Polling Anomalies

Is this a candidate-, rather than vibes-driven election?

8

Comments

Monday’s episode of The Daily, “Midterm Surprise: Democrats Are Surging in Red America,” features a conversation with host Michael Barbaro and NYT chief political analyst Nate Cohn. Despite the headline, it is no longer surprising that Democrats are polling well in states Trump won in 2024. What is surprising, at least to me, is that they’re doing that while simultaneously struggling in some Democratic strongholds.

To put the headline in context, though:

Cohn: [W]e polled the five red states where the Democrats seemed to have their best chance of winning, Ohio, Texas, Iowa, Alaska, and Kansas. All five of those states voted for Trump by at least 11 points in 2024. And we found the Democrats were really highly competitive in all five of them. We found them with meaningful leads in three of the states, Texas, Alaska, and Ohio. We found the race tied in Kansas, and we had the Republicans with a nominal one point edge in Iowa.

[…]

It’s extraordinary. I mean, to take a historic perspective on this, for the Democrats to even win one of these states would really be exceptional. It has been 16 years since a party has managed to win even one state that leans so much toward the other side in a regularly scheduled Senate election. So to contemplate that the Democrats have a serious chance to not just win two, but perhaps even more, this is something that we just haven’t seen in this era of polarized politics.

His analysis as to why this is happening isn’t novel:

This is an exceptional set of circumstances with a really unpopular Republican president, in some cases weak Republican candidates, and in all five cases strong Democratic candidates.

This is followed by a dive into each of those candidates. But, again, what prompted me to write about it is this:

Cohn: [I]n addition to the five red states that we just pulled, we also surveyed three blue to purple states, Maine, Michigan, and New Hampshire. And despite these enormous Democratic breakthroughs in the Red states, the races are still pretty competitive in these more Democratic-leaning states.

Barbaro: And why is that?

Cohn: Well, the answer ultimately comes back to candidate quality again. Donald Trump is extraordinarily unpopular in all of these states, but the Republicans have either nominated really strong candidates who are relatively moderate and have well-established bases of support in the state, or the Democrats have nominated someone who has the potential, at least, to alienate some voters who would otherwise vote for a Democrat.

Again, this is followed by a breakdown of the candidates in those races.

Intuitively, I find the explanation satisfying. Going back at least to the 2010 election, which I wrote about extensively in real time, Republicans have handicapped themselves by nominating extremist and/or wildly unappealing and/or unqualified candidates. It’s largely why Georgia has two Democratic Senators. And why Doug Jones briefly represented Alabama in the Senate.

But the political science literature is pretty clear: voters who go for “the man, not the party” are exceedingly rare and getting rarer. In our hyper-polarized era, voting for the other party’s candidate is just too costly. Everything from the control of the Supreme Court to transgender rights to abortion rights is at stake.

As Cohn himself notes, “there is and always has been a meaningful part of the Republican Party that has serious reservations about Trump, because there are Republicans who disagree with Trump on tariffs and disagree with him on immigration and disagree with his personal conduct and his moral character.” And yet they have voted for him, anyway, rather than hand the extraordinary powers of the presidency to Hillary Clinton, Joe Biden, or Kamala Harris.

While I rather hope Cohn is right, in that I think we’d be better off as a country if parties were punished for nominating extremist, disreputable, and unqualified candidates, I’m skeptical that we’re going to see that emerge all at once. Which makes me wonder if there isn’t some sort of systematic bias in polling relating to response rates.

If, in fact, there’s a blue wave that gives Democrats Senate seats in the likes of Texas, Alaska, and Kansas—and there are good reasons to think there will be—then it’s just highly unlikely that it will be counterbalanced by Republicans winning traditional blue states in the same cycle.

8 responses to “Polling Anomalies”

  1. I listened to that ep as well and considered writing about it.

    I find that Cohn is trying way too hard to make a both-sides kind of story out of it.

    Of the red states, for example, a huge reason that Iowa and Kansas are in play is due to one word: diesel.

    Yes, candidate quality matters (in both directions) in Texas, but the shift in Hispanics due to Trump’s immigration policies and failure to deliver on the economy is important as well.

    Alaska is small pop state with internal idiosyncracies and Peltola has won a statewide race there.

    Of the blue states, Maine has the whole incumbency factor (not to mention the botched nomination bit with Platner). I can’t speak to New Hampshire, and yes, El-Sayed has some specific headwinds to deal with.

    But this does not add up to “Dems are doing better in Red states, and Reps are doing better in Blue states in some equal way.” That forcing a narrative.

    Plus, while I do not think SC is going Blue, I would note that the GOP is spending money there.

    The clear story is a national anti-Trump wave that is affecting his party and that is not mitigated by some state-leve variations in a couple of Blue states.

    2
    1. @sltaylor @Jen: Yes, as much as we have nationalized our politics, there are still state-level idiosyncracies. And, honestly, I’m just not sure polling state-level races remains all that predictive. Surveys are increasingly self-selected samples and we have no idea how to do likely voter screens, given the volatility of people actually showing up to vote.

  2. Well, as a NH resident, I do not think that candidate quality is what is driving this particular Senate race’s competitiveness. I think there’s an undercurrent of fed-up-id-ness out there, and to have Pappas (former executive council, current congressman) running against Chris Sununu (he of the constantly running Sununu clan) is leaving people a little bit…cold.

    There is a hunger for something new, and when not presented with that as an option, people will return to type.

    A YouGov poll the other day had Pappas leading Sununu 51 to 42. That seems like a pretty big spread, I was surprised, but polling seems to be all over the place. Which is another caution, of course.

  3. Here’s my BlueSky rant about Cohn’s “recipe” headline that I commented on the post over the weekend.

  4. Red state candidate quality is an issue, but a major factor is that R’s are not improving the lives of the residents in the states that they control and a critical mass of voters are noticing that. It is voter dissatisfaction on local issues that is driving Dem success.

    The issue here in NH is R’s have nominated, the near perfect candidate in the senate race, John Sununu, his brother would have been perfect. Near universal name recognition, a sterling reputation as a Reagan R, he basically was in a witness protection program during the Felon’s years, so there are few examples felon adoration.

    Though this is an open seat, Papas carries the burden of incumbency in a throw the bums out milieu, since this seat has been a Dem for so long and Dems have dominated NH’s DC delegation for more than a decade.

    It will be close, but I expect Dems to retain this seat, the hate/disgust with the felon is just too great.

    1. Hahaha, I so internalized that, that I mentioned the wrong Sununu in my post. Yep, it’s John who is running.

      I make my own point by accident.

  5. Count me in the group that doesnt trust statewide polls very much, especially in the off year votes. I do think that in the ‘safe” states there is a tendency to nominate more extreme candidates and that can make the race close and risk losing if turnout is what determines the winners.

    Steve

  6. I posted this in today’s forum but it could just as easily be put here.

    Early voting starts 19 Oct. I just printed out my ballot. It has 107 choices. Besides the big ones of Gov, Senator, Representative, and Texas legislators, the majority of voting choices are judges, county positions, constable, school district budgets, etc. It really is too much and I think most will just vote on the first page of the ballots.

    I’ve actually seen people pick up the ballot for the first time on their way into the voting center and then start to look at it.

    No wonder why we are screwed up.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.