Yesterday morning, we passed 10,000 cases. This morning we’re at over 16,000. That’s a 60% increase. Yesterday’s midnight to midnight growth rate was 53%, that’s what I’ll mostly use in this post. The day before saw an increase of 47% The day before that was 38%. Obviously, the growth rate has been accelerating.
The good news is that it’s increasingly clear that this acceleration is due to increased testing. We have known for some time that there were far more infections than had been reported. Now that we have more testing, we’re finding more cases. This high growth rate phase should continue until our testing catches up to the actual cases. Right now, the growth rate of known cases is almost certainly higher than the growth rate of actual cases. Whenever one thing has a higher growth rate than another, math tells us it will eventually catch up if left long enough. So eventually, the number of known cases will be close enough to the number of actual cases that the growth rate of reported cases will slow back down. We don’t know how long that will take because we don’t know how many actual cases there are. But it’s almost certain that the reported cases are growing faster than the actual cases now.
The bad news is that we’re going to see some pretty startling numbers while we catch up. To give you an idea of what that might look like, let’s compare our earlier projection to one based on the most recent growth rate we’ve seen. When I made the projection back on the 14th, I used a doubling period of 3 days. That corresponds to a daily increase of 26%. Yesterday’s growth rate of 53%% corresponds to a doubling period of 1.6 days. Here are the numbers every three days for those rates. Notice that even the 3 day period numbers are a lot higher by April 1 just because we’re starting at 5000 higher than we originally projected today.
……….3/20…….3/23…..3/26…….3/29……..4/1
26% 15,000 30,000 90,000 180,000 360,000
53% 15,000 53,724 192,415 689,151 2,468,246
If we project a 60% growth rate out to April 1, we get over 4 million cases.
I want to be clear that I’m not saying that we will see 2 or 4 million cases by April 1. I’m merely pointing out that that’s where we would be if the growth rate stayed at 53% or higher until then. There are some reasons to think it might not. First, it’s not clear that we would have enough testing to detect that many cases by April 1. Second, It seems likely that we would catch up to the actual cases before then at such a high growth rate. This is especially true as the current containment and social distancing efforts are likely slowing the actual rate of spread. I will say that in the shorter term, we might very well see some numbers from that bottom row. For example, next Monday might reach over 50,000 even if April 1st is nowhere near 2 million. Here again, we just don’t know. But I think it’s important to recognize that numbers may get large much more rapidly in the short term, and to realize that it’s not Armageddon, it’s just the testing catching up.