> In fact during Apollo they even completely scrapped mathematical risk modeling because the results it always gave were basically 'you die.'

I've had a similar conversation with the "but if we really went to the Moon in 1969 why has it taken so long to be able to do it again" folk a few times.

The real answer is of course that we did it once, and realised that a project where about 99% of the failure modes are "astronauts turn into a rapidly expanding cloud of fried mince" and all of these failure modes are incredibly likely was not something we really wanted to do again.