Case Projections
Sorry for bad news on Father’s Day.
A few days ago, I posted projections for a few Texas cities (compared to DC, NY, and CA). Those projections were created on June 13. Now that 8 days have passed, I was curious how accurate they were thus far.
Projections are incredibly hard to estimate. And many epidemiologists are hesitant to post or discuss them amid the risk of being scrutinized and losing the public’s trust. However, while all of this is true, I also think it’s important point of discussion in order to change behavior. So here we go...
These figures include the original June 13 projections. I added cases for the past 8 days in each county (see red dots and lines). The projections look pretty darn accurate so far. Dallas projections may be slightly overestimated (but not by much). Harris and Tarrant projections look like they are UNDERestimated. By a lot.
The only thing that can change these projections are the variables that were used to create them: 1) county-level population density; 2) testing capacity; 3) 3-day average of social distancing; and 4) temperature and humidity over the past 2 weeks. The public can only help through social distancing. A few counties in Texas, including Dallas, have mandated facemasks. It will be another two-ish weeks to see if this had an impact on spread.
We should all keep a close eye on these numbers.
Love, your local epidemiologist