Polling wonks on Iowa results | Arkansas Blog

Tuesday, February 2, 2016

Polling wonks on Iowa results

Posted By on Tue, Feb 2, 2016 at 7:05 AM

Sam Wang of Princeton University writes: 

On the Republican side, Donald Trump and Ted Cruz switched places relative to polls. To compare the final polls with tonight’s counts, Trump underperformed by 26.5-24.3=2.2%, Cruz overperformed by 27.7-23.5=4.2%, and Rubio overperformed by 23.1-18.0=5.1%. The late swing for Rubio was visible in the final days of polling. All of this is well within the range of normal polling error in primaries. Multiple delegates went to Cruz (8), Trump (7), Rubio (7), and Ben Carson (3) – as expected.

It is premature to say that Trump is doomed. However, he does look a little less inevitable. It is certainly possible that he can crash from his high position in New Hampshire, South Carolina, and nationally. But I think a bigger risk to him is the possibility that tonight’s results will pressure Rubio’s lower-tier rivals to get out sooner rather than later. As I’ve written before, if the field gets down to three candidates after New Hampshire, that opens up a narrow route to stopping Trump. In short, tonight kept Marco Rubio’s chances alive.

On the Democratic side, meanwhile, Wang writes that the results were "substantively bad for Bernie Sanders." That seems to dramatically overstate the case. It was a good result for Sanders, seems to me — it's just that it's going to be exceedingly difficult to actually topple Clinton for the nomination. Here's Wang: 

After all the talk about hordes of Sanders supporters, in the end he only achieved a near-tie – 23 delegates for Clinton, 21 delegates for Sanders. Iowa is one of the most favorable states for him because of its ethnic composition. But it is not enough to win 50% of white Democrats. To have a chance overall, he needed a big win to (a) indicate that he can get enough white support to compensate for lack of support in nonwhite demographics in other states, and (b) create press coverage to boost him in the coming weeks. Outcome (a) didn’t happen. We’ll see about (b).

Nate Silver of fivethirtyeight, meanwhile, says that last night exposes Trump as a fringe candidate. That seems to be massively overstating the case (Trump finished a strong second, could easily win New Hampshire, and leads polling in multiple other states). Silver has been beating the bearish drum on Trump for months, so might be a teensy bit of confirmation bias here? 

On Monday, Iowa voters did something that Republican “party elites” had failed to do for more than seven months: They rejected Donald Trump.

Trump received 24 percent of the vote in the Iowa caucuses, placing him closer to the third-place candidate, Marco Rubio (23 percent), than to the winner, Ted Cruz (28 percent). Trump underperformed his polls, which had him winning Iowa with 29 percent of the vote, while Cruz and Rubio outperformed theirs.

Silver argues that Trump skipping the debate and his relative lack of a ground game may have been factors, but the big issue was the Donald himself: 

But there’s good reason to think that the ground game wasn’t the only reason for Trump’s defeat. Republican turnout in Iowa was extremely high by historical standards and beat most projections. Furthermore, Trump won the plurality of first-time caucus-goers.

There may have been a more basic reason for Trump’s loss: the dude just ain’t all that popular. Even among Republicans.

The final Des Moines Register poll before Monday’s vote showed Trump with a favorable rating of only 50 percent favorable against an unfavorable rating of 47 percent among Republican voters. (By contrast, Cruz had a favorable rating of 65 percent and Rubio was at 70 percent.) It’s almost unprecedented for a candidate to win a caucus or a primary when he has break-even favorables within his own party.

Silver says that Trump's strong national polls "might also be a mirage" and that "Iowa voters made Trump appear to be much more of a factional candidate along the lines of Pat Buchanan, who received 23 percent of Iowa’s vote in 1996, than the juggernaut he’s been billed as." 

Silver's colleague Harry Enten argues that Sanders "needed more than the tie he got with Clinton": 

Clinton will continue to be a favorite for the Democratic nomination if she continues to hold a large lead among nonwhite voters and basically breaks even with white voters, as she did in Iowa. Sanders, meanwhile, needs to cut into Clinton’s lead among nonwhites and expand his support among white voters beyond what he won in Iowa. If he does that, he’ll put himself in contention to win the nomination. If he doesn’t, he’ll continue to be an underdog.

Tags: , , , , , , , ,

From the ArkTimes store

Favorite

Comments

Showing 1-1 of 1

Add a comment

 
Subscribe to this thread:
Showing 1-1 of 1

Add a comment

More by David Ramsey

Readers also liked…

Most Shared

Most Recent Comments

Blogroll

 

© 2017 Arkansas Times | 201 East Markham, Suite 200, Little Rock, AR 72201
Powered by Foundation