Overall, stayed at 1.18 Outside New York, edged up to 1.23
Commenter Blink’s chart, showing the 7DDRR. Smooth, gradual decline, starting at the end of March, before you would expect the lockdowns to affect deaths, and lately becoming an agonizingly slow decline.
I think the commenter’s name was Dave.
Thanks Blink and/or Dave! I’ve added 7DDRR to my sheet as well.
Remember that as long as it’s above 1, the spread of the virus is increasing exponentially. This means that the lockdown is not sufficient to achieve its primary purpose.
Also remember that there’s really no limit to how bad a pandemic can be. If a pandemic killed almost everyone in North America, it wouldn’t be the first time.
This isn’t to say that the lockdowns aren’t causing a depression, because they seem to be causing the worst depression on record.
Interesting question. What would happen if everybody in the USA were exposed to the virus? There would be almost no deaths of people under 20. In fact, most of them wouldn’t feel anything. Those over 65 and with pre-existing conditions would be dropping like flies. So far, about half of COVID deaths are in nursing homes, where almost by definition people are old and sick. In between, a lot of people will be “asymptomatic” and a lot will have to stay home and rest. How many could get through with portable oxygen and not lying on their backs and how many would need hospitalization?
True, but for a significant number of people it seems to do permanent lung damage. The virus has already seen significant mutations, so people will probably get it again and again. If each case is the rough equivalent of (say) 20 pack-years of cigarettes in terms of lung health, it will become a much more serious problem before long.
According the NY State / NYT: antibody tests of 3,000 people in NY State at stores (selection bias?) indicates over 20% in New York City had the virus already, which is consistent with an population-wide fatality rate of around 0.5%.
The data on this NYC page shows that cases, deaths, and hospitalizations in the city peaked 17 days ago and have been going down fast ever since?
If those are right, the city could start safely easing soon.
I have a couple of further comments- the 7 day average does a good job smoothing out the number, because it essentially adjusts for the day of the week effect. But I believe most sites track reported deaths on a given day, not when they actually occurred. This means that these rates are lagging even more than actual deaths, implying that we are further from actual infection dates than indicated by extrapolating from the day of death, and may be further along the recovery curve than we think because if rates are dropping they won’t show up immediately.
In addition to tracking the curves, I have been looking at week over week daily rates, and for the first time since these curves started, we are seeing week over week declines in deaths. In fact 4 of the past 7 days have had lower absolute numbers than the week before:
This Thurs: 1911 vs last thurs 2136 (~10% decline)
Wed: 2108 vs last wed 2492 (~15% decline)
Tues, Mon and Sunday where higher than the week before but
Sat and Fri were also lower than the previous week (by small amounts)
Also, Easter weekend appears to have pushed the reporting spikes to Wed and Tuesday that week, which would have made last Tuesdays number a bit lower and Wed a bit higher than otherwise expected.
This is especially good news if the pattern holds because prior to this the reduction in the doubling rate was solely a function of the rate of increase of deaths slowing down but still increasing every week. If this continues, actual deaths are going down which is lost by simply looking at doubling rate curves.