Skip to main content

Basic Case Chart, October 22

By October 23, 2021Commentary

Dave Dixon’s latest version of just cases, date of specimen collection, seven day running average to correct somewhat for testing variation, lagged a week to allow for completeness of a day.  What fascinates me, if this is the peak for this wave, is that the rollover looks just as sharp as the prior wave.  I had expected vaccines to result in a somewhat blunter shape.  One explanation is that there really is a 20/80 rule at work–20%, or fewer, of people who get infected cause the vast majority (look, Jan, vast majority meaning 80% or more) of transmission.  When you run low on those “superspreaders”, the level of transmission drops dramatically.  Maybe there are other explanations as well, but something is up.  And while I think meteorological factors must play some role, the timing of the waves in Minnesota is just odd, so that they can’t be dominant factors.

Leave a comment