Share this page!

G. Elliott Morris Profile picture
Twitter logo
6 Oct, 4 tweets, 2 min read
Do you hear that sound? It’s the partisan non-response alarm going off
There can be non-response for vote choice in the 'independent' party category too, so just looking at the sample d/r numbers isn't quite the whole picture.

The +16 is probably a mix of both non-response and normal sampling deviation

If Biden is up by 8-9 like our model says then we would expect to see +16 polls in about one of every 30 samples, so this isn't like earth shatteringly weird
It's worth reminding y'all that our model (unlike others) adjusts for partisan non-response, so huge swings like this among polls that don't weight by party/past vote get discounted unless matched by polls that do have more stable political compositions

projects.economist.com/us-2020-foreca…

• • •

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

Keep Current with G. Elliott Morris

G. Elliott Morris 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 @gelliottmorris

G. Elliott Morris Profile picture
6 Oct
Trump is #actually pretty bad at politics. That's why his (net) approval rating has been historically stable and almost perennially low, and why he will very likely become the country's next elected one-term president (only the fifth since 1900). He's simply not good at this.
Trump's a minority pres who won the White House by drawing an inside straight in a few important states among a declining voting bloc. Since then, he has needed to expand his base to win re-election. But he has only ever catered to it—every time shrinking his electoral prospects.
If Trump wins again, his victory will be attributable in large part to years of trickery and attacks against higher election turnout. The simple fact of today's GOP is that it is a minority party, in power due only to electoral institutions that disproportionately advantage them.
Read 5 tweets
G. Elliott Morris Profile picture
6 Oct
Biden +11 in post-debate poll in Pennsylvania from Monmouth monmouth.edu/polling-instit…
Unlike the CNN poll earlier today, this one shows an increasing Biden margin among a sample that is actually more Republican than their July survey
Hard to overstate how bad polls are for the president right now.
Read 4 tweets
G. Elliott Morris Profile picture
5 Oct
I've gotten a lot of questions about our model's projection in Arizona today (probably because of the NYT poll). People think it should be bluer. Maybe! Our model has it closer to 50-50 (though on the verge of lean blue) than other models (eg 538) for a few reasons.
First, our model is picking up some real pro-Biden pollster-level and mode effects in many of the polls coming out of AZ. I remove them, Biden's current adjusted polling margin there increases from three to four percentage points, and same with the election-day prediction. Image
A second factor is that polls were a lot closer a week or two ago, and our model thinks that information is still valuable. And since Biden has only gained ~1 point or so nationally and in states like AZ over that period, the model doesn't want to adjust too much to newer data. Image
Read 8 tweets
G. Elliott Morris Profile picture
3 Oct
Early polling suggests the news that POTUS held a mostly-unmasked super-spreader in the Rose Garden last Sat, didn't get tested b4 the debate on Tues, held an in-person rally on Wed & tried to hide his own case until an aide's leaked on Thurs is going badly for the White House.
... not quite sure what else they could have expected
In times like these it helps to be especially mindful both that (a) Twitter is not real life and (b) that polls that don’t weight by party/past vote are prone to non-response, though the data I’ve seen (public & private) are pointing to real shifts, even if they’re only temporary
Read 4 tweets
G. Elliott Morris Profile picture
2 Oct
October 2, projected final election day vote for Clinton and Biden

National: Biden +8 (Clinton +3)

AK: -7 (-14)
TX: -5 (-13)
IA: -2 (-3)
OH: -1 (-3)
GA: 0 (-6)
AZ: 2 (-7)
NC: 2 (-2)
FL: 3 (-1)
NV: 6 (1)
PA: 6 (3)
MI: 7 (5)
NH: 7 (2)
WI: 7 (5)
MN: 8 (6)

projects.economist.com/us-2020-foreca…
this is estimate final results for all ballots
forgot the charts!
Read 4 tweets
G. Elliott Morris Profile picture
2 Oct
At the risk of stating the obvious, an incumbent president contracting a deadly virus is not something that’s in the training set for our election model. My prior is that this has the _potential_ to cause a 99th-percentile shift in the race, while still not being all that likely.
Our model explores a range of outcomes for potential poll change between now and Election Day. It’s constrained by the historical distribution of poll swings in the last month of last campaigns. 99th-%ile means a change that comes outside the margin of error of that distribution.
I need to think about this more. On the one hand, our model says that a poll swing from a president getting a deadly virus is a 99th-percentile (IE very rare) event. Surely that’s the case with Trump’s diagnoses? A 1-in-100 events? And, it might not even change anything.
Read 7 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!

Like this author's thread?

Get email notifications when new thread is out from this Author!

Check out other threads from this Author!

Read all threads by @gelliottmorris