The second wave of Covid
-
-
@jon-nyc said in The second wave of Covid:
@Horace said in The second wave of Covid:
I suppose there will be an increase in infection statistics here and there
How is this a prediction? This is already happening.
It's less a prediction and more defensive scaffolding against trivial claims that I'm wrong.
-
jon, before we play the game of "reframe Horace's predictions to maximize the likelihood that they'll be wrong", would you like to show some good faith and offer a prediction of your own that avoids the two pitfalls you've shown us here - obviousness or incorrectness?
-
I am going to be optimistic and say that like SARS, the COVID-19 will die out on its own and there will not be a second wave!
-
Actually you asked me to repeat them in late April. The ones that have come true seem obvious now but if you find the skepticism with which they were greeted in the original thread (in early April) you can see that’s just hindsight bias.
-
- active cases won't peak in April, despite models
- Effective reproductive rate stays above one
- People will think it's beat and relax and it'll go even higher
- when we look back at the end of the year, April will not be the month with the most cases or the most deaths.
The most interesting (to me) prediction is that even though these things will likely be true, our social reaction will be minimal. I think this has mostly to do with the demographics who are legitimately at risk. It is slowly seeping in that the vast majority of the folk who keep this country running are not in the high risk category and it has occurred to them by now that they can go about their business.
It would not strike me that your predictions were heterodox, but ok I trust that they are. I will await the results of our earlier disagreement about future national virus stats and whether they conform better with NYC or with the CDC model.
-
That was your proposal, I didn't take the other side of the bet. I gave a specific range that the IFR would fall into.
You think it's likely true that sometime this fall we'll see a month worse the April? How is that consistent with 'I suppose there will be an increase in infection statistics here and there'?
-
@jon-nyc said in The second wave of Covid:
That was your proposal, I didn't take the other side of the bet. I gave a specific range that the IFR would fall into.
You said the CDC model was obviously wrong, and were backed up by scientifically minded people who reminded us that science has the giggle test and NYC makes the CDC model fail that. I predicted that the CDC model would be closer to the reality of our future national statistics than the NYC numbers. That constitutes a literal and testable disagreement about future results.
You think it's likely true that sometime this fall we'll see a month worse the April? How is that consistent with 'I suppose there will be an increase in infection statistics here and there'?
I'm just trusting you that it's likely true. If it is true, it only makes my real prediction "bolder", which is that our social reaction will remain minimal.
-
@Horace said in The second wave of Covid:
You said the CDC model was obviously wrong
Yes.
I predicted that the CDC model would be closer to the reality of our future national statistics than the NYC numbers.
Yes you did, and you invited me to take the other side of the bet. I specifically declined to do so, rather I gave you my own prediction. I forget what lower bound I used, the upper bound was .75%.
-
earlier bit quoting Jon that Jon accidentally deleted
Ok then, you are riding that line between thinking a model is obviously wrong because it doesn't conform with a certain subset of the data, while allowing that the model is closer to the truth than that subset.
-
No, I don't disagree with the specific prediction about shelter-in-place.
I think they were superfluous because localities had closed all the places you'd want to go, and had enumerated exceptions for all the other reasons you'd leave the house.
So they were basically symbolic, and governors will be keen to avoid that particular symbolism, whereas three months ago they were keen to embrace it.
-
https://spectator.us/here-comes-second-wave-coronavirus/
The article looks at data compiled by The Washington Post and draws these conclusions:
Nationally, across the US it is possible to detect a downwards trend in the number of new COVID-19 cases and deaths. The epidemic appears to have peaked in early April and tailed off since then — just as it has in Europe. But that isn’t the whole story. Drill down state by state and there is a very mixed picture. Study the graphs which the Washington Post has helpfully put online and you can put states into three distinct groups. Firstly, there are those states where the epidemic curve is clearly in decline and has been for some weeks. There are 15 of these. Most obvious is New York but others include Massachusetts, New Jersey, Pennsylvania, Illinois. Away from the North East, Colorado, Wyoming, Louisiana, Nebraska and Hawaii also fall into this category.
Then there are the states where the first peak has not yet significantly declined. There are 30 states which fall into this category, including California, Texas, Georgia, Ohio, Iowa, Kentucky, New Hampshire, Maine, Michigan, Maryland, Minnesota. Mississippi, West Virginia and Wisconsin. Most of these states have rolled back restrictions on movement and forced closures of businesses but did so before there had been any appreciable fall in cases. Currently there is no obvious trend either way.
But then there are six states which, worryingly, are showing a detectable second spike. In Florida, for example, new infections fell from over 1,000 a day in early April to around 600 a day a month later. The state began to reopen on May 4. Since then, however, new infections have climbed back to their previous peak, reaching 1,419 on June 4 — the highest yet. It is a similar story in Arkansas, where cases had fallen from a peak of 402 on April 22 to 27 on May 15 before rising again to a new high of 450 on June 6. In Vermont the epidemic seemed all but over on May 12 when zero cases were recorded. For three weeks after that cases remained low. But over the past week there has been a rebound, with 72 cases reported on June 4. The other states with a discernible second spike are Alaska, Montana and Washington.
Is this a dire warning of what is to come? First it needs to be noted that none of these states had especially high peaks to start with. It would be a very different matter if New York started to show a second spike. Secondly, the number of reported cases does not necessarily reflect reality — it depends very much on the number of people being tested. It is a peculiarity of COVID-19 that there are huge numbers of asymptomatic cases — possibly as many as 80 percent fall into this category — and that most of these are going unrecorded. Thirdly, there has been no sign of a second spike in badly-affected European countries — even though most have reopened much of their economies and have lifted restrictions on personal movement. Across the world, the only nations showing any kind of second spike are those which, like South Korea and Singapore, succeeded in suppressing the number of cases very early on. And even they haven’t seen much in the way of spikes.Nevertheless, all eyes over the next few weeks are going to be on those US states which seemed to have suppressed the virus only for it to rebound.
