It sounds like you're arguing for a "Darwin of the gaps." If it's an unknown process, then how do we know it's a process?
Well that's the issue that ID keeps pounding away at. The process is unknown.
But that doesn't mean at the same time that you can decide how probable it is. If it's unknown, it's unknown.
I like the "Darwin of the gaps" phrase though. I'm going to steal it. ;)