I decided to dig into the numbers myself from this website that reports both positive and negative tests for each state:
https://covidtracking.com/api We'd been discussing in here whether the rise in daily cases simply reflects more testing. So I looked at a couple states and saw if there was a consistent pattern with # cases as we increase # tests. Unsurprisingly, more tests correlates with more cases, but the devil is in the detail.
For reference, I started with Illinois, where # cases and # tests went up together rapidly until May 1. But something interesting has happened in the past 2 weeks. The # tests has doubled, but the # cases has stayed relatively flat. Other states like CA and NY seem to follow this same trend recently.
Next, I looked at Texas. Their # tests has increased even more rapidly, doubling in the past week. Unlike Illinois, the # cases has gone up, about 50%. Florida is even more worrisome. In the past week, despite not increasing the # tests, the # cases has nearly doubled.
Lastly, I looked at Georgia. They've doubled the level of testing in the past two weeks but their # cases has actually decreased by about 25%.
Conclusion: who knows how well social distancing is working, and who knows how well these tests are being matched to symptomatic and high risk people versus random dudes getting tested on their way to wal-mart. But in general the number of cases don't directly scale with the number of tests (except early on, when testing is very low so only symptomatic people are getting tested), so we can still get some handle on how each state is doing even if the level of testing changes. But we need to look at the whole picture, not just # cases, to understand what is going on.
edit: If I have a few minutes to kill later on, I'll post some of the graphs in here. I'm guessing there's some website that's already doing this (plotting # cases vs # tests by state), so let me know if there is one.
www.instagram.com
abc7.com
