Though those numbers of course dont show the impact on future deaths (will they go down over the next year with the weakest culled), but at least the numbers are relevant, even if cherry picked to the worst hit area, it does show what the virus is capable of in a region with an older population and it's not pretty.

Also the absolute Madrid numbers (by far the biggest population of the examples cites) arent that bad (~50% higher than baseline deaths) but those are only one week peak numbers. I would be curious to see that data compared to say other catastrophes likeTsunami deaths in Thailand. Plus would be curious to compare the Bergamo numbers with Italy as a whole.

Interesting that there was really nothing happening until late February in Bergamo (below baseline deaths) and six weeks latter they peaked and now looks like they've gone totally past it (and the anecdotal serology tweet sent a couple days ago seems to agree).