A wicked early look at 2020 Senate and gubernatorial races

In recent posts, I began to take a wicked early look at the 2020 U.S. elections. First, I assessed the field of Democrats seeking to challenge Republican President Donald Trump in 2020. Then I turned to the 2020 presidential election itself, pondering how Democrats would potentially fare against Trump.

Now I turn my attention to

  1. The 34 elections for United States Senate (“Senate”) to be held in 2020.
  2. The three gubernatorial elections to be held 2019 (Kentucky, Louisiana, Mississippi) and the 11 gubernatorial elections to be held in 2020.

My goal is primarily to provide the view from 30,000 feet: what the “fundamentals” in each race reveal about the overall partisan landscape—and what the likelihood is Democrats will have the Senate majority in January 2021 (and cut into the Republican advantage in governor’s mansions) As such, I only briefly discuss actual or potential candidates in these races, other than incumbents seeking reelection.

“Fundamentals” are simply the sum of three values:

  1. The state’s partisan lean, measured by my 3W-RDM (weighted[1] three-election average of the difference between a state’s Democratic [minus Republican] margin in a presidential election and the Democratic [minus Republican] margin in the total national vote in that election).
  2. The estimated effect of incumbency (incumbent office-holders tend to receive a higher percentage of the vote than an open-seat candidate of the same party).
  3. The national partisan lean, as measure by the “generic ballot” question (variations on “If the election for were held today, would you vote for the Democratic candidate, the Republican candidate, or some other candidate?”)

**********

Just bear with me as I explain how I estimated the effect of incumbency for Senate and gubernatorial elections. As usual, unless otherwise noted, election data come from Dave Liep’s indispensable Atlas of U.S. Presidential Elections.

Senate. I first calculated an “expected margin of victory”[2] for each Democratic Senate nominee in the 35 Senate elections in 2018[3], the 34 Senate elections in 2016[4] and the 35 Senate elections in 2014[5]: a state’s 3W-RDM plus the national Democratic margin (minus Republican percentage of all votes cast) in that year’s elections. Using three elections years guarantees a minimum of two Senate elections from each state. The margins for the three previous Senate election years are:

2014 = D-5.8%

2016 = D+0.9%

2018 = D+9.9%

Next, I subtracted each actual margin (Democratic minus Republican) from the “expected” margin. I then calculated three averages of these differences within each election year:

  1. Races with Democratic incumbents
  2. Races with Republican incumbents
  3. Open-seat races (where expected margin is for party currently holding the office)

Within each election year, then, the effect of incumbency for Democrats is simply the first average minus the third average[6], while the Republican advantage is the second average minus the third average[7]. And the estimated effect of incumbency for each party is the weighted average (2018=3, 2016=2, 2014=1) of the election-year averages.

For Democratic Senate incumbents, the effect is +4.4 percentage points (“points”), and for Republican Senate incumbents the effect is +2.6 points. Somewhat arbitrarily, I divide these values by 1.5 for incumbents who have won a special election, but not yet served a full six-year term and by 2.0 for incumbents who were appointed to the seat and have yet to face the voters.

Governor. Complicating these calculations is that five states hold their gubernatorial elections in odd-numbered years; thus, in November 2019, Kentucky, Louisiana and Mississippi will elect a governor, as will New Jersey and Virginia in November 2021.

As a result, I analyzed data from two-year cycles: 38 gubernatorial elections in each of 2017-18[8] and 2013-14[9], and 15 gubernatorial elections in each of 2015-16[10] and 2011-12[11]; going back to 2011 guarantees at least two gubernatorial elections from each state (with New Hampshire and Vermont, which hold gubernatorial elections every two years, included four times[12]). The calculations were otherwise the same, except for calculating a four-cycle weighted average (4,3,2,1)[13]: for Democratic gubernatorial incumbents, the effect is +5.7 points, and for Republican gubernatorial incumbents the effect is +8.5 points.

That the effect of incumbency is stronger for governors than for Senators reflects how partisan Senate elections have become.

**********

Let us now turn to the elections themselves. I base the “national lean” of D+6.0 on generic ballot polls listed on FiveThirtyEight.com, which have varied between D+2 and D+9—and mostly between D+5 and D+7—over the last few weeks. While this value is broadly in line with the last four Senate election years (weighted average=D+4.3 points; unweighted average from last two presidential election years= D+6.5), it is much higher than the last four gubernatorial election cycles (weighted average=D-0.6 points; unweighted average from last two presidential election years= D-3.2).

2020 Senate elections. Republicans currently hold 53 Senate seats, with 47 held by Democrats (including Independent Senators Angus King of Maine and Bernie Sanders of Vermont, who caucus with Democrats), meaning that to win back the majority in 2020, Democrats need either to win a net four seats, or win a net three seats and win the presidential election (Democratic Vice President would break 50-50 tie).

