Lots of celebrating by Trump fans and despairing by Trump critics on social media today over this poll. It’s December, notes Philip Klein, and this guy’s now fully 20 points ahead nationally. And not only does he have a big lead, his lead’s gotten bigger: He’s up nine points since CNN last surveyed the country, thanks mainly to the decline of Ben Carson and Jeb Bush. His overall average in RCP’s tracker has now reached a new high of 30.8 percent. Trump critics have been sitting in the pumpkin patch waiting for the Great Pumpkin, i.e. Trump’s supposedly inevitable slide in the polls, for six months now. The CNN poll is evidence that the Great Pumpkin might not be coming.

Ron Brownstein was ahead of the curve in pinpointing the real split among Republican voters. It’s not “moderates” versus “conservatives” or “establishment” versus “tea partiers,” he argued in October, it’s blue-collar versus white-collar. Trump has consolidated a huge chunk of the blue-collar vote while the white-collar vote remains split among the rest of the field. Here’s a choice bit on that point from CNN’s new data. Advantage: Brownstein.

Republican voters are most sharply divided by education. Among those GOP voters who hold college degrees, the race is a close contest between the top four contenders, with Cruz slightly in front at 22%, Carson and Rubio tied at 19% and Trump at 18%. Among those without college degrees, Trump holds a runaway lead: 46% support the businessman, compared with 12% for Cruz, 11% for Carson and just 8% for Rubio.

That’s a good reason to think Trump’s strength will last into February. Realistically, no one’s going to cut deeply into his hold on blue-collar voters. Cruz is trying but the more I watch him, the more I think he’s ultimately a bad match for Trump fans. He’s trying to get to them entirely through immigration and tough talk about ISIS and the “Washington cartel,” but Trumpmania is more than that. Cruz sounds like a politician; he’s overtly evangelical, which, lord knows, Trump is not. And Cruz doesn’t pay nearly as much attention on the stump to blue-collar economic anxiety as Trump does. If you like Cruz, you like him because he’s a warrior for conservative dogma. If you like Trump, you like him because he talks a lot about lost jobs. That, even more than immigration, explains why he’s north of 30 percent now. If you disagree, go look at some of the eye-popping numbers Trump has banked when Republicans are asked who they trust most on the economy. Today’s CNN poll is no different:

econ

He leads big on immigration too, but not quite as big — and that issue, for all the media hype recently, isn’t rated quite as important by Republicans as the economy is. Ninety-one percent of GOPers say the economy will be extremely or very important to their vote next year; 78 percent say so of immigration. Republican opinion on Trump’s “deport ’em all” plan is also complicated. When you offer GOPers a binary choice between sending everyone home and not sending them home, there’s majority support for mass deportation (and heavy opposition among the rest of the electorate, in case you’re looking ahead to next year):

deport

But when you give them more options, the numbers change:

deport2

Just 21 percent of Republicans, a little more than half of the 36 percent who favor Trump for president, want to deport everyone. Even among Trump supporters, 55 percent say it wouldn’t be possible, no matter how much they want to do it. Trumpmania is bigger than immigration.

Further evidence: Despite endless rhetoric from Cruz and Marco Rubio about destroying ISIS, defending Israel, stopping Iran, and so forth, Trump crushes them when voters are asked who would best handle ISIS. Forty-six percent of Republicans say Trump; Cruz is second at 15 percent. Rubio, the hawkiest hawk in the top tier, is in single digits. How come? I think this too is a matter of blue-collar versus white-collar, not in terms of economics but in terms of culture. Trump sounds “blue-collar” and Jacksonian when he talks about ISIS: He’ll “bomb the sh*t” out of them, he’ll waterboard jihadis even if it doesn’t produce intelligence because they deserve it, and so on. Cruz doesn’t talk like that. He talks tough but the way a more traditional politician would, and plenty of voters don’t trust that sort of political-class tough talk after 15 years of post-9/11 war-on-terror meandering. Relatedly. here’s another eye-popping bit of data from yesterday’s PPP poll of New Hampshire:

-53% of Trump supporters are in favor of a national database of Muslims, to only 22% opposed to that concept. This is despite only 29% of Republicans overall supporting the idea to 44% who are against it. In fact supporters of all 13 of the other candidates are against this concept.

-49% of Trump supporters want to shut down the mosques in the United States, to only 18% against that idea. Overall only 25% of Republicans support doing that to 44% who oppose it. Again on this issue supporters of every single candidate other than Trump are against it.

-Additionally we asked voters about an assault weapons ban and only 20% of Trump voters support it to 66% who are opposed. We then asked about a ban specifically on assault weapons for Muslims, and 56% of Trump voters support that to only 22% who are opposed.

Cruz voters are more traditional conservatives and libertarians. Trump voters are Jacksonians. The Jacksonians are united behind one candidate. There’s Trump’s lead.

One more data point from CNN to suggest that Trump is for real:

bestchance

Every scenario that has him fading and ultimately collapsing against Cruz or Rubio involves Republican voters taking a hard look at their choices and deciding that there’s no way Trump can win a general election. If for no other reason, someone else must be nominated to maximize the party’s chances against Hillary. Look at that data again and tell me how and when that conclusion is going to be drawn.

Jeb Bush is at three percent now, by the way. Exit question via Logan Dobson: Should we take a poll seriously that assumes 48 percent of all registered voters are going to vote in a Republican primary? C’mon. How much of these results are genuine principled support for Trump and how much are random people saying “Trump” when asked a question because they’ve seen him on TV 24/7 since June and he’s the only Republican who sounds any different on any subject from all of the rest of them?

Update: Did CNN prime this poll to favor Trump? Hmmmm:

When you examine the poll, which was taken from Nov. 27 through Dec. 1, 2015 (which was a holiday weekend and is problematical all by itself in terms of who the respondents were), it is clear that the pollsters chose to ask five questions on the topic of illegal immigration prior to asking about the Republican nomination horserace. This is a bad polling practice that can skew the results.

As Pew Research has explained with respect to the ordering of the presidential approval question, “if the survey first asks about the economy and then asks about presidential approval, the respondent may still be thinking about the economy when answering the latter question … [and] if the respondent is only thinking about the economy because we brought up the issue, his or her response about the president may be biased by what we call a context effect: In this case we would be priming the respondent to consider the economy in an assessment of the president.”

This is essentially what CNN/ORC did with the five questions that were focused on illegal immigration.