If there's one thing we know for sure about Donald Trump, it's that he's a candidate for white people.

This would seem to be an almost insurmountable problem in an increasingly diverse America, but some are beginning to suspect — either with hope or fear, depending on whom you ask — that Trump could win a general election by pulling in large numbers of working-class white voters who are responding to his message of alienation, anger, and resentment. As The Wall Street Journal recently put it, "Trump's success in attracting white, working-class voters is raising the prospect that the Republican Party, in an electoral gamble, could attempt to take an unexpected path to the White House that would run through the largely white and slow-to-diversify upper Midwest."

Indeed, if Trump were to win the White House, this would seem to be the only way. But it's not going to happen.

The idea rests on a number of misconceptions, the first of which is that there are millions of blue-collar whites who would otherwise have voted Democratic, but who will vote for Donald Trump instead. As Chris Matthews said in January, "I think there's a lot of Reagan Democrats waiting to vote for him." The "Reagan Democrats" to which he refers were Democrats who crossed party lines to vote for Ronald Reagan in 1980.

The problem with this belief is that the Reagan Democrats are gone. Where did they go? They became Republicans. The phenomenon of Reagan Democrats was largely about race, the continuation of a process that began when Lyndon Johnson signed the Civil Rights Act and Voting Rights Act. Those socially conservative whites who had voted Democratic in the past shifted their allegiance, and they didn't go back.

You can argue, and many people have, that the alienation of the Democratic Party from the white working class is a serious problem for them, and it's part of what produces off-year defeats in years like 2010 and 2014. But because of the country's changing demographics, the white working class becomes a smaller and smaller portion of the voting public with each election, particularly in presidential election years when turnout is higher across the board. That's why Barack Obama could lose the white working class in 2012 by a staggering 26 points (62-36), and still win the election comfortably. So if you're going to argue that Donald Trump will ride these voters to victory, you'd have to believe that he'd do not just better than Mitt Romney did with them, but hugely better, so much so that it would overcome the advantages the Democratic nominee will have with other voters.

Consider, for instance, the Latino vote. Mitt Romney won only 27 percent of Latinos in 2012, an abysmal performance that convinced many Republicans that if they didn't "reach out" to this fast-growing segment of the electorate, they might be unable to win the White House any time soon. Latinos will be an even larger portion of the electorate this year than they were four years ago. Now think what will happen if Donald Trump, the man who made venomous antipathy toward immigrants one of the cornerstones of his campaign, becomes the GOP nominee. Not only would it be shocking if he got 20 percent of their votes, his nomination will almost certainly spur higher turnout among Latinos than we've ever seen before.

That's another problem with the blue-collar whites theory of a Trump victory: It rests on the idea that he'd bring out large numbers of those voters who don't vote often, but also requires that people opposed to Trump won't be similarly motivated to turn out. "I find it just so implausible that we could have this massive white nativist mobilization without also provoking a big mobilization among minority voters," political scientist Ruy Teixeira recently told The New Yorker. "It is kind of magical thinking that you could do one thing and not have the other."

Now let's talk about that Rust Belt. Even if you believe that Trump would do better in those states than recent Republicans have, it wouldn't be enough unless he was absolutely crushing the Democrat everywhere. The reason is that Democrats start in an excellent position in the Electoral College. In 2012, President Obama won reelection with 332 electoral votes, a cushion of 62 more than he needed. That means that if the Democratic nominee can hold most of the states Obama won — including swing states heavy with Latinos, like Florida, New Mexico, and Colorado — she could lose Pennsylvania (20 electoral votes), Ohio (18 votes), and Michigan (16 votes) and still be elected president.

I suspect that many people have been led to believe that Trump could ride white working-class votes to victory in the fall because he has performed particularly well with such voters in the Republican primaries. It only takes a moment to realize the problem with this logic. The people voting in Republican primaries are overwhelmingly, guess what, Republicans. Yes, there are Republican-leaning independents voting in those primaries, too, but they're mostly people who call themselves independent but consistently vote Republican. They're already in the GOP's camp; Trump would need them, plus a whole lot more.

That's not even to mention the moderate Republicans who are repulsed by Trump and would either vote for the Democrat, vote for a third-party candidate, or just stay home. Donald Trump's problem in the general would be that he has all kinds of voters who will oppose him, and be highly motivated to do so; he is easily the most unpopular candidate in either party. He might pick up a few extra votes from those who respond to his nativism and race-baiting, yet used to vote for Democrats. But there just aren't enough of them, and it won't be anything approaching what he'd need to win.