For a really large country like USA, state-level data is more accurate than country-level data when it comes to evaluating length of epidemic spikes. It is conc

For a really large country like USA, state-level data is more accurate than country-level data when it comes to evaluating length of epidemic spikes. It is conceivable to have rolling spikes that affect different states at different times, but when summed up give the illusion of a never ending surge. County data is even better, but it's harder to assess at a glance, probably more prone to cherrypicking. The capitalized quotes are from the article itself. I'm pretty confident they are fearmongering the specific point that delta is super hyper dangerous for young cohorts.