Table 1. 2020 Senate election overview

Name State Run

2020

3W-RDM INC Nat

Lean

Total Last margin First elected
DEMOCRATS
Edward Markey MA Yes 22.1 4.4 6.0 32.5 23.9% 2013
Jack Reed RI Yes 18.0 4.4 6.0 28.4 41.3% 1996
Richard Durbin IL Yes 14.7 4.4 6.0 25.1 14.6% 1996
Chris Coons DE Yes 12.5 4.4 6.0 22.9 13.6% 2010
Cory Booker NJ Yes 12.0 4.4 6.0 22.4 13.5% 2012
Jeff Merkley OR Yes 8.7 4.4 6.0 19.1 18.9% 2008
Tom Udall NM No 6.5 0.0 6.0 12.5 N/A
Gary Peters MI Yes 2.2 4.4 6.0 12.6 13.3% 2014
Mark Warner VA Yes 1.5 4.4 6.0 11.9 0.8% 2008
Tina Smith MN Yes 1.5 2.9 6.0 10.4 10.6% 2018
Jeanne Shaheen NH Yes 0.1 4.2 6.0 10.3 3.3% 2008
Doug Jones AL Yes -28.4 2.2 6.0 -20.2 1.7% 2017
 
REPUBLICANS
Susan Collins ME Yes 5.9 -2.4 6.0 9.5 37.0% 1996
Cory Gardner CO Yes 2.2 -2.4 6.0 5.8 1.9% 2014
Joni Ernst IA Yes -4.7 -2.4 6.0 -1.1 8.3% 2014
Thom Tillis NC Yes -6.0 -2.4 6.0 -2.4 1.6% 2014
David Perdue GA Yes -9.6 -2.4 6.0 -6.0 7.7% 2014
Martha McSally AZ Yes -9.7 -1.2 6.0 -4.9 Apptd 2019
John Cornyn TX Yes -15.3 -2.4 6.0 -11.7 27.2% 2002
Lindsey Graham SC Yes -15.7 -2.4 6.0 -12.1 15.5% 2002
Cindy Hyde-Smith MS Yes -18.5 -1.6 6.0 -14.1 7.3% 2018
Steve Daines MT Yes -18.6 -2.4 6.0 -15.0 17.7% 2014
Dan Sullivan AK Yes -19.2 -2.4 6.0 -15.6 2.1% 2014
Bill Cassidy LA Yes -22.2 -2.4 6.0 -18.6 11.9% 2014
Pat Roberts KS No -23.4 0.0 6.0 -17.4 N/A
Lamar Alexander TN No -25.8 0.0 6.0 -19.8 N/A
Ben Sasse NE Yes -25.8 -2.4 6.0 -22.2 32.8% 2014
Mike Rounds SD Yes -25.8 -2.4 6.0 -22.2 20.9% 2014
Tom Cotton AR Yes -28.2 -2.4 6.0 -24.6 17.1% 2014
Mitch McConnell KY Yes -28.7 -2.4 6.0 -25.1 15.5% 1984
James Risch ID Yes -34.2 -2.4 6.0 -30.6 30.7% 2008
Shelley Moore Capito WV Yes -35.5 -2.4 6.0 -31.9 27.7% 2014
James Inhofe OK Yes -38.1 -2.4 6.0 -34.5 39.5% 1994
Mike Enzi WY No -45.7 0 6.0 -39.7 N/A

At first glance, Democrats appear to have a significant advantage in the 2020 Senate elections (Table 1): of 34 Senate elections scheduled for November 2020, fully two-thirds (22) are currently Republican-held. And of those 22 seats, fully 73% (16) are potentially more vulnerable because they include…

Moreover, only one currently-Democratic seat appears particularly vulnerable as of now: Jones’ seat in deep-red Alabama (D-28.4); a reasonable estimate is that Jones would lose to a generic Republican by around 20 points. Even with the full effect of incumbency (+4.2), a repeat of Democrats’ strong overall performance in 2018 (D+9.9) and a pro-Democratic error of 5.4 points in 3W-RDM (the average miss over time), Jones would still be down about nine points to a generic Republican. Yes, Jones overcame similar odds in December 2017, but that was against a severely compromised Republican opponent.

And while first-term Democratic Senators Gary Peters of Michigan and Tina Smith of Minnesota (who won by double-digits in November 2018 after being appointed to replace Democrat Al Franken in December 2017) could be vulnerable—along with Jeanne Shaheen of New Hampshire and Mark Warner of Virginia, who narrowly won reelection in 2014—if Democrats only break even in 2020, as of now, they appear quite likely to prevail. They would join six safe Democratic incumbents (Ed Markey, Jack Reed, Richard Durbin, Chris Coons, Cory Booker[14] and Jeff Merkley) and a likely-safe open seat in New Mexico (with Democratic United States House of Representatives member [“Representative”] Ben Ray Luján a strong candidate to win the seat).

Ben Ray Lujan

2020 New Mexico Democratic Senate candidate Ben Ray Luján,

However, Democrats should not be banking on New York Senator Chuck Schumer switching from Minority to Majority Leader in January 2021 just yet. While as many as 16 Republican-held seats are arguably vulnerable, only two are in states that even lean Democratic: Maine (D+5.9) and Colorado (D+2.2). And while Gardner is clearly vulnerable (he underperformed by about four points in 2014, when he beat incumbent Democrat Mark Udall), even a slight improvement by Republicans in the total national Senate vote puts that seat at toss-up status, at best. And Collins has been winning statewide in Maine since 1996, including winning her fourth term by an eye-popping 37.0 points!

