2016 Republicans, New Hampshire, Poll Watch, State of the Race, Uncategorized

Forecasting New Hampshire: The Establishment Strikes Back

December 18, 2015

We’ve assumed New Hampshire would play an important role in choosing the leading establishment-friendly candidate.  No visible change in that.  If Marco Rubio does well and finishes well ahead of the governors, he’s likely to advance with full support from that section of the primary electorate.

If one of the governors can separate from the others, finish ahead of Rubio and close enough to the top (ideally winning the primary), that person at least gets the chance to replace Rubio if they can finish ahead of him in South Carolina.

For at least the past couple months, candidates have strategized in line with the above, with Chris Christie, John Kasich and Jeb Bush focusing almost entirely on the Granite State, while Rubio, who has a chance to make some noise first in Iowa, is holding back from committing to a particular state (something many insiders are criticizing).

In most of the country, outsider and/or insurgent candidates still have the edge.  Donald Trump is now averaging well over 30% in national polls.  Combined with Ted Cruz and Ben Carson, the group is at 61% in the Real Clear Politics average, fairly consistent with results from the past several months.

While Cruz has caught and surpassed Carson and Trump oscillates between a large lead and a small one, the total share doesn’t move much.  Iowa is in line with the rest of the country, giving the three a 64% share in the RCP average.  The only difference is Cruz is leading Trump by a thin margin.

Two months ago, New Hampshire was playing along with everyone else.  When the Boston Herald polled in mid-October, the insurgent three combined for 49%, a little lower than the national norm, but including Carly Fiorina brought the outsiders up to nearly 60%.

The Establishment Four (Rubio, Christie, Bush, Kasich) totaled a mere 24%, with the three governors combining for 18%.  Several other organizations (Bloomberg, PPP, CBS/YouGov) polled during the same window and got fundamentally similar results.

The four outsiders received 62%, 54%, 52% and 59%.

The four insiders received 22%, 34%, 30% and 24%

The three governors received 15%, 22%, 22% and 18%

No huge outliers in any of the results.  At that time, a governor would have needed to consolidate the support of all three to have a chance at contention, and an establishment candidate would need the votes of all four to have a solid chance to win.

Combining the four was not always enough to equal Trump by himself.  At the time, several of these campaigns pushed the idea that things would change as we got closer to voting time.  In most of the country, including first-voting Iowa, that simply hasn’t happened yet.

Ben Carson has completely collapsed, down to 6% nationally in a PPP poll released today.  However, the beneficiaries are Cruz and Trump, not Jeb or even Marco.  But in New Hampshire, where the governors have invested most of their time, it has changed, and in two ways.

First, according to a new Boston Herald poll, the establishment candidates are no longer trailing the outsiders as a group.  I’m taking the liberty of assuming this isn’t an outlier.  Their results were consistent with others back in October, and the second most recent poll, taken for WBUR about 10 days ago, has virtually identical results.

Trump/Cruz/Carson: 43%

Rubio/Christie/Bush/Kasich: 38%

You’ll notice Fiorina isn’t included in this calculation.  She scored at 6% with Boston Herald, 3% with WBUR.  If you include her with the outsiders, things have still appreciably narrowed.  That’s not where Carly belongs now.

Whether in New Hampshire, Iowa or nationally, her support overlaps the establishment candidates far more than the insurgents.  In September/October, she was another option for someone otherwise considering Cruz or Carson.  Now she’s more frequently shopped against Rubio or Christie.

I’m not adding her in yet as part of the establishment group, but including with the insurgents doesn’t fit now either.  It’s part of her current weakness, where despite strong overall favorability ratings, Fiorina doesn’t fit anywhere at the moment.  If several candidates fall apart, she would suddenly find herself viable.

This shift means it is far more likely an establishment-approved candidate defeats Trump in New Hampshire and moves forward with a win instead of a moral victory/decent second-place finish.

Before, it would have required consolidating almost all of the support on that side of the field.  No more.  Trump is at 27% and 26% in the two most recent polls.  Though a candidate would still need a majority, it’s much easier math now.

If Trump should lose to Cruz in Iowa, something both on-the-ground insiders and pollsters are saying is quite possible, a loss in momentum from being a loser would make this even more likely.  Though The Donald continues to dominate national polls (PPP has him at 34% today), he’s actually at his most vulnerable if you look at how the first two states might shake out.

Within the establishment-friendly group, the governors are in a much better position than two months ago.  In October, the best result for the trio was 22%, probably not quite enough to win even if two of the three dropped out and none of the support bled to Rubio.

Now they combine for 29%.  Still not easy, one candidate would need about 80% of this to have a chance to win, but the pie is growing too.  One more small uptick, and a governor could win without Rubio collapsing.

It’s hard to picture the candidate who is always among the very top tier in favorability doing worse than 8-10% on primary day, so this matters.  It was never realistic to expect one establishment candidate to grab 90% of the support in New Hampshire, with several camped out in the state.

Who has the edge?  Is Rubio or a governor more likely to win?  If a governor, which one?

Averaging together the last two polls:

Rubio 11.5%

Christie 11.5%

Bush 9%

Kasich 7.5%

I’d like to introduce you to my pal, Mr. Margin of Error.  These are all technically close enough to make almost any argument from the top line numbers.  Jeb is making a little progress.  Boston Herald has him at 10%, a level he reached once in October.  Besides an 11% result in September, it’s the best he’s done there since August.

