2020 New Hampshire Primary: Final Polling Update

[This poll was updated at 4:15 EST on February 11, 2020 to account for a Change Research poll (C) conducted February 8-9, 2020.]

At just after midnight EST on February 11, 2020, a total of 27 Democratic voters gathered in the small New Hampshire hamlets of Dixville Notch, Hart’s Location and Millsfield to cast the first votes in the 2020 New Hampshire Presidential Primary. Here are the tallies from those early votes (which, while fun to watch, have historically shown little relationship to the way the statewide electorate votes); I jotted them down as they were being broadcast live on MSNBC:

Table 1: 2020 New Hampshire Presidential Primary votes from three towns voting just after midnight

Candidate Dixville Notch Hart’s Location Millsfield Total
Klobuchar 0 6 2 8
Sanders 1 2 1 4
Warren 0 4 0 4
Yang 0 3 0 3
Biden 0 1 1 2
Buttigieg 1 0 1 2
Bloomberg 2 0 0 2
Steyer 0 1 0 1
Gabbard 0 1 0 1
TOTAL 4 18 5 27

The perhaps-surprising winner in these early returns is United States Senator (“Senator”) from Minnesota Amy Klobuchar, who had a very well-received debate performance in Manchester, New Hampshire on February 7; as the data in Table 2 suggest, she may have some momentum from that performance. Also, the name of former New York City Mayor Michael Bloomberg does not appear on the either the Democratic or Republican New Hampshire Primary ballot; these votes, as well as the only Republican vote cast in Dixville Notch, are write-in votes.[1]

To learn how I calculate candidate WAPA (weighted-adjusted polling average), please see here. I weight polls conducted partially (1.33 or 1.67) or wholly (2.00) after the Iowa Caucuses higher than prior polls.

Here is a photograph of a now-defunct ice cream parlor in scenic Portsmouth, New Hampshire in early September 2005; I was never the owner of this establishment:

Scan0046

**********

As of 2 a.m. EST on February 11, 2020, here is a breakdown of publicly-available New Hampshire Primary polls:

  • 72 since January 1, 2019
  • 58 since the 1st Democratic debate on June 26, 2019
  • 36 since the 5th Democratic debate on November 19, 2019
  • 28 since the 7th Democratic debate on January 14, 2020
  • 16 since the Iowa Caucuses on February 3, 2020
    • 15 starting February 4 or later
    • 1 (Monmouth University) on February 3

The final two tracking polls by Emerson College and the University of New Hampshire overlapped in time with each organization’s prior tracking polls, so I only used the most recent ones.

Table 2: Final New Hampshire WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Since

Iowa

Sanders 23.0 23.1 25.0 25.7 26.6
Buttigieg 16.7 17.0 18.8 19.4 21.4
Biden 15.7 15.2 13.9 12.9 12.1
Warren 14.0 14.2 12.8 12.5 12.7
Klobuchar 7.0 7.2 8.5 9.3 9.6
Gabbard 3.5 3.6 3.8 3.6 3.3
Yang 3.3 3.4 3.6 3.5 3.3
Steyer 2.5 2.6 2.9 2.8 2.6
Bennet 0.4 0.4 0.5 0.5 0.5
Bloomberg 0.3 0.3 0.4 0.2 0.1
Patrick 0.3 0.3 0.4 0.4 0.5
DK/Other 13.4 12.6 9.4 9.2 7.3

Besides Klouchar, the top two finishers in the Iowa Caucuses—Vermont Senator Bernie Sanders and former South Bend, Indiana Mayor Pete Buttigieg—also appear to have momentum going into New Hampshire. The likelihood is that Sanders will win the primary—though not by anywhere close to his 2016 winning margin of 60% to 38% over former Secretary of State Hillary Clinton—perhaps followed by Buttigieg.

By contrast, former Vice President Joe Biden and, to a lesser extent, Massachusetts Senator Elizabeth Warren appear to be fading somewhat. United States of House of Representatives member Tulsi Gabbard of Hawaii, entrepreneur Andrew Yang and businessman Tom Steyer remain clustered around 2.5-3.7 percentage points, while Colorado Senator Michael Bennet and former Massachusetts Governor Deval Patrick appear to be non-factors at this point.

We shall see.

Until next time…

[1] President Donald Trump received 15 votes in Hart’s Location and 16 votes in Millsfield, while former Massachusetts Governor William Weld received 4 votes and 1 vote, respectively. Mary Maxwell received 1 vote in Hart’s Location.

2020 Iowa Caucuses: How did my polling averages fare?

Given the extremely volatile polling for the 2020 Democratic presidential nomination following the conclusion of the Iowa Caucuses, I will not provide global monthly updates for next few months. Instead, I will focus on the first handful of primaries and caucuses: Iowa on February 3, New Hampshire on February 11, Nevada on February 22, South Carolina on February 20, the 14 Super Tuesday contests on March 3, and so forth.

Also: I now weight polls conducted partially after February 3, 2020 either 1.333 or 1.667 times higher, and polls conducted entirely after February two times higher, than polls conducted entirely before February 4, 2020.

On the night of February 3, 2020, I was sitting on my usual spot on our sofa, watching MSNBC and anticipating returns from that day’s Iowa Caucuses.

Iowa Visitor Center Sep 1990

Earlier that day, I had published my final WAPA (weighted-adjusted polling average) for the 11 declared Democratic presidential candidates, calculated four different ways (Table 1):

  • Using all 58 polls conducted since January 1, 2019
  • Using only the 45 polls released since the 1st Democratic debate on June 26, 2019
  • Using only the 21 polls released since the 5th Democratic debate on November 19, 2019
  • Using only the 15 polls released since the 7th Democratic debate on January 14, 2020

Table 1: Final Iowa Caucuses WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate
Biden 19.9 19.8 20.1 20.3
Sanders 18.4 18.8 21.0 22.7
Warren 17.1 18.1 15.6 15.6
Buttigieg 15.9 16.8 16.7 16.7
Klobuchar 6.9 7.3 9.1 9.7
Yang 3.0 3.2 3.6 3.9
Steyer 2.8 3.1 3.1 3.5
Gabbard 1.5 1.6 1.5 1.6
Bloomberg 0.4 0.4 0.6 0.5
Bennet 0.3 0.3 0.2 0.3
Patrick 0.0 0.0 0.0 0.1
DK/Other 13.8 10.6 8.5 5.2

Based solely on these numbers, one would reasonably draw the following conclusions:

  • United States Senator (“Senator”) from Vermont Bernie Sanders and Minnesota Senator Amy Klobuchar were rising in the polls heading into the Iowa Caucuses, as to a lesser extent were entrepreneur Andrew Yang and businessman Tom Steyer.
  • Massachusetts Senator Elizabeth Warren was declining in the polls.
  • No other candidate was moving in the polls one way or the other.

By 11:37 pm EST, however, I had grown tired of waiting for results other than successive waves of entrance polls, so I tweeted the following:

RIP, Iowa Caucuses (1972-2020)

I have defended their idiosyncrasies for decades, believing the retail aspects of campaigning there outweighed the low-turnout mischegoss of the process.

 No more.

 This is ridiculous.

 #IowaCaucuses #iowacaucus2020

I will not relitigate here the myriad problems the Iowa Democratic Party had with tabulating, validating and releasing three distinct measures:

  1. Initial headcount of support for each Democratic candidate (“Initial tally”)
  2. Post-realignment headcount of support for each Democratic candidate (“Final tally”)
  3. Allocation of “state delegate equivalents,” or SDE’s, the only measure ever previously reported

Moreover, my annoyance has abated since Monday night, primarily because I suspect these vote-reporting snafus revealed that the byzantine process of converting persons standing in rooms, then possibly standing in different parts of the room, into SDE’s has always been “riddled with errors and inconsistencies,” to quote a recent New York Times headline. And if this marks the beginning of the end of using caucuses to allocate delegates to each party’s nominating conventions, so be it; they are undemocratic, exclusionary and overly complex.

