Hey, what are you going to do in LA? The virus actually doesn’t grow exponentially (as in y=e^x or y=n^x) but actually all the growth cases I’ve seen are y=x^2 or y=x^3. Small detail, but the media keeps getting polynomial and exponential confused.

But good question… I’ve toyed with some diagrams in log space (good example would be deaths in NYC compared to deaths in LA) but what I’m trying to do here is be really clear so people don’t get further confused (the various media outlets are major contributors to that confusion). No one outside technical fields understands what a logarithmic axis means and when that LA data looks like it is getting close to NY data, people start to either get worried or start morality posturing. But you’re right. If I was just publishing this for you I’d do it in log space and show lines instead of power curves.

Take care!

Yes, Taehwan, clearly the data we’re working with is subpar and there is not a good strategy in place. I feel that the CDC messed up by not having a “shovel-ready” infection data strategy that they could roll out quickly. As I’ve observed, because they’re so slow, the Systems Engineering dept at JHU had to step up and take over the data role that the CDC should have owned. Lots of other issues, mostly regarding lack of systems thinking at the US level.