Previous Update: Covid 6/25: The Dam Breaks

General Viewpoint Reading: On R0Covid-19: My Current Model

This past week was not great on the Covid front, but it could have been a lot worse. I don’t mean that statement in a ‘worse is better because variolation‘ way. I do think this week was slightly better than expectations, and was looking decidedly better than that until Wednesday – earlier in the week I noted on Twitter that things looked remarkably good the last few days. If I had to guess, I’d still put this week at about par.

It seems odd to say this. We set multiple new records for infections. Positive test percentage set a new record yesterday. ICUs are full and running out of supplies in Houston and other places in the South. Measures taken to contain the situation were mostly pathetically weak.

When we run the numbers, they’re going to look rather grim.

The thing is, all of that was mostly baked into my model of what was likely to happen. We saw less rapid increase in cases than I expected. We saw even a little action to reverse opening measures before death rates skyrocketed, including in the Northeast, which was a pleasant surprise. We saw some additional support for and push for mask wearing from most people across the political spectrum, other than Donald Trump.

On the non-Covid front, things are not great. SlateStarCodex is still down. I’ve put my Aside on Other News at the end of the post.

Until then, let’s run the numbers. The charts now recombine New York into the Northeast and start on May 7. The graphs are unchanged, since it’s part of history.

 

Positive Tests by Region

DateWESTMIDWESTSOUTHNORTHEAST
May 7-May 1322419432563759156892
May 14-May 2022725427624034352982
May 21-May 2723979394184297737029
May 28-June 332200315045003933370
June 4-June 1035487246745573122693
June 11-June 1741976225107578717891
June 18-June 24662922679210722115446
June 25-July 1857613497416347216303

Deaths by Region

DateWESTMIDWESTSOUTHNORTHEAST
May 7-May 131082228815975327
Apr 23-291090206014424541
Apr 30-May 6775172312903008
May 28-June 3875166613872557
June 4-June 10743129712301936
June 11-June 17778104012071495
June 18-June 2483185912041061
June 25-July 18586581285818

Positive Test Percentages

DateUSA testsPositive %NY testsPositive %
May 7-May 132,153,7487.5%202,9808.2%
May 14-May 202,643,3336.0%246,9295.6%
May 21-May 272,584,2655.7%305,7083.5%
May 28-June 33,022,4695.1%417,9292.2%
June 4-June 103,252,8704.6%438,6951.4%
June 11-June 173,470,0574.6%442,9511.1%
June 18-June 243,629,4786.0%440,8331.0%
June 25-July 14,260,0047.2%419,6961.2%

The local situation in New York continues to be difficult to read. Two days ago we hit a new low with 0.84% infected for the state. Now we’re back up above 1%, and the weekly rate rose on net. That’s scary. A slow climb is still a climb until reversed. We do know it’s not a disaster, and the Northeast in general is stable for the time being.

Everyone else, I have some bad news.

The overall positive rate jumped again to 7.2% and once again hit a record on a Wednesday, hitting 8.5%. Clearly some states are reporting positive and negative results on different days, and it’s messing with the numbers, but the steady climb is clear.

The Midwest saw another, larger increase in positive tests. Given the inability to reverse conditions on the ground in a substantial way, looks like they’re on pace to get overwhelmed the same way the South and West are, only to do so slower.

Most of the South, and parts of the West, are rapidly becoming disasters. Texas, Florida, Arizona and California have the headlines. Florida almost doubled to just short of 50,000 cases. But the big four are not alone.

In the rest of the West, huge percentage jumps can be seen in Alaska, Idaho, Montana and Nevada, while Oregon, Washington, Wyoming and Utah continue to steadily get worse as well.

Most of the South is in deep trouble. This was a huge almost 50% jump in positive tests for the entire region. Positive test percentages are at crisis levels, if not as bad as the 50% we’re seeing in Mexico. I’ve been keeping my analysis mostly to the United States, but it’s clear that Mexico has it even worse than the worst states.

CDC Renews Application For Delenda Est Club

We already had the W.H.O. and the F.D.A., alongside old classics like Facebook, and new candidates like The New York Times. Now the Centers for Disease Control is feeling left out, and itching to ride the momentum of its botched Covid testing kits to get in on the action with this advice to colleges this fall:

To which I respond:

I have so, so had it with all instances of ‘it is unknown’ and ‘there is insufficient evidence’ as arguments for not doing obviously probably useful and definitely not harmful thing.

