Of course, it is a complicated metric, and this group explains that and explains why their model is structured to weed out certain biases. It is why I use this one, as I've been following them and they are fairly accurate.
Here is the complete text leading into your snip
"Calculating Rt from the reported number of reported cases is complicated. People are typically diagnosed after they have already spread the disease, and many are not diagnosed at all. As diagnostic guidelines loosen and testing availability improves, we expect to see more cases, though the underlying incidence of disease may or may not have changed. Lags in diagnosis, diagnostic delays, and changing diagnostic guidelines will all impact case reports, and bias estimates of Rt.
We can avoid these biases by estimating Rt from the number of new infections each day. We estimate new infections using a statistical model that combines information about reported cases, reported deaths, the percentage of the population vaccinated, disease stage duration, and disease severity and mortality risks."
Now yours:
Our infections metric takes into account the delays mentioned above, and includes individuals who haven't tested positive. Once we estimate the number of new infections each day, we can use that number to produce a more robust estimate of Rt.
Present-day estimates of Rt are highly uncertain, and can change dramatically over time.
We feel most confident about results for dates which are at least 14 days in the past. Additionally, Rt is easy to misinterpret. In many cases, we expect users will find our
Infections per capita metric to be more useful. See
here for a discussion of the pitfalls of Rt.
Could it be wrong? Sure and they, and I, have said as much. It has however been trending downward steadily and we are starting to see that in the data of daily case numbers.
If R was still well over 1 daily infection numbers would still be steadily rising. They are not. Could that change? As I have previously stated, sure, however, the data as it sits right now would indicate that things are trending the right direction.
Here is the per capita infection data they mention