Plus, the next four most vulnerable Republican incumbents (all finishing their first term)—Ernst, Tillis, Perdue and McSally—represent states averaging 7.5 points more Republican than the nation as a whole; this is why even in a very good Democratic year the fundamentals have these races “toss-up” at best. Moreover, while it is true that Ernst, Tillis and Perdue won in 2014 by an average of just 5.9 points (with McSally losing by 2.3 points in 2018)—a hair over the overall Senate Republican that year—all four now have the modest added advantage of running as incumbents in lean-Republican states. And where Democrats have a strong candidate to run against McSally—former astronaut Mark Kelly, husband of former Representative Gabby Giffords (D-AZ)—other strong candidates such as former Iowa Governor (and Secretary of Agriculture) Tom Vilsack and former Georgia House Speaker Stacey Abrams have ruled out running for the Senate in 2020.

Mark Kelly

Left to right: former Representative Gabby Giffords and 2020 Democratic Senate candidate Mark Kelly

Abrams and Vilsack are not the only high-profile Democrats choosing not to challenge vulnerable incumbent Republican Senators. Former National Security Advisor Susan Rice will not challenge Collins, while former Colorado Governor John Hickenlooper is running for president in 2020 instead.

Beyond those six races, Democratic chances to flip seats only get slimmer. Former South Carolina Democratic Party chair Jaime Harrison is formally exploring a bid against Senator Lindsey Graham. And while former Representative Beto O’Rourke (D-TX; running for president) and Representative Joaquin Castro (D-TX) passed on a run, Air Force veteran Mary Jennings “MJ” Hegar, who came within 3 points of defeating incumbent Representative John Carter (D-TX) in 2018, plans to run against Senator John Cornyn. Even with Democrats winning nationally by six points, however, the fundamentals suggest both Harrison and Hegar begin their races down around 12 points.

Jaime Harrison

2020 South Carolina Senate Democratic candidate Jaime Harrison

MJ Hegar

2020 Texas Senate Democratic candidate MJ Hegar

Mississippi, meanwhile, will see a rematch between Espy and Hyde-Smith as she seeks a first full term. But while he came within about seven points of unseating her in 2018, this will be a tough Senate race for Democrats to win, as the fundamentals have him down by 15.1 points—similar to the Democratic position against first-term Senators Daines in Montana (where outgoing Democratic Governor Steve Bullock is apparently running for president instead) and Sullivan (who only defeated Democratic incumbent Mark Begich by 2.1 points in 2014) in Alaska.

Mike Espy.jpg

2020 Mississippi Democratic Senate candidate Mike Espy

That leaves 11 Republican-held Senate seats which average 30.3 points more Republican than the nation. Even with three open seats it is very difficult to see how Democrats flip any of them. One intriguing exception, however, could be in Kentucky, where Senate Majority Leader Mitch McConnell (quite unpopular at home) is seeking a seventh term; if Air Force veteran Amy McGrath (who, like Hegar, came within three points of defeating an incumbent Republican Representative in 2018—in this case Andy Barr) were to run, she may be able to overcome the fundamentals showing a generic Democrat down 25.1 points to McConnell.

The bottom line?

While there are several plausible paths for Democrats to win back a Senate majority in 2020…

  1. Win presidency; Jones win in Alabama; win Maine and Colorado and one of Iowa, North Carolina, Georgia, Arizona
  2. Win presidency; Jones lose in Alabama; win Maine and Colorado and two of Iowa, North Carolina, Georgia, Arizona
  3. Lose presidency; Jones win in Alabama; win Maine and Colorado and two of Iowa, North Carolina, Georgia, Arizona
  4. Lose presidency; Jones lose in Alabama; win Maine and Colorado and three of Iowa, North Carolina, Georgia, Arizona
  5. Any of 1-4 above but substituting wins in even more Republican states such as Texas, South Carolina, Mississippi, Alaska and Kentucky.

…a great deal would have to go just right for Democrats in each scenario. In fact, it is easy to foresee anything from Democrats net losing a handful of seats (Alabama and some combination of Michigan, Minnesota, New Hampshire and Virginia) to winning a clear majority (holding Alabama, sweeping the six most vulnerable states and maybe even picking off South Carolina and/or Texas and/or Mississippi and/or Kentucky) is possible.

The silver lining for Democrats, though, is that forcing Republicans to invest money, time and resources in states like Alabama, Alaska, Arizona, Georgia, Iowa, Kentucky, Mississippi, North Carolina, South Carolina and Texas makes it that much harder for them to beat Democratic incumbents in Michigan, Minnesota, New Hampshire and Virginia.

2019-20 Gubernatorial elections. Republicans currently occupy governor’s mansions in 27 states, with Democrats occupying the remaining 23.

Three gubernatorial elections will be held in 2019, all in southern states averaging 23.1 points more Republican than the nation (Table 2). The lone Democrat is John Bel Edwards of Louisiana, and while the fundamentals have him down to a generic Republican by 10.5 points, he is generally popular with voters in his state and thus more likely than not to win reelection. By contrast, the only Republican governor to seek reelection this year—Matt Bevin of Kentucky—is the least popular governor in the country; still, the fundamentals have him beating a generic Democrat by a whopping 31.2 points. As for the open seat in Mississippi, the fundamentals have a generic Republican defeating a generic Democrat by 12.5 points.

This means that the likeliest outcome is no net change in partisan control of governor’s mansions in 2019—though that could mean the parties switch control in Louisiana and Kentucky!

