Political campaigns are an on-off switch. You won or you lost. Nuance can be overlaid — maybe you did better or worse than past politicians in the same race or better or worse than others running in the same cycle as you — but no race fits the contours of another precisely. Since hundreds of decisions led to the result, all of those decisions tend to be included as reasons for the win or the loss. The closer the race, the more that’s true.

If you’re lucky enough to win, there’s an additional impulse after the fact to argue that your win validates all of those decisions. You won because of your position on X, so therefore the electorate expects and hopes that you’ll enact X.

Those impulses are true of normal elections. In the case of the election of Donald Trump in 2016, the picture gets even more complex.

Trump’s path to the presidency went like this, in broad strokes. He announced his campaign and quickly became embroiled in a controversy over his comments about Mexican immigrants. That public fight did a number of things: Positioned him as “anti-P.C.,” endeared him to anti-immigration hard-liners and established him in opposition to the Republican establishment. The result was a strong core of conservative support that, in a splintered field of 17 Republican candidates, helped power him through the primary process.

Most candidates would then reposition themselves back toward the middle to appeal to the broader general-election voting base. Trump didn’t. But he had two other things working for him: a deeply unpopular opponent and a deeply partisan political moment. The former meant that a lot of people were forced to pick between two candidates they didn’t like, and more picked Trump. The latter meant that Republicans who were skeptical about Trump voted for him anyway.

He lost the popular vote by nearly 3 million votes. He earned a lower percentage of the vote than Mitt Romney had four years earlier and a lower percentage than any winning Republican since 1968, when the Deep South voted for the segregationist George Wallace. But he won Wisconsin, Pennsylvania and Michigan by about 78,000 votes, and so he won the White House.

Since his inauguration, Trump has repeatedly shown little inclination to do anything but continue to appeal to that central base of support that was with him from the beginning. In an interview with CBS’s Charlie Rose on “60 Minutes” on Sunday, former Trump campaign chief executive and administration strategist Stephen K. Bannon explained why Trump was taking this approach.

ROSE: You know that this White House leaks like nobody’s ever seen a White House leak. And that’s where the reporters are getting the story. And they’re getting the story about conflict between you and H.R. McMaster. They’re getting stories about conflict between you and Jared Kushner, and you and Ivanka Trump. They’re getting all these stories because people in the White House, including you, are leaking. You know that. And you have in fact said, “No administration in history has been so divided among itself about the direction about where it should go.” So I want to know from you, what’s the divide?
BANNON: The divide is, first off, President Trump and the way President Trump has always run his organizations, he will always take diverging views, I think that’s healthy. Because I think for an idea, a Darwinian environment for ideas is positive. Now, the one thing I disagree with is that I think there has been a divide in this administration from the beginning. It’s quite obvious. There’s one group of people that on the campaign, that said, “All you have to do is do what you said you were going to do in these major areas. Let’s punch out one thing after the other. You’re going to keep your coalition together, and we’re going to add to it over time as you’re successful.” There’s another group that has said, “Let’s compromise, and let’s try to reach out to Democrats, and let’s try to work on things that we can do together.”

Bannon obviously places himself in that first group.

These sentences are the key and worth looking at more closely: “Let’s punch out one thing after the other. You’re going to keep your coalition together, and we’re going to add to it over time as you’re successful.” Get the things done that you promised to do, and your base will stay loyal, it suggests. As you’re successful, the coalition will expand.

Bannon obviously wants Trump to follow through on his core campaign promises, particularly those related to immigration. This is what you were elected to do, he has certainly argued — and, no doubt, this is why you won.

He also needs to convince Trump that doing so will be politically beneficial. So: Do that, be successful and you’ll get more popular.

It’s just that … this isn’t how it works. Trump should see that by now.

If Trump is successful at delivering for his base, he is doing things that people not in his base disagree with, by definition. Maybe many of those other voters agree with some of what he’s doing, but most voters probably disagree with most of it. If someone believes that Trump’s position on ending DACA is wrong, they’re not going to suddenly approve of his presidency just because he successfully ended the program.

This isn’t theoretical. Since his inauguration, most of what Trump has done has been things that his base of support (including evangelical voters) agrees with. His approval rating has dropped by about 10 points. It was low to begin with; it’s lower now. Some of this has nothing to do with policy. Some of it does.

The nifty thing about this argument for those trying to steer Trump in a particular direction — which Bannon very much is — is that you can always claim that Trump hasn’t crossed everything off his list.

Remember this image? That’s Bannon’s to-do list on the whiteboard behind him from when he was in the White House.

We walked through how many of these items had gotten done when the images first emerged. Not many had then and not many have now, in part because many of them are awfully ambitious. (Notice, too, how many of these relate to immigration, as we mentioned above.) As long as most remain incomplete, someone employing the argument above can explain away low approval numbers. “You haven’t repealed Obamacare! That’s why your approval is in the 30s!”

It’s hard to argue otherwise. Trump was told repeatedly that his decision to stick to his base would cost him the election. It didn’t. Is it why he won? Well, it happened and he won. So telling him that he might do better by reaching consensus on things that are popular with a broader pool of Americans must necessarily compete with “I didn’t last year, and I am the president.”

Bannon knows this. Bannon leverages it. It may not be smart for Trump, but it is awfully smart for Bannon.