Why did Trump only get 24% of the republican vote in Iowa?

[Indus]

I don't see it being talked as much about.. what could it be?

Is it that we as a country are smarter than just giving into the xenophobic sentiment dressed up as "Make America Great Again"?

What do you think happened? What do you think is happening?

 

[Strk]

I don't see it being talked as much about.. what could it be?

Is it that we as a country are smarter than just giving into the xenophobic sentiment dressed up as "Make America Great Again"?

What do you think happened? What do you think is happening?

Cruz's campaign is plenty xenophobic, so pretty sure it isn't that

 

[Sonikku]

Evangelicals.

 

[shady28]

I don't see it being talked as much about.. what could it be?

Is it that we as a country are smarter than just giving into the xenophobic sentiment dressed up as "Make America Great Again"?

What do you think happened? What do you think is happening?

Ignoring the partisan political attacks and thinking about the first question logically -

The answer is that Trump never polled to win Iowa by any significant margin in the first place.

From Jan 13th.

Iowa Poll: Cruz holds 3-point lead as Trump attacks

Ted Cruz sits atop the Republican pack in Iowa with just 19 days until the caucuses, but Donald Trump is just 3 percentage points behind, a new Des Moines Register/Bloomberg Politics Iowa Poll shows.

The main surprise in the Iowa GOP caucus was Rubio, who siphoned votes from pretty much every other candidate.

From the same Jan 13th article :

In third place with likely GOP caucusgoers is Marco Rubio, with 12 percent, and Ben Carson is huddled close behind at 11 percent.

 

[Strk]

Ignoring the partisan political attacks and thinking about the first question logically -

The answer is that Trump never polled to win Iowa by any significant margin in the first place.

From Jan 13th.

Iowa Poll: Cruz holds 3-point lead as Trump attacks




The main surprise in the Iowa GOP caucus was Rubio, who siphoned votes from pretty much every other candidate.

From the same Jan 13th article :

Maybe it was all his Jesus segues in the debates? He was definitely pushing hard to get those out.

 

[Fanatical Meat]

Honestly I think Trump did pretty well in Iowa considering they pick the odd balls

 

[shady28]

Maybe it was all his Jesus segues in the debates? He was definitely pushing hard to get those out.

Of the 3, Trump has the least Evangelical cred. Rubio is a Catholic, no idea how that resonates with Evangelicals.

The polls that look at likely caucus participants are turning out to be far more accurate than general polls. If you look at polls that focus on likely caucus participants, Cruz has a slight advantage over Trump. Trump needs to get new participants in the caucus' to win, same problem Bernie has but I think Bernie is better at turning out the vote.

 

[zinfamous]

Nutter evangelicals always come out for their favoritist nutter evangelical in Iowa.

This really isn't a mystery.

 

[dullard]

Evangelicals.

Iowa Republicans are about 60% Evangelicals. They tend to vote solely based on religion. Trump is not a religious person and in general does not appeal to Evangelicals (I don't think he goes to church, mispronounces simple names in the Bible in a way that indicates that he doesn't even know much about religion, etc). So, in reality, he was splitting the remaining 40% of the Republican vote. And he did pretty well in a crowded field to get a majority of that remaining 40%.

New Hampshire will be the opposite. Very few Evangelicals there.

The great thing to politicians about Evangelical votes is that you only have to pretend to support their religion. You don't have to actually do anything about it. So, on that side of the coin they are an easy voting block to obtain. But, you actually have to attempt to appeal to them to begin with. Trump never really tried. Cruz spent pretty much his entire campaign in Iowa courting solely the Evangelicals.

 

[fskimospy]

Ignoring the partisan political attacks and thinking about the first question logically -

The answer is that Trump never polled to win Iowa by any significant margin in the first place.

From Jan 13th.

Iowa Poll: Cruz holds 3-point lead as Trump attacks




The main surprise in the Iowa GOP caucus was Rubio, who siphoned votes from pretty much every other candidate.

From the same Jan 13th article :

Why did you not use the more recent Register poll from right before the election which showed Trump up by 5, which was actually a pretty significant lead? Very misleading post.

http://www.desmoinesregister.com/st...rump-reclaims-lead-latest-iowa-poll/79562322/

 

[Gonad the Barbarian]

Honestly I think Trump did pretty well in Iowa considering they pick the odd balls

I agree, and also considering he's never run a campaign before. Both Cruz and Rubio have an existing machine that have managed successful senate campaigns.

 

[mxnerd]

Why? Because people realized that he is just a clown.

 

[dullard]

Why did you not use the more recent Register poll from right before the election which showed Trump up by 5, which was actually a pretty significant lead? Very misleading post.

http://www.desmoinesregister.com/st...rump-reclaims-lead-latest-iowa-poll/79562322/

Until you consider that EACH candidate was +- 4% in that poll (meaning a +-8% swing when comparing two candidates). That poll was 95% confident that Trump would be between 24% to 32% and that Cruz would be between 19% and 27%. Thus that poll you linked had a range of Trump up by 13% all the way down to Cruz up by 3%.