Table 2. 2019-20 Gubernatorial election overview

Name State Run

2019/ 2020

3W-RDM INC Nat

Lean

Total Last margin First elected
2019 DEMOCRATS
John Bel Edwards LA Yes -22.2 5.7 6.0 -10.5 12.2% 2015
2019 REPUBLICANS
Phil Bryant MS No -18.5 0.0 6.0 -12.5 N/A
Matt Bevin KY Yes -28.7 -8.5 6.0 -31.2 8.7% 2015
2020 DEMOCRATS
John Carney DE Yes 12.5 5.7 6.0 24.2 19.2% 2016
Jay Inslee WA No 12.1 0.0 6.0 18.1 N/A
Roy Cooper NC Yes -6.0 5.7 6.0 5.7 0.2% 2016
Steve Bullock MT No -18.6 0.0 6.0 -12.6 N/A
 
2020 REPUBLICANS
Phil Scott VT Yes 27.7 -8.5 6.0 25.2 14.9% 2016
Chris Sununu NH Yes 0.1 -8.5 6.0 -2.4 7.0% 2016
Eric Holcomb IN Yes -16.3 -8.5 6.0 -18.8 6.0% 2016
Mike Parson MO Yes -15.9 -4.3 6.0 -14.2 Succ 2018
James Justice WV Yes -35.5 -8.5 6.0 -38.0 6.8% 2016
Doug Burgum ND Yes -29.4 -8.5 6.0 -31.9 57.1% 2016
Gary Herbert UT No -33.1 0.0 6.0 -27.1 N/A

Looking ahead to 2020, two states currently governed by Democrats, Delaware and Washington, are all-but-certain to remain in Democratic hands, with Governor John Carney poised to reprise his nearly-20-point win in 2016 and a Democrat (state Attorney General Bob Ferguson?) heavily favored to succeed Governor Jay Inslee (running for president instead).

Equally certain to remain in Republican hands are West Virginia and North Dakota (where James Justice—who switched parties after winning as a Democrat in 2016—and Doug Burgum will seek reelection), as well as Utah, where Governor Gary Herbert is term-limited from seeking reelection.  The fundamentals in these states have Republicans ahead by 32.3 points over a generic Democrat.

That leaves six races which could be competitive—although Governors Eric Holcomb of Indiana and Mike Parson (who became governor in June 2018, following the resignation of Eric Greitens, just elected in 2016) of Missouri—are ahead in the fundamentals by 14-19 points.

North Carolina Governor Roy Cooper barely defeated Republican incumbent Pat McCrory in 2016, and while the fundamentals have him beating a generic Republican by 5.7 points, this race would be a pure toss-up in a neutral partisan environment. Montana is another story, though, with Bullock retiring after two terms (and 16 consecutive years of Democratic governors); the fundamentals suggest a generic Republican would win back the governor’s mansion in Helena by 12.6 points (and that is with Democrats winning by six points nationally).

That only leaves two New England Republican governors who just won reelection last year, but who the fundamentals see as highly vulnerable: Phil Scott, who won by nearly 15 points in deep-blue Vermont (D+27.7), and Chris Sununu, who “only” won by 7.0 points in swing-state New Hampshire. If they did not lose in 2018, though, it is unlikely (though not impossible) they will lose in 2020.

The bottom line?

As of May 2019, the 14 gubernatorial elections in 2019 and 2020 will most likely result in a net gain of 1 (with Republicans winning the open governor’s seat in Montana) governor’s mansion, expanding their overall lead to 28-22—but races this year in Kentucky and Louisiana, and next year in Indiana, Missouri, New Hampshire, North Carolina, and Vermont could yet surprise.

Until next time…

[1] The most recent election is weighted “3,” the 2nd-most recent election is weighted “2” and the 3rd-most recent election is weighted “1.”

[2] That is, relative to the Republican candidate. I excluded data from special elections such as the December 2017 Senate election in Alabama.

[3] For the California Senate election, I used the total votes for Democratic, Republican and all-other-party candidates in the June 5, 2018 “jungle primary.” For the Mississippi special Senate election, I used the results from the runoff election on November 27, 2018. For the Maine and Vermont Senate races, I counted as “Democratic” votes those cast for Independent Senators Angus King and Bernie Sanders, respectively, since each man caucuses with the Democrats (and there was no Democratic Senate nominee in Vermont); in Maine, I counted the Democratic votes as “other.” Notably, counting votes for King and Sanders as “other” (and Democratic votes in Maine as “Democratic”) only changes the national Democratic margin from +9.9 percentage points to +9.4.

[4] For the California Senate election, I used the total votes for Democratic, Republican and all-other-party candidates in the June 7, 2016 “jungle primary.” For the Louisiana Senate election, I used the results from the runoff election on December 10, 2016.

[5] I excluded the Alabama Senate race in which Republican incumbent Jeff Sessions ran unopposed.

[6] These values were +0.9% in 2018, +6.5% in 2016 and +10.6% in 2014.

[7] These values were +2.6% in 2018, +3.6% in 2016 and -0.7% in 2014.

[8] I counted the 2018 Alaska gubernatorial election as a Democratic open seat after Independent Governor Bill Walker suspended his reelection campaign on October 19, 2018, throwing his support to Democratic nominee Mark Begich.

