So, how risky is a Thanksgiving gathering?

Let’s throw some math at the question…

It’s not a conscious thing, but these days, our brains are constantly running risk assessments:

  1. What’s the likelihood that I get exposed to COVID?
  2. What’s the likelihood that I get infected?
  3. How bad will it be if I do get infected?

Unfortunately, “the science” hasn’t been providing us with much useful “data to follow” on those questions.

So far, the best data is on question #3: How bad will it be if I do get infected?

CDC: 95% survivability rate if over 70 … higher with no symptoms, no co-morbidities or younger.


Those are pretty good odds, right?

Today, let’s look at question #1 — the likelihood of getting exposed to the virus, say, at a Thanksgiving gathering.


To start, let’s “reframe” the question and ask What’s the probability that at least one person at the gathering is infected?

The pivotal base case data variable is what’s called the “prevalence rate” — the percentage of people in the relevant population who are currently infected.

The best guesstimate that I’ve found is that the national prevalence rate is between 1% and 3% … most likely about 2%.


Let’s work with the 2% number and, for simplicity, assume that we have a T-Day gathering of 5 people drawn randomly from the national population.

What’s the likelihood that at least 1 member of the group is currently infected (and presumed contagious)?

Again, we have to reframe the question and ask: What’s the probability that none of the 5 are infected?

If there’s a 2% chance that a person is infected, there’s a 98% chance that they’re not infected.

So, for a 5-person group, the probability of no one being infected is a “chain probability”:  98% raised to the 5th power — 98% x 98% x 98% x 98% x 98% = 90.4%.

That means that there’s a 9.6% chance that at least one person is infected (100% – 90.4% – 9.6%) … and, possibly contagious.

Got it?

Let’s expand across the prevalence rate range and vary the group sizes:


For example, if you’re drawing a group of 10 from a high prevalence population, then there’s a 1 in 4 chance (26.3% to be precise) that at least 1 person in the group is infected.

Is that a risk worth taking?

Maybe yes, maybe no.

Having a possibly infected person at your party doesn’t necessarily mean that you’ll get infected.

It depends on your immune system, the infected person’s contagion severity, and the nature of your exposure.

For example, your risk is higher if:

  • You’re in the vulnerable class: old with co-morbidities
  • The infected person is drawn from a higher prevalence population, e.g. a known hot spot locale (think: spiking city or college campus)
  • The infected person is highly contagious (i.e. studies indicate that a symptomatic infectee is 2 or 3 times more contagious than an asymptomatic infectee)
  • You are in prolonged close contact with the infectee
  • Nobody is wearing a mask


So, the way I’m thinking about the risk:

Stay away to protect yourself if you’re vulnerable.

Stay away to protect others if:

  • You’re symptomatic
  • You work or play in a COVID-hostile environment
  • You’re a unbridled hugger, kisser or close talker

Otherwise, given the survivability numbers, I’m willing to take the risk.

Happy Thanksgiving!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: