Thread by @ChrisPolPsych

Why did so much of the pundit/journalist class convince itself that a red wave of epic proportions was on its way, only to be surprised by the actual outcome? Was this really something that was so impossible to see coming? A thread! (1/n)

Much has been said about the proliferation of cheap GOP polls in key states in the last few weeks of the campaign. This was undoubtedly real -- and a factor that distorted the more naive poll aggregators. (2/n)

But the higher-quality polls were largely on target, and most of the averages still showed a relatively close race even with the crap polls included -- the generic ballot was nowhere near as dire for Ds as it was in say, 2010. (3/n) gelliottmorris.substack.com/p/the-pollster…

Special election results also suggested results better for the Democrats than the final pre-election pundit consensus implied. (4/n) fivethirtyeight.com/features/yes-s…

So, there were plenty of legit empirical signs that a wave might not be in the offing, even if you were hesitant to play up the potential role of Dobbs backlash. So, why did the consensus land off target? (5/n)

My working conjecture is that it has to do with pundits & journalists (and political sophisticates in general) *overestimating* how they differ from normies on some dimensions while simultaneously *underestimating* how they differ on other dimensions. (6/n)

With respect to the first, the failure 'to see Trump coming' in 2016 broke a lot of brains among commentators, generating an almost-traumatic sense that they were libs who didn't grok what the 'real America' thought (even though Trump did not win a majority of votes). (7/n)

A lot of this seems to have been subsequently filtered through both academic and popular discourse on the primacy of partisan polarization and sorting -- and longtime right-wing tropes about liberal elitism, which many pundits seemed to internalize post-2016. (8/n)

Of course, polarization and sorting *are* real, and pundits, journalists, etc *are* more socially liberal than the average American. (9/n)

But it's easy to get sucked into the Burroughsian language-virus aspect of polarization-and-culture-war discourse. It becomes taken for granted, with no sense of the limits of the phenomenon. (10/n).

For a lot of politically-aware people, 'we live in a polarized country & everything is culture war' has practically become a *valued* meaning structure, beyond its real explanatory utility. (11/n)

This is what brings us to the overestimation-of-divides part of this. Journalists, pundits, and the politically aware have absorbed the polarization literature as an explanatory frame, but they have not attended to what @ykrupnikov & @ryanbq refer to as the 'other divide' (12/n)

This is the divide between the most politically-engaged segment of the population and those of modest or low engagement -- which is *much* of the electorate. (13/n) cambridge.org/core/books/oth…

The mass public is not uniformly polarized - most of what we characterize as polarization is most descriptive of a narrow segment of the population. This holds for both belief polarization and affective polarization - dislike of political outgroups. (14/n)

Indeed, a big take-home of Krupnikov & Ryan's work is that normies find deep involvement in politics to be weird and off-putting. (15/n)

When you think about folks who are both hyper-aware of polarization and allied phenomena (culture war, education polarization) but under-aware of the engagement gap, you can see how it might be easy for them to mis-characterize the mass electorate. (16/n)

If many commentators see themselves as strangers in a strange land with respect to the general public post-2016 and view this solely in terms of the culture-war frame, they seem likely to miss aspects of real-world mass politics. (17/n)

If middle America seems 'different' from you and you've internalized the polarization frame & the right-wing critique of the Liberal Elite, you are well on your way to assuming that much of the public is like the most batshit right-wingers on your Twitter feed. (18/n)

But much of middle America does not follow politics that much at all compared to the most engaged stratum of the population - including journalists, etc. They are not invested in ideology or an ever-changing culture war. (19/n) press.uchicago.edu/ucp/books/book…

Again, an important point of Krupnikov & Ryan's work is that normies find deep political involvement and its correlates to be weird and even counter-normative. (20/n)

This is crucial: the average person does not like weird ideological shit. Commentators have become so invested in the polarization narrative and their own sense of difference from middle America that they have missed the possibility that lots of GOP stuff is weird. (21/n)

Perseverating on stuff that is hyper-salient to the engaged (quickly changing gender norms, debates over how to talk about race), they have missed how little of this language-virus stuff trickles down or registers at the mass level at all. (22/n)

Moreover, neglect of the engagement gap causes commentators to miss the fact that as much or more right-wing as liberal stuff comes off as scary or weird to normies. (23/n)

Jan 6 was scary. Election denial is a few steps removed from thinking the moon landing was faked. The idea that school administrators are putting out litterboxes for cat-identifying kids is tinfoil hat shit to normal people. (24/n)

Overt bigotry is also more off-putting and uncouth to normal people than we assume under the culture-war frame & the internalization of right-wing propaganda.

In sum, our 'other biases' maybe led many of us to underestimate the public's willingness to punish extremism. (25/25)