Using 2008, 2012, and current polling data, Brownstein divides the Republican electorate into four roughly equal groups divided by religion and education: college-educated Evangelicals and non-Evangelicals, and non-college-educated Evangelicals and non-Evangelicals. In the last two cycles, he says, educational levels weren't that important in determining candidate preference....Note the portion I've highlighted. Which candidate did non-Evangelical blue-collar Republicans support last time? The boss guy. Which candidate do they like this time? The boss guy. The workers liked the bosses.
But in this cycle, education levels -- generally an indicator of class -- are pretty big differentiators. In particular, Ted Cruz is leading strongly with college-educated Evangelicals, while Cruz and Donald Trump are in a close battle for blue-collar Evangelicals. Brownstein views this competition as potentially decisive. Yet he also notes that Trump's very best group among the four is blue-collar non-Evangelicals -- the "opposite corner" of the party from Cruz's stronghold, and a group that was part of Romney's coalition in 2012. There's really only one quadrant of primary voters that the Establishment candidates are (collectively) dominating, and that's the college-educated non-Evangelicals.
Obviously, secular blue-collar Republican voters don't always back a boss -- in 2008, the GOP coalesced around John McCain. But he was known as a war hero. That made him an alpha male.
What secular blue-collar Republicans don't seem to want, especially this year, is a harried schlub (Jeb Bush, John Kasich), a petulant teenager (Rand Paul), or the college intern (Marco Rubio). They might have gone for Chris Christie, but Trump stole his nasty-boss act. (Also, Bridgegate turned Christie into a humbled beta at just the point in the 2016 cycle when he needed to seem like an alpha.)
Call this false consciousness if you want, but GOP establishmentarians should have understood it a long time ago. They need a boss. They're running a lot of clerks.