[9] I counted Walker as a Democrat in 2014 Alaska gubernatorial election (though counting him as “Other” would have made little material difference). I counted the Rhode Island gubernatorial election as a Democratic open seat although outgoing Governor Lincoln Chafee was an Independent (who briefly sought the 2016 Democratic presidential nomination).

[10] For the 2015 Louisiana gubernatorial election, I used data from the runoff election held November 21, 2015.

[11] Because incumbent Republican governor Bobby Jindal easily cleared the 50% threshold on election day 2011, for the 2011 Louisiana gubernatorial election, I used the sum of all votes cast for the candidate of each political party (Republican, Democrat, Other) that day.

[12] West Virginia is counted three times because it also held a special gubernatorial election in 2011.

[13] Democratic incumbency “advantage” was +2.0% in 2017-18, +6.3% in 2015-16, +5.7% in 2013-14 and +18.9% in 2011-12; the corresponding Republican values were +17.3%, -3.4%, +10.3% and +5.1%.

[14] Or whoever replaces him, should he become the next president or vice president of the United States.

May 2020 update: Democratic presidential nomination and general election polling

It has become a tradition that for my birthday my wife and daughters present me with a new Maine (usually lighthouse-themed) wall calendar from Down East.

May 2019 calendar.JPG

Last year was no exception; here is the May 2019 entry on my “Maine Lighthouses” calendar.

I use this photograph to introduce what I anticipate to be a monthly updating of two recent posts:

**********

I begin with the 2020 Democratic nomination polling, assessed by my NSW-WAPA (national-and-state-weighted weighted-adjusted polling average). “WAPA” is a within-nation or -state polling average for any candidate listed in any poll released since January 1, 2019, weighted by pollster quality and number of days to a given primary or caucuses. The NSW weights are, in rank order, Iowa Caucuses and New Hampshire Primary (weight=5), Nevada Caucuses and South Carolina Primary (4), a time-weighted average of all post-South Carolina nominating contests (2) and national (1).

Since the initial post, there have been

  • 16 new national polls (including 4 Morning Consult tracking polls)
  • 2 Iowa Caucuses polls
  • 3 New Hampshire Primary polls
  • 0 Nevada Caucuses polls
  • 1 South Carolina Primary poll
  • 2 Texas Primary polls
  • 1 poll each for the California, Pennsylvania and Wisconsin Primaries

… for a total of 27 new polls (and 86 in 2019).

Three of the 16 “new” national polls were conducted wholly (ABC News/Washington Post, January 21-24, 2019; CNN/SSRS, March 14-17, 2019) or partially (USC Dornsife/LA Times, March 15-April 15, 2019) before March 31, 2019. I also fixed a few formula errors in my Excel workbook, so comparisons of polling averages between the previous post and this post should be taken with a modicum of salt.

One other change is that an additional 10 candidates were included in at least one national or state-level poll,[1] including former Alaska United States Senator (“Senator”) Mike Gravel, Speaker of the United States House of Representatives (“Representative”) Nancy Pelosi, actor Dwayne “The Rock” Johnson and President Donald Trump, bringing the overall total to 53!

As for candidacy declarations, former Virginia Governor Terry McAuliffe announced on April 17, 2019 that he would NOT seek the 2020 Democratic presidential nomination. But four other candidates declared their candidacy since the previous post: California Representative Eric Swalwell, Massachusetts Representative Seth Moulton, former Vice President Joe Biden and Colorado Senator Michael Bennet (with Montana Governor Steve Bullock hinting at a declaration later in May 2019).

And that still leaves former Georgia House Speaker Stacey Abrams, New York City Mayor Bill de Blasio and Gravel as possible candidates, for a total of 25.

As of May 2, 2019, here is the relative position of those 25 Democrats.

Table 1: National-and-state-weighted WAPA* for selected 2020 Democratic presidential nomination possibilities

Candidate National IA NH NV SC Post-SC NSW-WAPA
Biden 27.7 25.8 21.3 26.0 34.5 26.7 26.6 (-1.9)
Sanders 18.6 18.0 22.1 23.0 15.3 21.2 19.8 (-1.0)
Harris 8.2 8.3 7.6 9.0 10.6 9.5 8.8 (-1.3)
Warren 6.5 7.8 7.3 10.0 6.3 8.4 7.8 (-0.4)
O’Rourke 6.6 5.3 4.1 10.0 7.3 9.1 6.7 (+0.5)
Buttigieg 4.3 6.3 8.7 5.0 3.1 6.2 5.9 (+3.5)
Booker 3.2 4.1 3.4 2.0 8.2 2.7 4.1 (-0.1)
Klobuchar 1.7 3.6 2.6 2.0 1.0 2.2 2.3 (-0.4)
Yang 0.72 0.40 0.80 3.0 0.68 1.0 1.1 (+0.3)
Gabbard 0.65 0.41 0.79 2.0 0.74 0.56 0.89 (0.0)
Gillibrand 0.65 0.51 0.90 1.0 1.4 0.54 0.88 (-0.2)
Castro 0.91 0.98 0.07 1.0 0.68 1.5 0.76 (+0.2)
Abrams 0.21 0.00 0.00 0.00 3.1 0.17 0.62 (+0.6)
Delaney 0.33 0.90 0.45 0.00 0.00 0.17 0.35 (+0.1)
Hickenlooper 0.68 0.32 0.13 0.00 0.68 0.37 0.30 (+0.1)
Swalwell 0.17 0.28 0.29 0.00 0.23 0.17 0.20 (+0.1)
Ryan 0.18 0.19 0.58 0.00 0.00 0.00 0.19 (+0.2)
Inslee 0.46 0.42 0.00 0.00 0.00 0.24 0.15 (-0.0)
Messam 0.00 0.00 0.29 0.00 0.00 0.00 0.07 (+0.1)
Bullock 0.20 0.13 0.00 0.00 0.00 0.00 0.04 (0.0)
Bennet 0.03 0.13 0.00 0.00 0.00 0.00 0.03 (0.0)
Gravel 0.00 0.00 0.00 0.00 0.00 0.11 0.01 (0.0)
de Blasio 0.18 0.00 0.00 0.00 0.00 0.00 0.01 (0.0)
Williamson 0.11 0.00 0.00 0.00 0.00 0.00 0.01 (0.0)
Moulton 0.03 0.00 0.00 0.00 0.00 0.03 0.00 (0.0)
DK/Other 17.8 16.3 18.8 6.0 6.2 9.1 12.4 (+0.7)