As for which states “should” come first in future presidential nominating processes, I am currently agnostic.

Three days later, we finally have near-final results from the Iowa Caucuses (Table 2):

Table 2: Near-final Iowa Democratic Caucuses results, February 3, 2020

Candidate Initial Tally Final Tally SDE’s
Biden 15.0 13.7 15.8
Sanders 24.8 26.6 26.1
Warren 18.4 20.2 18.0
Buttigieg 21.3 25.0 26.2
Klobuchar 12.7 12.3 12.3
Yang 5.0 1.0 1.0
Steyer 1.7 0.2 0.3
Gabbard 0.2 0.0 0.0
Bloomberg 0.1 0.0 0.0
Bennet 0.1 0.0 0.0
Patrick 0.0 0.0 0.0
Uncommitted 0.6 0.1 0.2

The following three tables list the arithmetic differences between each candidate’s final Iowa Caucuses WAPA and each of the three reported measures; positive values indicate better performance in the Caucuses than in the polls.

Table 3: Arithmetic difference between Initial Iowa Caucuses % of vote and Iowa Caucuses WAPA

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Mean

Difference

Biden -4.9 -4.8 -5.1 -5.3 -5.0
Sanders 6.4 6.0 3.8 2.1 4.6
Warren 1.3 0.3 2.8 2.8 1.8
Buttigieg 5.4 4.5 4.6 4.6 4.8
Klobuchar 5.8 5.4 3.6 3.0 4.5
Yang 2.0 1.8 1.4 1.1 1.6
Steyer -1.1 -1.4 -1.4 -1.8 -1.4
Gabbard -1.3 -1.4 -1.3 -1.4 -1.4
Bloomberg -0.3 -0.3 -0.5 -0.4 -0.4
Bennet -0.2 -0.2 -0.1 -0.2 -0.2
Patrick 0.0 0.0 0.0 -0.1 0.0
DK/Other -13.2 -10.0 -7.9 -4.6 -8.9

Initial tally. If the Iowa Caucuses were instead the Iowa Primary, this would have been the only vote reported. On this measure Sanders, Klobuchar and former South Bend, IN Mayor Pete Buttigieg averaged 4.5-4.8 percentage points (“points”) higher in the initial tally than in their WAPA. And the closer in time the polls were to the Iowa Caucuses, the more “accurate” the WAPA.

Warren (+1.8 points) and Yang (+1.6) also overperformed their WAPA in the initial tally, albeit by smaller margins. And for Warren, older polls were more predictive than recent polls.

By contrast, former Vice President Joe Biden did an average of 5.0 points worse in the initial Iowa Caucuses tally than his WAPA. Steyer and United House of Representatives Member from Hawaii Tulsi Gabbard (-1.4 each) also performed somewhat worse than their WAPA.

Table 4: Arithmetic difference between Final Iowa Caucuses % of vote and Iowa Caucuses WAPA

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Mean

Difference

Biden -6.2 -6.1 -6.4 -6.6 -6.3
Sanders 8.2 7.8 5.6 3.9 6.4
Warren 3.1 2.1 4.6 4.6 3.6
Buttigieg 9.1 8.2 8.3 8.3 8.5
Klobuchar 5.4 5.0 3.2 2.6 4.1
Yang -2.0 -2.2 -2.6 -2.9 -2.4
Steyer -2.6 -2.9 -2.9 -3.3 -2.9
Gabbard -1.5 -1.6 -1.5 -1.6 -1.6
Bloomberg -0.4 -0.4 -0.6 -0.5 -0.5
Bennet -0.3 -0.3 -0.2 -0.3 -0.3
Patrick 0.0 0.0 0.0 -0.1 0.0
DK/Other -13.7 -10.5 -8.4 -5.1 -9.4

Final tally. Only three candidates improved their vote totals after supporters of non-viable candidates shifted to a viable candidate (15% of attendees at a precinct caucus):

  • Buttigieg (+5,638 supporters; +3.7 points)
  • Warren (+2,238; +1.8)
  • Sanders (+2,155; +1.8)

These three candidates, as well as Klobuchar (-1,288; -0.4), performed better in the final tally than their WAPA, on average. As with the initial tally, WAPA using more recent polls was most predictive for Sanders, Buttigieg and Klobuchar, while WAPA using older polls was most predictive for Warren.

Biden, on the other hand, lost 2,693 supporters and dropped 1.3 points between the initial and final tallies; Yang and Steyer also lost considerable support between the initial and final tallies. For all three candidates, WAPA using earlier polls was most predictive.

Table 5: Arithmetic difference between Iowa Caucuses SDE % and Iowa Caucuses WAPA

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate Mean

Difference

Biden -4.1 -4.0 -4.3 -4.5 -4.2
Sanders 7.7 7.3 5.1 3.4 5.9
Warren 0.9 -0.1 2.4 2.4 1.4
Buttigieg 10.3 9.4 9.5 9.5 9.7
Klobuchar 5.4 5.0 3.2 2.6 4.1
Yang -2.0 -2.2 -2.6 -2.9 -2.4
Steyer -2.5 -2.8 -2.8 -3.2 -2.8
Gabbard -1.5 -1.6 -1.5 -1.6 -1.6
Bloomberg -0.4 -0.4 -0.6 -0.5 -0.5
Bennet -0.3 -0.3 -0.2 -0.3 -0.3
Patrick 0.0 0.0 0.0 -0.1 0.0
DK/Other -13.6 -10.4 -8.3 -5.0 -9.3

SDEs. The same pattern holds for SDEs as for final vote tally, with one minor modification.

  • Buttigieg, Sanders and Klobuchar outperformed their WAPA, with the difference decreasing with more recent polls
  • Warren outperformed her WAPA, with the difference increasing with more recent polls
  • Biden, Steyer and Yang underperformed their WAPA, with the difference increasing with more recent polls.

The bottom line. To evaluate these comparisons globally, I used the sum of the squared differences (“SSE”) between each WAPA value and the results value. Excluding “DK/Other,” Table 6 lists the SSE for each comparison; higher values indicate lower predictive power.

Polling period Initial Tally Final Tally SDEs
All Polls 136.5 240.5 224.9
Since 1st Debate 115.8 210.8 198.2
Since 5th Debate 88.3 190.4 168.0
Since 7th Debate 77.1 177.8 156.1

WAPA was most predictive of the initial tally, not surprising given that poll respondents are asked which candidate they planned to support upon arriving at the caucus site, and not about second or third choices. WAPA was also slightly more predictive of the distribution of SDEs than of the final raw tally of supporters, though neither was especially predictive.

For each reported measure, WAPA was more predictive the closer the polls were to the Caucuses; I will admit this rather surprised me, given the candidate-specific differences detailed above. One explanation is that including older polls, however low-weighted, masks late polling movement of the kind that occurred to Sanders, Buttigieg and Klobuchar.

For now, however, I will continue to report multiple versions of WAPA, if only to see if this pattern holds for later contests.

Now, on to New Hampshire!

Until next time…

The 2020 Democratic Iowa Caucuses: Final Update

[Eds. note: This post was updated at 4 pm EST on February 3, 2020 to reflect on final Iowa Caucuses poll.]

At 7 PM Iowa time (8 pm EST) on February 3, 2020, Iowans will gather in nearly 1,700 precinct-level meeting places to support their preferred candidate to be the 2020 Democratic presidential nominee. They will also participate in a range of party-related business that does not concern us here.

With the exit of former Member of the United States House of Representatives (“Representative” from Maryland John Delaney on January 31, 2020, there are now “only” 11 remaining declared candidates to be the 2020 Democratic presidential nominee. The 17 candidates who have abandoned this quest have done so with grace, class and dignity; I commend them for it.

