This was originally a Facebook post of mine from 2020-7-21, 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 62,749 new cases of the novel coronavirus, down just a shade from 62,987 a week ago (Covid Tracking Project). The graph looks a little bit like maybe things are leveling off. It's still a bit early to tell, but we've gone four days without breaking a new record in cases, so at the very least that's better than it could have been.
But there is another reason to suspect that maybe the spread is stalling a bit again. And that's the positive test rate. I've attached a graph that shows the positive test rate over time, and between something like June 10th and something like July 10, it doubled. It went from about 4.3% to about 8.5%. So even though we were doing more tests, a larger fraction of those tests were coming back positive. This is how we know the increase in spread was real, and not just a side effect of us doing more tests.
However, for the last ten days or so, the positive rate has been basically flat, with the rolling 7-day average figure staying between 8.5% and 8.7%. In other words, the last ten days haven't seen things improve, but they haven't gotten worse either. Presumably people are being more careful.
On the deaths side of things, we reported 1029 deaths today, up from 736 a week ago. The deaths keep marching up, with them trending up 13 of the last 15 days, measured week-to-week. If tomorrow we discovered a simple trick that stopped the virus from spreading to even a single person, ever again, it's entirely possible that deaths would keep climbing for the next few weeks, just from the people who've already gotten sick. So the deaths are bad news, but they're not a great indicator of what's happening right this moment.
The usual graphs are attached.
This page is released under the CC0 1.0 license, except for the graph on testing, which belongs to Johns Hopkins University.