The revelation that it was not the pilots of the Lion Air flight prior to the one that crashed in Indonesia who saved them (by killing off the trim circuits), but an extra pilot deadheading in the jump seat suggests there must be something pernicious in the programming making resolution by pilots on their own difficult.
The answer could lie in the fact that the impact of the program is incremental–something that the FAA people and overseas authorities never knew.
The effect on the trim was believed to be 0.6 degrees maximum. However, it was compound, in that at each juncture a further 0.6 degrees was added until an incredible maximum of 2.5 degrees was reached.
From an official certifications (both for FAA technical people and abroad), this would not have been acceptable, especially with reference to a single angle of attack sensor.
FOR THE PILOTS at the coalface the insidious incremental nature of the trouble would easily catch them unawares because:
First there would only be small nose-down inputs with which he or she would be sure they would be able to cope.
It would then gradually become more and more difficult though seeming still quite possible deal with.
Then without time to think about the need to kill the trim circuits (according to training for runaway trim), maximum downward trim would apply and the aircraft would be plunging, with the pilots too desperate in the remaining seconds to think. Hence the value of a third person.
If the MCAS program had applied the full down trim at the beginning the pilots would have had more time and height to deal with it. On the other hand why allow such extreme downward trim without double checks it is required?