Nate Silver

Nate Silver @NateSilver538

The media was also a few weeks slow in paying attention to the *rise* of Delta. It's basically just like coverage of polling during an election campaign: the narrative is always two weeks behind the actual news. — PolitiTweet.org

Nate Silver @NateSilver538

I saw lots of people tweeting this thread the other week but I don't think folks are really grappling with its argument, which is that a lot of you are fighting the last war when it comes to policy and all of the options now involve some version of "learning to live with" COVID. — PolitiTweet.org

Nate Silver @NateSilver538

Kind of crazy Australia still makes it extremely cumbersome to get the AstraZeneca vaccine if you're under 60 (they don't have many doses of other vaccines) at a time when they're battling Delta and half the country is under a strict indefinite lockdown. https://t.co/eDJeFTqkrY https://t.co/1jrWg2Kaft — PolitiTweet.org

Nate Silver @NateSilver538

"Who is morally culpable for COVID spread?" is a pretty different question from "What would have happened to the trajectory of the virus had a 50% more transmissible variant not emerged?". They're both important Qs but fusing them together probably doesn't help answer either one. — PolitiTweet.org

Nate Silver @NateSilver538

Not sure I'd agree that the vaccines are "not terribly effective" against Delta transmission. With that said, I agree that the NYT story's perspective is strange given countries with higher vaccination rates than the US (e.g. Israel, the UK) have also had issues with Delta. — PolitiTweet.org

Nate Silver @NateSilver538

@ZoeMcLaren It's not clear for me whether there is some fixed number of doses, as opposed to orders driving additional incentives to increase supply. Also politically it would be quite a mess for any US president if an avoidable wave were to occur because we hadn't ordered enough boosters. — PolitiTweet.org

Nate Silver @NateSilver538

Here's a paper with one hypothesis explaining why. https://t.co/AhSzDSrj1M — PolitiTweet.org

Nate Silver @NateSilver538

In the UK, Delta's transmissability advantage relative to Alpha seems to have declined. There was also this pattern with Alpha vs. the original strain. At first, it had a huge transmissibility advantage. But it diminished as more people were infected. https://t.co/2FaC7gtNir — PolitiTweet.org

Nate Silver @NateSilver538

(Rare) point of disagreement w/Gottlieb here in that sometimes saying "we don't know" is the best one can do IMO and that's better than providing false certainty. But it does seem like the efforts to model COVID spread have been wanting. https://t.co/ylLKHfMaWW — PolitiTweet.org

Nate Silver @NateSilver538

Delta is clearly very contagious but I wish we had better explanations for the rather abrupt turnaround in cases in India and now (fingers crossed) in the UK. — PolitiTweet.org

Nate Silver @NateSilver538

There are a lot of studies that control for confounders, other than confounders that are hard to control for, and in some cases that's probably worse than not controlling for confounders at all. — PolitiTweet.org

Nate Silver @NateSilver538

This case is more complicated than some of the others, there's a lot of conflicting data to evaluate, but in general the profession is way too comfortable with the notion of the "noble lie". — PolitiTweet.org

Nate Silver @NateSilver538

The discourse about booster shots has been weird lately and this helps explain why: Public health officials are worried it could deter people from getting their 1st and 2nd doses. Understandable, I guess, but I'm tired of their efforts at pop psychology. https://t.co/BDqqdse1Q8 https://t.co/xTI1VNfiJQ — PolitiTweet.org

Nate Silver @NateSilver538

It's reflective of certain type of (I would argue somewhat incoherent) liberal ideology that restrictions placed as the unvaccinated are portrayed as "coercive" whereas restrictions placed on *everybody* (e.g. lockdowns) generally were not framed that way. https://t.co/itqJmRmuOi https://t.co/ZKBC7xjldC — PolitiTweet.org

Nate Silver @NateSilver538

RT @davidshor: For the record the negative impact of gerrymandering is ~20X larger than the theoretical upper bound of a massively well fun… — PolitiTweet.org

