People seemed to like my writing about the lessons of the Virginia gubernatorial campaign before the votes are counted, so I’ll do the same for the Alabama senatorial race. I think the biggest lessons concern polling, but I’ll also get to the tawdry stuff.
As I write this, the RCP poll average is Roy Moore +2.2, which is still within a normal polling error of Doug Jones winning. Indeed, given the tumult of a contest featuring Moore defending against allegations of groping a 14-year-old girl while dating teens in his thirties, many public pollsters have thrown up their hands in a collective shruggie.
Much of the reporting on the wildly divergent polls has focused on how much of this can be caused by the underlying baseline assumptions one makes about the likely voter pool. Nate Silver also notes the modal effects — the differences between live-caller and automated polls — and tries to explain them, as well as how they affect pollsters’ demographic assumptions. And SurveyMonkey published a final update focusing on the difference between past turnout and self-reported intent to vote in this election.
Often the problem with the final rounds of polling in a campaign is “herding” — firms shading their results so as not to get to far out from the seeming average. That’s clearly not a problem here. Some firms will be right and many will be wrong.
This is a bit more forgivable, given the (ahem) unique circumstances in this special election (and specials are difficult to poll in general). Unfortunately, what most people will take away from this campaign is another layer of distrust in polling. Instead, it might be more useful to take this opportunity to understand that what the public wants from polling (a prediction) is more elusive than what a campaign wants from its private polling (strategies and tactics). If it all comes down to turnout — and it does — the lesson of the messy polling here is not who is going to win, but what campaigns have to get to win.
The reason most in the biz of polling or poll analysis still marginally favor Moore is a function of just how deep a red state Alabama is now. Even the sort of Democratic enthusiasm we’ve seen in 2017’s other specials may not be enough for Moore to lose. Republicans turned off by the Moore scandal staying home may not be enough. It might actually take those things combined with substantial GOP defections to Jones for Moore to lose.
Enough of those things could happen. The always controversial Moore has historically underperformed in past races and thus may be uniquely positioned to blow what would be an easy race for anyone else. But I wouldn’t bet a large sum on it.
Beyond polling, the immediate lessons of a Moore win would be about the power of partisanship. However much distrust the right may have of the establishment media, the initial reporting by the Washington Post was exceedingly careful, as was the vast majority of the follow-up by other outlets (the most controversial accuser was brought forward by Gloria Allred, not the media. And frankly, the additional writing in the accuser’s yearbook hurts her credibility mostly because of Allred’s mischaracterization of the evidence, not the evidence itself. Alabama voters aren’t supporting Moore because they distrust his accusers; they distrust his accusers because they are supporting Moore.
As an aside, for all of the discussion about hypocrisy the Moore campaign has produced, the major church leaders in Alabama are not openly allied with Moore. Elitism rears its ugly head!
Indeed, elitism may yet raise its ugly head if Moore wins, as Dan Balz lays out to save me the trouble:
“For Republicans, there likely can be no truly good outcome. If Moore wins, the party will have preserved the seat but will be saddled with a new senator under a cloud of allegations, including assaulting a teenager many years ago as well as a pattern of pursuing teenagers half his age when he was in his 30s. If he wins and is sworn in, he probably will face an ethics investigation that will keep the controversy alive until his fate is resolved and perhaps much longer than that. For the Republicans, it’s a hot mess.
If Moore loses, the GOP would be spared his presence in the Senate. But the result will have inflamed the anti-establishment forces led by former White House chief strategist Stephen K. Bannon, deepening antagonisms that continue to roil the party. A Jones victory also would tarnish the president, who has enthusiastically endorsed Moore and campaigned near the Alabama border Friday night in a display of that support. Additionally, a Jones victory would put the Republican majority at greater risk in 2018.”
Although there are insider-types who think there would never be a vote to expel Moore, I am less sure about that. An investigation can develop its own momentum and logic — and this story is already a lot less convoluted and seemingly supported than others. Senators watching this circus from afar may see it differently once the circus comes to town.
For example, would Moore choose not to cooperate with an investigation? Given Moore’s record, I wouldn’t rule that out, even though denying the legitimacy of the body to which you just got elected would seem to be bad tactics. Senators, even GOP Senators, might not take that well. And then there would be a whole host of individual considerations at play if it came to either a vote on the ethics committee’s recommendation or a vote to block consideration of the same.
What would not be individualized will be the way the Moore campaign will continue to fan the flames of intraparty conflict regardless of whether he wins. An investigation and possible expulsion would fan those flames almost as much as a Jones victory. If enough Alabama voters cared so little about the Senate as to nominate Roy Moore — a man whose unfitness was obvious well before the allegations of molestation — perhaps the most populist faction of the GOP should expect a similar disdain. Indeed, I suspect may of them would revel in it every bit as much as they wear “deplorable” as a badge of honor. That’s a lesson we don’t seem to be done learning yet.
PS: Consider sharing this post with the buttons below, as well as following WHRPT on Twitter. Thanks for reading and sharing.