To learn how I calculate candidate WAPA (weighted-adjusted polling average), please see here; for modifications, please see here.

Here is a photograph of the Iowa Visitor’s Center on I-80 in Rock Island I took on September 5, 1990:

Iowa Visitor Center Sep 1990

**********

As of 4:00 pm EST on February 3, 2020, here is a breakdown of publicly-available Iowa Caucuses polls:

  • 59 since January 1, 2019
  • 46 since the 1st Democratic debate on June 26, 2019
  • 22 since the 5th Democratic debate on November 19, 2019
  • 16 since the 7th Democratic debate on January 14, 2020

Before I present my final pre-Caucuses WAPA, however, here are some words of caution.

1. Caucuses differ in key ways from primaries

In the Iowa Caucuses, voters gather in a public space to publicly declare their support for a candidate. As in, they literally divide into groups of supporters for former Vice President Joe Biden, United States Senator (“Senator”) from Vermont Bernie Sanders, Massachusetts Senator Elizabeth Warren, former South Bend, IN Mayor Pete Buttigieg, Minnesota Senator Amy Klobuchar, and so forth.  A head count is taken; this initial tally is, in effect, what pre-Caucuses polls measure. This tally has never been reported.

However, any candidate whose supporters do not comprise 15%[1] of all voters at a caucus site is deemed not “viable,” and that candidate’s supporters now must choose a new, viable candidate (or, “uncommitted”). Representatives of viable candidates attempt to persuade their friends and neighbors to caucus with them; there are already stories of candidates attempting to form “alliances” in the days leading up to the Caucuses.

Once every candidate is viable, a final tally is taken. This tally has also never been reported. Why? Because the actual purpose of these precinct-level caucuses is to identify delegates to county-level conventions, which then identify delegates to the state convention, which then identifies the 41 delegates to the Democratic nominating convention to be held in Milwaukee, WI from July 13 to July 16, 2020. Thus, what has always been reported are a projection of what percentage of those 41 delegates will be pledged to vote for each candidate at the national convention; these are known as “state delegate equivalents,” or SDE’s.

This year, though the Iowa Democratic Party has announced it will report three values Monday night (or Tuesday morning):

  • The initial statewide tally for each candidate
  • The post-viability tally for each candidate
  • SDE’s for each candidate

It is thus conceivable, if not especially likely in my opinion, there will be three different “winners” of the Iowa Caucuses—or at least, confusion over the order of finish. What is still very likely, is that only three-five candidates will get any sort of boost out of Iowa.

One other way in which caucuses differ from primaries is much lower turnout, which is especially harder to forecast with any accuracy. For example, 171,517 Democrats participated in the 2016 Iowa Caucuses, while 653,669 Iowans voted for Democratic presidential nominee Hillary Clinton that fall; general election data from Dave Leip’s invaluable Atlas of U.S. Presidential Elections. Caucus turnout was thus 26.2% of general election turnout, roughly speaking. That’s year’s New Hampshire Democratic Primary—usually a relatively high-turnout event—had 253,062 participants, fully 72.6% of that state’s vote total for Clinton.

2. Absent poll numbers

There have been fewer polls of the Iowa Caucuses this year than in any election cycle since 2004: 19 in the previous month, compared to an average of 24 during the same period in 2008-16. But one poll’s absence is more glaring than the rest: the gold standard of Iowa Caucuses polls, the Des Moines Register Iowa Poll conducted by Ann Seltzer, which FiveThirtyEight.com rates A+. The Register, in conjunction with CNN, had planned to release its final Iowa Poll at 9 pm EST on February 1, 2020. However, due to the apparent absence of Buttigieg from the list of names read to at least one poll respondent, the poll was cancelled.

This poll would have had a weight of 0.987, edging out a Monmouth College poll conducted January 23-27, 2020 (0.977) and a Siena College/New York Times poll conducted January 20-23, 2020 (0.970), for highest weight overall. My Twitter feed is filled with rumors as to what the results of the poll would have been, from a massive surge by Sanders to a surge by Warren. I would not put much credence into any of these rumors.

Well, except for one thing. Three polling firms, Emerson College (A-), David Binder Research (C+) and Civiqs (C+) have conducted multiple polls of the Iowa Caucuses in the last three weeks. Using the most recent pair of polls, here are average changes ranked from highest to lowest:

  • Buttigieg  +2.0 points
  • Sanders  +1.7
  • Warren  +0.7
  • Bennet, Patrick  no change
  • Gabbard, Yang -0.7
  • Steyer  -1.3
  • Klobuchar -1.7
  • Biden -3.0

Finally, former New York City Mayor Michael Bloomberg has been excluded from eight of the 16 Iowa Caucuses conducted entirely after the 7th Democratic debate; he averages 1.1 percentage points (“points”) in the seven polls which include him.

3. The Des Moines Register endorsement

Iowa’s largest newspaper still commands attention, particularly among undecided or not-fully-committed caucus-goers. According to FiveThirtyEight.com, the last five Democratic candidates endorsed by the Register in competitive races—1988, 2000, 2004, 2008, 2016—saw Iowa Caucuses results a median 6.0 points higher than their polling average at the time of the Register endorsement. That value drops to 4.3 if you include only data from 2000 forward. The average value, meanwhile, was an increase of 8.5 points, but that drops to 5.3 when you exclude the astonishing 21.6-point increase for North Carolina Senator John Edwards in 2004. Still, even the 2.1-point increase for Clinton in 2016 could matter in a very close contest.

On January 25, 2020, the Register endorsed Warren. This was just six days after the New York Times endorsed both Warren and Klobuchar. As you see in Table 1 below, the race in Iowa has been close for months, though there is evidence Biden and, especially, Sanders have pulled slightly ahead. Still, Warren went up in those two recent polls, and another late poll—from Data for Progress (B+/C-), January 28 to February 2, 2020—gives her 19% of the vote. Thus, a minimum 4.3-point increase for Warren is highly plausible. If it comes primarily at the expense of Sanders and Biden, it could make the difference between Warren finishing in a close top two or a more distant fourth…or even fifth, behind Klobuchar.

Table 1: Final Iowa Caucuses WAPA for declared 2020 Democratic presidential nomination candidates

Candidate All Polls Since 1st Debate Since 5th Debate Since 7th Debate
Biden 19.8 19.7 19.9 20.0
Sanders 18.4 18.8 20.9 22.4
Warren 17.1 18.1 15.6 15.5
Buttigieg 16.0 16.9 16.7 16.8
Klobuchar 7.0 7.4 9.1 9.8
Yang 2.9 3.2 3.5 3.7
Steyer 2.8 3.1 3.1 3.5
Gabbard 1.6 1.6 1.6 1.6
Bloomberg 0.4 0.4 0.6 0.5
Bennet 0.3 0.3 0.2 0.3
Patrick 0.0 0.0 0.0 0.1
DK/Other 13.7 10.5 8.7 5.7

The bottom line is this: anybody who thinks they know what will happen in the 2020 Iowa Caucuses has absolutely no idea what will happen in the Iowa Caucuses. The polling, already very difficult to do in a multi-candidate race, is extremely close—with fully four candidates above the 15% viability threshold (and a fifth not far behind), at least statewide; we have almost no sense of caucus-goers’ “backup” choices, or even who will participate; and we are lacking the definitive poll of this race. Perhaps Sanders and Biden really do have a slight edge over Warren and Buttigieg, but I would not bet anything remotely of value on it; in fact, I think Warren’s chances of finishing first or at worst a very close second are understated.

We shall see.

Until next time…

[1] Without rounding, I believe.

January 2020 update: Democratic presidential nomination and general election polling

As of January 14, 2020—and the seventh Democratic presidential nomination debate in Des Moines, Iowa—there are only…

  • 20 days until the Iowa Caucuses,
  • 28 days until the New Hampshire Primary,
  • 39 days until the Nevada Caucuses,
  • 46 days until the South Carolina Primary, and
  • 49 days until 10 states vote on “Super Tuesday,”[1] including California, Texas and my adopted home state of Massachusetts.