The data in Table 1 suggest the following as of May 2, 2019:

  1. The top five candidates remain the same (and in the same order): Biden (26.6%), Vermont Senator Bernie Sanders (19.8%), California Senator Kamala Harris (8.8%), Massachusetts Senator Elizabeth Warren (7.8%), former Texas Representative Beto O’Rourke (6.7%).
  2. That said, more than half (53.6%) of potential Democratic primary/caucus voters prefer someone other than Biden or Sanders.
  3. The gap between Harris and Warren for 3rd place has narrowed from 1.9 percentage points (“points”) to 1.0 points.
  4. Surging from 8th to 6th place was South Bend, IN Mayor Pete Buttigieg, whose NSW-WAPA jumped from 2.4% to 5.9%—a remarkable 3.5-point jump.
  5. Buttigieg’s surge seems mostly to have come at the expense of Biden (-1.9), Sanders (-1.0) and Harris (-1.3). It also knocked New Jersey Senator Cory Booker (4.1%) and Minnesota Senator Amy Klobuchar (2.3%) down one spot to 7th and 8th, respectively.
  6. Biden’s relative decline masks the fact he has seen a mini-bump in the national polls (weighted much lower than state polls) since his April 25, 2019 declaration of candidacy—and that he had been dropping somewhat before that.
  7. The mini-surge for O’Rourke (+0.5) comes entirely from two April 2019 polls of the 2020 Texas presidential primary[2]; O’Rourke averages 23.5% in the two polls, just ahead of Biden (21.5%), Sanders (18.0%) and Buttigieg (11.5%).
  8. The only other candidate to make an appreciable move was entrepreneur Andrew Yang (1.1%), from 11th to 9th, knocking New York Senator Kirsten Gillibrand (0.88%) from 11th to 9th. To be fair, however, they are separated by less than ¼ of a point.
  9. Rounding out the top 10 is Hawaii Representative Tulsi Gabbard (0.89%), meaning five of the top 11 choices are women—all current members of the U.S. House or Senate.
  10. Former Secretary of Housing and Urban Development Juan Cástro (0.76%) saw his NSW-WAPA increase slightly, but he is still stuck in 12th place overall, just ahead of…
  11. …Abrams, who at 0.62%, is the highest-ranked non-declared candidate.
  12. The remaining 12 declared/potential candidates—Maryland Representative John Delaney; former Colorado Governor John Hickenlooper: Swalwell; Ohio Representative Tim Ryan; Washington Governor Jay Inslee; Miramar, FL Mayor Wayne Messam; Bullock; Bennet; Gravel; de Blasio; author Marianne Williamson and Moulton—continue to languish well below 0.5%.
  13. Even with a choice of 25 declared and potential candidates, 1 in 8 respondents still could not state a preference (11.9%) or preferred some other candidate (0.5%).

With all that, the fundamental pecking order remains the same for the 2020 Democrats (unlikely to change before the first Democratic presidential candidate debates on June 26-27, 2019[3]):  Biden and Sanders are the clear leaders, splitting just under half of the overall vote between them (46.4%), followed by Harris, Warren, O’Rourke and Buttigieg (29.2% total). Just behind these six are Booker and Klobuchar (6.2% total). Add Yang’s 1.1%, and 5 of every 6 Democrats polled prefer a total of nine candidates. While one of those nine men and women will probably (but not definitively) be the 2020 Democratic presidential nominee, the race is very much in flux with more than 15 months until Democrats convene in Milwaukee, WI on July 13, 2020.

**********

Little has changed since my first wicked early look at how various potential Democratic nominees fared against President Trump in hypothetical one-on-one matchups, both nationally and in a total of 12 states (AZ, IA, MA, MI, MN, NV, NH, NC, PA, SC, TX, WI; mean 3W-RDM[4]= D-1.9).

Weighting each Democrat’s WAPA (vs. Trump) by her/his NSW-WAPA shows Democrats ahead of Trump nationally by 2.8 points, down slightly from 3.1 points. Remove Biden’s 7.4-point margin against Trump, and the Democratic advantage drops to 1.1 points; Sanders, Harris, Warren, O’Rourke and Booker are each ahead of Trump by between 0.4 and 3.6 points, while Gillibrand trails by just 1.2 points and Buttigieg by just 3.4 points. All other tested 2020 Democratic presidential nominees trail by between 6.3 (Klobuchar) and 15.5 points (Messam).

