Share this page!

G. Elliott Morris Profile picture
Twitter logo
9 Sep, 3 tweets, 1 min read
I'm on the record as saying that the "red mirage" study that Bloomberg highlighted is wildly inaccurate. It assumes that mail-in ballots get counted at a rate of 15% per day. But in 2016, the vast majority in all the close states were counted by 12AM election night. 1/2
2/2 This doesn't mean that I don't think a mismatch between the election-night results and eventual outcome is impossible. If the race is close, that could happen even with a mild differential in mail v early/in-person counting rates. But it's not that _likely_ at all.
Yes, we should expect that counting will be slower bc of increased volume. But I will happily bet against anyone who thinks that it will be 15% a day. Most election workers in most states can tell you that's a rather remote possibility, even during covid.

• • •

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
7 Sep
About right
I'm not sure how much an "improving" economy will help Trump. For one thing, the most predictive measurement of growth in the past has been annual change in a variety of indicators, which is almost certain to still be negative in Nov. For another, this: thecrosstab.substack.com/p/polls-sugges…
There's also a poll sci lit Q here. We have observed almost 0 correlation between Trump's approval ratings and readings of economic growth AND consumer sentiment over the past 4 years. The hypothesis that bad economy = bad for the president might not work out this time.
Read 7 tweets
G. Elliott Morris Profile picture
30 Aug
This is important context for people sharing the recent decline in support for Black Lives Matter or drawing inferences from Kenosha:

“Today, 43 percent say the country will become less safe if Trump is reelected. Just 32 percent say it will get safer.” news.yahoo.com/new-yahoo-news…
If voters prefer Biden on safety, I’m not sure how the unrest in Wisconsin is supposed to be good for Trump. The “law and order!” messaging has been a flop all summer.
Per usual, the best strategy for Biden (and really any candidate who is up by 9 points) is probably to lay low and not doing anything the media can make a controversy out of
Read 5 tweets
G. Elliott Morris Profile picture
29 Aug
Some thoughts on this Morning Consult poll that everyone is losing their minds over. 1/5

morningconsult.com/2020/08/29/pos…
First, the change in the race from the last poll to this one (+10 to +6) is insignificant, especially considering the design effect on the data (online non-prob polls, esp from marketplace panels, have significant non-sampling error). So you can't conclude a bounce from it alone.
Second, it's just one poll, and other data might suggest something different. You should average the data together! There hasn't been any movement yet in the USC Dornsife poll, for example, though it has yet to have a sample entirely post-RNC.
Read 5 tweets
G. Elliott Morris Profile picture
28 Aug
I will offer an olive branch. There is a lot of value in incorporating context and prior judgment or "subjectivity" in statistical models (that's why being Bayesian is so great!). A lot of what Dave says in this thread is reasonable. Let me just respond to add my point of view.
I want to avoid being argumentative here. Hopefully, in exposing people to both points of view, we can learn something from the disagreements between our methods.
First, just to quantify differences and put them in context, let's say Dave gives Biden a roughly 67% chance of winning today (between hsi 60 and 75). Our model has him at 88%. That's a big difference! However, I think it's actually pretty easy to explain.
Read 24 tweets
G. Elliott Morris Profile picture
26 Aug
OK, I keep seeing this convention hot take that hashtag #ACKSHUALLY the way for Trump to win is to target his convention to loyal Republicans and increase their turnout by 10 points or more over Ds (which is totally possible under the theory). Here's why that's wrong. 1/n
Trump would need to drive nearly impossibel levels of differential turnout to close a 9-10 percentage point gap in the polls based on turnout alone.

Here is one table that shows a hypothetical election where Biden is up 9.5 and 65% of all Ds, Rs and Is turn out to vote.
Now let's crank Republican turnout up to 70%, 5 points higher than Dem or Indie turnout. Already this is a pretty large turnout differential for presidential elections, and I think it's unlikely a convention could cause this on its own.

This nudges Trump up from -9.6 to -7.4
Read 7 tweets
G. Elliott Morris Profile picture
20 Aug
#NEW I did some math & wrote a piece about what trouble with the USPS and vote-by-mail actually means in November. The takeaway is that above-average VBM usage & ballot rejection rates could cost Biden 0.6 points in vote margin on average. Thread economist.com/united-states/…
I was particularly irked by some of the less-thought-through analyses on vote-by-mail going around in recent weeks. No, the effects aren't large enough to decrease Biden's vote margin by 4 points except in the most extreme cases of rejection rates and partisan VBM composition.
But focusing on the doomsday scenario probably obscures the danger.

If VBM usage is high & mostly Democratic (polls put the composition of postal ballots at 70-80% Democratic right now), & ballot rejection rates are slightly higher than average, Biden could easily lose 1-2pts.
Read 14 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