Conversation

One challenge of COVID-times is that everyday activities vary widely in riskiness, but it's hard to make decisions based on that while living one's life. So we mostly adopt blanket policies. This fascinating project offers a more nuanced understanding of marginal risk. twitter.com/microcovid/sta

This Tweet is unavailable.

Andy Matuschak

@andy_matuschak

"Can I have an indoor dinner party with six asymptomatic friends?" How should I think about this? One way: I'm ~10x more likely to get COVID from that activity as from grocery shopping. If I feel OK about shopping for groceries 1/wk, a dinner once per quarter might not be crazy.

9:46 PM · Aug 29, 2020Twitter Web App

Replying to

The microCOVID project suggests an interesting angle: set a risk budget (eg 1% chance of infection / year). If 1 µCOVID = 1/million chance of infection, I can "spend" 192 µCOVID / week. Then it provides a calculator to help estimate the risk of everyday activities, in µCOVIDs.

Andy Matuschak

@andy_matuschak

Aug 29, 2020

Knowing my "budget," and using the calculator, I can stop worrying about masked park hangouts, since they're around 1 µCOVID. I could do that every day and "spend" only a fraction of my budget. Outdoor dinner? ~100 µCs. So once a week at most.

Andy Matuschak

@andy_matuschak

Aug 29, 2020

It's important not to be too literal about any of this, of course. The estimates are inexact; you can't aggregate "average risks" when tail risks are what matters. But I find this a helpful way to understand relative risks. I think it'll help me make less blanket self-policies.

Friedrich 해민 Markgraf

@fzwob

Aug 30, 2020

Replying to

@andy_matuschak

Still doubling your risk from shopping though, so a better question to ask yourself would be „would I be OK with going grocer shopping twice as often“. Still a great project/tool! Just shows reasoning about this is hard.