This is utterly insane. You do not get to not model the physical world and wait for peer review. You are literally the Centers for Disease Control. You are saying that when there is an infectious disease that can have asymptomatic carriers who infect people, you’re not sure that testing more people won’t reduce risk of further infections.

And since you’re not sure that this is helpful, you do not recommend doing it, which is effectively recommending not doing it.

What is going on? Are they worried that colleges will overload the testing system?

Then say that this would overload the testing system. 

It’s masks all over again. We believed (falsely) that we didn’t have enough masks and they’d be hard to get, so the W.H.O. and most authorities got together to lie to everyone and say masks didn’t work. And now, when we have masks for days, half the country remembers and won’t wear one, and that’s the only reason we’re still in this mess.

The C.D.C. presumably thinks we don’t have enough tests, so it’s saying tests don’t work, so colleges won’t use up the precious tests on nineteen year olds.

If that is what they mean, then they should say that. And I’m fine with it. I don’t approve of Alabama students throwing Covid parties to see who gets infected, unless they’re doing safe and proper variolation, regardless of how important they think it is to later safely attend Crimson Tide games. But I do think it’s reasonable to say that young people are low risk and broad-based testing of them can’t be a high priority. I also think it would be reasonable to tell everyone to come two weeks early and keep everyone in their dorm rooms during that time while delivering them pizzas. Maybe they’d actually do the reading.

Keep Talking To Your Area Man About Lagged Exponential Growth

Exponential growth keeps going until something stops it. Yesterday we had a record high of cases, and the highest positive test percentage we’ve had since May 9. Numbers are jumpy, so it’s hard to tell how rapidly things are still getting worse, but for now it is clear they are still getting worse in terms of infections. Hospitalizations are also rising. Deaths will follow.

At what point does the death rate staying down start to get seriously weird?

June 18 starts the surge in positive tests that represents the full second wave. June 23 represents when it accelerates. My default assumption has been one week to test positive, and about two weeks after that to see the average death.

That would mark the surge in deaths to start around July 2. In other words, today, with things picking up speed on July 7.

So no, this isn’t weird. Not yet. But if there is no spike in the next seven days, then that’s pretty weird. If that actually happened, I’d look more carefully at hospitalization data, which I usually disregard as not worth the trouble. But mostly I’d be terribly confused. The infection fatality rate seems to clearly have fallen, but why would it have fallen so much so quickly now that a surge in infections doesn’t kill more people? Quite the tall order.

Still Alive

It’s clear that death rates have fallen.

The question is why. Different explanations have very different implications for what happens next and what you should do personally.

Explanation 1: Better Medical Care

This is the best possible news in any non-overwhelmed area. If it’s better medical care, we can expect that to continue, so the death rate drop gets sustained and applies to your risk. It also means we’re learning to be better, so we can expect to keep learning, and deaths likely fall further and faster in such worlds.

Explanation 2: Nursing Homes are Protected

This is a very good thing to do. Nursing homes by all accounts were a huge part of why death rates in the first wave were so high. But if this is the explanation, then the drop in death rates is only good news insofar as less people are dying. It’s not good news in the sense that your behavior can adjust. Before, the death rate was high in general but probably lower for you. Now, the death rate is lower in general, but your rate hasn’t changed. And thanks to lower death rates, you’re now going to be more likely to act like a Covidiot.

Explanation 3: Young Getting Infected

Mirror image of protecting nursing homes. It doesn’t change your risk if you get infected. All it means is that less people die for a given level of infection. Which is good, but you shouldn’t change your behavior other than being more wary of younger people, especially in their 20s.

Explanation 4: Fraud

This is the worst explanation. We noted before that there’s been a rise in ‘mysterious deaths’ and many states seemed to be trying to cook their books to justify re-openings. They were facing popular pressure, and also fiscal pressure to get tax revenue up and unemployment costs down. What if the death rate drop is in large part due to fraud? I’d like to see more careful looks into this. Obviously this is very bad news all around. It means our society is breaking down more than we expected, and also we can’t trust the numbers, and also people are actually still dying.

