Challenge accepted. No models: simple over-performance (House - pres) by caucus. This way we don't have to argue over the definition of "moderate". In sum: Blue Dog > Blue Collar > New Dems > The Squad Candidates who want to win in tough districts would do well to mimic over-performers.
Caucus membership _is_ a model and one that has a huge selection effect on the outcome
isn't the fundamental issue here that it is not surprising to me that kamala harris did poorly in blue dog districts, and i'm not sure if that tells me much about how different non-kamala-harris house candidates in those districts. i'm just not convinced this is the right baseline?
What about House 2022 - Biden 2020? But, these quibbles are beside the point: we need candidates who will win/over-perform in these purple districts, and folks like MGP and Golden are showing us the way! There's a reason Jamie McLeod-Skinner isn't a member of Congress
here is a list of dem losers in close districts in 2024. all of them are to the right of the median democrat in bonica's composite score, and several of the big losers are _very_ moderate (peltola). how many of them do you think would have won by moving more to the center?
(and to answer your question, i think blind comparisons to past presidents results suffer from the same problem -- it ignores all the other factors that may explain the differences in house and pres results that aren't "moderation" — which is one reason we built our WAR)
I don't pretend to know all these candidates individually, but (on avg) they would have lost votes by mimicking The Squad. I bet they would've won more votes by being loud about heterodox popular (sometimes local) issues. Can't win every race of course, but I want to maximize our chances.
another crux of the debate here is that people are pretending that squad vs cuellar is the choice dems are making in close districts, when the reality of the challenger data is taht it's golden vs cartwright
That's correct. Squad vs Cuellar is the debate held amongst certain consultant/org circles. But when push comes to shove, the *campaign* consultants working in purple districts understand all of this and there are revealed preferences. Totally agree with you there.
But it all comes to a head when orgs push for progressives in purple seats. These primary unforced errors stay with me.
this sparked a good idea. i think we may be able to answer more of the questions we want to get at by looking at changes in house vote share as a function of changes in ideological score when a primary defeat happens. Anecdotally, 2022 was a tough year and Chavez-DeRemer has a high WAR
12/14 have positive WAR. Almost all of them outperformed. These races were only close because of their moderation for the most part. Mary Peltola got 46% of the Alaska vote compared to Harris/Walz at 41%!
I intentionally include 2022 because the premise of "be super progressive & win more votes" is that the base will turn out. House candidates have more sway over turnout in midterms years, so the Squad should have over-performed more in 22. Didn't happen. Spreadsheet: docs.google.com/spreadsheets...
Bonica's ally in this discussion, @gelliottmorris.com, is closer to the mark. Over-performers take heterodox, popular positions. That way they aren't seen as "the usual Democrat". (In contrast: corporatist positions like Sinema's siding with Pharma against lower drug prices is a very bad idea.)
That begs the question: what's popular? I'm often reminded on this classic chart from 2016 showing that cross-pressured voters are (on avg) economically liberal and socially conservative.
One piece that I'd love to see more discussion here is why "climate" and "energy" have been consistently thrown into the category of helpful to take heterodox positions. They're not social issues (or, I guess if they are that raises the second order question of what is social vs econ).
And to be clear, I'm talking about energy/climate qua IRA not opposition to fraking. Though of course that then raises the question of why anti-climate is "good" rather than like Senimas corporatist heterodoxy. Feels like there's just a lot of handwaving on that issue in particular.
Happy to answer! It's because the far left side of the issue will increase cost of living (for a good, crucial cause of course). There are "all of the above" approaches that both increase clean energy and lower bills -- but caution is warranted bc the stereotype puts Dems on the wrong side of voters
Makes sense. Seems like then there is a way to highlight clean energy support while loudly opposing certain energy restrictions. And maybe opens door to characterizing GOP anti-climate obsession (anti-solar/wind) as leading to cost increases for purely ideological reasons (GOP's version of "woke")
That helps explain why the Blue Collar Caucus does so well. Candidates who display their solidarity with the working class on economic issues send an excellent signal to voters. Also, per the theory (Zaller), less-engaged/less-educated Americans are a key swing demo.
One place where I do agree with @adambonica.bsky.social is that when you take popular, heterodox positions, be *loud* about it. Remember Manchin's shooting of the cap & trade bill? www.youtube.com/watch?v=xIJO...
In meme form, I'm curious if Bonica believes that Manchin would have done just as well if he curated his brand as a progressive. eg, Feingold was well-known and well-loved by progressives, but that didn't help him over-perform H Clinton in 2016. And 1pp of over-performance would have been enough!
See today's Slow Boring by @mattyglesias.bsky.social for the larger picture. One key idea is that the biggest benefits may come from our 2028 nominee who can enlarge the Dem party's tent. www.slowboring.com/p/moderation...
Finally, none of this analysis would have been possible without the herculean and consistent efforts of @the-downballot.com and @drewsav.bsky.social to determine presidential vote by CD. Thank you everyone who does that work!
Ok, but based on polling about access to abortion and birth control as well as gay marriage, it wouldn’t seem that the majority of Americans are all that socially conservative. I think most people also still favor a separation of church & state. So are you only looking at other social issues?
So, 2022 is exactly like 2026. #gaslighting is ugly, no matter who does it I watch 2025 town halls. Things are different is an understatement
These numbers are actually very similar to the more complex WAR based models, which also give about a 3% benefit for going from least to most moderate, not too far from the 3.9% here. The question is whether that 3-4% is small or big, imo it's pretty big in today's world of straight ticket voters.
Isn't there some survivorship bias here? The Blue Dogs all represent 50/50 or even lean-red districts. If they are in Congress they are over-performers by definition.
Only some. To wit: why can't the Squad over-perform? Why are there no progressives representing Trump districts?
I think this is a big issue. Obviously ppl in close districts that *win* are going to be overperformers. Also candidates in swing districts are way more likely to self identify in a moderate caucus. Don’t see how you can draw actionable takeaways from this.
The Squad wins and they aren't over-performers. Where are the progressive over-performers in purple districts who ID with the Squad? What's stopping them? (Other than the general election electorate obviously...viz, Kara Eastman)
Your data convinces me that ppl in close districts that win over perform the national ticket (obvious) and are more likely to self-identify as moderate. What else are you concluding?
I'm also concluding that the electoral strategy used by the Squad does not win net votes (_maybe_ more base votes, but fewer swing votes) and thus is a poor electoral strategy for use by purple-state candidates.
Ok thanks, I appreciate the thoughtful engagement. I’d be interested to see data to this effect. I don’t think this methodology allows for a causal assessment like that.
The anecdata (eg, Eastman, Gill, McCleod-Skinner) also indicate that progressives perform poorly in swing districts -- but would love more systematic data there.
all due respect your original post compared blue dogs to the squad - but just showed that ppl winning in close districts (by definition) outperform. Now you are asking a different question: would the squad perform better by moderating? Also interesting, but not what you originally looked at.
And the original post included the Blue Collar coalition -- don't sleep on them! I think it's interesting that candidates who glom onto Blue Collar/worker/econ issues over-perform in their run-of-the-mill districts.
Yes. Interesting but could just be correlation, which is not what yglesias is suggesting.