William and Mary added another 200 tests and ten new positives to their COVID-19 dashboard count on Monday.
I’m not quite sure what to make of that, but I’m going to take a guess. While it superficially looks like bad news, I’m guessing that it really isn’t. I’m guessing that this most recent Monday figure should best be compared to last Monday.
What’s the reasoning behind that? In a nutshell, I’d bet that Monday has three days’ worth of bad news rolled into it.
First, it sure looks like they’re scheduling their “census” testing as routine, Monday-to-Friday work. After that first big round of tests, there’s now a well-established pattern of processing a couple of hundred tests a day. And, apparently, there’s no testing over the weekend. (Otherwise, you’d see a big lump of those weekend test results show up at some point.)
Second, by contrast, I’d bet that the testing of symptomatic students is done on an as-needed basis. If somebody walks into the student health department and says, I think I have COVID, I don’t think they’re going to tell them to come back in a few days to get a test. And, I’d bet that any contact-tracing testing is also done on an as-needed basis, rather than scheduled M-F.
If I’ve guessed right, then what you see on Monday is one day’s worth of routine, student census testing, and maybe three days worth of on-demand tests for people who are symptomatic or need to be tested due to known exposure (contact tracing testing).
The upshot is that when Monday’s testing comes in at four times the positivity rate of the past few days, I don’t think that’s reason to panic. Yet. It might signal that the outbreak is not yet contained. Might.
But if I had to bet, I’d bet it’s an artifact of the testing schedule. If I’m right, then Monday’s data reflect three days’ worth of testing-for-cause, compared to one day’s worth on the other four days of the week.
I guess we’ll find out tomorrow.
Addendum: Somewhere in here, I need to acknowledge the possibility that this might be occurring purely by chance. Duly acknowledged. With counts this low, you could get some significant-looking day-to-day changes that are merely the luck of the draw. But I’m going to take them at face value, and ignore the possibility that I’ve just wasted my time (and yours) trying to explain totally random variation.