One of the weird things about the ‘electability’ discussion in the Democratic primary is that there’s no recognition of what an accurate assessment of electability would require: a near perfect knowledge, months after the primary of how voters, most of whom you haven’t met and whose motivations you don’t really understand will vote in particular states. It’s not a trivial thing to estimate, and most people (possibly including me!) will not be very good at it. Some will be dreadful at it.
Osita Nwanevu raises another good point about electability (boldface mine):
But received wisdom about electability is powerful precisely because it defies reason and is resistant to critical scrutiny. Like many of the other concepts that shape electoral punditry and political discourse—charisma, qualification, momentum, authenticity—electability is a shibboleth of a political mysticism that “tickles the brain” only because it cannot fully engage it—a drab, gray astrology, maintained by over-caffeinated men…
This is a discourse incapable of producing anything beyond recursive guesswork—hypotheticals within suppositions that send us pacing in circles over questions that no election can actually resolve. The victory or defeat of any given candidate does not foreclose the possibility that they might have performed differently under slightly different circumstances and cannot tell us conclusively whether another candidate might have done better or worse. The 2016 election race drew us close, but not close enough, to understanding this. Any politically engaged person today can rattle off a list of factors that might have tilted the race: Russian interference, irresponsible coverage of the Clinton email scandal, Trump’s omnipresence on cable television, James Comey’s eleventh-hour machinations, the Clinton campaign’s inattention to the Rust Belt. Yet the politically engaged have also taken to believing that electability is a stable and perhaps even measurable quality innate to the candidates themselves. This belief persists despite the victory, in that election, of a man who was widely considered one of the most unelectable candidates ever to seek the presidency. Now many of the sages who rendered that judgment have reconvened to tell us Donald Trump can only be beaten by someone matching a profile—white, male, moderate—that has not won Democrats the presidency in 24 years.
It might work this time around. It also might not. All we can be reasonably sure of is the persistence of a dynamic that Trump’s nomination and election brought into relief—given partisan polarization, and assuming the absence of a strong third-party challenge, just about any candidate from one of our two major political parties can reliably expect to win the support of about half the electorate. Different camps within the Democratic Party have put together plausible theories on what might put one candidate or another over the top in the states and regions necessary to prevail in the electoral college. But these are hermetic arguments that could run up against a variety of competing factors—from unforeseeable world events to the state of the economy to the competence of each campaign organization—once the general election leaves the world of abstraction. The extremely early relevant numbers that we have, the candidate favorability and head-to-head matchups, don’t tell us anything more than what we should already know: We are in for a close race, and the leading Democratic candidates are competitive with Trump.
I’m not as quite as negative on the whole concept as Nwanevu, if one’s statements are hedged and qualified, since we do need to identify potential weaknesses of the different candidates. Furthermore, as he notes, what issues will rise to prominence is not cast in stone, and that can affect the electability arguments. But if we make these arguments, we should be far more modest about knowing what is and is not electable.