Donald Trump will almost certainly be the delegate leader heading into July's Republican National Convention – but that doesn't mean he'll win the nomination outright. (Peter Stevenson/The Washington Post)

In the Washington Post-ABC News poll released earlier this month, white voters prefer Donald Trump in a hypothetical November matchup against Hillary Clinton by nine points. That's not a huge margin by today's standards, but it probably would have been enough for Gerald Ford to have defeated Jimmy Carter in 1976. Ford lost by two points, leading with whites by four points, according to exit polling. Had he led by nine, Ford would have won.

But that was 1976, when whites were about 90 percent of the electorate. In 2012, white voters were less than three-quarters of the electorate, as they'll probably be this year. And for any Republican, including Donald Trump, a nine-point lead among whites won't be enough.


Former Mitt Romney strategist (and avowed Trump opponent) Stuart Stevens looked at the math for the Daily Beast last week. "In 1980, Ronald Reagan won 56 percent of white voters and won a landslide victory of 44 states. In 2012, Mitt Romney won 59 percent of whites and lost with 24 states," Stevens wrote. He continued: "The simple truth is that there simply aren’t enough white voters in the America of 2016 to win a national election without also getting a substantial share of the non-white vote."

You can see the evolution of the electorate over time in exit polling.


A few things to notice here. First of all, notice that white voters consistently prefer the Republican Party and that nonwhite voters don't. Or, at least, they don't now that Asian American voters have switched from backing Republicans (in 1992) to backing Democrats (since 2000). In 2012, Asian American voters were even more supportive of Barack Obama than Hispanic voters were.

Notice, too, how the percentage of the overall Democratic vote that is from white voters (the darkest blue) has shrunk over time. In 2012, less than 60 percent of the votes for Obama came from white voters, according to a Washington Post analysis -- whereas almost 90 percent of the support for Mitt Romney came from whites. Both parties have seen the share of their support from nonwhites increase, but for the Democrats, that increase has been much more substantial.


After Romney's loss, the Republican Party decided to focus on outreach to nonwhite voters in an attempt to shift that light-red line, above. Stevens's point is that Trump faces unique challenges with outreach to black and Hispanic voters that other Republicans wouldn't.

Because, again: The white vote doesn't go as far as it once did. If you compare the margins of support for the GOP among white voters with the results of the election, you can see that clearly. In 1980 and 1988, the Republican candidate received 20 points more support from white voters than the Democrat did. In 1980, the GOP won by 10 points; in 1988, it won by eight points. In 2012, Romney again won whites by 20 points -- and lost by four points overall.


The Post's Greg Sargent looked at how the racial split would work at a state level. Trump may do better with white voters than Romney in some places -- such as in Rust Belt states, where his message on trade might resonate --but he'd have to do significantly better with white voters than Romney did to make a dent.

Trump's argument is basically that he'll increase turnout among white voters who are enthusiastic about his candidacy. Maybe. That turnout increase appears to have helped him win some state contests in the Republican primary season so far. But even if Romney would have increased white turnout by 20 percent nationally, still winning the same percentage of that vote, he would have barely tied Obama in the popular vote.

In short, then: If Trump does half as well with white voters as Romney did -- and if he's not able to make inroads with nonwhite voters -- there's simply no way he can win.