I think the difference is just, that SpaceX does big steps and big RUDs, which make it hard to pinpoint the causes, while Blue Origin tries smaller steps and less RUDs for the sake of isolating problems better.
Yes, but SpaceX has put a lot of effort into sensors and telemetry so they do have a LOT of diagnostic data even in a catastrophic RUD. They were able to pinpoint the locale of the in-flight strut failure on CRS-7 with that data. They sorted out AMOS-6 with that data. AMOS-6 in particular was an unusual failure caused by thermal buckling of the tank liner that was partially due to tank filling procedure. No amount of incremental testing would have likely caught that failure as it was due to the time-dependent interaction of several systems.
If SpaceX was having RUDs for basic underengineering, that would be one thing, but their RUDs have (to date) caught some legitimately difficult gremlins. I fear it will be a long time before Blue Origin gets to that level of intimate experience with their hardware.
