The question of how President Trump is faring with the American electorate is obsessional for everyone interested in politics.

The problem is there’s no reliable way to measure it, and we’re almost four months from the midterm election that will function as a partial national referendum on this question. So we’re all reduced to reading tea leaves and fighting over points of interpretation.

The tiniest changes in polls are studied with Talmudic focus, even though we are repeatedly and wisely warned that such shifts are “noise” rather than “signal” — they only look like they mean something. We are also warned to avoid believing in little pieces of data that “confirm our priors” — which is to say, information that supports our own beliefs. But “confirming our priors” is exactly the point of the exercise.

We don’t want bad news. We want comforting news.

It is unsatisfying to Trump fans and foes alike that his overall approval rating ranges between 40 and 43 percent, and has for months. The foes think it should be lower, while the fans think it’s higher and that somehow the truth is being suppressed. And the number is neither good enough nor bad enough to point clearly to the future midterm results.

The ultimate question is: Will there be a blue wave that’ll break across the country and sweep Democrats into control of the House of Representatives the way a red wave swept the GOP into the House in 2010?

Political history tells us there should be a pronounced shift at the very least — that the “out party” always gains in the midterms, especially when the party in power controls not only the presidency but Congress. And the number of seats Democrats need to flip to attain the majority is relatively small — only 23.

That modest shift would not constitute a “wave,” a phenomenon in which everything breaks to the benefit of one party. In a wave, most competitive races, no matter where they are, fall like dominoes in one direction.

This happened for Republicans in 2010, when they netted 63 new seats and took the House of Representatives — and for the GOP in 2014, when they netted nine spots in the upper chamber and took control of the Senate.

A wave would net the Democrats 30 to 40 seats in the House and might also give them unexpected gains in the Senate or at least limit their downside damage there (the Senate races this year are unusually tough for Democrats).

Through 2017, they had every reason to think they’d see a wave. Democratic candidates performed very well in special elections, and Democratic voters turned out everywhere to give their candidates the best possible shot to win.

And Trump had a disastrous first year that saw his poll numbers sink into the low- to mid-30s after self-inflicted wounds like his horrid reaction to the Charlottesville white-power riot. His numbers declined not because Democrats hated him more than they had, but because Republicans pulled away from him.

Then that stopped, for a variety of reasons. Trump is now viewed positively by anywhere from 80 to 90 percent of Republicans, and that has brought his polling back to low but not disastrous levels.

Every time Trump creates a controversy — from the separation of children from their parents at the border to the NATO-Putin horror show of the past two weeks — everyone waits to see whether Republican voters will lower the boom on Trump and push his numbers down back to Charlottesville levels.

It hasn’t happened all year and it’s unlikely to happen in the wake of the past week, because Republican poll respondents are onto the game and — whatever they might feel about Trump playing footsie with Russia — they’re not here to provide comfort to Democrats and liberals that November will deliver them from Republican control of the House.

What the approval rating doesn’t tell you is whether they’ll bother lining up for a down-ticket Republicans because they like Trump or whether they don’t like Trump enough to bother.

Liberals and Democrats detest and fear and loathe Trump in a way they have never loathed another Republican president (and they’ve loathed them all). They cannot believe anyone feels differently, and they are terrified of a Trumpian future in a way they have never been terrified before.

They rallied in 2017 to resist and organize and raise money and deliver messages, and they did so with great success. But you can sense a growing anxiety rising from Trump’s homeostasis around in the low 40s.

Trump’s victory knocked everybody for a loop in 2016, and no one really knows whether he possesses some kind of magical understanding of American politics now that will help him generate another unexpected result in his favor.

If that were to happen — I don’t think it will, but who knows — I think we would see the non- and anti-Trump voters in the United States descend into nothing less than a collective panic the likes of which we’ve never seen.

The consequences of that panic might be a deep demoralization. Or they might bring about an even more determined radicalization on the left.

The only thing that’s easy to predict is what happens if there’s a wave and Democrats attain a comfortable majority in the House: impeachment by summer 2019.