Nate Silver

Nate Silver @NateSilver538

@joshtpm Yeah, the quality of data from state to state seems to vary a lot, as does how comprehensive they are being in trying to count COVID deaths. So this might be something more for down the road when data quality has improved. — PolitiTweet.org

Nate Silver @NateSilver538

I'm not sure that lockdowns themselves are likely to kill people. They may actually save lives in the short run (e.g. fewer traffic deaths). However, failure to seek medical care (or not being able to receive it) could be a big contributor to mortality. — PolitiTweet.org

Nate Silver @NateSilver538

It might be useful to look at excess mortality in states/counties that *don't* seem to have had bad COVID-19 outbreaks, but which engaged in lockdowns anyway similar to states/countries that did. That would help you to separate the impact of lockdowns from missed COVID deaths. — PolitiTweet.org

Nate Silver @NateSilver538

p.s. if you're wondering about whether Cuomo is being this *specific*, yes he is. https://t.co/8KWFiPnq07 — PolitiTweet.org

Nate Silver @NateSilver538

I actually think there's likely to be a fair amount of consensus over what Cuomo says is his explicit strategy: gradually open stuff up, but stop/reverse when either the spread starts meaningfully increasing (R > 1.1) OR there are signs of hospital/ICU strain (>70% capacity). — PolitiTweet.org

Nate Silver @NateSilver538

RT @maggiekb1: The differences in how men and women respond to viral infections is a complicated (and not always easily predictable) mixtur… — PolitiTweet.org

Nate Silver @NateSilver538

Given what we seem to be learning about outdoor transmission, this seems like a way to possibly erode public support for lockdown measures while not necessarily getting a huge amount of mileage in terms of public health. https://t.co/Yoh1dIwhjt — PolitiTweet.org

Nate Silver @NateSilver538

Meanwhile, most European countries are talking about opening up on timelines equally or slightly sooner than many US states, including in cases where they had bad outbreaks. — PolitiTweet.org

Nate Silver @NateSilver538

Most attempts to put partisan frames on the "when are states re-opening?" question don't hold up super well to scrutiny. There doesn't seem to be that much difference between blue states and red states' plans, especially adjusting for the number of cases in their states, etc. — PolitiTweet.org

Nate Silver @NateSilver538

Seeing a story retweeted into my feed a lot that I think it's 50:50 whether it's actually true or not. I don't want to sound snobbish but often best to be wary when highest-quality news outlets aren't reporting a story that other outlets are reporting. — PolitiTweet.org

Nate Silver @NateSilver538

@JamesSurowiecki So when I see Swedish authorities expressing surprise at the death toll, I wonder if they're being facetious. Their fatality rate as a share of the number of infections they *claim* to have (they claim 30% infected in Stockholm) is actually fairly low. — PolitiTweet.org

Nate Silver @NateSilver538

@JamesSurowiecki A herky-jerky response could conceivably be worse. But, of course, delaying infections allows you to benefit from new drugs/therapies that are developed, a vaccine, etc. — PolitiTweet.org

Nate Silver @NateSilver538

@JamesSurowiecki I don't think it's a bet on reducing deaths. It's a bet that herd immunity is inevitable and/or too costly to avoid, therefore you get it over with as quickly as possible but with the constraint that it isn't so quick as to overwhelm hospitals and nursing homes. — PolitiTweet.org

Nate Silver @NateSilver538

@EsotericCD I'm a hot dog guy *and* a Michigan guy, but I don't particularly like Detroit-style Coney Island hot dogs. — PolitiTweet.org

Nate Silver @NateSilver538

@JohnJHarwood @mattyglesias In theory, if you do a month with an R of 0.85 then you could cut down on new infections by about 60%. — PolitiTweet.org

Nate Silver @NateSilver538

@JohnJHarwood @mattyglesias It's positive, yes, but it's not *much* below 1, which implies that you can't relax social distancing that much without it maybe going over 1. — PolitiTweet.org

