2016 Democrats, 2016 Republicans, Iowa, Predictions, Uncategorized

Forecasting Iowa: Final Predictions

February 1, 2016

The waiting ends tonight.  My stalling ends now.  For comparison, you can check the Real Clear Politics polling average here and the FiveThirtyEight estimate there.  Traditionally, the final Des Moines Register/Iowa poll is at least as good an indicator as anything else.

All of those measures are currently aligned.  They each suggest victories for Hillary Clinton and Donald Trump.  If you are reading only to see if I disagree, you can stop now.  I think they will both win.  If you’re curious how, by what margin, and where everyone else falls, keep reading.

Democrats (State Convention Delegate Allocation)

Hillary Clinton 49.8%

Bernie Sanders is not Barack Obama.  Unlike the 44th president, he does not have the support of a sizable chunk of the Democratic establishment.  Obama’s edge in support extended beyond the youngest voters or newest voters.  He had an edge with regular caucusers too.

The huge increase in turnout was due to enthusiasm for nominating the first female president, or the first African-American president, or Iowa favorite John Edwards.  Obama was the most popular of a much stronger group.

In 2015, Joe Biden was a possible savior.  In 2008 he was an asterisk.  I think Hillary will win.  I think her third place finish was more impressive than this victory would be, unless the margin is far greater than I’m envisioning.

Obama, Edwards and the afterthoughts were all Democrats.  Sanders isn’t.  This may seem obvious, but the majority of caucus participants are registered Democrats (in fact non-Dems need to change their registration to participate.)  Not only does Hillary have virtually the entire national and state party behind her, but many voters are loyal Democrats.

This has enabled a stronger closing argument than listening to it makes it seem.  Hillary is a bad campaigner.  She’s not good with a crowd.  She isn’t likable unless you directly identify with her or her life story.  But she’s riding the donkey to the finish line, mentioning her husband and the current president at every turn.

If you’re a committed Progressive Uber Alles, not enough.  If you’re too young to remember the Clinton economy, not enough.  But the most frequent caucus participants are long-time Democrats of a certain age.  That age is elevated enough to remember the original New Left.

When they and the movement were young, there was great promise.  Then Nixon and Reagan each won 49 states in their re-election campaigns.  These voters remember all too well that when times were a changing they were moving right, at least at the presidential level.

Dog whistles matter.  Two of them will help Hillary at the finish line.  First, the State Department refusal to release 22 Top Secret emails, something negative overall for Clinton, is a short-term positive.  Loyal Democrats will first view this as partisanship, another attack on their presumptive standard bearer.

Second, Sanders got statewide coverage for his rally/concert Saturday night, in which the campaign channelled its inner Woody Guthrie.  It was a neat moment.  And a reminder of unfulfilled promise for thousands of liberal-leaning seniors.

I have strong doubts Hillary can make this approach stick long enough to get elected.  Voters don’t like hearing about what can’t be done.  Protecting what you have sounds better when you feel like you have enough.

But it was probably enough for now.  She can worry about the rest tomorrow.

Bernie Sanders 48.9%

So close.  I’m assuming the weather won’t be good enough to make it easy to caucus, nor bad enough to keep too many voters home.  Given that it’s February in Iowa, he can’t blame the snow if he falls short.

I’m assuming he gets more than his share of O’Malley defectors.  I’m assuming he gets more than his share of undecideds.  Hillary has made as good of an argument as she can.  If she hasn’t won someone over yet, she probably won’t.

A Sanders win wouldn’t shock me by any means.  I’m rooting for it.  As I’ve previously disclosed, I’m not a Clinton fan.  A Bernie win makes the next few weeks/months infinitely more interesting for me.  My sister (a Berner volunteer coordinator) would be very happy.

But I can’t make the math work.  His poll numbers are slightly worse than two weeks ago, not better.  Registration numbers do not indicate a surge in new voters.  I’m sure plenty of voters will register day of, but likely not enough.  There’s a way to argue Bernie actually gets a few more caucus votes, but the state convention delegate allocation is what counts.

Distribution of strength aids Clinton with this.  I think there’s at least a 20% chance he gets more votes, but she gets more delegates.  It’s that close.  FWIW, the same thing happened in 2008.  Hillary got a few more votes than Edwards, but he won more state delegates.  All we heard was she finished third.

Martin O’Malley 1.3%

Consolation for O’Malley.  In 2008, future Vice President Joe Biden received 0.9% of the Iowa caucus vote.  Under party rules, it’s almost impossible for a non-competitive candidate to score much better than this.  Bill Richardson landed at 2.1%, Christopher Dodd at 0.0%.

O’Malley’s number is as high as I’ve given him because he does have pockets of collegiate support, meaning he will hit the 15% threshold in at least a few precincts.



