I want to believe. But I don’t.
“When you see a number this different, it means you might be right on top of a shift in the campaign. What you don’t know yet is if the change is going to take place or if it is a momentary ‘pause’ before the numbers snap back into place,” [pollster Bill McInturff] said…
In addition to Trump’s decline in the GOP race, the new NBC/WSJ poll shows a nine-point drop in the percentage of GOP primary voters who can see themselves supporting the real-estate mogul — from 65 percent in January to 56 percent now.
The highest candidate scores on this scale: Rubio (70 percent can see themselves supporting him), Cruz (65 percent), Carson (62 percent), Trump (56 percent), Kasich (49 percent) and Bush (46 percent).
And in hypothetical one-on-one match ups, Trump trails both Cruz (56 percent to 40 percent) and Rubio (57 percent to 41 percent). In January, Trump was ahead of Rubio (by seven points) but behind Cruz (by eight points).
Rubio’s at 17 percent, good for a six-point lead over Kasich for third. Jeb Bush is at, er, four percent. Is there … any reason to believe this poll is accurate, given all the data this week out of South Carolina showing that Trump is ahead comfortably? Let’s ask the nerds, people who study polls for a living:
Not that I can think of https://t.co/HcRg4vm9Xl
— Nate Cohn (@Nate_Cohn) February 17, 2016
@allahpundit Well, I can't think of any poll to suggest (1) Trump is way down; (2) Cruz is one benefiting. But everyone else cld be herding.
— Taniel (@Taniel) February 17, 2016
To believe that WSJ/NBC is right, you need to believe that basically everyone else is wrong, including the new Bloomberg poll of South Carolina released tonight showing Trump more than doubling Cruz’s share of the vote there, 36/17. The two other national polls taken since New Hampshire show Trump at 35 percent in one case, good for a 15-point lead over the field, and at 39 in the other, good for a lead of an even 20 percent. Of the last five polls of South Carolina, none has Trump lower than 33 percent or Cruz higher than 19 percent. If WSJ/NBC is picking up a “shift in the campaign,” no one else is.
The only way to defend this result, I think, is to zero in on the fact that it was conducted entirely after last Saturday’s debate, at which Trump got into some trouble by insisting that George W. Bush “lied” about WMDs in Iraq. The two other national polls were conducted mostly before the debate. That’s where the idea comes from that WSJ/NBC might be picking up a sea change in the race that other polls have missed. Maybe Trump has finally, finally, finally gone too far. It’s true that lots of Republicans, including those at Trump’s own rallies this week, complained about his Iraq comments — but people who say they’ve switched their vote because of it have been hard to come by in media reports. PPP’s poll of South Carolina found Trump leading after the debate even among Republicans who rate Dubya positively. Today’s new Bloomberg poll was conducted starting on Saturday, the day of the debate, and ending yesterday and still has Trump cruising to victory. Quote:
Trump did not appear to be hurt by a highly combative South Carolina debate performance Saturday, with his level of support staying between 30 percent and 40 percent on each of four nights of polling, with highest scores coming on Sunday and Tuesday nights.
The head-to-head numbers in the WSJ/NBC poll are especially noteworthy. Does anyone think Trump hurt himself at the debate so badly that, despite waltzing to a win in New Hampshire on a night when Rubio finished fifth, he’s suddenly lost a net of 23 points to Rubio in a one-on-one race? If you believe PPP, Trump leads Rubio in South Carolina head to head, 46/45, despite endless blather from righty pundits that Trump is doomed if he ends up as a binary choice with Mr. Electable. Cruz, meanwhile, has seen his favorable rating tank in multiple South Carolina polls lately thanks to his increasingly nasty war with Trump, yet suddenly WSJ/NBC has him running Trump off the field in a two-man race. That’s exceedingly hard to believe. Essentially, to square this encouraging new national poll with all of the data coming out of SC, you need to believe that something’s happening with Republicans across the country that’s mysteriously bypassing South Carolina — which, let me remind you, is supposed to be more friendly to Cruz than the rest of the country because of all the southern evangelicals there. How do you square it? How is this poll not obviously an outlier?
Exit quotation from Cruz, seizing on some good news where he can find it: