Do you wonder why Donald Trump thinks he’s wildly popular beyond the nefarious precincts of those enemies of the people, the liberal news media? The president himself offered a succinct talking point on that subject near the beginning of his very non-succinct February 16 press conference.
A new Rasmussen poll, in fact – because the people get it – much of the media doesn’t get it. They actually get it, but they don’t write it. Let’s put it that way. But a new Rasmussen poll just came out just a very short while ago, and it has our approval rating at 55 percent and going up.
True to form, Trump cherry-picked the most favorable polling data available, and ignored the rest. But he could not do that — short of just, well, making stuff up, which is always an option for him — if it were not for an unusually wide range of findings in the polling universe about public attitudes toward the 45th president’s job performance so far.
As Nate Silver notes, Trump’s recent approval ratings vary from a high of 55 percent (with 45 percent disapproval) in the aforementioned Rasmussen poll to a low of 39 percent (with 56 percent disapproval) in a survey from Pew Research. The differences are most likely the result of a combination of sampling and survey techniques. Trump consistently does better with narrower samples. Rasmussen claims to be measuring likely voters, even though we are more than a year and a half away from the next national election. Pew is sampling all adults, a significantly larger universe than those who will ultimately vote in that next election. Rasmussen is also famously a robo-pollster, which means he’s only reaching the half of the electorate that has land lines. Pew utilizes a traditional live-interview methodology, which is generally thought to be more accurate, but that some theorize can be misleading with respect to highly controversial politicians like Trump. (This is the “shy Trump voter” theory.)
While polls like Rasmussen’s have a poor reputation and polls like Pew’s are considered closer to the gold standard (FiveThirtyEight’s pollster ratings give Raz a C+ and Pew a B+), we are in a period of great uncertainty about polling quality. And as it happens, the final 2016 national poll from Rasmussen pretty much nailed Clinton’s popular-vote margin over Trump, while the final Pew poll (conducted two weeks out, to be fair) showed Clinton up by six points.
So with all this confusion, is Trump justified in just citing whichever polling results he wants? No, not really. Most observers who are interested in approximating the truth go with polling averages. At the moment, RealClearPolitics’ average of recent polls places Trump’s job approval ratio at 45/51. It’s also important to pay attention to trends. As it happens, since Trump bragged about his Rasmussen numbers, his approval ratio in that tracking poll has deteriorated from 55/45 to 50/50, the worst ratio of his brief administration.
It is an entirely different question how much these numbers matter as a predictor of the next election. While the party of the president almost always loses House seats and more often than not loses Senate and gubernatorial seats in midterms, and less popular presidents usually lose more than popular presidents, variations in the landscape can make it very tricky to lay odds. The Senate landscape in 2018 is insanely pro-Republican. GOP control of the upper chamber could very well survive even a Democratic electoral tsunami. Since all House seats are up in 2018, GOP control there is significantly more vulnerable, but thanks to gerrymandering and superior “efficiency” in the distribution of voters, Democrats will have an uphill battle to win the net 24 seats necessary for a flip in control — and with it the ability to thwart the Trump/GOP agenda. Nate Cohn appears to think it’s too much of a reach even if Trump’s approval ratings stay roughly where they are today.
Democrats might not take the chamber with a victory on the scale of their huge win in 2006, when they gained 30 seats, or on the scale of the Republican sweep in 2010, which garnered 63 seats. With so many Republican seats safely out of play, a similarly impressive win might still leave the Democrats short of House control.
In 2006 and 2010, Mr. Bush and Mr. Obama had approval ratings near or above 40 percent on Election Day. So if you had to make a rough guess, you would probably say that Mr. Trump’s approval rating would probably need to be even lower for House control to become a true tossup.
So while it is hard to deny that Trump is amazingly unpopular for a new president, unless his approval ratings trend farther down the way even those of popular presidents typically do, his party may not suffer the kind of humiliation Democrats experienced in 2010. For all the shock Trump has consistently inspired with his behavior as president, there’s not much objective reason for Republican politicians to panic and begin abandoning him based on his current public standing. But in this as in so many other respects, we are talking about an unprecedented chief executive, so the collapse some in the media and the Democratic Party perceive as already underway could yet arrive.