Explanation 5: Better Testing

It would be easy to say that no, this isn’t it. Positive test percentages didn’t drop that much, and now are rising. Certainly any results from the second wave will face effectively worse testing, as capacity is being overwhelmed faster than it can be added. But there’s a version of this that can sort of work. Perhaps as we increase capacity, we still find mostly the same positive test percentage but get to test people who are less sick and/or less vulnerable, and they therefore get detected but don’t die. I don’t think so, because this would show up in changes in other statistics, and again this won’t apply going forward. But I can’t quite entirely rule out some effect here.

Explanation 6: Selection for Vulnerability

Perhaps how likely one is to be infected given exposure correlates highly to one’s risk of death, so death rates should drop over time as the most vulnerable are already infected? But given the low infection rate in the areas with new cases, doesn’t seem like effect can be big enough here. This would be unclear news, as it might change your perceptions of risk (if I’d be at risk I’d already have it!) or not (I’ve been avoiding infection so I’d be safe so far regardless). On society-wide level it would be good news.

What else is there to consider?

Bar the Doors

Indoor restaurants and bars are, I believe, the biggest super spreaders that we have any control over, now that we’ve lost control over religious services. Going to large indoor religious services, especially with lots of singing, seems like an actual worst case scenario. Just yesterday we learned there was a choir at a mega-church that had infections, then went ahead and performed in a mega-church, and did so for the Vice-President. So we’re pretty insane on that front.

But here’s some interesting restaurant data. Restaurant dining activity by state:

This does not differentiate indoors versus outdoors. New Jersey for example has a big spike in activity but indefinitely postponed indoor dining. New York City also isn’t dining indoors any time soon. That’s a big relief to me, as growth rates locally are on the brink and I expected indoor dining to take us over the edge. We also saw a few closures in southern California, and a few rollbacks in places in the South.

It wasn’t anything like what is needed. But it was something! And it happened without waiting for the bodies to start piling up.

I don’t think that it’s possible to open indoor dining, even at reduced capacity, without substantial herd immunity or a wave of infections. It’s possible it can be done in a very careful, limited way, but you definitely can’t do bars, at all. At a bare minimum, this is going to eat up far more of the flex space than it is worth, space that could be used for schools and offices and socializing outdoors and so on.

If you go to an indoor restaurant in the United States without a positive antibody test or two weeks isolation time in both directions, let alone go to a bar, you’re being grossly irresponsible. Period. Don’t do it.

Great Minds Think Alike

Jenny Mason is the sister of Jon Finkel, history’s greatest Magic player. Her analysis of Covid-19 asks many of the right questions that are rarely asked, and mostly comes up with similar answers to my own. I recommend reading it.

Her central points are as follows.

  1. There has not been a substantial second wave anywhere there was a strong first wave. This implies that herd immunity, as I’ve noted here, is likely playing a big role.

2. Sweden didn’t come out of this as the hero, but things were nowhere near as bad as the critics predicted for it, and cases managed to peak and then steadily decline. I don’t put as much stock in this as she seems to. I view locking down by law as not that relevant to how locked down, mask wearing and otherwise distant and responsible an area really is. But it’s certainly interesting that things were only bad and not nightmarish.

3. Protesters don’t seem to be getting infected much in areas with previous waves, like Seattle and New York City, despite all the obvious close contact. I’ve noted this too. My main conclusion was that being outdoors was a bigger factor than we thought. Since then, we’ve seen some statistical changes, like the share of infections coming from young people in Minnesota, that has made it seem like the protests were spreading things directly just fine, thanks, in places without a previous wave. But as usual there’s lots of other explanations and the protests weren’t that large.

4. Household infection rate is shockingly low. She notes: There have been quite a few studies estimating the household transmission rate for COVID, and they are consistently below the 38% rate for the flu: 11%17%17%30%. For something this infectious, that’s weird. Super spreaders are part of the answer, but would have to be very extreme to be enough.

5. Kids are hard to infect, and find it hard to infect others, and almost never die from it. Not clear to me that she ties this into the rest of her thesis, but this is still true and it’s still weird. If anything I continue to update towards kids not being carriers worth worrying about.

6. Herd immunity thresholds from disease-acquired infection are lower than from vaccination. I wrote On R0 a while ago to point this out and keep pointing it out but it feels like shouting into the wind. If anything I think her discounts here are way too small, and herd immunity threshold is much lower than people think.

7. This week’s actual physical science news about Covid-19, which shows that many people who test negative for antibodies are still effectively immune. Huge if true! Still has to be peer reviewed and all that, so we haven’t seen the paper, but this is essentially the best possible news. It potentially means a lot more infections than we thought and a less deadly virus than we thought, and puts us much closer to immunity. Again, we haven’t verified yet, but huge if true.

