Superspreader Individuals or Situations

Megan McArdle offered this thought in a thread commenting on an elaborate analysis of probabilities (too elaborate for me to even try to follow):

Over the long term, I think this stuff matters less; if you remain open, in any moderate population center, a superspreader will find you. But early in the epidemic, a random walk will make some places look smart and others look dumb, purely by chance.

— Megan McArdle (@asymmetricinfo) May 4, 2020

I can venture to comment on this, however.  The assumption here seems to be that "superspreading" is a function of an individual (think Typhoid Mary perhaps, but not necessarily asymtomatic).  That's certainly been my understanding from the past. 

But in the context of this new pandemic, I ran across an interesting report by someone who tried to assemble worldwide reports of mass contagion and then to analyze common features. I may have mentioned this before.  The features were crowds plus intimate contact and/or a lot of vocal activity--cheers, shouts, etc. 

One would think we could rely on people to avoid such situations, although when you look at the rallies protesting against lock-downs you have to wonder.  But in principle avoiding such situations is easier than identifying potential superspreaders.  It's likely unknowable currently to determine the proportion of total infections occurring from superspreader individuals, versus crowd contagion, versus individual contact.