There are a lot of ways to lose a midterm—getting crushed among independents, for instance.

But one way is with a low turnout from your base (e.g. Dems in 2010) and I tend to think that won't be a problem for Dems in 2022 because their voters will see the stakes as existential.
By "existential" mean that many Democrats will think that "democracy itself is on the line" between the Big Lie (and the fact that Congressional majorities could make it easier for the GOP to overturn the 2024 election) and the increased attention to voting rights.
Why the minority party tends to gain at the midterms is a complicated question, but much of it is simply that voters want to check the party in power. However, if the minority party could use a good midterm to permanently entrench itself, the calculation is a lot different.
Also, the fact that Republicans *aren't* rebranding themselves when losing parties almost always do is significant. It's possible that the midterm gains that the minority party typically makes are premised upon doing so.

fivethirtyeight.com/features/why-t…

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Nate Silver

Nate Silver Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @NateSilver538

9 Apr
This is a long, interesting thread on how much more transmissible B.1.1.7 (aka "the UK variant") really is.
One thing I'd add is that contact-tracing data from the UK found an advantage of ~33% (as compared to the 50-70% that you sometimes see cited elsewhere) and I can imagine contact-tracing is more robust than methods based on statistical extrapolation.
assets.publishing.service.gov.uk/government/upl… Image
As a non-expert on this stuff, I don't necessarily trust my judgment to sort out competing expert claims.

However, I do find it interesting when there's a range of expert opinion and only a certain portion of that range tends to make it into popular media coverage about COVID.
Read 5 tweets
3 Apr
Good to see that COVID deaths in the US, after a hiccup, have resumed a decline, even as cases tick up again. ourworldindata.org/covid-deaths
Since deaths lag cases, maybe they'll start rising again. Or maybe not since we've vaccinated lots of seniors. I'm too chickenshit to make concrete predictions lol. But the fact that deaths are still declining *for now* seems like important context if you're writing about surges.
I also think people should be wary about drawing parallels to past surges. There might seem to be a certain inevitability: cases rise slowly at first, then quickly, then deaths rise too. But those came in a world without 3+ million vaccine doses being delivered daily.
Read 4 tweets
2 Apr
In many respects I find the COVID situation less worrying in NYC than in other places like Michigan.

In NYC, cases steady, hospitalizations & deaths declining. You'd like to see cases declining too; Hopefully that happens soon as more people get vaxxed.

www1.nyc.gov/site/doh/covid…
There was a lot of confusion caused by the fact that NYC didn't report data for several days, which created the misleading impression of a huge spike when it reported the backlogged data. News organizations should have done a much clearer job of flagging this.
On the other hand, there's a fairly rapid acceleration in cases in Western New York (i.e. Buffalo, etc.) That's going to affect the statewide numbers if it continues. forward.ny.gov/covid-19-regio…
Read 4 tweets
31 Mar
So here's how this came out. I have no idea about the "right" answer—I'd probably have chosen the popular 41K-65K bucket if forced to pick—but I'm glad there are sizable numbers in all 4 buckets because I don't trust highly confident predictions about COVID trajectories.
On the one hand, the case for optimism is sort of obvious. At some point, barring immune-escape variants, it seems likely we'll hit a threshold where rising levels of vaccination/immunity simply wins out over other factors, as in Israel right now.
But I wonder if people are underestimating the lags. It takes ~2 weeks from your 1st shot to have much protection at all & 5-6 wks to count as "fully vaccinated" with a 2-dose vaccine. Plus there are reporting lags. A "new" case today may actually have been acquired 1-2 wks ago.
Read 4 tweets
31 Mar
At some point if there's excess demand for vaccine in urban areas and excess supply in rural areas, that starts to reflect a policy failure of federal and state allocation formulas.
Most of this is presumably vaccine hesitancy being greater in rural (i.e. generally redder) areas. But I suspect also there are asymmetries caused by more people coming from the country to the city for vaccines than the other way around.
Example of that in New York, for instance:

amny.com/news/several-s…
Read 4 tweets
25 Mar
Our new pollster ratings are up!

They've been updated to reflect the results of the 2020 general election + the GA runoffs.

Also a shiny new interactive. Fivey Fox makes a cameo appearance.

Here's the link. I'll discuss some key findings in this 🧵.

projects.fivethirtyeight.com/pollster-ratin…
Note that we now have pages for individual pollsters. So you can see exactly which polls made it into the rating for each polling firm. Basically this means every poll within 3 weeks of an election since 1998!

projects.fivethirtyeight.com/pollster-ratin…
Or if you want to go even deeper, you can find the entire database on GitHub. We strongly encourage people to use this database for academic research, etc. A LOT of hours of gone into building and maintaining it.

github.com/fivethirtyeigh…
Read 17 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!