behind forbidding one and allowing the other?
Derate is philosophically the same as unbolting the original engine and bolting on another, lower thrust motor. So why would we not adopt the same attitude to takeoff.
whereas Flex/assumed temperature cannot take that into account.
My take is that it is a case of philosophical navel gazing. Flex is only a money-saving animal, rated is what you are gifted by the OEM. (Or what you pay heaps for to get the gift for a derate option).
a pilot by his own mishandling can find himself outside the certified territory
Certification stuff is a set of presumptions, one of which is that the crew is competent to do whatever is required to be done. There are no guarantees in certification, just probabilities. Generally, things go reasonably fine as evidenced by a very low mishap rate overall. If, on the day, the pilot is incompetent, the bets are off.
firewalling the levers poses no risk
Danger, Will Robinson. Cavalier firewalling can introduce some undesirable dynamics into the handling. I was involved in a fatal mishap investigation years ago where the crew lost it principally due to firewalling the good motor. My take is, leave the throttles where they are unless you are sure that you are going to hit the hill.
Again, none of this stuff comes with an iron-clad guarantee that it is all going to work according to Hoyle ..... all you can do is load the dice in your favour.