8. Points to analysis that suggests many more cases than we expected, even after the adjustments I’ve been making, and that serosurveillance will give underestimates of infection rates. Consider this part two of the seventh point. To extent that these are big effects, they’re big news.

Jenny then concludes that areas that have already suffered a wave likely are already at herd immunity, including New York City, Seattle and Sweden. Her theory on kids is that this comes from cross-immunity that comes from other diseases.

Again, I think her article is worth a read, and have a similar picture to hers. She goes farther with it than I do. I don’t think anyone has hit true ‘herd immunity’ in the sense that New York City could act like there was no virus and it wouldn’t have a problem. Nor do I think that such a level is obviously sufficient to make a protest not spread the virus, assuming there are people who come already infected – R0 is not a constant number at any given time. I do however agree that (partial) herd immunity is doing a ton of work.

Overall, all of this is potentially very good news, as it means we are likely to reach herd immunity faster, and with fewer infections and even fewer deaths, than would otherwise have been necessary.

What the Hell is Up With the Stock Market?

It’s bizarre. The economy is in shambles. Infections are exploding across the country, with all but a handful of areas looking primed for trouble. There has been civil unrest that has increased greatly the focus on Marxist causes and calls for redistribution. Crime is up. The Democratic presidential candidate, who promises to raise taxes and otherwise be big-business unfriendly relative to his opponent, has a double digit lead. States, basically all of them, are on the verge of budgetary collapse with devastating consequences and no sign of a fix.

Yet the stock market continues to go up. What the hell is going on?

Will Eden asked this on Twitter a few days ago. Eliezer Yudkowsky replied that the only thing that makes sense and doesn’t violate the EMH is a drop in real interest rates, because TIPS spreads don’t indicate recovery.

I don’t think that’s right. I do think real interest rates dropping is part of the story. It’s especially helpful in the sense that if a company won’t make money for a while, low interest rates allow the net present value of its profit flows not to care all that much.

There are also other parts.

The biggest part is that the economic effects of Covid-19 shifted and will continue to shift business and profits away from small business towards large business. Stocks represent Big Business. Even small stocks, even if they’re not Big Business, are still big business.

How much of that effect was the direct effect of the need to distance and shut down benefiting business with cash reserves and scale, and hurting those who benefit from locality and loyalty and routine? How much of it is that in a time of crisis, reliability of major corporations is valued more highly, and they are better able to transition to doing more things online? Versus how much of it is that the government’s actions outright subsidized Big Business in contrast to small business?

I think this story is mostly about the direct effects of the crisis, and the government’s interventions were not central.

One could argue that big business would have been in deep trouble if not for the intervention of the Fed to stabilize certain markets in March. Without that, cascading failures would likely have occurred, and preventing that limited many big business failures, while many small businesses still failed. That’s true, although I view such interventions as obviously necessary and good and to the benefit of all. It’s preventing a systemic failure. I have friends who view this as more hostile slash unfair. Either way, that’s over now, and all it did was prevent one type of bad thing from happening.

One could argue that big business got $500 billion, scaled up, in secret. Arnold Kling, who often has insightful things to say, argued that this was de facto the end of American capitalism, as the government was effectively in charge now. Again, some of my friends were likewise treating this as a really huge deal. I don’t see it that way. It’s not that much money in the grand scheme, and it’s loans that have to be paid back. If companies could otherwise raise money, and most could, all this does is save them the change in cost of capital. While in some ways it seems like a big number, in my mind this whole thing is not all that big an effective number.

Mostly this was saving insolvent slash illiquid companies in certain areas like air travel, combined with an opportunity for massive corruption. It’s absurd that we agreed to supervision of how the money got distributed, then this money was distributed without any transparency whatsoever, and everyone mostly shrugged at that. It would be surprising if the bandits involved did not make out like who they are. But in terms of adding this kind of value to the market, it doesn’t make sense. For most companies, prices clearly don’t reflect ‘saved from bankruptcy by a government loan and now has possible value if things recover.’ By contrast, small business got massive grants that took the form of forgivable loans. It was botched in many ways, as you would expect, but a gift is much, much better than a loan.