Nate Silver @NateSilver538

180-190 million Americans have already received at least one vaccine dose and we'll easily top 200 million soon especially once vaccines are authorized for children under 12. There's a good chance we'll need boosters at some point. If not hopefully we can send these doses abroad. — PolitiTweet.org

Nate Silver @NateSilver538

RT @AlexSamuelsx5: Do you see yourself as a political anomaly? Often, certain voting groups are often portrayed as monolithic even though… — PolitiTweet.org

Nate Silver @NateSilver538

@seeglazek @JamesSurowiecki I mean I tend to agree that the 90%+ efficacy numbers seem unlikely vs Delta. But if you look at the consensus it would seem to suggest ~80% for 2 doses of mRNA which is sort of my working assumption at the point. — PolitiTweet.org

Nate Silver @NateSilver538

@seeglazek @JamesSurowiecki I don't know how to make sense of it all but I *don't* think the way to do it is by eyeballing the data or by relying on anecdotes from one's non-heterogeneous peers. — PolitiTweet.org

Nate Silver @NateSilver538

@seeglazek @JamesSurowiecki Yeah but there are other studies including some in peer reviewed journals that just came out today and show much higher efficiency. This feels a bit like polling analysis where people prefer the data points that suit their priors. — PolitiTweet.org

Nate Silver @NateSilver538

RT @elena___mejia: FiveThirtyEight is looking for a data viz intern to join us in the fall! Come work with us! And please DM me with any qu… — PolitiTweet.org

Nate Silver @NateSilver538

@seeglazek @JamesSurowiecki There are a lot of complications and confounders, some of which are counterintuitive including e.g. that not all people are equally likely to be tested. And there are rigorous studies designed to deal with those. I don't think this is a case where eyeballing the data adds value. — PolitiTweet.org

Nate Silver @NateSilver538

@seeglazek @JamesSurowiecki Why is it not consistent? 45% of the UK population is not fully vaxed (and those unvaccinated people clustered in certain age ranges and geographic areas and so can spread to one another) and Delta has an incredibly high R0 among the unvaccinated. And yeah AZ < mRNA. — PolitiTweet.org

Nate Silver @NateSilver538

@seeglazek @JamesSurowiecki What happened in Provincetown isn't that relevant because infections are clustered so it's ~n=1. Nearly every beach town in a temperate climate had lots of visitors on July 4. What happened on Fire Island? On Martha's Vineyard? In Bar Harbor, ME? On Long Beach Island, NJ? — PolitiTweet.org

Nate Silver @NateSilver538

@seeglazek @JamesSurowiecki The studies are all over the place in re: how effective the vaccines are against mild infections from Delta, from mid 60s through low 90s. I'm pretty skeptical that people's evidence from their social circles can meaningfully distinguish between the higher and lower estimates. — PolitiTweet.org

Nate Silver @NateSilver538

Also wish one of these stories interviewed someone with a breakthrough infection who was like "I understood the risks were low but not zero, I thought it was worth it to participate in this activity that was important to me, and having gotten unlucky doesn't change anything." — PolitiTweet.org

Nate Silver @NateSilver538

@seeglazek It's consistent because infections are clustered so it's not as though you're drawing from a random sample. — PolitiTweet.org

Nate Silver @NateSilver538

In a country with 162 million fully vaccinated people, most of whom have resumed some version of their pre-pandemic lives, you'll have your fair share of breakthroughs. And of course some of them will be clustered, just like other infections are. — PolitiTweet.org

Nate Silver @NateSilver538

Part of what I'm critiquing is articles like this one that draw an implicit tension between the statistical evidence of fairly low (and almost always non-severe) rates of breakthrough infections with anecdotal examples of where such infections did occur. https://t.co/alKn8GCqit — PolitiTweet.org

Nate Silver @NateSilver538

A big part of "why are we hearing about so many more breakthrough infections right now?" is simply that there's more community spread in the US, period. Cases are up about 4-fold from the early summer trough, so there are a lot more opportunities for rare spillovers to occur. — PolitiTweet.org