BTW Utah

Record breaking day today for new cases in Utah (at 1,960)

Of course, Thursday was a record breaking day too (at 1,543).

As was Wednesday (at 1,498).

*************************************

But, it's the end of the week, and records on those days can be sort of normal as labs catch up on backlogs before the weekend.

*************************************

So it's probably much better to look at the number of new cases over the last 7 days (to mitigate the effects of day of the week patterns). Like so:

• Today was a record breaking day for cases over the last week (at 9,525).
• Of course,  Thursday was a record breaking day too (at 9,055).
• As was Wednesday (at 9,001).
• And Tuesday (at 8,773).
• And Monday (at 8,641).
• So gosh it's a good thing last Sunday and Saturday did not set 7 day records! But last Friday did (at 8,594).
• And the Wednesday before that.

*************************************

Now, I will be the first to tell you not to pay much attention to record numbers in trending time series. Records are part of what you should expect to see. 

But you shouldn't see a lot of them. And clustering of them is highly unusual. Further ... they should get rarer with the passage of time (just like, say, breaking home run records in baseball).

So let's set out a really simple forecast of what we should expect to see over the next week. Our 7 day growth rate has been a little on the high side of 10%, so I assumed 10% to be conservative. And I get:

• Saturday: total cases at 102,627; new cases at 1,118
• Sunday: total cases at 103,833; new cases at 1,207
• Monday: total cases at 105,118; new cases at 1,285
• Tuesday: total cases at 106,307; new cases at 1,189
• Wednesday: total cases at 107,807; new cases at 1,499
• Thursday: total cases at 109,504; new cases at 1,697
• Friday: total cases at 111,660; new cases at 2,156

Due note that I'm not doing anything sophisticated here. Anyone with a spreadsheet and minimal knowledge could do this. 

What's unusual would be acting surprised at the numbers, as Governor Herbert seemed on Friday. There's a word for that: blinkered. And I kinda' like Gary Herbert, and have been very pleasantly surprised by his whole term in office.

*************************************

In local news ... Wednesday was the first time I've seen a line at our testing center by the soon to be abandoned "old movie theater". 

OK, yeah, I couldn't resist that little bit of snark.

*************************************

Back to the data: most people are also not aware that Utah updates it's updates, but quietly. They have a lovely dashboard, which only shows the new numbers on a daily basis. But in the background, they update the old numbers as lost, forgotten, misplaced, and whatever results come in. 

And they don't show you that very readily.

But Wikipedia does. Funny thing those updates there: they always shows that Utah Dashboard has underreported its numbers. In statistics, making errors that are usually in one direction is called bias. It's a bad thing. 

So yeah, the Utah Dashboard is biased towards underreporting the severity of the numbers. 

Having said that ... not by much ... don't panic. It's just that people who really care about getting the data right make a point of getting rid of bias. I'll be waiting.