Here is an updated assessment of the relative position of the 12 remaining declared candidates. Since the previous update, three candidates exited the race. Former Secretary of Housing and Urban Development Julian Castro dropped out on January 2, 2020 then strongly endorsed United States Senator (“Senator”) Elizabeth Warren of Massachusetts. Eight days later, spiritual advisor Marianne Williamson ended her bid. Finally, New Jersey Senator Cory Booker ended his campaign on January 13. The 16 candidates who have abandoned their quest to be the 2020 Democratic presidential nominee each exited with grace, class and dignity; I commend them for it.

To learn how I calculate candidate NSW-WAPA (national-and-state-weighted weighted-adjusted polling average), please see here;[2] for modifications, please see here.

And, of course, here is the January 2020 lighthouse photograph in my Down East 2020 Maine Lighthouses wall calendar; my wife Nell never forgets come the holidays.

Jan 2020 lighthouse.JPG

**********

Table 1 below aggregates data from all national and state-level polls publicly released since January 1, 2019 (as of midnight EST on January 14, 2020), including:

  • 298 national polls (including 53 weekly Morning Consult tracking polls and 33 weekly YouGov tracking polls)
  • 42 Iowa caucuses polls
  • 43 New Hampshire primary polls
  • 14 Nevada caucuses polls
  • 34 South Carolina primary polls
  • 76 Super Tuesday polls[3]
  • 81 polls from 22 other states.[4]

There are 588 total polls, up from 558 last month. Table 1 now splits post-South-Carolina polling into Super Tuesday and post-Super-Tuesday; the values are broadly similar.

Table 1: National-and-state-weighted WAPA for declared 2020 Democratic presidential nomination candidates

Candidate National IA NH NV SC Super

Tuesday

Post-

SuperTues

NSW-WAPA
Biden 28.3 19.8 20.7 27.0 36.2 26.5 28.1 25.6
Warren 16.2 18.1 16.9 17.7 12.5 17.7 18.3 16.6
Sanders 16.8 16.1 18.0 19.7 11.8 16.3 15.7 16.5
Buttigieg 6.3 15.6 11.8 6.2 4.5 6.2 6.6 9.5
Steyer 0.6 2.2 1.9 4.4 3.9 0.4 0.2 2.6
Klobuchar 1.6 4.4 2.6 1.6 1.0 1.4 1.1 2.4
Yang 2.1 2.2 2.7 3.0 1.4 1.3 1.8 2.2
Gabbard 1.0 1.5 3.2 1.2 0.9 1.0 0.7 1.7
Bloomberg 1.0 0.2 0.4 0.1 0.4 0.3 0.4 0.31
Bennet 0.3 0.3 0.2 0.3 0.2 0.3 0.1 0.27
Delaney 0.3 0.4 0.3 0.00 0.3 0.2 0.2 0.25
Patrick 0.06 0.00 0.07 0.03 0.1 0.2 0.00 0.04
DK/Other 25.4 19.2 21.2 17.1 24.1 28.2 25.4 22.0

Only 30 polls of the 2020 Democratic nomination process have been released since the previous update—14 national (most by Morning Consult or YouGov[5]), four each from Iowa and New Hampshire, two from Nevada, and one each from South Carolina, California, Connecticut, Massachusetts[6], New Mexico and Wisconsin—so it is not surprising there is little change in the overall contours of the race. Former Vice President Joe Biden remains the nominal frontrunner (25.6, unchanged from last month), primarily because of his 23.7-percentage-point (“point”) lead in South Carolina, also unchanged from last month. In Iowa and New Hampshire, however, the two candidates battling for second place—Warren (16.6, down from 17.0) and Vermont Senator Bernie Sanders (16.5, up from 16.0)—are much closer to first place. The fourth leading contender—South Bend, Indiana Mayor Pete Buttigieg (9.5, up from 9.1)—is also very strong in these two early states. These four candidates alone account for more than two-thirds (68.1%) of declared Democratic voter preferences.

This more-inclusive version of NSW-WAPA slightly overstates the gap between Biden and Warren; only examining polls conducted entirely after June 26, 2019, when the first round of Democratic presidential debates ended, Biden drops to 24.6 and Warren rises to 17.5; Sanders is at 16.3 and Buttigieg is at 9.7. And examining only the 64 polls conducted since the fifth Democratic debate on November 20, 2019,[7]  Biden holds steady at 24.5, while Sanders rises to 18.3, Warren drops to 14.3 and Buttigieg jumps to 11.2.

Looking only at Iowa and New Hampshire, meanwhile, shows an effective four-way-tie. Using only the post-first-debate polls in Iowa shows Warren at 19.2, Biden at 18.9, Buttigieg at 16.4 and Sanders at 15.8. Examining only the six most recent polls shows Biden at 19.3, Sanders at 18.0, Buttigieg at 16.9 and Warren at 15.2. Given Iowa’s byzantine caucus rules—at each caucus, participants divide into candidate support groups, reapportioning themselves until every group has at least 15% of participants—it is likely these four candidates will divide between them the vast majority of “votes” counted on February 3.

As for New Hampshire, using only the post-first-debate polls shows Biden at 19.7, Warren at 18.3, Sanders at 17.7 and Buttigieg at 12.2. Examining only the seven most recent polls shows Sanders edging Biden 19.4 to 19.1, with Buttigieg (14.6) and Warren (14.5) trailing a bit behind. Of course, a top two finish in the Iowa Caucuses would very likely boost a candidate’s New Hampshire Primary percentage.

Still, the overall message from the most recent polling, sparse though it is, is that Biden and Sanders have reestablished themselves as the clear Top Two, with Buttigieg and Warren fading slightly. With all that, presidential primary and caucuses polls historically differ from voting results by as much as 10 points, so determining the order of these four candidates, particularly in the first two contests, with any precision is something of a fool’s errand.

In the next tier are four candidates with NSW-WAPA between 1.7 and 2.6 still looking for a chance to rise into the top four: billionaire activist Tom Steyer, Minnesota Senator Amy Klobuchar, entrepreneur Andrew Yang and United States House of Representatives member from Hawaii Tulsi Gabbard.[8] Steyer is now battling Buttigieg for fourth place in Nevada and South Carolina, while Klobuchar is hoping for a surprise in Iowa as the most recent polls boost her support to 6.3. Steyer and Klobuchar will join the Big Four on Tuesday’s debate stage, making it the first Democratic debate with no candidates of color. The top eight candidates total just over three-quarters (77.0%) of declared Democratic voter preferences.

While the remaining four candidates divide 0.9 points between them, none seems close to ending their campaign soon; indeed, former New York City Mayor Michael Bloomberg is essentially betting his campaign on winning the Florida Primary on March 17, assuming no clear front-runner has yet emerged. Perhaps of greater interest are “Don’t Know/Other” values ranging from 17.1 to 28.2. Incorporated in these values are both residual support for former candidates and genuine undecideds; not knowing which of the remaining candidates these voters will ultimately support makes NSW-WAPA values even more “wobbly.”

Returning to the debates, Morning Consult, Quinnipiac and YouGov conducted national polls of the 2020 Democratic presidential nomination both before (but after the November 2019 debate) and after the December 2019 debate. Candidates whose polling average increased by at least 1.0 points are “Don’t Know/Other” (+1.7 points), Klobuchar and Yang (+1.0 each), with decreases of the same magnitude for Buttigieg (-1.0) and Biden (-1.3).

**********

On January 9, 2020, FiveThirtyEight unveiled its first-ever model of a presidential nomination process. They use many more inputs than I do; comparing the two sets of values could be a useful lesson in whether a “simpler” approach—in no way do I consider my NSW-WAPA a model—yields similar results to a more complex one. I strongly encourage you both to contrast our respective methods and to compare my state-level WAPA to their projections.

