I am not a betting man, but if I were, I’d bet on The Big Wave this time out: The Republicans take the Senate.
I have lived through three major Republican wave elections. The first one in 1994 tolled the beginning of my political awareness. The most recent one in 2014 represents, to my mind at least, the zenith of recent GOP performance, a cycle where they won pretty much every close race and nearly ran the table (ah, Virginia Senate . . . the one that got away). But in between those two came the 2010 election, the first midterm after Barack Obama took office. You may remember it as an exhilarating repudiation of Obamacare and Obama’s fiscal policies (this was the “Tea Party” electoral cycle, after all). And you would have good reason to: The Republicans added a whopping seven seats in the Senate and scored a 63-seat gain in the House (their best performance since before the Great Depression).
I remember it far more bittersweetly, however, as an electoral cycle where the GOP frittered away winnable Senate seats in states such as Nevada, Delaware, Colorado, and even Alaska — if you consider Lisa Murkowski to be an unreliable ally at best — and lost its chance to retake control of the chamber. The damage was done exclusively in the primaries, when Tea Party-inspired voters rejected more electable candidates in favor of ones who were either incapable of running competent races or simply too far out-of-step with their state’s electorates.
Which brings us to the fourth Republican wave election I am about to experience. Will it be more like 2010, as several flawed Republican Senate candidates come close but fall agonizingly short? Or will it be like 2014, where a rising tide lifts everyone to victory and even puts unexpected races in play?
The current polling (and the current state of polling) allows for arguments to be made on both sides: It is a Choose Your Own Adventure story of prognostication. On the one hand, the polls (including the latest New York Times/Siena numbers) seem to suggest that, despite massive national headwinds favoring the GOP, questionable candidate selection in key Senate races will prevent Republicans from capitalizing. On the other hand, the massive problems affecting the polling industry are no secret to anyone following politics, the two blatant examples being Trump’s 2016 and 2020 races, where polling averages wildly understated Republican support from the top of the ticket on downwards. And yet on the third hand, the polling averages were mostly accurate in 2018, the most recent midterm we have as a comparison, and notably one where Donald Trump was not on the ticket to drive “hidden” supporters to turn out. Which way to go?
I am not just politically conservative; I am dispositionally so. As a reflexive pessimist, I like to joke that the best way to avoid disappointment is to set your expectations so low that any overperformance is an “upside surprise.” So it is with some trepidation that I find myself believing we are facing an election more like 2014 than 2010.
My reason for this is actually quite simple: Polling is broken, broken more than ever, and it is breaking down at an accelerating rate. The primary culprit is the vast dive in response rates to pollsters, particularly among right-leaning voters and independents. The reasons for this cliff-dive in response rates can be debated (I don’t think it’s any particular mystery: a plunge in trust in the mainstream media and anything perceived as MSM-adjacent, broadly speaking), but the empirical fact remains that it is becoming harder and harder to “find” Republican voters when pollsters make calls, which leaves them adjusting, attempting to properly weight extremely small demographic subsamples, and ultimately applying their own preconceptions as to what the electorate might “look like” come Election Day.
In a transparently Democratic year such as 2018, that sort of methodology (and it is a true methodology, not just ad hoc, at least for any reputable pollster) can return polling results that track fairly well with the outcome, since the sorts of respondents most likely to sit through a phone survey (i.e. Democrats and Democratic-leaning voters) are also the most likely to vote. But in 2016 and 2020, the obvious social taboo against supporting Trump (and, by association, the Republican Party) clearly cratered response rates among GOP-leaning voters, and it seems highly unlikely that 2022 — as a Republican wave year — is not going to follow that trend, given that most major public pollsters have explicitly said they are not altering their 2022 methodology to reflect the huge polling miss of 2020. I am not a betting man, but if I were, I’d bet on The Big Wave this time out: The Republicans take the Senate, and it’s more likely to be with 52-54 seats than by 51.