Stagnant polls, ‘hidden’ voters, and Trump’s path to reelection

In the two weeks since the Republican convention there has been a burst of Trump optimism. A “bounce” of sentiment. The national polls seemed to narrow, continued unrest was assumed to inure to Trump’s benefit, “shy” Trump voters would deliver victory, and the bookies moved the odds towards Trump.

But all that change in sentiment appears to have been a mirage, perhaps borne of boredom with a race that has not changed — fundamentally — in over a year. The fact is that Trump is still a decided underdog. As of Labor Day, the RealClearPolitics ballot test average is 49.9 percent to 42.8 percent for Biden, an advantage of 7.1 percent, down modestly from a 9.5 percent advantage on Labor Day, 2019.  The tightening might be due to a couple of outlier polls, an Emerson poll with only a two-point difference and a Rassmussen one-point poll.

Entering 2020, Trump had a simple strategy: He could lose Michigan and Pennsylvania and still win if he held everything else. In case Wisconsin flipped to the Democrats, he had to find 10 electoral votes — hence the investments in Nevada (six votes), New Hampshire (four) and New Mexico (five), with a speculative play for Minnesota (10).

Today Minnesota looks to be a better bet than Wisconsin, and Trump is polling within five points in Pennsylvania. The Trump puzzle changes a fair amount. If we put Michigan in the Biden column, Trump can still win by losing Wisconsin and Arizona if he holds Pennsylvania. Alternately, Trump could win by losing Michigan, holding Wisconsin and Pennsylvania, gaining Minnesota and then lose a combination of Arizona and either Georgia or North Carolina.

There are plenty of other permutations opened up by a Trump win in Pennsylvania, but they are much less plausible. Conceivably Trump could still win without Florida, Ohio, or Iowa; however, those states are likely trailing indicators — meaning that if they are gone it is likely other more competitive states are also lost. If Trump loses Ohio, for example, he has likely lost every other midwestern state — an insurmountable deficit.

Regardless of how Trump tries to put the winning jigsaw together, it won’t matter if he can’t win the battleground states. And right now he is polling behind in every single one. Florida and North Carolina are close — and some polls have put Trump ahead — but the average for every state, whether on the RealClearPolitics site or the 538 site put Biden ahead.

It should be noted that the polls are likely underestimating the true level of Trump support. While some news outlets and pundits stubbornly claim that there are no hidden Trump voters, both a recent study and circumstantial evidence supports that there is a small reservoir of secret Trump voters. What the deniers refuse to acknowledge is the fact that if there is a systemic error in polling that is smaller than the polling margin of error, that number is extraordinarily difficult to flush out. If Trump’s true support is underestimated by two or three points (the Cloud Research study estimates three points), that is within the margin of error of most polls — although Pew Research thinks the average margin of error is much higher than reported.

At three points, Trump would still trail the popular vote but would go from a loser to a winner in Florida and North Carolina and would be within two points in Pennsylvania and Arizona, which would make him a winner if he could pick up those states, according to the 538 site. That’s a thin thread, to say the least, and time is not only running out, early voting is about to start.

Trump could catch up — he did against Clinton in 2016. But he is facing significant headwinds that are making it more and more unlikely he will be able to pick up undecided voters. In 2016, Clinton was the de facto incumbent. With 63 percent of voters thinking the country was on the wrong track, Trump had the advantage with undecideds.

In 2020, Trump is the incumbent and 65 percent think the country is on the wrong track.

Not only that, the issue on which Trump is focusing — crime and the riots — is not moving the needle for him. As for crime and violence, only 37 percent of voters and 34 percent of independents have confidence in Trump to handle the issue, with 53 percent of all voters and 51 percent of independents not having confidence. The numbers are worse for the coronavirus with Trump’s deficit among all voters at 33 percent confident and 56 percent “uneasy” (30 percent to 55 percent for independents).

Biden does not do much better, but he does outperform Trump.

On crime and violence, Biden’s confidence deficit is 39 percent to 43 percent among all voters and 29 percent to 47 percent among independents (nearly equally bad). With respect to the coronavirus, Biden is behind 40 percent to 43 percent (31 percent to 47 percent among independents).

The bottom line for Trump is that he needs to do more than convince voters Biden cannot handle the protests and violence — where the public agrees with Trump. Unless Trump can show that he can re-establish peace in the cities, picking up undecided voters will prove impossible. The slice of “shy” Trump voters is not enough to put him over the top.

NOTE: This post has been updated from the original to correct an error in electoral college math.

Keith Naughton, Ph.D., co-founder of Silent Majority Strategies, is a public affairs consultant who specialized in Pennsylvania judicial elections. Follow him on Twitter @KNaughton711.