Over the past few days we’ve seen record high reported testing (over 629,000 on Friday) and the lowest daily death count since March 26 (358 yesterday).
I’ve been tracking 15 states, representing about 207 million of population, or about 2/3 of the United States. Of these, I model that 9 states have passed peak known active cases, and 6 states are still increasing. I’ve not been able to find a pattern based on social distancing or mandates. Why, for example, is California growing so much faster than Georgia? One factor is that 4 of the 6 increasing states double count cases (more on this in the individual state discussions). A more powerful commonality I’ve noticed in the 6 increasing states is that they have less active cases as a percentage of population than the 9 states had when they peaked at their maximums. This has me thinking that there might be a critical mass of COVID where the disease naturally peaks. So let’s take a look at this.
Here are the 9 states I’m following that have already peaked. This chart shows active cases at the peak as a percentage of the respective state’s population. I think there are two things interesting about this. First, I’m actually surprised about how tightly this clusters. Although from high to low it’s about 10 to 1, the data ranges from 0.03% to 0.35%. But why aren’t we seeing any states where it gets to, say, 1%? We’ve been told at various times that the disease propagates easily, and eventually we will all get it. So why are we seeing an upper bound of 0.35%?
The second interesting observation is that the high population density states are clustered to the left, with less dense states on the right. I’m now speculating that the level of peak active cases is dependent on total population and population density.
The horizontal line on the chart is the average peak as a percentage of the population for the states that have already peaked. I have left NY and NJ out of the average, as they are very high density outliers – NY because of NYC, and NJ as a whole has the highest density of any state in the Union. The Average of all 9 states is 0.14% of population, while the average excluding NY and NJ is 0.09% of population. The horizontal line is drawn at the 0.09% level.
Now, here is the interesting thing. Here are the 6 states I’m modeling that are still increasing. I’ve drawn the same horizontal line on this chart at the adjusted average above of 0.09%. I find it fascinating that, with the exception of Arizona, these states have not yet reached the level of COVID that the improving states experienced – and Arizona is close.
Could it be that these states will reach a similar level of COVID prevalence before beginning the decline? I think there might be something to this. I’m going to project peak active cases for these 6 states based on the adjusted average of 0.09% of population, and paint this line on the modeled active case graph for each of these states. It will be interesting to see what happens.
I’ll report again on Thursday…
As always, feel free to send me your questions about my assumptions, methodology, or modeling in general.
- Likely date of active case peak (Chalke modeling): April 10
- Likely date of peak deaths (IHME): April 16 (last revision on June 15)
- Total Test Results reported today: 449,448 (high)
- Total Pending tests reported today: 1,619 (extremely low)
- National reported case Growth Rate today: 0.89% (extremely low)
Shane Chalke Interviews
Groom Ventures has agreed to host a website that will archive my daily reports, and supplement with other commentary. John Groom worked at one of my companies back in the day, and is an excellent writer. The website is: www.howmuchrisk.com For those of you that post my daily report on Facebook, let me suggest you link to this site, as the direct Facebook posts don’t seem to copy the graphs.
Here is the national picture of active cases – It’s been rising over the past 6 days, primarily driven by FL, TX, and AZ. I’m still modeling less than 150,000 known active cases left. This number is high though, as many states routinely double count cases.
Here are the new reported cases nationally. Long, slow decline – dropping fast in many areas, and increasing in some.
Here are the daily death reports. This is looking far better than expected — We’ve had two days in a row of well under 500 deaths. This should be headline news.
On to the states. Virginia is looking great, now down 37% from the peak. The grey line is the average peak for states past peak (w/o NY and NJ). If NC peaks around that level it will fit in nicely with the other states. New reported cases have fallen 3 days in a row here, so it might be happening soon.
Arizona is all in the news lately. It is the one state I model that is already above the average peak as a percent of the population. AZ is currently at 0.10% of the population. It appears to show signs of leveling over the past few days. Arizona’s data is exaggerated, since they count specimens tested rather than people, and it’s not uncommon to get multiple tests if you’re sick – each time you’re tested you’d show up as a new case. Even worse, Arizona counts positive antibody tests as new cases. Therefore, Arizona is not as bad as the data makes it appear.
South Carolina is currently sitting right about on the average peak (the blue line). South Carolina also double counts cases, as they treat each positive test as a new case. Worse still, until June 11th, SC counted positive antibody tests as new cases.
Washington peaked at a surprisingly low 0.03% of the population, so it might not be over yet. It’s been increasing for 9 days, but lately shows signs of leveling off. The Covid Tracking Project recently recast a lot of cases to previous dates (you might remember this has been a significant problem with Washington data), so my curve looks a bit different, but presumably more accurate.
Florida has risen since May 28th. It’s interesting that Florida is still far below the average peak, so based on this it could have more to run. To reach the average peak, we’d need to see nearly 50% more active cases. Let’s hope not – we’ll watch it.
Both NY and NJ are in the end game, with a small percentage of COVID compared to their peaks. NY is now down 92% from their peak – I model only about 5,000 cases remaining. NJ is down 90% with about 2,600 active cases remaining. This is very small for these state’s populations, and especially small given where they’ve been.
California is now behaving as it should, cresting now. IHME has moved back their peak daily death projection to June 25th. California is still far below the average peak, but looks pretty flat lately. California data is exaggerated by the fact that they report all positive tests as cases. With testing more ubiquitous, some people with the disease are tested multiple times.
So it’s been tedious to continually adjusting for Massachusetts’ method of reporting historical cases. At the beginning of June, MA reported nearly 4,000 historical but newly discovered cases, which skewed the data. Now that we’re 2 weeks past this event, I’m now modeling based on the data as reported. That’s what that bump in cases in early June is all about on the graph. It’s not real. If something like this happens again, I may have to do the work of adjusting for it, but for now, I’m letting it flow through. MA is another state that double counts cases by reporting each positive test as its own case. MA is down 87% from the peak.
Georgia has been in the 4-6K range for 2 months. Here again, the numbers are exaggerated. Georgia counts each positive test as a case. To make matters more distorted, until May 27th GA reported positive antibody tests as new cases. Nonetheless, GA has seen a day to day decline in new cases for a couple of days now. GA is a state that peaked well below average as a percent of the population, so could have more to run. We’ll see.
…And here is Michigan. The COVID Tracking Project recently cleaned up Michigan’s data, which was driving me crazy. I’m now modeling less than 1,000 active cases left in Michigan.
Continued, almost straight line, decline in Pennsylvania. PA is now down 78% from the peak.
Texas is rising slowly but steadily. Texas is still well under the average peak (shown in blue). Texas is a low-density state, so I expect it to peak below the blue line. Texas also reports positive tests as cases, so is doing some level of double counting.
And finally, here is Colorado. Colorado is one of the states that has had aberrations in their data. I’m modeling that they are 74% below peak, but I’ve shied away from any conclusions in this state due to the data irregularities.
So that’s it for today. The numbers are very small as a percentage of the population. Unless you’re in a high density area, your chances of contracting COVID are very small. However, even though the probability is very small, that doesn’t help if you’re the one catching it. Everyone please continue to be as cautious as feel necessary.
–Shane Chalke, FSA