One instructive comparison, however, is to compare FiveThirtyEight’s nomination probabilities to NSW-WAPA divided by the sum of NSW-WAPA for declared candidates only. The latter value is not, strictly speaking, a probability, though I treat it as such when I assess Democratic chances against President Trump. Table 2 displays these comparisons as of 8:18 pm EST on January 13, 2020.

Table 2: Comparison of FiveThirtyEight nomination probabilities to declaration-weighted NSW-WAPA

Candidate FiveThirtyEight Weighted NSW-WAPA
All polls Post-debate 1 Post-debate 5
Biden 39% 32.9 31.2 29.8
Sanders 22% 21.1 20.7 22.2
Warren 13% 21.3 22.2 17.3
Buttigieg 9% 12.2 12.4 13.6
All others 16% 12.5 13.5 17.1

Both approaches imply it is far more likely than not—on the order of 5:1 or 6:1 in favor—one of Biden, Sanders, Warren and Buttigieg will be the 2020 Democratic nominee for president. And both models see Sanders with a roughly 1-in-5 chance of being the nominee. Where the approaches diverge, however, is on the relative chances of Biden and Warren. The FiveThirtyEight model is much more bullish on Biden’s chances of winning a majority of pledged delegate (what their model actually estimates), while NSW-WAPA is much more bullish on Warren’s chances. NSW-WAPA is also slightly more bullish on Buttigieg’s odds.

**********

Turning to polls of hypothetical 2020 matchups between proposed Democratic presidential nominees, Biden would beat Trump nationally by 7.6 points, Sanders by 4.6 points and Warren by 2.8 points—each value about 0.3 points lower than last month, while Buttigieg would lose by 0.3 points; Bloomberg, based on 12 polls—11 released in the last three months, would win by 1.3 points. The other six candidates for whom I have matchup data would lose by between 2.9 (Klobuchar) and 8.9 (Gabbard) points, although these numbers are misleading, as they are primarily based upon data from pollster Harris X, who tend not to push undecided voters to choose, making for unusual polling margins.

Weighted by a rough estimate of the likelihood of winning the nomination (NSW-WAPA/.778), the 2020 Democratic nominee would beat Trump by an average 3.2 points, just over the median Democratic presidential margin (+3.0) in the previous six presidential elections, which include three elections with an incumbent seeking reelection and three elections with no incumbent. Excluding Biden and Sanders, however, decreases the margin to -0.3 points, with the caveat from the preceding paragraph.

However, it is the Electoral College which decides who wins presidential elections. Examining polling averages from the 28 states[9] for which I have at least one hypothetical match-up poll and comparing them to my partisan-lean measure 3W-RDMyields a median national popular vote win by the Democrat by between 2.9 (excluding Biden and Sanders) and 4.8 points.[10] I will address state-level returns in much more detail in an upcoming post.

Until next time…

[1] Alabama, California, Colorado, Massachusetts, Minnesota, North Carolina, Oklahoma, Tennessee, Texas, Virginia

[2] Essentially, polls are weighted within nation/state by days to  nominating contest and pollster quality to form an area-specific average, then a weighted average is taken across Iowa (weight=5), New Hampshire (5), Nevada (4), South Carolina (4), time-weighted average of subsequent contests (2) and nationwide (1). Within subsequent contests, I weight the “Super Tuesday” states twice subsequent contests. As of this writing, I have at least one poll from, in chronological order, Maine, Michigan, Mississippi, Missouri, Ohio, Washington, Arizona, Florida, Illinois, Georgia, Wyoming, Wisconsin, Connecticut, Delaware, Maryland, New York, Pennsylvania, Indiana, Oregon, Montana, New Jersey and New Mexico.

[3] Primarily California (33), Texas (19) and North Carolina (8)

[4] Primarily Wisconsin (14), Florida (12), Pennsylvania (9) and Michigan (8)—not coincidentally, the four states President Donald J. Trump won in 2016 by the narrowest margins, making them among the most-polled overall.

[5] Including a few YouGov polls released prior to December 20, 2019 only just posted by FiveThirtyEight

[6] Including one conducted by Evan Falchuk and Lou DiNatale between October 23 and 25, 2019

[7] Distribution: National (34), Iowa (6), New Hampshire (7), Nevada (2), South Carolina (3), California (6), Wisconsin (2) and Texas, Illinois, Connecticut and New Mexico one each

[8] Examining only the most recent polling: Steyer 4.3, Klobuchar 3.3, Yang 2.7, Gabbard 2.2…and Bloomberg 1.0. All other candidates are between 0.15 and 0.18.

[9] Pennsylvania, New Hampshire, Wisconsin, Michigan, North Carolina, Texas, Iowa, Arizona, South Carolina, Minnesota, Nevada, Massachusetts, Florida, New York, Kentucky, Maine, Ohio, North Dakota, California, Alaska, Washington, Colorado, Missouri, Utah, Virginia, Montana, Connecticut, Georgia, New Mexico

[10] The mean—heavily skewed by some extremely Democratic-leaning polling in Utah—ranges between 3.5 and 6.2.

December 2019 update: Democratic presidential nomination and general election polling

With the sixth Democratic presidential nomination debate set for December 19, 2019 in Los Angeles, California, here is an updated assessment of the relative position of the now-15 declared candidates. Since the previous update, four candidates exited the race: Miramar, Florida Mayor Wayne Messam on November 20, former United States House of Representatives Member (“Representative”) Joe Sestak of Pennsylvania on December 1, Montana Governor Steve Bullock on December 2, and, rather surprisingly given that she had qualified for the sixth debate, United States Senator (“Senator”) from California Kamala Harris on December 3. The 13 candidates who have abandoned their quest to be the 2020 Democratic presidential nominee each exited with grace, class and dignity; I commend them for it.

To learn how I calculate the value I assign to each candidate, NSW-WAPA (national-and-state-weighted weighted-adjusted polling average), please see here;[1] for recent modifications, please see here.

And, of course, here is the December 2019 lighthouse photograph in my Down East 2019 Maine Lighthouses wall calendar.

Dec 2019 lighthouse.JPG

**********

Table 1 below aggregates data from all national and state-level polls publicly released since January 1, 2019(as of 1:00 am EST on December 19, 2019), including:

  • 284 national polls (including 50 weekly Morning Consult tracking polls and 30 weekly YouGov tracking polls)
  • 38 Iowa caucuses polls
  • 39 New Hampshire primary polls
  • 12 Nevada caucuses polls
  • 33 South Carolina primary polls
  • 73 Super Tuesday polls[2]
  • 78 polls from 20 other states.[3]

There are now 558 total polls, up from 488 last month.

