Your First Ward Roving Reporter – Some truly strange data

Gird your loins Dear Reader! Because today you’re gonna get a real lesson in pandemic statistics and general causality! And the reason for this impending classroom session is, other than the news is not nearly as bad as it seems, I’m not quite sure what to make of yesterday’s Illinois coronavirus data.

Please allow me to explain!

It starts with yesterday’s 1,105 news cases. While that was certainly the largest 24-hour jump since we started tracking coronavirus data, remember, the number of newly infected people isn’t nearly as worrisome as the speed at which the disease is progressing.

The obvious answer to this semi-sudden expansion is, since March 24, the Illinois coronavirus testing capacity has doubled from 2,000 to 4,000 test kits a day. And if you look at our brand spanking new table, at first glance, the increased testing could be the case jump culprit.

Date     Total Cases     % Increase     New Cases    Prevalence   Total Tested

3/18     288

3/19     422                  46.5 percent         134             1 in 14    (2,000/day tested)

3/20     585                  37 percent            163             1 in 12

3/21     753                  29 percent            296             1 in 7

3/22     1,049               39 percent            896             1 in 2

3/23     1,285               28 percent            236             1 in 8

3/24     1,535               22.5 percent         250             1 in 8

3/25     1,855               21.5 percent         320             1 in 6

3/26     2,538               37 percent            683             1 in 6     (4,000/day tested)

3/27     3,026               19 percent            488             1 in 8

3/28     3,491               15.4 percent         465             1 in 8

3/29     4,596               31.6 percent      1,105             1 in 4               27,762

We crept up from a 1.3 percent mortality rate to 1.4.

But if you look a little deeper, while more test kits might be a factor, it ain’t the whole story.

As for the new table, I’ve added the number of new daily cases, the prevalence of the coronavirus among those tested, and the total number of Illinoisans tested. In retrospect, I shoulda been tracking the latter all along because, as the number of people tested increases, so will the number of cases.

“But Jeff! What the heck is ‘prevalence?’” I’m glad you asked!

Epidemiologists often apply the terms “incidence” and “prevalence” when discussing a pandemic. Incidence is the likelihood that someone in a specific population will contract the disease, while prevalence is the number of people within a specific population who HAVE the disease.

Prevalence is typically expressed in decimal ratios, but I’ve simplified it so our collective statistic addled heads won’t summarily explode. On March 28, of the 4,000 people tested, 465, or one in eight actually had the virus.

As you can see, the prevalence rate appears to be incredibly consistent – until yesterday.

But before I offer some explanations for that, please understand that our prevalence stat is fraught with peril. The Governor noted that the Illinois coronavirus testing capacity had doubled from 2,000 to 4,000 as of March 22, but considering the lag time involved in getting test kits in the right hands, I semi-arbitrarily chose March 26 as the 4,000 test start date.

Don’t forget that “testing capacity” does not necessarily equate to the number of people actually being tested, either

And I used the term “semi-arbitrary” because that March 26 date best fits the previously consistent data. With the exception of March 22, the prevalence rate has unerringly fallen between 1 in 6 to 1 in 8. But then yesterday, 1 in 4 tested inexplicably had the disease, which really doesn’t make much sense.

The possibilities include:

  • We’re getting better at testing the right people
  • The latest batch of tests are less accurate
  • The latest batch of tests are more accurate
  • It’s the kind of statistical blip that occasionally occurs
  • The disease has become more contagious

I have to say, I was worried about the 1 in 4 stat until I internally exclaimed, “Aha! It’s been five or six days since Mayor Lightfoot fumed over the overcrowded lakefront, and that’s squarely within the coronavirus incubation sweet spot.

But that theory was blown to bits by Chicago’s newest numbers.

Even after last Monday’s lakefront frivolity, the Second City reported a scant 93 new cases (1,975 total) yesterday. But there’s no way the entire state (including Chicago) could see a 31.6 24-hour percentage jump, while the eminently denser City of Chicago comes in at one-fifth that, or just 6 percent!

The only possible explanations for this disparity are:

  • Chicago isn’t getting their fair share of test kits
  • Chicago is underreporting their numbers
  • Testing is less accurate in Chicago
  • Testing is more accurate in Chicago
  • The disease has become less contagious
  • Something in the Chicago River imbues a unique coronavirus immunity

Though I’d love to go with the latter for all the obvious comedic reasons, one of my sources said Chicago is, indeed, getting the short end of the testing kit stick. But despite these more difficult and disparate numbers, when all is said and done, our trendline holds true!

Again, let’s harken back to the stock market for a perfect example of what I speak. Please note the dual “resistance” and “support” trendlines inserted on the stock chart below:

Trendlines

The resistance line is drawn across the peaks, while the support line is drawn across the valleys. Had I created that graphic, I would’ve also included the central trendline drawn between the other two. Among other potential factors, savvy technical analysts only consider the emergence of a different trend when the peaks remain above resistance or the valleys consistently sit below support.

Right now, our Illinois “resistance” trendline points include three percentages, 39, 37, and 31. And the fact that those peaks are, for now, decreasing in magnitude, massively reinforces the current trend, or the continued flattening of the Illinois coronavirus curve.

I would expect some sort of case number pop when Chicago gets a better handle on testing, but the odds of even that city undermining what we’ve seen so far are pretty slim. And once we integrate that inevitable bump in case data, unless something else changes, our trend will likely remain in force.

To wit, despite yesterday’s bump, we’re still number eight in statewide cases.

Meanwhile my rather intelligent friend Bill Wright (please don’t tell him I said that) duly noted that the national numbers are getting better, too! The number of new U. S. cases has actually dropped the last two days as has the number of coronavirus deaths. The shift is small, but it does portend better days.

Put more simply, it’s not nearly time to bet against the current Illinois numbers. What do I like to say? That’s right! Keep up the good work!

 

 

3 thoughts on “Your First Ward Roving Reporter – Some truly strange data

  1. You made my head hurt

    • Jim,

      Sadly, I knew I would, but the naysayers just don’t get it so I thought I’d fuckin’ map it out. And it worked because I’ve only received one negative comment today.

      Woo hoo!

  2. Do you take into consideration the amount of time it takes to get a test result back? My friend got their results after 5 days, which the Dr told her was the turnaround. If testing capacity increased on the 26th (ish), then it makes sense we would see a spike of new cases 5 days later. Right?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s