The point of this post is to estimate what fraction of the Fairfax County adult population is likely to be walking around in an infectious state. That is, for every 100 adults you encounter outside the home, what is the likelihood that you encounter at least one person who is capable of infecting you with COVID-19.

**The upshot is that, for Fairfax County, we are approaching the point where 1% of the adult population, at any one time, is walking around in an infectious state**. Right now, my best estimate is 0.8%, but give it another week or two, and at the current rate of growth, that will be 1%.

This is a pretty strong argument for continuing the ban on large public gatherings, particularly church services where singing will occur. Singing generates as much aerosol as coughing, and all it will take is one infected person who is a “superemitter” to generate a significant cluster of new cases. See Post #679 for a summary of this issue.

**This is an update to Post #624.** And when I compare the current and prior estimates, in the roughly four weeks since I did that, the number of disease carriers circulating in the population has doubled.

**The methods are laid out in Post #624.** Take the count of newly discovered cases for the past four days, inflate that to account for case growth over time, and add in a factor for likely cases that will never have an official diagnosis of COVID-19. And that should give you a reasonable guess for the fraction of the adult population that is current capable of spreading the disease, but doesn’t yet know it.

This does not account for individuals who never have symptoms, and yet may spread the disease for some extended period of time. I think the jury’s still out on that one. The pre-symptomatic population very definitely sheds virus and spreads infection. It’s not clear that the people who never have any symptoms do the same.

There are only two things I would change, from my prior analysis. Those are the labeling of my “best case” and “worst case” scenarios. These were estimates of the number of persons who had COVID-19 but were never counted as officially diagnosed with it.

**In the month since I did that least estimate, there have been several studies of the fraction of various populations with antibodies to COVID-19. ** Based on those, it’s pretty clear that in the typical case, the cumulative number of person who have had COVID-19 is, in fact, several times larger than the cumulative number of persons who were formally diagnosed with it. **My “best case” scenario now appears to be far too optimistic, and I’d say that my “worst case” scenario is about the middle of the pack, in terms of the results of these studies. ** I have remove and relabeled the estimates accordingly.