All graphics are 7-day averages. Today’s raw reporting for each metric is in the tweet below.
(Data retrieved directly from state dashboards. Not every state updates its numbers daily.)
UNITED STATES
Today’s raw reported metrics:
- Tests: 1,209,792 (-34,995)
- Cases: 35,488 (-4,435)
- Deaths: 536 (-303)
- Currently Hospitalized: 36,470 (-375)
- Currently in ICU: 7,369 (-114)
(+/- compared to same day last week for Tests/Cases/Deaths & yesterday for Hosp/ICU)
The first thing you might notice is the lower positive testing percentage, including in the historical data. This is due my backdating (mostly) testing data for about 35-40% of states to include antigen tests where we report a state’s antigen-positive cases.
I could have tried to standardize by shrinking the numerator and removing antigen cases (and matching confirmed cases + molecular tests, but we’ve gotten used to the case counts where they are, so I chose to increase the denominator and include antigen tests instead.
Realize that there are limits to comparisons, (e.g., different types of tests yield different positive testing percentages). But because we use % percentage more frequently than just tests or cases in a vacuum, my intent is to match cases/tests to the best of our ability.
Also, where a state only reports confirmed cases, we stick to the tests that match if able. So I didn’t just increase all tests. If a state counts antigen-positive cases in our data, I added antigen tests where available. Just trying to match up each state's fraction—that's it.
Give me a bit longer to clean up the presentation on the raw data (and add notes where there were data dumps or removal of data that needs to be factored into any daily data). I’ll add a link to it in this thread.
I know a guy who has a team of folks who gather the same data from the state dashboards, but in leaner fashion and with minor adjustments to testing data to try and glean a more accurate positive testing % on a state-by-state basis.
I am wiped from sorting this data. Here's a to-be-improved version of the raw data (note the highlighted areas where I address pulling out large data dumps): docs.google.com/spreadsheets/d…
Oh, and "I am wiped" is code for "no regionals tonight."
But I'll provide a couple tidbits: (1) every region is now below 5.5% positive with the updated data, and (2) California is at an impressive 1.87% positive today (7-day average).
• • •
Missing some Tweet in this thread? You can try to
force a refresh
All graphics are 7-day averages. Today’s raw reporting for each metric is in the tweet below.
(Data retrieved directly from state dashboards. Not every state updates its numbers daily.)
UNITED STATES
Today’s raw reported metrics:
- Tests: 1,459,570 (+48,238)
- Cases: 50,782 (-8,576)
- Deaths: 1,059 (-612)
- Currently Hospitalized: 36,845 (-690)
- Currently in ICU: 7,483 (-86)
(+/- compared to same day last week for Tests/Cases/Deaths & yesterday for Hosp/ICU)
Trend lines still looking good overall. We’re still seeing some acceleration in the percentage decrease in deaths. And while hospitalizations and ICU declines have slowed a bit, they’re maintaining double-digit drops week-over-week for several weeks now.
Did you reply similarly to JHU as you did me here, or to the many (very prominent) accounts who linked to JHU’s data, when JHU put out its daily % positive for every single state by using CTP’s “totalTestsResults” while including antigen positives in case counts?
I'm sure you cannot have missed the ubiquitous use of the two columns in CTP's CSV/API (which CTP used for its topline daily data updates) to calculate and blast state-by-state positivity—including by media and policymakers—despite admonitions from CTP.
In GA, and FL, and AR, and TN, and TX, and LA, and MS, and others…this results in a massive error. Just saying, “We warned you against comparing states” while venerated institutions are doing just that and putting it out into the world like a shockwave is willful blindness.
But certainly CTP is and was aware that legions of people used its API/CSV data, and those that did almost invariably used the two columns circled in red. A simple, “watch out, this isn’t a good comparator” doesn't stop the world from using data derived from these columns.
I don't get why cases/tests didn't match up (either include probable & antigen or just confirmed & molecular) in the columns that pretty much everyone used (and which everyone knew that everyone used). I get that cases and tests are separate beasts, but we know people pair them.
All graphics are 7-day averages. Today’s raw reporting for each metric is in the tweet below.
(Data retrieved directly from state dashboards. Not every state updates its numbers daily.)
UNITED STATES
Today’s raw reported metrics:
- Tests: 1,955,408 (+188,288)
- Cases: 65,127 (-3,450)
- Deaths: 1,482 (-306)
- Currently Hospitalized: 37,535 (-792)
- Currently in ICU: 7,569 (-108)
(+/- compared to same day last week for Tests/Cases/Deaths & yesterday for Hosp/ICU)
Percent positive is falling nicely. Remember though, this was expected when we started adding in antigen tests in those states where we reported probable cases (which are almost all antigen positives). This weekend, I’ll provide a look at the “old” % positive just to compare.
I just realized this morning a large error in several states in the way that @COVID19Tracking (and now our team) reported many states testing totals compared to the way it presented cases.
It's another denominator problem, folks.
🧵🧵🧵
When available, CTP (and our team) reported probable cases as the preferred metric, which is defined as follows:
Poking around on the state dashboards, the overwhelming majority of probables come via positive antigen test.
But when it comes to reporting tests, we (and CTP, in its list of test totals that everyone uses) exclude antigen tests!
Data retrieved directly from state dashboards, and not every state updates its numbers daily.
All graphics are 7-day averages, but today's raw reporting for each metric is in tweet below.
UNITED STATES
The United States reported the following raw counts—
Tests: 1,191,529
Cases: 53,531
Deaths: 1,715
Currently Hospitalized: 39,841
Currently in ICU: 7,985
Quite a few small-ish states reporting larger death counts today: MO - 134, MN - 140, OH - 169, and OK 167. Of MN’s 140 deaths, 138 were part of an audit and occurred “over the course of the last year” and will be removed from today’s count. From the MN website: