The Homa Files

This morning, the WSJ reported that — based on their analysis of data from Johns Hopkins University:

In the 24 hours ending at 8 p.m. Eastern time Thursday, 4,591 people have died from Covid-19,

The prior record was 2,569 on Wednesday.

Worldometer – my primary source for C-19 fatality numbers – reported 2,179 Daily New Deaths.

So, based on 2 credible sources, C-19 deaths yesterday either increased by 2,022 (to 4.591) … or, they decreased by 390 (to 2,179).

That difference between estimates certainly falls into the category of “statistically significant” … especially hours after a government plan is announced to re-open the country for commerce.

I have greater confidence in Worldometer’s data analysis, so I’m going with their 2,179 DND number (with more explanation of the black bar to follow later):

My conclusion: Yesterday’s Daily New Deaths declined slightly from the prior day. … but the 3-day moving average is trending up slightly.

But, there’s a looming question:

What the hell is going on?

=============

First, a confession.

I’ve been wrong on something all along.

In a prior post, I said:

Counting deaths — while a bit macabre — is a more reliable process than counting, say, the number of infected people.

Cutting to the chase, I’ve concluded that the most reliable number being reported is the number of COVID-19 related “Daily New Deaths”.

I certainly didn’t expect that , in mid-stream, the CDC would change the definition of countable deaths.

But, they did and NYC (following the CDC’s guidance) added 3,700 new deaths to their totals — they are an estimate of the accumulated number of “deaths possibly related to Covid-19” that had not been previously reported.”

The implication: new reporting of fatalities is inconsistent with the data from C-19’s inception until April 14.

For a complete discussion, see:
April 14-15 Change in US Data

=============

In essence, Worldometer analysts decided to book NY’s 3,700 newfound deaths on April 14 and restart their data series.  Here’s their daily chart:

Apparently, WSJ analysts booked the 3,700 differently.

On my DND chart (above), I simply backed out the NYC 3,700 (the black bar) to restore visual continuity.

What a mess.

Stay tuned. I’ll be trying to unscramble the rest of  the numbers I’d been reporting on the daily update..

Related Posts:

Flattening the COVID-19 curve

How will we know when we’ve turned the COVID-19 corner?

This entry was posted on April 17, 2020 at 7:25 am and is filed under COVID-19 Tracking. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.