19 comments

[ 4.7 ms ] story [ 25.8 ms ] thread
Great resource.

I'd love to see per-country graphs for new infections, recoveries and deaths.

Looking at the world graph it appears that the infection rate is slowing down on March 18th. I hope it's true!

It's been that way every day for the past week, presumably due to reporting latency.
Aside from that, it's very important to remember that this dashboard shows confirmed cases, not infections. Trends and patterns are as much about testing/reporting as about the virus itself.
Total number of confirmed infections is a misleading metric. Criteria for testing constantly changes everywhere. Test kits become more (or less) available. Hospital resources are allocated differently based on changing situation, etc. Two random examples:

* March 15th, Southern Finland - "Coronavirus testing to focus on health professionals. … people returning from trips abroad will not necessarily be tested, let alone all of the people suffering flu symptoms." https://yle.fi/uutiset/osasto/news/hus_coronavirus_testing_t...

* UK NHS now recommends simply staying at home and not even contacting them in mild cases. https://www.nhs.uk/conditions/coronavirus-covid-19/self-isol...

They update it throughout the day, check back at midnight for the final count for the day.
I have a cron job making graphs from their data,

https://tonybox.net/posts/covid19/

Cool. Maybe fit a logistic function instead of an exponential, though :)
How do you do that if the growth rate hasn't started to slow yet?
It's called a semi-log plot. X is linear time, Y is log population count. The slope of exponential-like acceleration phase indicates the growth rate.

PS: this type of data is much better modeled by relatively simple ordinary differential equations because the rate of change of a communicable disease is proportional to the number infected.

I setup a 301 redirect, directly to the Johns Hopkins dashboard using a common nickname for coronavirus:

https://TheRoni.com

Really cool work! Would be interesting to fit a logistic as well to see if any case curve flattening can be seen. I plotted the points for Italy and it seems like the growth rate is decreasing.
Hug of death on the images
This covid-19 sheet was publised recently by John Wilkinson / critical care Northampton https://my.visme.co/projects/4d80743j-covid19-sheet#s1
This page has some usability issues:

(1a) It requires me to horizontally scroll rather than re-flowing anything to fit my browser window width (1114 pixels).

(1b) It defeats the ability to scroll left and right using arrow keys. On some pages, left clicking in some text will restore that, but not in this case.

(2) It defeats the ability to vertically scroll using Page Up and Page Down keys. Arrow key up and down still work as do Home and End. And like in (1b), left clicking doesn't restore it.

May all that is just a limitation of the visme.co site, though.

Why is there only one case in Russia?

As well India which has a bigger population and like Russia is closer to China then any North or South American country.

This virus is suspicious!