This is a live view of the most important COVID-19 metric: acceleration rates of positive tests, hospitalizations, and deaths.
Source: The Covid Tracking Project
Key Insight: For every week, Hospitalization rate decreases by about 9.19%.
From My Original Tweet #
The most important number regarding Coronavirus is a number none of the tracking sites or governments will tell you: acceleration. Lower acceleration of hospitalization and deaths is good. It means the number of new incidents yesterday was smaller than the new incidents from the day before.
I have been tracking this since April 1, so April 2 is the first day available.
Note: I call this “acceleration,” but I am not using the actual acceleration formula. That would confuse too many people, as the swings in numbers and in positive/negative would seem crazy. But the formula ((d1-d0)/d0, where d is day) is reasonable proxy for acceleration.
UPDATES:
A Nobel laureate agrees that acceleration is the most important metric in figuring out where we’re going. This LA Times story about his methods is very well worth reading. Note: the article is from March 23. A lot has changed since, but everything Dr. Levitt said then turned out to be correct.
Although the number of daily deaths had increased, the rate of that increase had begun to ease off. In his view, the fact that new cases were being identified at a slower rate was more telling than the number of new cases itself. It was an early sign that the trajectory of the outbreak had shifted.
As we know now, Dr. Levitt was far more accurate than Christopher Murray’s failed IHME model. And it’s why acceleration is such an important measure.
Because of this, I’ve automated my daily change and acceleration tracker. The data are refreshed every day between 5:00 and 6:00 p.m. Central time.
Finally, people have asked if I get paid to do this. No. I’m doing this because the media and academia are either too stupid or too corrupt to provide information that’s actually important and useful. Policy makers and the public make poor decisions because of lack of useful data. Somebody has to fill the void.