Table 1: National-and-state-weighted WAPA for declared 2020 Democratic presidential nomination candidates

Candidate National IA NH NV SC Post-SC NSW-WAPA
Biden 28.1 19.9 20.8 27.3 36.3 27.2 25.7
Warren 16.0 18.3 17.6 18.8 12.6 18.2 17.0
Sanders 16.5 15.5 17.5 19.0 11.7 16.2 16.0
Buttigieg 6.2 15.1 11.1 6.1 4.5 6.4 9.1
Steyer 0.5 2.1 1.5 3.1 3.1 0.3 2.1
Yang 2.0 2.0 2.6 2.7 1.4 1.4 2.1
Klobuchar 1.5 4.0 2.0 1.3 0.9 1.3 2.1
Booker 2.1 2.1 1.7 1.5 2.8 1.5 2.0
Gabbard 1.0 1.4 2.9 1.1 0.9 0.9 1.6
Castro 0.9 0.5 0.2 1.0 0.2 1.1 0.55
Williamson 0.3 0.1 0.4 0.3 0.5 0.2 0.30
Delaney 0.3 0.5 0.4 0.00 0.3 0.2 0.28
Bennet 0.3 0.3 0.2 0.4 0.2 0.3 0.28
Bloomberg 0.7 0.2 0.2 n/a 0.3 0.3 0.21
Patrick 0.05 0.00 0.05 n/a 0.1 0.1 0.04
DK/Other 23.6 18.0 20.8 17.4 24.2 24.4 20.6

The race continues to follow the same pattern. Former Vice President Joe Biden remains the nominal frontrunner (25.7, down from 26.2), primarily because of his 23.7-percentage-point (“point”) lead in South Carolina, essentially unchanged from 24.0 last month. However, he is less strong in Iowa and New Hampshire, where the two candidates battling for second place—Massachusetts United States Senator (“Senator”) Elizabeth Warren (17.0, down from 17.3) and Vermont Senator Bernie Sanders (16.0, up from 15.8)—are much closer to first place. And this more-inclusive version of NSW-WAPA overstates the gap between Biden and Warren; only examining polls conducted entirely after June 26, 2019, when the first round of Democratic presidential debates ended, Biden drops to 24.6 and Warren rises to 18.2; Sanders is at 15.8. Rounding out the Big Four, overall and in the four earliest states, is South Bend, Indiana Mayor Pete Buttigieg (9.1—up from 8.1, and 7.1 the month before). These four candidates account for more than two-thirds (68.0%) of declared Democratic voter preferences.

Looking only at Iowa and New Hampshire, meanwhile, shows an even tighter race, especially when only post-first-debate polls are considered. Using these more recent polls, Iowa is effectively a four-way tie, with Warren at 19.7, Biden at 18.9, Buttigieg at 16.0 and Sanders at 15.0. There is a similar scrum in New Hampshire, with Biden at 19.5, Warren at 19.4, Sanders at 17.0 and Buttigieg at 11.4.

In the next tier are five candidates with NSW-WAPA between 1.6 and 2.1 who are running out of chances to rise into the top four: billionaire activist Tom Steyer, entrepreneur Andrew Yang, Minnesota Senator Amy Klobuchar and New Jersey Senator Cory Booker—essentially tied for 5th place—followed by Hawaii Representative Tulsi Gabbard. Other than Booker, these candidates rose in the last month, particularly in the early contests. However, only Steyer, Yang and Klobuchar will be joining the Big Four on Thursday’s debate stage, despite protests by Booker and former Secretary of Housing and Urban Development Julián Castro, who remains mired around 0.6. The top nine candidates total more than three-quarters (77.7%) of declared Democratic voter preferences.

While Castro and the remaining six candidates divide just 1.7 between them, nobody else seems close to ending their campaign soon. Indeed, Castro missed the December 9 deadline to run in 2020 against Texas Senator John Cornyn. Meanwhile, the upsurge in NSW-WAPA for “Don’t Know/Other” reflects lingering support for Harris.

Returning to the debates, seven pollsters—six nationally[4] and one in South Carolina[5]–conducted polls of the 2020 Democratic presidential nomination both before (but after the October 2019 debate) and after the November 2019 debate. Simple average differences in polling percentage (South Carolina poll results weighted four times national results) show measurable gains for Buttigieg and Steyer (+1.7 points each), Gabbard (+0.9) and Booker (+0.6), as well as measurable declines for Biden (-2.9) and, especially, Warren (-4.1). Warren’s decline was even sharper after weighting averages by pollster quality (-5.9) and number of days between polls (-7.9): her decline was steepest in higher-quality polls conducted farther apart in time. This decline is reflected in the decline in Warren’s NSW-WAPA from 19.1 last month to 18.2 now using only polls conducted after the first Democratic candidate debate. By contrast, “Don’t Know/Other,” Castro, Sanders, Booker and Klobuchar saw higher increases in support, albeit still small, with higher-quality polls conducted farther apart in time.

**********

On December 13, 2019, FiveThirtyEight unveiled its own 2020 Democratic presidential nomination polling aggregations. We use different aggregation methods, so while values we estimate for each candidate are broadly similar, there are clear differences, most notably in the relative standings of Buttigieg and Warren in Iowa and New Hampshire. I encourage you to compare our methods and results, especially since I base much of my own methods on their example.

Here are four key areas where our aggregation methods differ.

First and foremost, I combine state and national polling averages into a single overall value: NSW-WAPA; FiveThirtyEight does not. Similarly, when a state has not been polled for a while, FiveThirtyEight adjusts each candidate’s standing in that state by the change in their national standing, or “secular trend.” I do not do this. The first four states are polled often enough to make such adjustment unnecessary there, and I use time-and-quality weighted averages of a) the 10 Super Tuesday states and b) all subsequent states, mitigating the need for secular trend adjustment. Moreover, while I think such adjustment is necessary for general election polling, particularly the closer the election is, I am skeptical state-level primary polling moves in tandem with national polling, if only because there is no “national presidential nomination primary” to assess.

That secular trend is apparent in how much worse Warren is faring in Iowa and New Hampshire in FiveThirtyEight’s models compared to mine, likely due to her polling decline since the last debate, as well as the corresponding stronger position of Buttigieg in those states.

Second, the way I weight elapsed time since a poll was conducted compared to the way FiveThirtyEight does means NSW-WAPA is slower to detect sustained movement in a candidate’s standing. Still, it caught the major movements: the decline in support for Biden and former Texas Representative Beto O’Rourke since the start of the debates, the rise of Harris after the first debate and her subsequent decline, and the slow steady ascent of first Warren (with a recent decline) then Buttigieg.

Third, FiveThirtyEight adjusts for the tendency of some pollsters to release more favorable results for some candidates, on average, relative to others. I do not, simply because I had not performed or seen the calculations. Also, adjusting for this mathematical “bias” in my general election polling aggregates rarely alters final aggregates very much. However, if I can obtain the bias estimates FiveThirtyEight uses, I may consider using them.

Finally, FiveThirtyEight assigns a higher weight to polls with a larger sample size; I do not weight by sample size at all. For one thing, this would give inordinate influence to Morning Consult tracking polls, whose average sample size in 2019 has been 14,562—while their pollster rating is B/C (which I code 2.5/4.3 = .581). There is in fact a slight inverse relationship between average sample size and pollster quality: r=-0.20 for 24 polling agencies who have conducted two or more national Democratic nomination contest polls in 2019.[6] Moreover—and here my epidemiological training comes into play—increasing sample size makes an estimate more precise (i.e., smaller margin of error) but does not reduce any systematic bias; the latter is reduced through the sort of adjustment discussed in the previous paragraph. Simply put, I see no valid methodological reason for giving more weight to polls with higher sample sizes.

**********

There has been a slight pro-Trump trend in recent polls of hypothetical matchups between potential 2020 Democratic presidential nominees and President Trump I may address in a later post. For now, though, here are the averages: Biden would beat Trump nationally by 7.6 points, Warren by 3.1 points and Sanders by 4.9 points, while Buttigieg would lose by 0.3 points and Booker by 0.7 points[7]; Bloomberg, based on eight polls—seven released in the last two months, would win by 1.2 points. The other eight candidates for whom I have matchup data would lose by between 2.9 (Klobuchar) and 9.3 (spiritual advisor Marianne Williamson) points, although these numbers are misleading, as they are primarily based upon data from pollster Harris X, who tend not to push undecided voters to choose, making for unusual polling margins.

Weighted by a rough estimate of the likelihood of winning the nomination (NSW-WAPA/.794), the 2020 Democratic nominee would beat Trump by 3.4 points, broadly in line with the median Democratic presidential margin (+3.0) in the previous six presidential elections, which include three elections with an incumbent seeking reelection and three elections with no incumbent. Excluding Biden and Sanders, however, decreases the margin to -0.2 points, with the caveat from the preceding paragraph.