Boeing’s experience was instructive. Boeing, which seems to have lost its ability to make airplanes that can properly fly and become a moral maze, and which has no customers who want a new airplane for a while even if it can make them, and was already bleeding money, still managed to get private financing in order to avoid any potential strings that might be imposed on it by a Biden administration. The stock went from ‘seems like it’s only worth anything because bailout’ to then turning that bailout down.

Much bigger than all of this direct subsidization was the simple, straightforward Federal Reserve intervention to do reasonable things and do standard monetary policy in response to a real shock. Which is their job. If anything they should have done more, and/or adopted level targeting.

I think that, in turn, is dwarfed by the way the landscape happens to work out in this situation, to the benefit of big business. If one believes that This Too Shall Pass, and we’ll definitely be recovered a year or two from now, and we’re past the scary ‘things might fall apart before that’ stage, then this all can make sense – they’ll come back stronger than ever, with more market power and more sales, and make that much more money.

I also strongly believe that the EMH, the Efficient Market Hypothesis, is false. I’ve believed that for a while. I also believe this crisis has strongly shown it to be false. As I’ve noted, markets seem to be reacting to news far slower than those paying attention can figure it out. Markets predictably, on a given day, for a while, reacted in the wrong direction to news. News that looked bad to the market was good news, because it meant a better reaction that might help, whereas those paying attention already knew how bad things actually were from data they’d had for a week or two, and were busy processing the real news. There was also a reversion effect for a long time, that probably still holds somewhat, that each day’s move partly reverses the next day, and for at least a while a momentum effect, where what happened early in the day was reinforced later in the day. For March and April it was clear things were highly predictable slash wrong on such scales.

Based on my trading experience, I am highly confident that such inefficiencies were real, and that my old firm made a lot of money exploiting them.

One can also look at Robinhood and other potential new sources of capital. What Matt Levine calls the Bored Markets Hypothesis seems strong, that those who have few other ways to gamble and interact and get excitement are looking to the market, and mostly they end up long and pushing stocks up. I haven’t looked at the magnitude here, so I’m doubtful this is a big impact outside of a few weird stocks (e.g. Hertz) but I definitely believe that a bunch of inflows (people buying stocks for the hell of it) can drive prices up.

This and everything else here is not investment advice. That said, if I had spare cash on the sidelines, I certainly wouldn’t be in a hurry to buy into what looks like a historically overpriced market with an overhang of multiple huge crises still to be handled. But I also wouldn’t have done that a month ago or two months ago.

Markets can stay insane for a long time and all that, and very long term either things are going to go very, very badly in ways that make stocks the least of my worries, or you’ll want to have been long the whole time rather than staying out entirely.

As for ‘why aren’t you rich?’ and ‘how much money did you make off this?’ the answer is that I’m rich enough that I have made a conscious decision to do other things with my life and to avoid trading. The things I know how to predict here are short term, on the order of hours or days, rather than months or years. To start down that path would prevent me from focusing on anything else, and doing the trades involved in the face of the fees and taxes and interest rates I’d have to pay as an individual with little setup would be tricky. Whereas in the long term, I am confident the market isn’t thinking right, but that doesn’t mean I can beat it by enough that I should be constantly trading and both letting that eat my attention and also paying massive capital gains taxes to do so.

Aside on Other News

Other news seems rather terrible.

The central topic of online conversation, other than Covid-19 itself and the continuing adventures of the Orange Man, now seems to be cancel culture’s latest targets, the dangers of cancel culture, and the cultural forces that are trying to use this moment to paint America as History’s Greatest Villain, to tear our country apart and set us at each others’ throats. Almost all the talk of potentially constructive object-level action, to address issues like stopping state-sanctioned murder, has been effectively sidelined by calls for the only action our society still seems capable of, which is symbolic action against scapegoats.

Everyone is rushing to take actions they think will make them less likely to be the next scapegoat, despite being, in the eyes of those doing the scapegoating, eternal scapegoats by nature of who they were when they were born. The more intensely one tries this, the more one’s attempts merely grant the attackers power and highlight one as a juicy target.

Meanwhile, these symbolic actions have, as I’ve noted before, crippled any legs that a response to the Covid-19 crisis had legally, rhetorically or practically. You can try to claim that symbolic risky action against the outgroup good but actual living life risky action bad, but, well, good luck with that. There’s no trust left in ‘experts’ or institutions, and there’s no will. We’re out of state capacity. The thing will mostly burn until it burns out.