Nate Silver @NateSilver538

@JohnJHarwood @mattyglesias I would guess that it varies between 0.6 and 1.0 in states that are attempting social distancing. Mostly toward the higher end of that range (e.g. 0.85 or 0.9) as the states lower than about 0.8 usually have something unique going for them that may not be replicable elsewhere. — PolitiTweet.org

Nate Silver @NateSilver538

@mattyglesias In some ways, we're in the most complicated scenario from a policy standpoint, which is NOT meant to be confused with the worst-case scenario. If social distancing worked super duper well (gets R to 0.5) or only mediocrely (R only down to 1.2) we'd have a clearer path to follow. — PolitiTweet.org

Nate Silver @NateSilver538

@mattyglesias You could do an essay about how "flattening the curve" is similar to "Medicare for All" in this respect. I don't know if FTCers were deliberately tolerating ambiguity as a consensus-building strategy as some M4Aers were, but it meant different things to different people. — PolitiTweet.org

Nate Silver @NateSilver538

@kmedved @CountingBaskets The other thing that kind of swayed me a bit is that ICU entries seem to have declined for a couple of weeks running in Stockholm (while remaining ~steady in the rest of Sweden). That's certainly an interesting data point. — PolitiTweet.org

Nate Silver @NateSilver538

@kmedved @CountingBaskets Also, 30% is their claim for Stockholm, not Sweden. Stockholm is like 40% of ICU entries in Sweden but a considerably lower share of the population. — PolitiTweet.org

Nate Silver @NateSilver538

Between the *current* wave of COVID-19 remaining bad in many states, the possibility of either a long plateau or a second wave, and states going back and counting deaths they missed originally, any projections that have 5-figure death tallies for the US this year seem optimistic. — PolitiTweet.org

Nate Silver @NateSilver538

Some of this may be states going back and counting deaths that they missed originally. But still, since a lot of deaths *were* missed, you can expect to see more days like this where states try to catch up to the *true* number of fatalities. — PolitiTweet.org

Nate Silver @NateSilver538

A soberingly high death tally on a day that brought some good news on other fronts (e.g. clinical trials). Death tolls have been bouncing around lately as states seem to be reporting them in batches; 479 in PA today, for example. (PA typically reports 75-100 per day.) — PolitiTweet.org

Nate Silver @NateSilver538

US daily numbers via @COVID19Tracking: Newly reported deaths: Today: 2,700* Yesterday: 2,198 One week ago (4/22): 2,037 Newly reported cases: T: 28K Y: 24K 4/22: 28K Newly reported tests: T: 230K Y: 202K 4/22: 314K Share of tests positive: T: 12% Y: 12% 4/22: 9% * New high — PolitiTweet.org

Nate Silver @NateSilver538

@asymmetricinfo It does, but that helps to explain why things got so bad in NYC. You likely had 2-3 weeks (very late February and early-mid March) of very rapid spread before people started wising up, which roughly corresponds to the average length of an ICU stay. — PolitiTweet.org

Nate Silver @NateSilver538

There's also been a hell of a price. Sweden has death rates (importantly: as a share of the overall population, *not* of infections) 3-7x higher than their neighbors. And I hesitate to think what would happen if we tried this in the US with a less healthy population. — PolitiTweet.org

Nate Silver @NateSilver538

So what's misunderstood about Sweden 1) They ARE doing a fair amount of distancing even if some high-visibility stuff (e.g. restaurants) remains open. 2) ICU rates from COVID-19 are not all *that* high especially if you have a healthy population that treats its old people well. — PolitiTweet.org

Nate Silver @NateSilver538

By contrast, NYC likely had relatively unmitigated spread at (very rough guesses) an R of 2.5 to 3.0 (maybe even higher) for 2-3 weeks. That can be enough to overwhelm ICUs, even if the share of people who require ICUs is relatively low as a % of all infections. — PolitiTweet.org

Nate Silver @NateSilver538

Why aren't Stockholm's ICU's overwhelmed? Well, because ICU's being overwhelmed is a function of the *flow* and Stockholm's flow is never that high. Sweden *is* doing a lot of distancing, just not enough to get R < 1. It's the original "flatten the curve" strategy. — PolitiTweet.org