Comparing available state-level results,[8] which decide presidential elections via the Electoral College, to my partisan-lean measure 3W-RDM implies Democrats would win the national popular vote by between 3.7 (excluding Biden and Sanders) and 5.8 points. Most encouraging to Democrats should be polls from North Carolina (R+6.0), Georgia (R+9.6), Arizona (R+9.7) and Texas (R+15.3), which show Democrats either even (Georgia) or within five points of Trump; on average, they imply a national Democratic lead of 8-9 points, confirming strong opportunities for Democrats in the southeast and southwest.

By contrast, polls from Democratic-leaning Nevada (D+2.0) show Democrats anywhere from 0.7 points ahead to 2.1 points behind, implying Democrats would lose nationwide by 1.3-4.1 points. And while Democrats are 2.1-5.5 points ahead in the swing state of Michigan, which Trump won by 0.16 points in 2016, their position is…wobbly…in Florida (R+3.4), Pennsylvania (R+0.4) and Wisconsin (D+0.7), all of which Trump won narrowly in 2016.

On balance, however, Democrats remain extremely competitive in a wide range of swing and Republican-leaning states against an incumbent president, with the most likely nominees—Biden, Warren, Sanders, Buttigieg—beating him by a combined 4.8 points.

Until next time…

[1] Essentially, polls are weighted within nation/state by days to  nominating contest and pollster quality to form an area-specific average, then a weighted average is taken across Iowa (weight=5), New Hampshire (5), Nevada (4), South Carolina (4), time-weighted average of subsequent contests (2) and nationwide (1). Within subsequent contests, I weight the 10 March 3, 2020 “Super Tuesday” states (Alabama, California, Colorado, Massachusetts, Minnesota, North Carolina, Oklahoma, Tennessee, Texas, Virginia) twice subsequent contests. As of this writing, I have at least one poll from (in chronological order) Maine, Michigan, Mississippi, Missouri, Ohio, Washington, Arizona, Florida, Illinois, Georgia, Wyoming, Wisconsin, Delaware, Maryland, New York, Pennsylvania, Indiana, Oregon, Montana and New Jersey.

[2] Primarily California (31), Texas (19) and North Carolina (8)

[3] Primarily Wisconsin (13), Florida (12), Pennsylvania (9) and Michigan (8)—not coincidentally, the four states President Donald J. Trump won in 2016 by the narrowest margins.

[4] Morning Consult Tracking, Monmouth University, Fox News, YouGov, IBD/TIPP, CNN/SSRS, Quinnipiac University

[5] YouGov

[6] The correlation increases to -0.43 if you exclude Morning Consult.

[7] Although this matchup has not been assessed since late September.

[8] From 27 states: Pennsylvania, New Hampshire, Wisconsin, Michigan, North Carolina, Texas, Iowa, Arizona, South Carolina, Minnesota, Nevada, Massachusetts, Florida, New York, Kentucky, Maine, Ohio, North Dakota, California, Alaska, Washington, Colorado, Missouri, Utah, Virginia, Montana, Connecticut, Georgia.

Emerson College polls: Post, and ye shall receive

Five days ago, I wrote about the sharp Republican shift since September 1, 2019 in polls conducted by Emerson College of hypothetical 2020 presidential election matchups between President Donald J. Trump and possible Democratic nominees. Emerson College is a high-quality pollster, rated A- and showing no partisan skew in recent elections according to FiveThirtyEight.com’s pollster ratings. What distinguishes Emerson College from other similarly high-quality pollsters assessing these matchups, though, is that the results they release have 0% choosing undecided or an unnamed candidate; this is likely, at least in part, because approximately 70% of Emerson College polls are conducted using automated voice response (aka “robopolling”).

I Voted sticker

When I first wrote in June 2019 about Emerson College polls of these hypothetical matchups, it was to compare their results with those of HarrisX, a C+ pollster with a strong Republican bias of 1.3 percentage points (“points”). I calculated then that “[HarrisX”] polls have an average of 28.0% undecided between the named Democrat and Trump (or would choose a third-party candidate); I estimate these voters would break roughly 7-4 in favor of the Democratic nominee.” That rises to 2-1 if polls assessing former Vice President Joe Biden or United States Senator (“Senator”) from Vermont Bernie Sanders are excluded.

However, as though in response to my November 25, 2019 post, Emerson College just released results of hypothetical 2020 presidential election matchups in New Hampshire. They assessed five possible Democratic nominees: Biden, Massachusetts Senator Elizabeth Warren, Sanders, South Bend, IN Mayor Pete Buttigieg, and entrepreneur Andrew Yang; they surveyed 637 registered voters between November 22 and November 26,  2019.

Actually, the FiveThirtyEight.com poll tracker lists two sets of results: one with 0% “undecided/other” and one with between nine and 12% “undecided/other” (mean=10.8%); note that the former results are based upon 547 voters, 90 fewer than the latter results. We can thus directly assess how those roughly 1 in 9 voters would vote if “forced” to choose; results may be found in Table 1.

Table 1: Comparing Emerson College Polls of Hypothetical Match-ups Between President Trump and Specific 2020 Democratic Presidential Nominees in New Hampshire

Democrat With Undec/Other Without Undec/Other %Undec/Other %Trump by Undec/Other
Biden D+4 D+5 12% 45.8%
Warren D+4 D-2 10% 80.0%
Sanders D+7 D+3 9% 72.2%
Buttigieg D+8 D+6 11% 59.1%
Yang D+5 D-2 12% 79.2%
AVERAGE D+5.6 D+2.0 10.8% 66.7%

Poll results with a non-zero percentage choosing “undecided/other” show the hypothetical 2020 Democratic presidential nominee beating Trump in New Hampshire by an average of 5.6 points, slightly higher than the 4.6 points I calculate using all polls released since January 1, 2019. Moreover, all five Democrats lead Trump by between four and eight points. Once “undecided/other” voters are forced to choose, however, that average drops sharply to 2.0 points; Warren and Yang now trail Trump. That is because, on average, “undecided/other” voters choose Trump 2-1; this masks a curious divide, though: when offered Biden or Buttigieg, the split is nearly even, but when offered Warren, Sanders or Yang, the split is closer to 3-1.

While not strictly an apples-to-apples comparison, “undecided/other” voters breaking 2-1 for Trump is an exact reversal of my June analysis, when analogous voters broke nearly 2-1 against Trump. This reversal feels…implausible…if only because for years, the general rule of thumb was that these voters would ultimately break toward the non-incumbent, perhaps by as much as 2-1, as the incumbent was such a well-known quantity that opinions were basically fixed: a voter already knew if they approved of her/him or not. Still, that “rule” no longer seems to apply, so the best thing to do is to continue to aggregate all polls as best we can.

At the same time, the Emerson College results, when applied to other recent polls of hypothetical 2020 presidential election match-ups, are…odd. For example, consider this comparison of national SurveyUSA (A rating, D+0.1; 3,850 registered voters [RV]. November 20-21, 2019) and Emerson College polls (1,092 RV, November 17-20, 2019):

Table 2: Comparing National November 2019 SurveyUSA and Emerson College Polls of Hypothetical Match-ups Between President Trump and Specific 2020 Democratic Presidential Nominee

Democrat SurveyUSA Emerson College SurveyUSA

%Undec/Other

%Trump by Undec/Other
 

Biden

52%

D+13

49.5%

D-1

9% 155.6%
 

Warren

49%

D+7

50%

D+0

9% 88.9%
 

Sanders

52%

D+12

50.5%

D+1

8% 137.5%
 

Buttigieg

48%

D+7

48%

D-4

11% 100.0%
 

AVERAGE

50.2%

D+9.8

49.8%

D-0.5

9.2% 120.5%

In the four SurveyUSA polls, the hypothetical 2020 Democratic presidential nominee tops 50% on average, with a nearly-10-point lead over Trump and only about 1 in 11 choosing “undecided/other.” By contrast, the four Emerson College polls—with every respondent forced to choose either Trump or his Democratic opponent—show a dead-even race, with Trump edging the Democrat by 0.5 points, on average. But for these two sets of polls to exist simultaneously, not only would every single (or, in the case of Warren, 89%) “undecided/other” voter have to break for Trump, but a small proportion of Democratic-leaning voters would also have to switch to Trump as well.

