2. If you do an analysis of the number of flights per year, and the number of uncontrollable engine/airframe fires - particularly those that end in a pax evacuation, I suspect that you would find that it is a discussion that isnt worth having (emphasis added)
Actually, the situations that require the most discussion / training are the low frequency, high risk ones.
When something happens
often Recognition Primed Decision Making (we've done this many times, therefore we know what to do) makes it unlikely that we will flamingo up (like the c*ck up specified above only worse).
When something happens that we have never dealt with before, there is no RPDM and therefore it is far more likely to be flamingoed up. This is why training / discussion should be concentrated here.
The other half of the equation is high risk. If you try to cover all the low frequency events without regard to their merit will result in an astronomical training program doomed to failure. An event that could lead to fatalities is the epitome of high risk, and therefore requires training.
This is why discussion and recurrent training should be concentrated on
high risk, low frequency events like fires, and not relegated to "it won't happen to me". They do happen, and because they don't happen that often, we need to think about them.