Grounded

It's like November 2016 all over again.  With each morning a fresh sprig of existential dread as I check the overnight news on my phone.  Today I learned that all of Italy, not just the northern provinces, is now on full quarantine, as their total number of cases approaches 10,000.  The financial markets got clobbered again yesterday – the S&P losing 8% and SAN.PA dipping below 80 (my online sell order for 85 thus blithely ignored).  My better half insists it's time to buy gold.

My company’s “non-business-critical” international travel ban is still on, so I continue to work from home at least until the end of March.  Working from home is what I do anyway when I’m not traveling, so it’s not too a big cramp in my style, so to speak.  Ask me again in four weeks.

The left-leaning media outlets in the US cannot resist engaging in some schadenfreude around Trump’s arrogance and ill-preparedness for all this.  There are 729 confirmed cases in the US as of this morning, but since test kits have been in short supply, one suspects that this is a gross underestimation, kind of a negative ascertainment bias. 

The Economist had a good piece in their “Graphic Detail” section (the one in the back near the obituary) analyzing travel patterns between other countries and China to predict how many cases a country should have versus the reported cases.  The US seemed to be underestimating only slightly, while Russia (“nothing to see here!”) should have many times more cases than they’re reporting.

Seeking succor in probability theory, I spent some of my Saturday fiddling with Excel to calculate the number of cases per capita by country.  A problematic exercise given the testing issues mentioned above, but I thought this might provide some context and clarity.  Here it is, sorted by cases per 100,000 inhabitants (cases as of Mar 10 am CET):

My Excel defaults to European decimal style, so the number for Italy is one-point-five cases per 100,000 inhabitants, not one thousand five hundred.  So one could look at this and conclude that there’s nothing really to worry about: you would have to shake hands with around 65,000 (100,000/1.5) Italians to encounter one infected person.  The current situation in the US – only one case per 14 million people – seems like nothing to lose sleep over.

Of course, this kind of logic crumbles under the slightest scrutiny.  In the Italian scenario above, there’s no way of knowing whether it will be the 65,000^th or the 1^st person you encounter who’s infected.  And of course it’s not the number of confirmed cases; it’s the number of carriers, most of whom are asymptomatic, out there, somewhere.  The numbers rise faster than we can keep track of them.  In the time it took me to write, edit, and post this entry, the number of cases in Switzerland went from 374 to 485.  My alma mater just told all its students not to bother coming back from spring break.  I don’t believe I will be seeing the inside of an airplane for a while.