A strict apples-to-apples comparison in national standings is complicated, though, by a significant change I made to my Excel workbook. I adjust general election polling margins by pollster bias (or what FiveThirtyEight.com calls “mean-reverted bias”): how much, on average, a pollster over- or underestimates Democratic election margins. The pollster HarrisX dominates national presidential “trial heat” polling (while not releasing a single state-level poll), including every publicly-released matchup between Trump and Delaney, Gabbard, Gravel, Hickenlooper, Inslee, Messam, Moulton, Ryan, Swalwell, Williamson and Yang. FiveThirtyEight has not yet estimated HarrisX’s pollster bias. However, Harris Interactive has a pro-Republican bias of 1.5 points, and in a recent article, Nate Silver notes that HarrisX conducted a poll for Scott Rasmussen. Rasmussen Reports/Pulse Opinion Research also has a 1.5-point Republican bias. Therefore, I decided to add 1.5 points to every Democratic margin reported by HarrisX. This boosted the weighted-adjusted average of Democrats against Trump by 0.4 points.

Finally, the only significant state-level change is that Democrats overall are beating Trump in Wisconsin by 7.4 points, up from 6.1 points, on the strength of a set of Zogby Interactive/JZ Analytics polls (Biden, Sanders, Harris, Warren, O’Rourke, Buttigieg) conducted April 15-18, 2019.

And the good news for Democrats remains that, collectively, the state-level polls imply a national margin of 6.0 points over Trump, twice what the national polls suggest.

Until next time…

[1] And that excludes New Hampshire Senator Maggie Hassan, included in an alternate version of a New Hampshire Primary poll conducted by the University of New Hampshire, April 10-18, 2019.

[2] Change Research, 4/18-4/20/2019; Emerson College, 4/25-4/28/2019

[3] They will be held over two nights to accommodate 20 (of at least 21) candidates, with no more than 10 appearing each night. Criteria for obtaining one of the 20 available debate slots may be found here.

[4] My measure of how much more or less Democratic a state votes relative to the nation, using a weighted average of a state’s presidential voting compared to the national popular vote in the three previous presidential elections.

A wicked early look at the 2020 U.S. presidential election

With the 2018 midterm elections behind us, and the field of Democrats seeking to challenge Republican President Donald Trump in 2020 taking shape, let us turn to the 2020 presidential elections, scheduled to take place (technically, to conclude) on November 3, 2020.

Previously, I addressed the wicked early relative standings of the 44 Democrats listed in at least one public pollreleased since January 1, 2019 by:

  1. Calculating each candidate’s polling average (WAPA[1]; weighted by time to election and pollster rating), both nationally and within any state with available 2019 polling.
  2. Calculating a single average (NSW-WAPA[2]) for each candidate by weighting their WAPA in early primary/caucus states (Iowa, New Hampshire, Nevada, South Carolina) more than in later states, in turn weighted more than national polls.

I similarly approach the 2020 presidential election: mixing “fundamentals” with public polls released since January 1, 2019 asking voters to choose between a named Democratic candidate and Trump.”[3] “Fundamentals” are simply expected margins between the Democratic and Republican presidential nominees in a state (the Electoral College [EC] elects the president of the United States) based upon my 3W-RDM measure (weighted average arithmetic difference between state and national Democratic margins in the last three presidential elections[4]). This measure is not perfect: it has “missed” (in either direction) by an average of 5.4 percentage points (“points”) over the last six presidential elections, with 67 misses (22%) higher than 5.4%.

Nonetheless, “fundamentals” (3W-RDM plus hypothesized 2020 Democratic national popular vote margin) provide a useful baseline against which to assess wicked early polling data. This allows me to avoid such “macro-level” indicators as the state of the economy or Trump’s approval ratings advocated by political scientists in (solid) volumes like this:

Forecasting Rosenstone.JPG

**********

Let me begin with the metaphorical (literal?[5]) elephant in the room: incumbent presidents who seek reelection win about two-thirds of the time, according to Yale political science professor David R. Mayhew[6].

All else being equal, then, Democrats have an uphill battle to defeat Trump in 2020. However, all is not equal, as both “fundamentals” and polling suggest. To calculate the “fundamentals” for each state (and District of Columbia [DC]), one needs

  1. That state’s 3W-RDM and
  2. The Democratic margin in the national popular vote.

While #2 is difficult to estimate more than 18 months in advance, there may be historic guidance.

Just bear with me while I review 60 years of presidential election results. As usual, election data come Dave Liep’s indispensable Atlas of U.S. Presidential Elections.

Over the last 10 presidential elections (1980-2016), the Democratic presidential nominee has lost the national popular vote by an average of 1.0 point. However, because Democrats have fared better in presidential elections since 1992, perhaps more weight should be given to more recent elections. One way to do this is to weigh the 1980 margin “1,” the 1984 margin “2” and so forth through a weight of “10” for 2016; this yields a Democratic edge of 1.7 points. This is broadly similar to taking the mean of the 10- , 5- and 3-election averages (D+1.9). Even better for Democrats is a weighted average of the the last five presidential elections (D+2.9), not surprising given that Democrats have won the national popular vote in four of those elections (while only winning the EC twice).

