[GreedClarifies]

I see that many have noted that this is not inflation adjusted, which would be a good start.

But more importantly, this needs to be as a percentage of the total infrastructure (or real estate to make it easier to compute) in the area.

IIUC as a percentage of total real estate value the amount of damage has been going down over time, as one would expect.

[flavius29663]

When you look at nooa data, it's not even a record, adjusted or not, 2005 and 2017 were worse by a lot. https://www.ncei.noaa.gov/access/billions/

[MattSteelblade]

It's not in total damage, but number of events, i.e. 25 this year so far.

[gruez]

>But more importantly, this needs to be as a percentage of the total infrastructure (or real estate to make it easier to compute) in the area.

>IIUC as a percentage of total real estate value the amount of damage has been going down over time, as one would expect.

You could plausibly tell a story with the absolute numbers though. For instance, if people are increasingly building and/or moving to disaster-prone places, that would still be bad, even if damage as an absolute % is lower. However, after reading the article and skimming the site it seems like the general narrative that they want to push isn't that, and is instead something along the lines of "climate change is real and is causing so much harm, look at all these disasters!". I believe in climate change and think it's causing serious problems, but I'm also against shoddy reporting, even if it's for an agenda that I support in principle.

[prometheus76]

Yeah but then you might not get that spicy headline...

[VincentEvans]

Wouldn’t adjusting historical values for inflation actually increase the magnitude of the numbers? Eg $100 in 1980s is $300 in 2023 (or whatever, not real numbers) - the number is larger, not smaller.

[flint]

A bit of battlefield preparation for the coming massive insurance rate increase for eastern coastal cities? perhaps.

[alistairSH]

Coming? It’s already started (FL and GA).

[phkahler]

Also, why the number above some arbitrary value and not just the total dollar cost?

[bell-cot]

If there are several different metrics which you could plausibly report for a situation, then you use the one(s) for which the measured values are the best fit to your agenda.

[occamrazor]

Because total dollar cost is dominated by major hurricanes, which are rare, therefore any underlying trend would be invisible under the stochastic noise.

Frequency at a moderate severity level is much more stable and can give better insight.

As other posters already pointed out, the severity of the events should be adjusted for inflation, building characteristics, and total exposure.