Hopefully this does not all turn out the way many survivors of countries destroyed by communist takeovers are warning us about. It’s up to us to ensure that does not happen.

The New York Times is playing a central role. Its attack on Scott Alexander, which Twitter reports is going to be a true hit piece attempting to not only doxx Scott but also completely falsely label Scott a ‘white supremacist’ is not an outlier. It is their new mode of doing business. I have joined the movement to #GhostNYT. No clicks, links or business of any kind. If their reporters come knocking, they get nothing. If and when the situation is resolved to Scott’s satisfaction, I’ll consider softening that position.

Please join me and #GhostNYT. Also please sign the petition against doxxing him.

Meanwhile, China has taken advantage of its opportunity and forced on Hong Kong a beyond over the top National Security Law. This law claims jurisdiction over all people of the world, everywhere. They have taken the position that they can and have made speech, anywhere, by anyone, illegal if it is adverse to China’s control of Hong Kong. By writing the previous sentences, I have in theory made myself a criminal in China, subject to arrest were I ever to visit there. I am happy to see Britain and hopefully Australia are preparing to offer a place for citizens of Hong Kong to go. We should do the same. Of course, the United States has done nothing of substance to respond to this on any level.

We all feel the chilling effect of all this on both speech and action. This section was rewritten multiple times to attempt to properly balance my seething rage at what is going on with the need to be prudent.

In March, April and May, I felt unsafe and stressed because of Covid-19. Now I feel unsafe mainly because of other things.

Despite slash because of such worries, I didn’t feel I could write this post without this aside. So I wrote it.

But let’s not discuss it here further.

As always, please keep the comments to Covid-19 related discussion that is relevant to forward-looking analysis. Who is to blame and who should be scapegoated need not interest us here, except insofar as it is important to who to believe and trust, and what actions or events come next.

New Comment
13 comments, sorted by Click to highlight new comments since:
There has not been a substantial second wave anywhere there was a strong first wave. This implies that herd immunity, as I’ve noted here, is likely playing a big role.

Iran had a big outbreak (much bigger than official numbers) and now has a clear second wave. Some of this is in different cities but I haven't found a careful analysis.

Have you looked at the mobility data? Most (maybe all) of the places in Europe that had a strong first wave kept mobility low (esp. public transport and workplace) and have adopted strict social distancing in public spaces and some level of masking. I haven't seen any good evidence that herd immunity is playing more role than we would expect (e.g. 15-30% reduction in R in worst hit places like NYC, Lombardy, London, Madrid).

I think there's pretty good evidence that most adults are susceptible from the huge outbreaks in prisons, meat plants and hospitals with bad PPE.


2. Sweden didn’t come out of this as the hero, but things were nowhere near as bad as the critics predicted for it, and cases managed to peak and then steadily decline.

This seems wrong-headed. Sweden has been terrible compared to ALL its geographic neighbors (which are very similar culturally and started from the same position). They have a very high death rate and a poor rate of testing. They have suffered substantial economic damage despite not locking down. AFAIK, the hospital system "survived" the peak in part because they did not treat the most vulnerable. As well as low testing, they have very low mask use, and so they are poorly set up for the end of summer (when people will be inside much more). I also think Sweden is pretty different from the US. I'd expect compliance with testing, contact tracing and isolation to be high in Sweden, while it seems compliance in NYC is not high.


4. Household infection rate is shockingly low.

Why is it shocking? If R=2.5 with zero social distancing and tons of superspreader events, it's just not that infectious. We also know there is variability in infectiousness across individuals (overdispersion). SARS-1 had a household secondary attack rate of 8%, H1N1 flu had rates that varied across studies but they are mostly lower than 38% (e.g. 15-20% is common).

Regarding children, this meta-analysis finds adults have a 1.4x higher risk of infection. Not a huge difference. Kids are much more likely to have mild symptoms or be asymptomatic.

[-]Zvi120

1. I don't think mobility data correlates well with risk taken - it's easy to screw up and limit the wrong things (get stuck together indoors), or to limit the right things without changing mobility much (move outdoors). It's indicative of trying to do anything at all, at least. I've talked extensively over many posts about why I think herd immunity is a bigger deal than people think, especially in On R0.

2. Sweden did badly, but it's important to notice that it did far less badly than a naive model would expect it to do. Why did things end up getting contained when they did? Why wasn't it much worse? Pointing out that Norway did better doesn't change the need to answer that. This is not me saying Sweden is a model, it's a control group and we need to understand its data.