While I’m still convinced he’s virtually dead in terms of nomination chances, Bush is doing well enough in New Hampshire and making enough forward progress to noticeably impact the primary.  Can he win it?

Not unless he makes significant additional progress.  As we know, many voters have not made up their mind.  Even those who say they have may change it depending on how the candidates do in Iowa, what sort of attack ads they see over the next several weeks, events in the world, whatever.

There are ways to see who is closer to picking up additional support.  The Boston Herald poll was extra detailed in providing background data.  In addition to listing how candidates are doing as a second choice, it broke favorability out between very favorable and somewhat favorable.

Being viewed as very favorable is obviously better.  In a race with tons of candidates, it’s unlikely you’ll get many votes from people who only view you somewhat favorably.  The data shows the average candidate gets more than 80% of their top line support from voters who view them very favorably.

It’s also insulation against attack ads and incoming fire from other candidates and media.  If you view Marco Rubio very favorably, his back-and-forth with Ted Cruz will likely make you even more determined to support him.  For voters who feel that strongly about Cruz, it would benefit him the same way.

Though many voters like both Cruz and Rubio, it’s more common for a voter to find one very favorable and the other somewhat favorable.  This means their current conflict may actually wind up helping both, by making their very favorable voters more likely to vote for them, while sacrificing the somewhat favorables who would have chosen someone else anyway.

So how are the candidates doing on the very favorable index?

Trump 30.3%

Cruz 26.4%

Rubio 25.1%

Christie 21.5%

Fiorina 16.8%

Kasich 14.9%

Bush 14.8%

This is a really good proxy for the current ceiling of each candidate.  There’s always room for improvement over the next several weeks, but look at this as a current best-case scenario.  The other measure is combining first and second choices:

Trump 34.9%

Rubio 27.9%

Cruz 27.1%

Christie 25.7%

Bush 22.1%

Kasich 14.9%

Fiorina 13.2%

You’ll notice a very strong correlation between very favorable and first or second choice.  Kasich is an exact match, and several of the candidates are very close.  If you compare the two measures, it gives you an indication of who is doing better in the polls than their favorability would suggest and who has more room.

Fiorina +3.6%

Kasich  0%

Cruz -0.7%

Rubio -2.8%

Christie -4.2%

Trump -4.6%

Bush -7.3%

Jeb has no chance of winning New Hampshire unless he improves his very favorable rating.  He’s basically maxing out his top line support right now.    If the election were held next week, he’d need to hope for a divided field, get all of his first and second choice voters to pick him, and get a big chunk of his votes from voters who only view him somewhat favorably.

He still has 7 weeks, but lots of work to do still.  In case you’re wondering how he’s being more efficient translating favorability to polling support than the others, it’s likely because he’s made a good case for himself as having Commander-in-Chief skills.  A CNN New Hampshire poll taken a couple weeks ago had him rated more highly in that measure than his polling support.  It caught up.

Kasich has the worst odds of any governor.  He’s got the worst top-line polling and nothing in the underlying data is showing he has significantly more room.  His best results were over the summer when voters didn’t know him as well.  At this point, he’s a spoiler, and is now running anti-Christie ads.

He’s not likely to drop out before New Hampshire votes, but the establishment cause is helped if he does.  However, Rubio would probably like him to stay in, as Christie is his biggest competition and Kasich is correctly targeting him by comparing their respective records as governor.

It looks like Christie is about halfway there.  In order to win, he needs a very favorable score in the 25-30% range.  It won’t wind up higher with candidates attacking him.  He’s at 21.5%, part way between the laggards and goal.  Rubio, Cruz and Trump already sit in that range.  He’s made significant progress in the past few months, but the knives are out now.

Rubio is looking pretty good.  There are a million things that could throw him off, from doing (relatively) poorly in Iowa to his lack of ground game being as much of a problem as insiders think.  But, his underlying numbers show he could easily catch Trump.

He also has a huge amount of voters who view him somewhat favorably.  Convert a few of those and his odds are very solid.  Lots of variability.  He could finish first, he could finish fifth.  I would say his chances are better than any single governor, but are at best 50/50 against beating all of them.

The other notable conclusion is the upside for Cruz.  Iowa is his state, not New Hampshire, so winning would help tremendously, and finishing third would still look good.  The gap between him and Trump when looking at very favorable scores and combining first and second choices is slim.

If he defeats Trump in Iowa, and swings a few Trump supporters (many of whom think highly of Cruz), he can beat him in New Hampshire.  This is true whether or not Rubio or a governor finish ahead of them.  If Cruz bests Trump in both early states, The Donald is a goner.

Several weeks to go.  Much will change.  As of today, at least four candidates (Trump, Cruz, Rubio and Christie) appear to have reasonable chances of winning New Hampshire, with Jeb and Fiorina holding on to dim hope and Kasich mostly in the way.

One thought on “Forecasting New Hampshire: The Establishment Strikes Back

  1. This article was a lot of work weaving all the names, percentages, and how those numbers reflect any changes in the last month or so. It will be interesting to see if your predictions for the candidates are fulfilled and if the primary votes to come bare out your thoughts and possibilities for each candidate.

    There is a lot here to digest!


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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s