Donald Trump 28.7%

I think The Donald does it.  Not by much, but think he wins.  He’s ahead in every Iowa poll.  Maybe you really can make something so by saying it over and over.  This is an achievement.  Even if he falls just short, it’s still an achievement.  It really shouldn’t make any difference whether he wins by a couple points or loses (we’re talking about a few thousand voters), but it will, at least in the short run.

As long as his voters don’t completely vaporize, he will have proven he can turn out voters who normally don’t participate in this stage of the process.  It gets easier from here.  Primaries require less commitment, the weather is better in most of the early voting states, and many make it easier for Independent voters to play.

On average, Trump is up 5 points in the polls.  Figure Cruz gets a larger share of the undecideds.  Assume his harvesting operation turns out his voters as well as everyone figures.  That’s still not quite enough.  The DMR/Iowa poll relies entirely on already registered voters.

They still have Trump up 5 (28/23).  If a decent amount of Trump voters take advantage of being able to register at the caucus, he’ll win by more than expected, not less.  There are some weather concerns, but the blizzard won’t hit most of the state until Tuesday.

Cruz needed to take advantage of Trump’s absence to make a strong closing argument in the debate.  He didn’t.  There just aren’t enough reasons for voters split between Marco and Ted to pick Cruz, for him to get a high enough percentage.  He’s making a negative argument.  Don’t vote for Marco, because of amnesty.

Voters considering Rubio want a positive message.  Between failing to make one in the debate and going negative in ads, Cruz isn’t going to swing enough Rubiophiles to him.

If Ted is stuck in the 20s, Trump is fine as long as a moderate amount of his support shows.  The debate gambit helped this.  While it increased the amount of voters who are opposed to him, it helped lock down the already favoring.  The DMR/Iowa poll indicates he has the highest percentage (71%) of surveyed supporters who are definitely in.

Ted Cruz 26.4%

If he wins, not a surprise.  The leading evangelical-friendly candidate normally over-performs his final polls.  If he does slightly better, and Trump does slightly worse, he finishes ahead.  It’s a simple formula.

A large victory margin would surprise me, but Cruz 29%, Trump 27%, very plausible.  I’m almost starting to think Ted is better off losing (as long as it’s close).  Hillary Clinton may get away with cynically stitching together enough primary voters to win.  She can make an argument for hiring her the way an attorney argues a case.

Strong legal arguments win over judges.  Cruz built his career on doing this.  Primary voters are more like juries.  Emotion and likability are important too.  Also, you can pick some of the jurors, but not all of them.

Cruz is optimizing his pitch for his jurors, but he’s going to need to win a few others over.  Arguing against Trump isn’t enough.  Talking about Rubio’s immigration apostasy isn’t enough.  It probably won’t even work in Iowa.  If it does, it was by the slimmest of margins.

Even South Carolina Republicans are more secular than Iowans.  He doesn’t have the same endorsement support down there.  Unless Nikki Haley or Tim Scott endorse him before the contest, he’s on his own.  An Iowa win would be due to the most impressive traditional ground operation of any GOP candidate.  That’s great for a caucus, less helpful in a primary.

Hillary may need to argue to a jury, but the judge is in her pocket.  Cruz has no chance of gaining a majority of establishment support.  Hillary has all of it.  An outsider/insurgent candidate needs a bold, large, coherent vision (yes, it’s tough to do all that at once.).  Ted is offering up talking points, albeit strong ones.

Not enough, even if the final tally says it was.

Marco Rubio 16.6%

This is almost exactly where the available data shows he is.  His support is a mile wide and an inch deep.  There’s a scenario where many voters who have him as a second choice decide to move his way at the last minute.  If that happens, he winds up over 20% and makes things real interesting.

There’s another where having a lower than average percentage of committed voters hurts him, pushing him down near 10% and Carson slips past him.  When in doubt, assume neither extreme.

He’s making a big push over the weekend, both in advertising and appearances.  Rubio looks and sounds confident, the media is saying nice stuff (in his lane, that’s not a bad thing) and he’s effectively playing the god card where applicable.

He’s very comfortable talking about his faith, but does so in a way more appealing to secular ears than Cruz.  Ted is the son of a preacher man.  Marco was born and raised Catholic, spent some time in his youth as a Mormon, switched back to Catholicism, and simultaneously attends an evangelical church.

I expect this diversity of faith is a feature going forward, not a bug.  It’s a template for what Rubio needs to do if he actually wants to win.  On this topic, he’s descriptive and passionate without offending.  It’s the version of Marco his cheerleaders regularly picture.

He’s not there yet on enough subjects.  Watching the last debate, I thought he missed an opportunity to move past Cruz.  Then everybody said he won and I thought I was wrong.  I wasn’t.  He did fine, but only well enough to remind voters they like him.  To beat Trump, to beat Cruz in states that are Ted-friendly, he needs to add more texture.

