This week:
Last week:
I just want to highlight, again, the fact that since you're comparing the day-of total with the backfilled total from the previous week, what you're labeling as 5k fewer cases than last Thursday is actually a 2.5k increase over last Thursday.
I know, I know, you're just doing it the same way you've always done it, but it bears showing that that method carries with it an artificial depression of the rate change. And in this case, it's obscuring the fact that we've most likely passed the inflection point into the winter surge.