Thread by @NateSilver538

US daily numbers via @COVID19Tracking:

Newly reported deaths:
Today: 1,651
Yesterday: 1,743
One week ago (4/25): 1,818

Newly reported cases:
T: 30K
Y: 34K
4/25: 36K

Newly reported tests:
T: 265K
Y: 305K
4/25: 271K

Positive test rate:
T: 11%
Y: 11%
4/25: 13%

The positive test rate, which I believe is a better metric than the raw number of + tests, continues to show slow, incremental improvement. But I do mean *slow*. It's consistent with a nationwide R of around 0.9, where R>1 means the epidemic is growing & R<1 means it's shrinking.

That is a national average, however. There are places like NY where there's a reasonably steep decline (R of perhaps 0.7-0.8). In turn, there are likely to be other places where infections are still growing (R of perhaps 1.1 to 1.3) and it isn't just an artifact of more testing.

Overall, there isn't a lot of room for error. A few states have had clear, sustained improvement and may have room in their R "budget" to relax restrictions. But for many others, even slight changes could bring R >1. Or R may be >1 already.

That's not to say it's necessarily up to policymakers. I don't care about the state capitol protests, which are an overplayed story. But it's clear from e.g. mobility data that people are moving around more and abiding less strictly by social distancing.

I wish we knew more about what this activity consisted of. Taking a nice, long drive to a non-crowded state park should be pretty safe, for instance. But pressure points seem to be outdoor activity in more crowded spaces, and small-to-medium-sized social gatherings.

Also, I wish we knew more about how much these activities have the potential to increase R. Some evidence seems to suggest that outdoors is safer than indoors, for instance. We'll find out, I guess, because I don't expect most people will stay inside by themselves all summer.