Counties in CA

I know cases are skyrocketing in many places around the country. I’ve been wanting to present ALL the data everywhere, but it just doesn’t make sense to do it that way. So I’ve stuck to my plan, which is to reorganize what I’m doing in such a way that it helps make sense to all of you out there. You don’t need me to repeat the headlines blaring at you from everywhere. I feel my obligation is to subject the underlying data to some analysis, to help people assess just how bad the situation is in various localities.

Part of the reason I felt a need to reorganize my data is that I wasn’t covering some of the localities that I felt needed covering. What I did made sense a couple of months ago. But now there are a couple of things that have changed my perspective: 1) we have hundreds of local epidemics, not one national one; 2) comparisons of disparate locations are no longer as useful as they used to be; and 3) we’re not getting the information or the analysis we need from the CDC or many public agencies.

I’m not a reporter, or a scientist, or an epidemiologist. But I have received dozens of messages from people telling me that I am one of their only sources of information about the pandemic. This both scares me and fills me with a sense of obligation to help people make sense of this catastrophe that has befallen us.

So today I’m going to bring you up to date on the reorganization of county data, with hopes that you’ll comment freely on the changes, because I plan to implement this methodology with everything I’m doing.

One of the things that bothers me about typical media presentations is that they tend to concentrate on a single metric, e.g., cases, or fatalities, or testing, etc. In fact, none of these by themselves describe the situation. To the extent you focus on a single metric, you ignore other measurements that may be equally or even more important.

But the idea of a single metric is very appealing to our brains that clamor for an orderly story. So I’ve set about developing one, that I call the Z-score, named after my dog, Zorgi. The Z score is an index that includes the following individual measurements over a 30 day period:

  • Change in daily cases
  • Change in case doubling days
  • Change in daily fatalities
  • Change in the rolling 7 day fatality count
  • Change in fatality doubling days
  • Change in daily patients
  • Change in ICU patients
  • Ending HUR from CV19 only
  • Ending IUR from CV19 only
  • Change in daily testing (LA, OC, SD only)
  • Change in the positivity rate (LA, OC, SD only)
  • Change in Rt (States only)

My first attempt at assigning a Z score was purely based on gut feel for what was going on. Underneath that, I was mentally analyzing various factors to come up with a score. But I’m as prone to error as anyone else. It would be far better, I realized to come up with a programmatic way to assign a Z score, based on logical rules that any one of you could examine.

It’s not that I expect you to accept my interpretation at face value. But I think it might be valuable for you to know, for example, why I might assign LA county a Z score of 10.6, while SD county only has a 1.3. I don’t want you to think, “Oh, he lives in SD county, so he just wants to make himself feel better by assigning a lower score.” No, I’d rather you be able to look at the component of the score, critique them as appropriate, and then I can modify them if necessary. In the long run, the end result might be a Z score that everyone finds useful.

Tonight is my first cut at this. I’m still tweaking everything, so I’m not ready yet to post the rules that go into assigning a number to each metric. But at least you can look at the value, see the numbers that went into it, and judge for yourself whether it makes sense.

One thing you’ll notice right away is that the Z score is no longer on a 1 to 10 scale. Instead, I’ve left it open ended. I think the maximum score is somewhere around 25, but I haven’t figured that out yet. A score like that is probably what New York City would get at the height of the crisis.

Today’s tables and graphs:

  • CD4 – Final Z Scores for 10 Counties
  • CD1 – County Dashboard – Cases & Fatalities
  • CD2 – County Dashboard – Hospital Data
  • CD3 – Z Score Calculations
  • CVS1 – Imperial County Vital Stats Chart
  • CVS2 – Kern County Vital Stats Chart
  • CVS3 – LA County Vital Stats Chart
  • CVS4 – Orange County Vital Stats Chart
  • CVS5 – Riverside County Vital Stats Chart
  • CVS6 – San Bernardino County Vital Stats Chart
  • CVS7 – San Diego County Vital Stats Chart
  • CVS8 – San Luis Obispo County Vital Stats Chart
  • CVS9 – Santa Barbara County Vital Stats Chart
  • CVS10 – Ventura County Vital Stats Chart
  • VS3 – Southern California Snapshot

Tomorrow, if I have enough time, I’m going to try to do this for the states I’m covering. If I run out of time, I’ll just post the Snapshots with no commentary. Saturday, I’m seeing my son and daughter and grandchildren and their families for the first time in months. We’re going to be super cautious, and we really believe we’ll be OK. That means no update on Saturday, but I should be back on Sunday.

Stay well & healthy, everyone!

Many of you have generously offered to give me money, coffee, burritos, and other material rewards. I really appreciate that. If you’re still so inclined, I’d urge you to donate to the Equal Justice Initiative.

Leave a Reply

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

 

This site uses Akismet to reduce spam. Learn how your comment data is processed.