Continuing to Rise

The daily case numbers have continued to rise at the national level.  In my last post I mentioned a large spike in the data from Texas.  However, national numbers have continued to rise even since that time.  We are about 1,000 cases per day higher than we were just after that spike a week ago.  Here’s the national graph.

Overall, about two thirds of states have rising numbers and many are at or above their summer peaks.  Colorado is about at the same peak as during the summer surge.  So are Arkansas, Kansas, Massachusetts, Missouri, North Carolina, Oklahoma, and Rhode Island.  Kentucky is higher than it was during the summer, as are Minnesota, Montana, Nebraska, North Dakota, Puerto Rico, South Dakota, Utah, Wisconsin and Wyoming.   New York and New Jersey have both started to rise after being steady through the summer.  Montana, South Dakota, Utah, Wisconsin, and Wyoming all have alarming rates of increase over the past couple of weeks.  Here is Utah’s graph to show what I mean, I’ve chosen it as an example because I have a lot of family and friends there.

With that said about the overall trends, it’s worth noting that there are some oddities in the data.  

First, the daily reporting has been more chaotic than usual.  I’ve already mentioned the large spike in Texas on the 21st, but other states have seen similar, though much smaller, spikes.  The last two weeks haven’t followed the usual 7-day cycle that we have seen in the past.  I don’t have an explanation for why this is, and it may not even need an explanation.  There’s always a certain amount of noise in data, and the amount of noise in a particular stretch can change at random, a kind of meta-noise if you will.  Still, the last two weeks have been noisier than usual, and it’s worth keeping that in mind in interpreting the numbers.

Second, the shape of the rising curve is unusual.  The spring and summer surges both started with exponential growth – a curve that is cupped upwards, growing faster and faster.  The curve of the last two weeks is cupped slightly downwards.  Here again, I don’t have an explanation for what circumstances may be causing this.  The type of curve typically indicates slowing growth; we’re used to seeing it near the end of a surge, not at the beginning.  Of course, it might not mean anything, just an artifact of the noise that I’ve already noted.

So the upshot is that we’re definitely increasing nationally, and some states are increasing alarmingly.  I don’t know what this means for the future.  Here is Excel’s best guess for the next ten days, but the uncertainty in this projection is higher than usual.  For example, by October 7, Excel thinks that cases could be 130,000 higher or lower than this and that deaths could vary by 6,000 either way.  That’s a lot of variation for ten days.


Leave a Reply

Your email address will not be published. Required fields are marked *