Marco does well enough to continue plenty intact, but not enough to make him a favorite yet.  It’s there, but does he channel it in time?

Ben Carson 8.6%

Taking the easy way out and splitting the difference.  DMR/Iowa Poll has him at 10%.  Figuring some of his voters would rather throw in with Cruz than see Trump win once they actually get to the caucus gives you an estimate of 6-7%.

There is a world where he finishes third or fifth, but this way, I’m only off by one spot at most.

Rand Paul 7.1%

He’s my pick to over-perform expectations.  No poll has him this high.  It’s an 80th percentile (or higher) outcome according to FiveThirtyEight.

His support is concentrated in pockets, some among college students, some not.  This is bad if you’re trying to win, great if you want to avoid embarrassment and register an adequate result.

Cruz is hitting Trump regularly on Eminent Domain.  This also helps the libertarian in the race.  It’s not enough to push Rand into double-digits, but it helps a little.  He also did well in the debate and is likely picking up some late momentum under the radar.

Jeb Bush 3.6%

He’s actually below this number in the DMR/Iowa Poll.  I’m betting on his ground organization, most specifically him having enough precinct captains and assistant captains to make up more than half of this percentage just with them.

That’s right, because I believe in the power of Jeb! I’m giving him a whole 3.6 percent.  I feel bad for him and worse for his investors.  His strategists ran the wrong campaign for their man.  I don’t blame them for failing to anticipate the contours of 2016.  I do blame them for failing to adjust.

Chris Christie 2.4%

Should.  Have.  Run.  In.  2012.

Rick Santorum 2.1%

This is my bold prediction.  His consensus finish is last.  Unfortunately for Rick Santorum, you don’t care why I’ve chosen to slot him here.

John Kasich 1.6%

This is ever so slightly below his averages.  Figure on a couple people throwing in with Rubio at the caucus to push back against a precinct speech for Cruz or Trump which managed to offend a few establishment-friendly voters in the room.

Mike Huckabee 1.5%

He’s ahead of a few of the people above him on my list in polls.  My sense is Santorum supporters are more loyal.  This is based only on gut feeling and watching too much C-SPAN.  After a candidate does a town hall, the cameras follow him/her talking to supporters and doing interviews.

It appears the micro slice of non-Cruz/Carson supporting values voters like Huck, but feel more strongly about Santorum.  Snuck in my reasoning after all.

Carly Fiorina 1.4%

In an alternative universe, she could have finished in the top three.  You can shuffle the bottom five or six candidates in any order, with equal chance of accuracy.  Watching her after a recent town hall, I got the impression she knows she’s cooked.  Unlike Santorum, I don’t think she finds any dignity in trying.  She’s a competitor.  Losing sucks.

UPDATE (8:55am Pacific):

My comment about being more bullish on Rand Paul than any poll is no longer correct.  A newly released survey from Opinion Savvy has him at 9%.  They also have Trump, Cruz, and Rubio in an effective 3-way tie.  The Donald leads by 1 point, so his streak is intact.

Another new entry from Emerson has similar results for the top three, with Trump ahead of Cruz by 1 and Rubio very visible in the rear-view mirror.  I’m sticking with my guesses, but nothing like brand new polls from organizations that didn’t previously survey Iowa to strike fear into the heart of the overconfident blogger.

Emerson recently released a poll in New Hampshire that gave Jeb Bush his best numbers (18%) since the pre-Trump era, so they have a bit of an establishment-friendly candidate lean.  If they’re correct, the RNC and most elected GOP officials are going to feel better than they figured in another week or two.

New polls on the Democratic side are exactly in line with previous results.  They show Sanders microscopically weaker than two weeks ago.



2 thoughts on “Forecasting Iowa: Final Predictions

  1. All of these percentages are fine, but it’s rather uncourageous of you to predict to only one decimal point. 28.70%, 26.40% and 16.60% exactly? Whatever… Grow a pair, man! Come on, make some specific predictions!

    In any event, throw out the percentages – I don’t care about that. Percentages don’t elect candidates, electoral votes do! How many electoral delegates do these numbers imply for each candidate, and no rounding this time!



    1. That’s the great irony. I have no clue. If my vote share numbers are correct, think Trump and Cruz would each get 8-10 delegates out of the 1237 they need for nomination. There are 30 total to distribute. The whole exercise is a beauty contest.

      On top of that, it takes months to actually assign them at the state convention. By the time that happens, people who earned delegates tonight will have dropped out and their delegates may move to the survivors. This is different from the primary states and some caucus states where nothing happens to re-assign delegates until the national convention. It’s a cluster.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s