The broader point I'm making is this: hurricanes in the SE US and the Caribbean are, to a large degree "expected" disasters. As such they can be planned for, and damage can be minimized. And, I believe what you want to say is that they aren't all that "bad" anymore because of proper disaster preparation.
I would agree that hurricanes shouldn't be all that devastating in effect. But the data shows, quite convincingly IMHO, that we aren't doing as good as we could be doing in actually preparing for them.
When it comes to hurricanes that equates to selecting evacuation routes and locations to shelter the storm etc. If needed they may help create such shelters, but that stuff really seems to be working well. What happens after the storm is largely a question of how well other organizations have done their job. Aka what percentage of homes are habitable translates into how many people FEMA needs to find shelter for.
However, that last bit is a real question. Do we ramp up FEMA’s ability to build temerity houses or reduce the need for such temporary houses? I think it’s clear FEMA could do a much better job in disaster aftermath, but for hurricanes I think that’s largely solving the wrong problem.
PS: Also that extrapolation on death rates for Maria is extremely questionable as it’s been rising in Porto Rico for years. The increase from 2013 to 2014 is larger than the increase for 2016 to either 2017 or 2018. https://www.statista.com/statistics/580903/death-rate-in-pue... You can assume the only thing bad that happened was a hurricane, but the data really doesn’t support that.