GA Tech Release Calculator to Estimate Risk of Covid-19 Exposure at US Events

AFAICT, the site computes the risk that at least 1 person out of a group of N people will carry the virus. It does this by estimating the percentage of the population likely to be carriers.

I think it overestimates the number of carriers. For SC, it is estimating 400K carriers, or about 9% of the population. I think that estimate is absurdly high. For comparison, SCDHEC is reporting 350K cumulative cases (58K reported).

The number of carriers out in public is not the same as the number of active infections. If you know you're sick, you're going to stay home and not be at an event. So what we really want to know is how many of the active infections are presymptomatic. We can do this by forecasting the number of cases over the next couple of weeks, and work backward using a distribution for the incubation period (very roughly: 0% 0-2 days, 25% 3-4 days, 25% 5 days, 25% 6-7 days, 25% 8-11 days). Then multiply by an under-reporting factor (which may have been 20x or more early on, but is likely much smaller now that we have widespread testing).

More information about the project can be found here: https://cos.gatech.edu/news/georgia-tech-researchers-release...

The site appears to be suffering from the influx of users right now.