> except for that ~20% reduction in R at best And following up to note the above is wrong--assuming heterogeneous spread (most patients infect no one, but a fe

> except for that ~20% reduction in R at best And following up to note the above is wrong--assuming heterogeneous spread (most patients infect no one, but a few super-spreaders infect many), contact tracers will naturally find those super-spreaders disproportionately, with disproportionate benefit. I still doubt that's the dominant difference, though.