The final result of Cruz up by 4% wasn't too far off. It was just at the edge of what was predicted for Trump and just past the expected range for Cruz.

That is about as good as you can realistically expect in a difficult to poll situation (Iowa is early before many people made up their minds, it is a caucus not a typical vote, it had many candidates, etc).

 

[Phokus]

Evangelicals.

And inferior ground game to Cruz

 

[cbrunny]

Iowans love Canadians too.

 

[DrPizza]

Evangelicals.

"I'm a Christian first and..."

/thread.

 

[fskimospy]

Until you consider that EACH candidate was +- 4% in that poll (meaning a +-8% swing when comparing two candidates). That poll was 95% confident that Trump would be between 24% to 32% and that Cruz would be between 19% and 27%. Thus that poll you linked had a range of Trump up by 13% all the way down to Cruz up by 3%.

The final result of Cruz up by 4% wasn't too far off. It was just at the edge of what was predicted for Trump and just past the expected range for Cruz.

That is about as good as you can realistically expect in a difficult to poll situation (Iowa is early before many people made up their minds, it is a caucus not a typical vote, it had many candidates, etc).

While I agree that polling caucus states is difficult you are understating the improbability of this result, assuming a valid sample. The margin of error reported for these polls is a maximum margin of error which stands for what it would be if a candidate polled 50%. The actual MOE for these candidates is considerably lower.

More importantly, if you're looking for the probability of the Trump v. Cruz result you don't take the two margins of error and add them together, you calculate the difference of proportions between the two's numbers and get a confidence measure from that. I'm quite confident this result was out of the margin of error for that.

 

[dullard]

While I agree that polling caucus states is difficult you are understating the improbability of this result, assuming a valid sample.

The problem is the "valid sample". That poll started Jan 26. Polls that started Jan 29 were closer to the actual result with poll results of Trump ahead just by 1% rather than 5%:

http://fivethirtyeight.com/features/iowa-teaches-pollsters-to-poll-until-the-end/

As people hadn't made up their mind yet, how can you have a valid sample? Older polls even by just a few days were not valid in this case.

 

[fskimospy]

The problem is the "valid sample". That poll started Jan 26. Polls that started Jan 29 were closer to the actual result with poll results of Trump ahead just by 1% rather than 5%:

http://fivethirtyeight.com/features/iowa-teaches-pollsters-to-poll-until-the-end/

As people hadn't made up their mind yet, how can you have a valid sample? Older polls even by just a few days were not valid in this case.

Well that's something else you could talk about, and I agree that also makes polling difficult. That's different than saying the results between Cruz and Trump were within the margin of error though, and shady should not have tried to use the Register poll to say that.

Every sample is only valid for the time in which it was taken. That's just the inescapable nature of polling

 

[buckshot24]

Cruz "stole Iowa", that's why.

http://www.usatoday.com/story/news/...donald-trump-ted-cruz-iowa-caucuses/79748926/

 

[feralkid]

Trump is maxed out, voter-wise.

He's not attracting any new supporters.

Some of his current supporters have grown weary of his schtick.

In other words: He's a Loser.

Which is too bad, he still had so much of the GOP left to destroy.



.

 

[dullard]

I'm quite confident this result was out of the margin of error for that.

http://fivethirtyeight.com/features/is-the-polling-industry-in-stasis-or-in-crisis/

Polls of presidential primaries and caucuses are another matter entirely; they haven’t been much good. The average error for presidential primary polls since 2000 has been 7.7 percentage points — about twice as large as for presidential general elections. The polls were especially bad in the 2012 Republican primaries, when they missed by an average of 8.7 percentage points.

Polls of primaries are not very accurate. When the polls average about an 7.7% error, a 5% lead in one poll is not enough to claim that a person is actually in the lead. Despite their headline +-4% claim, you have to basically double it due to so many factors that aren't built into the statistics of the poll (and thus are in addition to the 4% headline MOE). This result was fairly typical.

 

[Pocatello]

Trump should had gone to the last debate, instead he showed he was being a spoiled kid who couldn't take the heat.

 

[BurnItDwn]

Stupid is as stupid does.

 

[fskimospy]

http://fivethirtyeight.com/features/is-the-polling-industry-in-stasis-or-in-crisis/

Polls of primaries are not very accurate. When the polls average about an 7.7% error (despite their headline +-4% claim, you have to basically double it), a 5% lead in one poll is not enough to claim that a person is in the lead.

I agree that primary polling is poor, but you wouldn't just double their stated margin of error if you wanted to correct for it. That's just not how the math works.

Anyways, back to the original point: shady' statement was misleading.