This is actually a good example, because it demonstrates how "rational" people consider (based on the debacle we just went through) probabilistic predictions to equate to binary truth. Except the problem is: if that's how one thinks, one can't see the error in it.
Epistemology and non-binary logic are not just complex, they're counterintuitive.
However in this case, your excellent "and if not why not qualifier" could possibly save the day...but then it comes down to whether the judge doesn't get it wrong.
But who is qualified to be the arbiter of Truth when it comes to the reasoning?
Truth and agreement at the abstract level where we're working right now is easy, but the move to the object level is anything but (even though it seems easy).
I don't think that has to be the case. There are a lot of objective standards you could apply and tests you can do.
For example, as a somewhat bad example, ask people if they trust vaccines or not, and if not why not.