4. It's shocking because those people are having very intimate contact over extended periods of time in indoor situations, often sleeping together, touching, sharing food and cooking, over periods of days of infectiousness, etc etc. It's certainly not the naive or public perception of such risks if precautions aren't taken. And it then requires an explanation for why we can't contain this thing with relatively light countermeasures.

On the children, the metastudy doesn't seem like a methodology I'd think would produce good results on such matters. I'm open to evidence that difference is that small but it definitely seems like the vector is mostly harmless...

I've talked extensively over many posts about why I think herd immunity is a bigger deal than people think

I understood the argument as "there'll be herd immunity faster in specific locations (e.g. subway riders or people under 20 in some neighborhood)". The logic makes sense but I'd guess the effect is small, due to population mixing / small-world network effects. Young people are probably getting infected more but they are still far from HI everywhere and they are probably well mixed. I haven't seen any positive empirical evidence for your view over my take (big first wave --> people take precautions more seriously and have slower reopening + 20-30% drop in R due to fewer susceptible).

There's Google/Apple style mobility (which actually records amount of time spent in work/home/retail/public transit) and questionnaires that ask for "number of contacts per day". People have used both to model cases/deaths and they are both pretty useful. Some papers (China) and UK. The point is that we know you can predict spread using these proxies for contact. So you can actually see if the amount of predicted contact is lower in NYC, London, Madrid and Lombardy vs. places that didn't have a big first wave (e.g. LA, Miami, Phoenix). And the predicted contact was lower in the former places. (But I haven't done a careful study).

2. Sweden did badly, but it's important to notice that it did far less badly than a naive model would expect it to do. Why did things end up getting contained when they did? Why wasn't it much worse?

Public transit use was down 55% in Sweden at peak and is still at -7%. Norway was down 65%. Swedes stopped going to the cinema and other high-risk venues were way down. Without a formal lockdown, there was a huge change of behavior in Sweden. I'd guess Swedes were aware that all the countries around them had tighter restrictions and much lower death tolls. So they acted to reduce risk. (People in the UK also reduced risk more than was required by government.) So I don't see any mystery in Sweden. The real mysteries: Vietnam, Thailand, Cambodia, Laos and Indonesia. And I'm surprised how well the SF Bay has done.

4. It's shocking because those people are having very intimate contact over extended periods of time

Agree it goes against the naive model. But if you take seriously that 20% of people do 80% of infecting (or maybe a bit less than that), then it's likely that a decent proportion are essentially not infectious. Also note that many household members are younger children, who are harder to infect.

This was mostly critical comment. But I'm happy that you're writing these updates and strongly upvoted the post!

@LW team I can't see some of the images in this post because they're embedded from Twitter and I have Twitter blocked. Can haz fix?

Should be fixed now, I think

isn't D':

Looks like I missed one. How about now?

you did it!!! thank!

[-][anonymous]30
The infection fatality rate seems to clearly have fallen, but why would it have fallen so much so quickly now that a surge in infections doesn’t kill more people?

and

Explanation 3: Young Getting Infected

My current leading hypothesis is that the difference in IFR between different demographics is substantially greater than previously reported. If under-30 groups see far more asymptomatic cases and very-low-symptom cases than older folks, that would likely cause a substantial underreporting of total cases in those demographics specifically (even more so than in the general population), which would lead to relative overreporting of their IFR - and the reported IFR for those groups is already much lower than for older groups.

This would also partly explain the ongoing confusion about whether there are tons of asymptomatic cases or not. There may be, but only in some demographic groups.

I have joined the movement to #GhostNYT. No clicks, links or business of any kind. 

Maybe we need something like SciHub for people accessing NYT articles in a way that doesn't give them business?

It seems like as of today deaths are still not spiking. Good news? Do we have stats on hospitalization rates by age group? If it's something like 0.1% of <30 yo infected need ICU care, that suggests everyone(without comorbidities) under 30 could get infected more or less at the same time and we'd still be fine, assuming higher risk groups stay isolated.

Of course this plan is not helped by propaganda being eaten up by old people that the virus is over, masks are evil and other info hazards.

[-]Zvi40

Holiday is messing with reporting so I'd wait 2 days before drawing conclusions.

After that... Huge if true. Question in my mind is, do we trust it to be real...