Noting the 3 week periodicity of recent cases

We became aware early of the weekly periodicity of reported cases and deaths. THis appears to be an artifact of reporting. THere is a spike every 7 days in both reported cases and reported deaths.

I note now from here: United States COVID - Coronavirus Statistics - Worldometer a periodicity of about 3 weeks in the reports. I would venture a guess that these have something to do with Halloween, Thanksgiving, and Christmas; but I’m not really convinced. I don’t see a similar spike coming up related to new years, and halloween didn’t seem to be a popular holiday to break quarantine.
Anyway, has anyone else noted this? It could be just noise…3 data points is not a strong trend…but any thoughts on what is happening, other than randomness?

If I’m understanding what I think you’re saying correctly, that looks like it’s clearly Thanksgiving, Christmas, and MLK Day. (Not “new infections that took place because of those holidays,” to be clear, but “days when reporting is lagging behind actual cases because it’s a federal holiday and lots of people aren’t at work.”)