Nate Silver

Nate Silver @NateSilver538

@mattyglesias I think that is mostly true but maybe also a bit too literal. I'm not sure in the early days of discussion about curve-flattening if people were necessarily being so precise about exactly what they meant. — PolitiTweet.org

Nate Silver @NateSilver538

This has been a good listen. There are a lot of little communities that I'm a part of online, and for some reason the NBA stats community has been responsible & smart in how they've stepped up to cover COVID-19. — PolitiTweet.org

Nate Silver @NateSilver538

@jbarro Yeah, these things are a little bit of a mixed bag. On the one hand, if they're now able to test people with slightly lower risk, the rate of positives will go down even if infections haven't. On the other hand, it might be a good sign if more testing capacity has opened. — PolitiTweet.org

Nate Silver @NateSilver538

Last two days of reporting in New York have seen "only" 32% and 30% of tests come up positive. That's still a lot but considerably down from a peak of 50% on March 31. https://t.co/MtU6toy8TO — PolitiTweet.org

Nate Silver @NateSilver538

RT @edyong209: @NateSilver538 Also relevant, from here https://t.co/FedbvDTKkZ And related paper: https://t.co/DCgdBjnCl9 https://t.co/HH6… — PolitiTweet.org

Nate Silver @NateSilver538

Political support and overall tolerance for social distancing is also fairly high now. It may not always be. So that may be another argument for "keep doing it now, while you can, recognizing that it may not be as feasible to maintain this in the long run." — PolitiTweet.org

Nate Silver @NateSilver538

This seems like an under-discussed point re: reopening. You probably want to get the overall caseload down by keeping strict-ish current measures in place longer, so that if cases start growing again (or even remain flat, i.e. an R around ~1) it's from a lower baseline. — PolitiTweet.org

Nate Silver @NateSilver538

Testing capacity probably varies by something like 10x across different major Western countries. It also probably varies by something like 3x-5x across different US states. So you could easily have some places that are detecting 1.5% of their infections, and some that detect 15%. — PolitiTweet.org

Nate Silver @NateSilver538

Taken as a whole, those studies seem to suggest it's possible—but hardly certain—that the ratio of total infections to detected cases is higher than the ~10x ratio commonly cited. But the thing is there's not just going to be one universal ratio; it will vary from place to place. — PolitiTweet.org

Nate Silver @NateSilver538

There are at least ~a dozen studies already that attempt to measure seroprevalence (how many people were infected) for COVID-19 in various parts of the world. Will likely be dozens more in a few weeks. I'm seeing analysis of e.g. the Santa Clara Co study that take it in a vacuum. — PolitiTweet.org

Nate Silver @NateSilver538

OK, here's one fairly explicit comparison to the polling world. It's usually going to be better to do a meta-analysis of all studies, including thinking about biases that could systematically affect all of them, than to spend a huge amount of energy picking apart any one study. — PolitiTweet.org

Nate Silver @NateSilver538

Seems like the optimism and pessimism biases on Coronavirus Twitter are roughly in balance right now. This condition probably will not hold, though. Possible the mood has shifted from too-pessimistic to too-optimistic when the evidence hasn't really changed that much. — PolitiTweet.org

Nate Silver @NateSilver538

If you want to criticize another model, by far the best way is to build your own model, which is what these U. Texas folks have done. https://t.co/9874o0VQxZ — PolitiTweet.org

Nate Silver @NateSilver538

To be clear, I think the green is the least-worst option and that Sweden/blue is mistaken, because green buys you time to get a vaccine, treatments, or mass testing/tracing. But when you talk about "flattening the curve", be clear whether you mean the green way or the blue one. — PolitiTweet.org

Nate Silver @NateSilver538

The green curve is great except—unless you eradicate the it completely or get a vaccine, most people are still susceptible and cases will start increasing again once you relax too much. Perhaps in practice you wind up with several waves of distancing being turned off and on. https://t.co/MeqvkzvLr4 — PolitiTweet.org

