Cost of having a large event in the realm of COVID-19 outbreak

I decided to write a short post explaining how to estimate the cost of a large event in terms of new COVID-19 infections —to consider whether it makes sense to cancel it or not.

Let’s assume we’re talking about a large contest for kids — let’s say, we expect to see ~250 kids and ~250 parents, all of them are from random parts of California.

As of today, there are 79 confirmed COVID-19 cases in California; combined with ~10 explosion of the number of cases in the U.S. over the last week and ~ a week of a no-symptom period for the infected ones, the multiplier for yet unconfirmed cases might be 10x or more. So ~800 people in California are expected to be contagious today.

The chances a random Californian is contagious is ~ 800/40M = 0.002% — seems small, but if you’re in a room with 500 people, there is a 1% chance one of them is contagious.

Let’s assume a single infected person infects 50 other people on such an event in average (that’s a rather pessimistic estimate, though if these people are always working in small & changing sub-groups, it’s totally possible). Taking into account a 1% probability of this scenario, the expected number of new infections is 50 * 1% = 0.5.

Now, let’s factor in the families — if you’re infected, it’s almost certain your whole family will be infected too. And in our example case:

  • the factor for the number of families per participant is 0.5 (most likely attendees are parent + child)
  • an average family size with at least one child is ~4.

So we get 0.5 (the expected number of infections on the event) * 0.5 (number of families per participant) * 4 (average family size)= 1 new infection as the expected cost of the event, or, taking into account the mortality rate, 0.04 human lives — not counting any chain infections (except for the family members).

In other words, it’s an event with ~1/25 chance of fatality.

CTO @ ServiceTitan.com, creator of https://github.com/servicetitan/Stl.Fusion