Shy Trump Voters And Not So Shy Black Voters
Larry Schweikart has an interesting article at undercoverdc , but I'm going to link to it also at FR, because I found it easier to read at FR--see which suits you best:
The article is obviously is obviously about the polls, and it basically covers two topics (plus a bit more). Some of this has been touched on previously, but Schweikart goes into a fair amount of--readable--detail.
The first topic is how much weight to give the polls in light of the by now well recognized phenomenon of "shy" Trump voters--the extreme reluctance of Trump voters to level with pollsters. Can the polls be simply ignored? Schweikart argues that ignoring the polls is realistic:
I think the key answer is Donald Trump. There is growing evidence that Trump voters are more committed than ever to casting a vote for Trump, yet far more reluctant to tell anyone their intention. In 2016, Trafalgar solved that with the question, “Who do you think your neighbor will vote for?” vs “Who do you plan to vote for?” But the experts at Trafalgar say even that question is not revealing the larger numbers of Trump supporters—some of whom will tell the pollster at the beginning of the interview they plan to vote for Trump but won’t say so on record.
In other words, dismissing the head-to-head polls is realistic given the incredible difficulty the pollsters are having getting the Trump supporters to “come out of the closet.”
But the "shy" Trump voters is more of a side show in the article, which is largely devoted to the remarkable shift in Trump's favorability among black voters--who by contrast to whites have not been at all shy in expressing their favorable attitudes toward Trump to the pollsters. Schweikart approaches the question of how this could affect the 2020 election not only with numbers drawn from 2016 to the present but also with some historical research that I think you'll find fascinating.
I won't try to paste it all in--check it out. Here are two teasers:
A rule of thumb has been that Democrats need at least 90% of the black vote to win. If that’s the case, Biden isn’t even in the same zip code as a victory.
Because if [the black Trump vote] is even remotely in that ballpark [Trump's approval among blacks], there aren’t enough suburban Karens in the world to save the Democrat Party in November.
Oh, the extra topic, barely touched on, is that Trump is doing very well among Hispanics, as well.