Which, again, seems…implausible.

As always, caveat emptor.

Until next time…

About those recent Emerson College polls…

I first wrote about Emerson College polls here, using the fact their polls of hypothetical 2020 matchups between a Democrat and President Donald J. Trump force respondents to choose a candidate (i.e., have 0% “other/undecided”) to assess Harris X polls, which often have very high proportions “other/undecided.” At the time, I concluded “other/undecided” Harris X polls respondents likely would vote for the 2020 Democrat presidential nominee roughly 2-1.

When I wrote that post in June, margins reported by Emerson College polls were broadly in line with those reported by other, non-Harris-X polls of these hypothetical matchups. Since the end of August, however, they have taken a sharply Republican turn compared both to previous Emerson College polls and to all other polls, based upon analyses of my WAPA (weighted-adjusted polling average).

Table 1 clearly demonstrates this pro-Republican shift.

Table 1: Polling margin for specified 2020 Democratic presidential nominees over President Trump, Emerson College vs. All Other Polls and January-August 2019 vs. September-November 2019

Biden–National
Emerson All Other
Jan-Aug 7.24 8.35
Sep-Nov 0.29 8.03 Difference
-6.95 -0.32 -6.63
Warren–National
Emerson All Other
Jan-Aug 1.52 1.82
Sep-Nov 1.29 4.66 Difference
-0.23 2.84 -3.07
Sanders–National
Emerson All Other
Jan-Aug 4.28 4.99
Sep-Nov 0.12 5.10 Difference
-4.16 0.11 -4.27

While the post-August-2019 Republican lean of Emerson College polls for the three current polling leaders for the 2020 Democratic presidential nomination—former Vice President Joe Biden, Massachusetts United States Senator (“Senator”) Elizabeth Warren and Vermont Senator Bernie Sanders—is apparent, it is most notable for Biden. Thus, while in all other polls, Biden’s WAPA declined modestly from 8.35 to 8.03, in the Emerson College polls it dropped from 7.24 to essentially even! Put another way, there was a 6.63 shift Republican in Emerson College polls for Biden starting around September 1, 2019. The shift was similar for Sanders—4.16 more Republican in Emerson College polls and no meaningful shift in other polls. Warren, meanwhile, had a sharp increase in her WAPA—from 1.82 to 4.66—in non-Emerson-College polls after September 1, 2019, while dropping slightly in the Emerson College polls. Overall, across all three candidates, there was a 4.66 pro-Republican shift in Emerson College polls after September 1, 2019.[1]

I Voted sticker

**********

Before addressing what could have caused this shift, let us address what did NOT cause this shift. On November 5, 2019, FiveThirtyEight.com updated its pollster ratings, which I use in calculating WAPA. Emerson College was upgraded from B+ to A-, making it one of the highest-quality pollsters regularly assessing the 2020 presidential election. Moreover, Emerson College’s “mean-reverted bias”—how much more or less Democratic its average polling margins are compared to other pollsters in the same (already-concluded) election—barely changed, shifting from D+0.1 to R+0.0 (which I code as D-0.025). Finally, mean sample size of Emerson College polls—still of registered voters—dropped only slightly after September 1, from 1,120 to 1,043.

But what about other pollsters? If anything, the mix of pollsters assessing the 2020 presidential election improved after September 1, 2019, from B-/B to B, albeit with a slightly more Democratic skew (D+0.2 to D+0.3).

Moreover, here are the averages of unskewed pollster-average margins for Biden, Warren and Sanders across all pollsters with at least a B+ rating,[2] before and after September 1, 2019:

  • Biden: 9.3 percentage points (“points”) to 10.8 points
  • Warren: 2.4 points to 6.6 points
  • Sanders: 5.8 points to 7.5 points

Clearly, both the more recent Emerson College polls and the lower-rated pollsters are finding much closer races between each of these three candidates and Trump than are the higher-rated pollsters. And while I removed the skew from these margins, the higher-rated pollsters have a mean skew of D+0.5—so take these averages with a modicum of salt.

Nonetheless, polls conducted by pollsters similar in quality to Emerson College—albeit with some small percentage of “undecided/other” voters—show the three leading Democratic candidates increasing their hypothesized margins against Trump by 1.5 to 4.2 points after September 1, 2019, while the Emerson College polls show declines in support from 0.2 to 7.0 points.

So what gives?

There are two broad categories of possible, non-mutually-exclusive explanations.

  1. Starting September 1, Emerson College pollsters adjusted how they weigh their samples by various demographic factors, such that their polls skew sharply more Republican relative to other high-quality pollsters.
  2. Emerson College pollsters, perhaps because they force respondents to make a choice, are capturing a genuine pro-Republican shift in the electorate other high-quality pollsters are missing.

Taking each possible explanation in turn…

Emerson’s sampling methodology did not change after September 1, remaining a combination of Interactive Voice Response (i.e., “robocalls”) of landlines—but not cellphones—and an online voter panel. However, their sample weighting, based upon 2016 turnout, did change subtly—from age, region, income, and education in their August 2019 national polls to age, [interview] mode, party registration, ethnicity and region in their November 2019 national polls.

Given how strongly education is now associated with partisanship, especially among white voters, that could account for at least some of the difference. I am perplexed, however, how interviewing mode is associated with turnout—other than landline-users tending to lean Republican (older, whiter, less urban). For that matter, I am not clear why a poll of registered voters would adjust for turnout at all. Adjust for the relative proportions of these groups in the universe of registered voters, sure—but adjustment for participation rates (i.e., turnout) of various groups in 2016 seems more appropriate for a likely voters model. Still, without seeing the raw data, I will not even speculate how these changes in sample weighting would affect publicly-released polling margins.

There is, meanwhile, an argument to be made that the universe of “decided” voters has drifted Republican in recent months, especially with the announcement by Speaker of the United States House of Representatives (“House”) Nancy Pelosi on September 24, 2019 that the House would formally begin an impeachment inquiry into Trump. While polling at first showed approval for this action higher than disapproval, the difference is closer to even now as House Republicans have rallied behind a president of their party.

Also, while Democrats fared well in recent governor’s races in Kentucky, Louisiana and Mississippi, the final actual margins were an average 2.8 points more Republicans than the final WAPA, perhaps reflecting a substantial Republican bias among “other/undecided” voters—essentially what the Emerson College polls might show. Moreover, the A+-rated Siena College/New York Times Upshot (R+0.3), recently released a set of state-level polls that also show leading Democrats faring less well against Trump than the consensus of other pollsters in those states.

While some combination of these two possible explanations—a change in sample weighting and an actual pro-Republican shift in the electorate—probably accounts for the clear shift towards Trump in recent Emerson College polls, it is entirely possible their last three national polls are showing these shifts purely by chance; even the best sampling strategy will be well wide of the mark at times (this is, in fact, the basic logic behind polling aggregation). However, three such wide sampling “misses” in the same direction and of the same size are extremely unlikely, though far from impossible.

In the end, the best those of us who track election polling can do is throw every publicly-available poll into the analysis, weighting, averaging and adjusting as best we can—all the while remembering that even the best estimates are just that, estimates.

Until next time…

[1] Values were similar for South Bend, Indiana Mayor Pete Buttigieg and California Senator Kamala Harris, but with each having only one post-August-2019 Emerson College poll assessing support versus President Trump, I excluded their polls from the analysis.

[2] IBD/TIPP, CNN/SSRS, Fox News, NBC News/Wall Street Journal, ABC News/Washington Post, SurveyUSA, Quinnipiac University. There is no post-August-2019 NBCNWSJ polling for Sanders.