Travis did this:
It’s death by reported date, not occurrence, fwiw
Nice, but I’d rather see excess deaths, since “COVID deaths” seems a bit too biased, one way or the other depending on the state.
And, yeah, reporting date really screws things up, bunching deaths on Mondays and Tuesdays.
What colors would work best for you? 
1 Like
Last of ethnicity/race breakdown for now
a graph:
NY and NYC have much higher excess mortality than the other major states
And CA was not.
So far.
That first wave – the only real COVID wave that hit NYC so far – was huge excess mortality in general. No other state/location that I have data for (in the U.S.) had anything that bad.
Global data set:
links to this paper:
https://science.sciencemag.org/content/371/6527/347
Just noticed the CDC made a new page (in addition to the ones there before)
Hmmm, I gotta set up my embedding stuff better… or have a default image when I don’t have an image for a story.
With my old eyes, I’m not fond of the color choices here, but here’s some slice-and-dice for U.S.:
Thanks, very nice overview of how the pandemic has hit the US. Oddly, my older eyes like those colors – I found the graphs pretty intuitive to read and the colors helpful in rapidly interpreting them.
In case people want to hear my critique of the dataviz:
Some of the graphs (such as the stacked column for % COVID deaths by race/ethnicity) worked just fine… others, not so much:
If you’re going to use a black background, then you need to not use a dark purple. Is that too much to ask?
Interesting. That whole page was much brighter on my Mac than it is on this work PC I’m using right now. I see your issue here. But on the Mac it actually looked really good. The color gradations clearly showed the relationships, and all the lines were easy to see.
Wondering what “share of cases” and “share of deaths” is. I mean, there are nine lines, none are higher than 7%, so, not a percentage share of the total.
Just keep it simple: excess deaths per 100000. Yes, that means explaining what the “expected” is.
I’m going to watch your critique, so I don’t get ninja’d by you.
You are correct: LA County is nearly twice as populous as the second-most populous (Cook).
I agree: take population by county, use the same circle-size chart format, and it would look about the same.
A few more videos:
