The growth is still exponential, but we may be starting to see a change in the Log-Log plot of new cases vs. total cases. Note that the death rate does not seem to be dropping out of exponential growth at all at this point, it does appear to be slowing, however.
But the early indicator I'm looking at is the trend on "predicted cases" which is the forecast based on the given day's numbers at a selected growth rate, projected out to a target date, in this case, Easter.
As my buddy Isaac pointed out, the flat line is the one that is tracking best leading up to the target date is the better prediction. As we see above, the red line, the growth behavior for 22% growth keeps coming down with each new day's data, while the yellow line, 14%, and even the green line, 10%, tend to be predicting what the next few days are going to look like. These have both been consistently predicting around 750 thousand cases (three quarter million) by Easter.
The thing that we don't want to see is the death rate and the growth of cases rate to be accelerating or staying at the same high value over long spaces of time, especially after State Governors have taken action to curtail contagion. Happily, it doesn't look like that's happening, rather the opposite. How significantly to the opposite remains to be seen. We have to monitor the numbers. A risk is a backlash or a resurgence of contagion, like when people start coming out of quarantine and resuming their old routines, if they do that any time soon.
Note that the Idaho cases may have already dropped out of exponential growth.
The number of cases and deaths are "small" so the plot is noisy, but this is looking like the kind of behavior we want to see, i.e., the new cases line dropping straight down. So Governor Little may have made the right decision at the right time in shutting the State down on March 25th.
By the way, the Idaho new case growth rate does appear to be dropping substantially, somewhat better than the USA rate.
No comments:
Post a Comment