In New York following Sandy this happened (emphasis added, and for more background go here):
The New York State Department of Financial Services and Governor Cuomo informed the insurance industry that hurricane deductibles should not be applied. The decision was based on the fact that the storm did not sustain hurricane-force winds while over New York.The State and Governor were incorrect in that judgment, according to the NHC. What that implies is that a bunch of homeowers (technically/legally) may owe insurers a refund (if they have already been paid for losses) or are owed less in payments under their insurance policies (if they have not been paid).
In practice, there is unlikely to be a large public demand for enforcement of the hurricane deductible, and insurance companies may decide that it is not worth their effort, as most had their losses covered by reinsurers. Will any reinsurers take a closer look? I'd guess it is unlikely, and if they did we might never hear about it anyway.
As a case study in the application of "evidence-based policy" you won't find a better one than the hurricane deductible. Sometimes "evidence" matters, sometimes it does not. Sometime we care about whether it matters, sometimes we do not.