2020-9-15

This was originally a Facebook post of mine from 2020-9-15, and is archived here as a curiosity. Minor changes to formatting, as well as basic copy-edits, may have been made in the transition from Facebook post to web format.

Today, the United States reported 35,445 new cases of the novel coronavirus, up from 22,548 a week ago (Covid Tracking Project). If you look at the seven-day rolling average, that's at 37,049, up from 36,649 a week ago.

We reported 1031 deaths today, up from 362 a week ago, and the seven-day rolling average is at 844, up from 724 a week ago.

Both the case counts and death counts show a sharp upward move the last few days on the rolling average graphs. Since the normal motion is for moves in cases to be followed by moves in death about a month later, a simultaneous move in both is most likely going to be an issue with the data, not an actual issue with sick people. And since we're exactly a week behind the artificially low numbers of the Labor Day weekend, it stands to reason that we'd be seeing what looks like a bump upward as data collection gets back to normal.

Only after another week or two should we have a better idea about whether cases are still really dropping. But hospitalizations are still down week-over-week. Right now we're reporting about 30,300 hospitalized patients, compared with about 32,300 a week ago. And the testing positivity rate, as measured along a rolling average by JHU, is down from 5.2% to 4.8% over the past week. So some indicators are still tracking in the right direction.

The usual graphs are attached.


This page is released under the CC0 1.0 license.