I don’t really agree. If I have two choices to make a long distance trip, drive or fly, it is safer to fly. If I’m going to the grocery store, there’s no option to fly, so using those type of trips in the calculation doesn’t make sense.

If we talk about the safety of cars vs planes, we should really only be considering trips of a distance where planes are a viable option. Even then a trips per crash seems like a far worse metric than miles per crash. You want to account for complexity of the trips still.