Nate Silver @NateSilver538

Conversely, other countries are attempting something like the green curve. They *are* trying to limit the total number of cases... for now, anyway. https://t.co/K7rzcxYa3a — PolitiTweet.org

Nate Silver @NateSilver538

Basically, Sweden is attempting the "flatten the curve" strategy from the blue curve on this famous graphic. They are not really trying to limit the total number of cases. They *are* trying to spread them out. So they're pursuing the blue strategy, not the red one. https://t.co/nBTe2d6lDi — PolitiTweet.org

Nate Silver @NateSilver538

That doesn't *necessarily* mean Sweden's health care system will eventually be overwhelmed. I have no idea about that either way. It does mean that a large % of the Swedish population will eventually get COVID-19. In fact, that's what they're counting on: herd immunity. — PolitiTweet.org

Nate Silver @NateSilver538

In the medium-to-long run, though, "fairly flat but slightly growing" (Sweden) winds up being a LOT different than "fairly flat but slightly shrinking" (likely where most of the US and Western Europe are with current lockdown measures). — PolitiTweet.org

Nate Silver @NateSilver538

Sweden *is* doing *some* social distancing and they have other advantages (e.g. low # of persons per household). So maybe their R is 1.2 or 1.3 or something. In the short run, an R of 1.2 may not look *that* different from one of, say, 0.8 or 0.9. Both will be fairly flat. — PolitiTweet.org

Nate Silver @NateSilver538

Put another way, the differences between Sweden and other countries will probably become more obvious over time, assuming everyone maintains their current course. — PolitiTweet.org

Nate Silver @NateSilver538

Seeing some "Sweden's outcomes aren't actually *that* bad compared to other countries" sentiment. But here's the thing: Sweden's deaths are still increasing at a pretty sharp rate (though note week vs. weekend reporting issues) which is NOT true in most of the US and Europe. https://t.co/LLjNucYx7q — PolitiTweet.org

Nate Silver @NateSilver538

@thecity2 @kmedved @tylercowen I think later on it evolved into "get R<1 until we can figure out what to do later", which is sort of the de facto policy in most non-Sweden Western countries. This *does* strive to limit the total number of cases, but the issue is there's no clear exit strategy. — PolitiTweet.org

Nate Silver @NateSilver538

@thecity2 @kmedved @tylercowen Flatten the curve has meant a lot of different things. In some (not all) early usages, people were talking about not necessarily limiting the total number of cases but spreading them out over longer to reduce burden on the health care system. This is sort of what Sweden is doing. — PolitiTweet.org

Nate Silver @NateSilver538

@kmedved @tylercowen Although maybe that's too pessimistic on what R=0.8 looks like? If there's a 5% aggregate attack rate, what percentage of people are currently infectious? — PolitiTweet.org

Nate Silver @NateSilver538

@kmedved @tylercowen Yeah, there's a more sophisticated tier of arguments that revolve around the medium-to-long-run sustainability of current measures, the value of buying time to develop better interventions and collect better data, etc. — PolitiTweet.org

Nate Silver @NateSilver538

A related hypothesis: I think people are often *really* arguing about 3 when they're ostensibly arguing about 1 or 2. — PolitiTweet.org

Nate Silver @NateSilver538

This is not meant to sound one-sided. The same holds in reverse if you take a particularly conservative/cautious posture on re-opening. It just seems like people are talking past one another a lot. — PolitiTweet.org

Nate Silver @NateSilver538

If you favor a more aggressive posture on reopening it might help to clarify which of these—not mutually exclusive—you agree with: 1 The health costs will be less bad than we assume 2 The economic/social costs are worse than we assume 3 We should be less willing to trade 2 for 1 — PolitiTweet.org

Nate Silver @NateSilver538

Is anybody collecting all of the serological studies in one place? — PolitiTweet.org