Share this page!

Lakshya Jain Profile picture
Twitter logo
11 May, 4 tweets, 1 min read
The question is whether it’s worth spending heavily in a race you’ll likely lose, or whether it’s better to go all-in on the three battleground races in PA/WI/NC for Democrats.
Spending in Ohio has advantages, including forcing the GOP to spend some money there that could go elsewhere. And it’s really about widening the field of play. But another complication is that the more nationalized this gets, the tougher it is for Ds to win.
Ultimately, if Tim Ryan is to pull it off, the election has to avoid being nationalized. There’s just one too many Republicans in Ohio for Democrats to win a high-turnout Senate election here. He should be backed with funds, but the race getting nationalized would be a problem.
In case it isn’t clear I have absolutely no idea where I fall on this question either. I change my mind by the day. I think they should spend, I just think that it’s...not necessarily as crystal clear as “$$ = better chance” after a certain point here.

• • •

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

Keep Current with Lakshya Jain

Lakshya Jain 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 @lxeagle17

Lakshya Jain Profile picture
13 May
2014 was a non-voting midterm -- Democrats got screwed over by differential turnout. But again, turnout among whites was decently correlated with education.

The problem? College whites were significantly more Republican than they are now. Image
Regression plots 2014_votes/2012_votes, weighted by county CVAP. I picked the South, because the educational and racial polarization makes it a fairly insightful case to analyze. The states I considered: North Carolina, South Carolina, Georgia, Alabama, Mississippi, Louisiana.
There's another angle to this too -- Hispanics still lean heavily D, and their turnout tends to dip in midterms. That counters the educated voter difference that Democrats have in some states, and they'll need to work heavily to make sure they aren't hurt badly by this.
Read 4 tweets
Lakshya Jain Profile picture
11 May
At what point do you go from serious candidate to statewide joke?
There's this myth among several Republicans that John James is an exceptional candidate because he gets them more of the Black vote.

The problem with this narrative is that it is provably false. He got 80 (!) more votes than Trump in the city of Detroit, which is ~80% Black.
James outran Trump in areas like Kent County, which were far more ancestrally friendly to Republicans and have been swinging towards Democrats. This is an area where he'll probably need to do a lot better than expected in 2022 if he is to beat Whitmer.
Read 4 tweets
Lakshya Jain Profile picture
10 May
I seriously think one of the biggest blind spots in political forecasting is the average pundit's unawareness of how uninformed the median voter actually is. Most people don't live and breathe electoral politics. Just because *you* know something doesn't mean everyone else does.
"shifting this state legislature/county commissioner race from likely to tilt D because of a minor scandal where the legislator lived outside his district for 5 years" bro. absolutely *nobody* cares. sorry, but the median voter doesn't even know who the hell that even *is*.
There are very few events of saliency that meaningfully make it to people, and once a party has a certain image, it is very, very difficult to shed it because of how few events even reach the eyes of most voters.
Read 4 tweets
Lakshya Jain Profile picture
9 May
I do want to explain this in more detail, so here’s a thread:

1) this bot is centered around the 2012/16/20 elections. In that set, Democrats win the popular vote, so your median environment projected onto 2024 demographics is D-leaning. But the PV doesn’t choose the winner!
2) Democrats have seen their support vanish in many rural areas but have seen pretty consistent suburban swings towards them. So the bot naturally thinks this will continue. The way the GOP wins the PV is by suburban reversion, which, if the UK showed anything, isn’t likely.
3) The GOP coalition wins are driven by a pretty structural bias in the electoral college and one that doesn’t look like it’ll vanish for a while. It’s remarkably efficient. And do long as Texas stays redder than the nation, I don’t see that advantage going away.
Read 5 tweets
Lakshya Jain Profile picture
5 May
You get the Hispanic reversion people have been discussing in this simulation, but a map like this really shows how even slight slippage with the white vote electorally dooms Democrats because of the racial balance of the United States and the electoral college.
That's not something anyone likes to hear, but the reality is that if the Democrats want a majority coalition in the near future, they probably need to continue making some serious gains with the educated white vote, especially if Hispanic margins continue to shrink.
The reality, as @Nate_Cohn mentioned, is that the diversification of the United States hasn't necessarily resulted in the overwhelmingly pro-Democratic electorate everyone thought it'd give. But their small gains with the white vote have outsized effects.
Read 4 tweets
Lakshya Jain Profile picture
2 May
Model update -- @bot_2024 V3.0 is out! @notkavi and I have completely rebuilt it to use more generic data across elections and better model partisanship and turnout on a demographic level, which should give us more interesting and realistic coalitions for 2024. Details in thread! Image
This model projects 2024 demographics from the available 2012, 2016, and 2020 demographic data. It then takes in county-level demographic data for 2012, 2016, and 2020 and uses it to construct underlying demographic features for the electorate.
We use pretty interesting demographic features commonly available across the 2012/2016/2020 electorates, such as CVAP with racial breakdown, poverty, income, religion, and education. That also helps us get more information about what's happening with the underlying electorate.
Read 11 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 emails when @lxeagle17 has new unrolls!

Check out other threads from @lxeagle17!

Read all threads by @lxeagle17

:(