It seems to me you're conflating the probability of x happening in the future as assessed from some previous point in time versus the probability of x happening in the past as assessed from the present. Without clarification the wording is ambiguous but I think generally interpreted to mean the former. It would be quite odd to go with the latter as it's rather tautological.

> All of this is the long way around the barn to say that after the hurricane, the probability that Sandy hit Manhattan is definitely 100%

Nobody talks like this in real life. I can't even believe someone bothered to write multiple sentences to try and justify it.