Is the country on the right track or going in the wrong direction? A question that has been popular in political polls since it was introduced by pollsters working for Reagan in the late 1970s. It’s an odd question, because it lacks parallelism (perhaps should be right direction vs wrong direction), but it’s an interesting way to track voter behavior. This was, as are so many important things, alluded to in the opening of an episode of the West Wing. Unfortunately, I couldn’t find the clip on YouTube, but it was some pretty typical Sorkin writing that gives the characters an opportunity to teach the viewer about something.

Setting aside the question structure, the main point is that I’m fascinated by polling data. I’m especially fascinated by data that don’t agree with themselves. Rasmussen has been a pretty kind place for Trump. In the Rasmussen’s Trump approval tracking as been the most favorable for him, and currently has him at a 51% approval (which is quite a bit higher than his average of 44.6%). The differences between Rasmussen and others might have something to do with the differences in tracking polls that are using adults, registered voters, or likely voters. Trump does better in the likely voter polls, and especially better with Rasmussen’s likely voter model. But that’s a tangent.

In the very same survey, however, Ramussen has only 31% saying that the country is on the right track, and 64% saying we’re going in the wrong direction. So a decent group of respondents approve of Trump and believe the country is going in the wrong direction…under his leadership. Those are the folks I find really interesting. Even if you don’t blame Trump for the direction, you’ve still got to think that he hasn’t succeeded in changing the direction to the right track, right? Sure, it might not be his fault, but he still didn’t turn it around during his first time, but at the same time they approve of the job he’s doing…which, logically, can’t include putting us on the right track, because they say he hasn’t done that.

This odd group has existed before. In 2012, for instance, the margin wasn’t as great, but 42% said that the country was on the right track, but Obama had an approval rating (based on Gallup) of about 47%. Granted, these aren’t data from the same poll, so it’s not the same as the internal comparison of Rasmussen respondents, but there’s still some contradiction in those numbers.

Why does all this matter? Well, as I said, I find it interesting when polled respondents appear to contradict themselves, but it’s also worth noting that the incumbent party has not held the White House with “right track” numbers in the low 30s since the question was popularized with pollsters in 1980. There were two cases where the number was at 39%. One of those came with a flip in party (2000, which gets an asterisk for sure) and the other came with the party holding the White House (1996). So it sure seems like anything under 39% is trouble, 39% on the nose is uncertain, and above 39% is good news for the candidate of the incumbent party. When Trump won his first seat, for instance, only 31% of Americans said that the country was on the right track, so a Trump win was predictable based on that number alone.

But Trump made history last election. He was the first president since Eisenhower to be elected without previously holding office, and he’s the only president other than Hoover who neither held office nor was a career military officer. It was also arguably the largest upset in the days of modern polling (although, as I’ve said a bunch of times, the polls really weren’t all that wrong). But, he upheld the historical precedent of flipping the party in the White House if the right track question response is below 39%. Still, unless that answer changes between now and the election, if he manages to win in November, he will again make history by keeping the same party in the White House with the lowest wrong track/right direction rating, and that statement is based on his 31% from Rasmussen. A Trump win would be even more striking when considering that the average response of the question is 27.3%.

Of course, there’s only one poll that matters…the official one on November 3rd.