I'm saying that isolated outbreak locations themselves don't seem to be accelerating exponentially. As though there's something outside mitigation measures that's slowing spread. Obviously if you aggregate the entire US, the curve will look like it's plateauing.

Here's his explanation: https://www.youtube.com/watch?v=8aHrx68IT7o

"isolated outbreak locations themselves don't seem to be accelerating exponentially"

I don't understand what this means. Different locations are different. Specific areas are plateauing or accelerating, there is a divergence, and the result is that overall, the epidemic is accelerating again but at the same time, the location of the growth has rapidly shifted.

You may assume it's causally unrelated to lockdowns if you like, but the growth has shifted away from the developed cosmopolitan countries and states that locked everything down to the populous, not so developed countries that could not or would not lock down and initially seemed to be doing better, and this has already happened before the reopening of the original group has really picked up steam. The wave is already surging before we even get to the effect of reopening the economy fully in western Europe and the US.

Meaning localities that have an outbreak follow a progression. If someone goes from one city to the next and carries disease, they're effectively seeding a new outbreak. By the logic you're saying, we're all just Chinese case numbers? It'd help to watch the video, cause I think you're fundamentally misunderstanding what I'm saying.

I'm saying that, for instance, China, Brazil, and the US didn't follow the same trajectory and continue to be different. And because of the diversity in trajectories (for whatever reason, as people continue to speculate) the location of the majority of new cases has kept changing.

Your first sentence sounds like you are claiming there is a uniformity that to me obviously doesn't exist.

I feel like people latch onto something that confirms their prejudices and then studiously avoid acknowledging current events that rapidly show it to be an error. Like, lots of poorer countries had much fewer cases per capita than western Europe and the US, but now the new cases are shifting to India, Pakistan, etc.

Actually maybe this is the best way to explain (especially on this site) - imagine the possibility of a spread with an extremely high initial first derivative of the number of cases, with a constant negative second derivative. I think there's a good chance this could be at least more accurate than what's commonly thought compared to all of these very severe exponential models.