16 comments

[ 0.21 ms ] story [ 1023 ms ] thread
I quickly hacked a website together last weekend based on the COVID Tracking Project data because I wanted to see how much testing was being done and where.

I think that it's pretty interesting to sort by the columns and browse the data.

The site's not great on mobile, but I'll be working on that this weekend.

http://VirusTracking.net

...and I put the wrong URL in the submission. The correct URL is http://VirusTracking.net

If a moderator could change the URL that would be great.

Sure thing. Fixed now.
(comment deleted)
I really like that you can see the doubling rate.

Would love to be able to drill down by county but that would likely be a PIA to do as would most likely require finding and integrating lots of different data sources.

I've got a bunch of improvements that I want to make on the site (mobile version coming soon!) and once it's looking good, I might try to tackle some drilled down hyper-local data.

Thanks for the suggestion, I've added it to the list.

Cool site, but can you fix the numbers for Virginia? Currently shows only 19 confirmed infections, but e.g. on [1] it's 604. Spot checked some other states and the others look correct.

1. https://studylib.net/coronavirus#country-us

Same for Maryland. Current number tested is much higher.
Yeah, I've seen that Virginia's numbers are strange. I'm just using the data from the Covid Tracking Project[1], and they seem to be scrubbing the data pretty well, but for some reason they haven't gotten Virginia's right.

I'm trying to not source the data myself, since they seem to have a good methodology that they generally follow. I'll ping them and see if that have an explanation.

Another issue is that I need the historical daily data for Virginia since the site is showing the time series, so I will have to find that specific information too.

Thanks for checking out the site, and stay tuned for some improvements to it.

[1] https://covidtracking.com/

Virginia's data has been fixed now.

I was also able to hack together a mobile version of the site this weekend, so you might check that out too, although the desktop version makes it easier to compare states.

Unfortunately this, just like many others, is inaccurate because it is only based on reported cases and so far America hasn't tested very many people
Yes, they're only reporting on cases they know about. They're not reporting on cases they don't know about. Thank you Captain Obvious, once again you save the day!
Maybe could make a site that tracks the Bayesian probability of infection rate? (not 100% sure if I'm using those words right)
If you read the introductory information at the top of the site (pasted below), I totally agree with that the amount of testing is VERY important.

It's why I show the daily testing numbers and also calculate the RATE of testing for each state, so there's at least a reference for how states are doing comparatively.

---- Why is having accurate stats on the number of tests so important?

By definition, a confirmed case is one confirmed by a positive test for COVID-19. If your state isn't doing much testing, then you won't have many confirmed cases. However, there could still be a high number of people with coronavirus in your state, they just haven't been counted yet.

Yeah, even corrected for tests/capita, Cuomo/De Blasio have really botched the situation in New York.