Based on recent electoral history, then, a Democratic national popular vote margin of between 1.5 and 3.0 points is certainly plausible; the 2.1 points by which Democrat Hillary Clinton beat Trump in 2016—an election with no incumbent seeking reelection, which Mayhew found were 50-50—falls squarely in this range.

But incumbency does need to be taken into account. In the last 10 presidential elections in which an incumbent president election sought reelection (including 1964 and 1976, when elevated Vice Presidents Lyndon Johnson and Gerald Ford, respectively, first sought election in their own right), that incumbent won by an average of 7.7 points. Once you exclude the landslide reelections of Dwight Eisenhower (1956, 15.4 points), Johnson (22.6 points) and Richard Nixon (1972, 23.2 points), however, the average drops to just 2.2 points. And that latter figure drops to a loss of 0.5 points if you exclude Ronald Reagan’s 1984 reelection landslide (18.2 points).

An even better way to assess the effect of incumbency, though, is to compare initial and reelection margins. Starting with Eisenhower in 1956, there have been eight such election pairs:

Table 1. Presidential margins, initial and reelection, 1952-2012

President Initial Election Reelection Change
Year % Margin Year % Margin R – I
Dwight Eisenhower 1952 10.9% 1956 15.4% 4.6%
Richard Nixon 1968 0.7% 1972 23.2% 22.5%
Jimmy Carter 1976 2.1% 1980 -9.7% -11.8%
Ronald Reagan 1980 9.7% 1984 18.2% 8.5%
George H. W. Bush 1988 7.7% 1992 -5.6% -13.3%
Bill Clinton 1992 5.6% 1996 8.5% 2.9%
George W. Bush 2000 -0.5% 2004 2.5% 3.0%
Barack Obama 2008 7.3% 2012 3.9% -3.4%
Average, all   5.4%   7.0% 1.6%
Average, GOP only   5.7%   10.7% 5.0%

Since 1956 (Table 1), elected presidents won reelection by an average 1.6 points more than they won initially; this would put Trump at just 0.5 points down in 2020. However, when you examine only the five Republican presidents who sought reelection over this time span (Eisenhower, Nixon, Reagan, Bush 41, Bush 43), the margin increase jumps to 5.0 points; this would put Trump 2.9 points ahead in 2020. That said, the last three elected presidents to seek reelection actually saw an average drop of 0.8 points in their margins; the decline is similar (-0.6 points) for the last three Republican presidents to seek reelection (Reagan, Bush 41, Bush 43).

In other words, a longer analytic time frame implies Trump would win the 2020 national popular vote by 1.0 to 7.7 points, while a shorter analytic time frame implies a Democratic win of 1.7 to 2.9 points—a range of R+7.7 to D+2.9.

Well, that was not very helpful, was it?

**********

Actually, this range provides a baseline against which to assess 2020 presidential election polls asking respondents to choose between a named Democrat and Trump.

Table 2: “Fundamentals”-based 2020 state-level Democratic margins 

State/DC 3W-RDM EV D-7.7 D-2.9 D-1.0 D+0.0 D+1.7 D+2.9
131 +12.0-+82.0 175 +≥4.3 +≥9.1 +≥11.0 +≥12.0 +≥13.7 +≥14.9
OR +8.7 7 +1.0 +5.8 +7.7 +8.7 +10.4 +11.6
NM +6.5 5 -1.2 +3.6 +5.5 +6.5 +8.2 +9.4
ME +5.9 4 -1.8 +2.0 +4.9 +5.9 +7.6 +8.8
MI +2.2 16 -5.5 -0.7 +1.2 +2.2 +3.9 +5.1
CO +2.2 9 -5.5 -0.7 +1.2 +2.2 +3.9 +5.1
NV +2.0 6 -5.7 -0.9 +1.0 +2.0 +3.7 +4.9
MN +1.5 10 -6.2 -1.4 +0.5 +1.5 +3.2 +4.4
VA +1.5 13 -6.2 -1.4 +0.5 +1.5 +3.2 +4.4
WI +0.7 10 -7.0 -2.2 -0.3 +0.7 +2.4 +3.6
NH +0.1 4 -7.6 -2.8 -1.1 +0.1 +1.8 +3.0
PA -0.4 20 -8.1 -3.3 -1.4 -0.4 +1.3 +2.5
FL -3.4 29 -11.1 -6.3 -4.4 -3.4 -1.7 -0.5
IA -4.7 6 -12.4 -7.6 -5.7 -4.7 -3.0 -1.8
OH -5.8 18 -13.5 -8.7 -6.8 -5.8 -4.1 -2.9
NC -6.0 15 -13.7 -8.9 -7.0 -6.0 -4.3 -3.1
GA -9.6 16 -17.3 -12.5 -10.6 -9.6 -7.9 -6.7
AZ -9.7 11 -17.4 -12.6 -10.7 -9.7 -8.0 -6.8
212 -15.3-

-45.7

164 -≥23.0 -≥18.2 -≥16.3 -≥15.3 -≥13.6 -≥14.8
TOTAL DEM EV   182 191 245 259 279 279

1 DC, HI, VT, CA, MD, MA, NY, RI, IL, CT, DE, WA, NJ

2 TX, SC, MO, IN, MS, MT, AK, LA, KS, NE, SD, TN, AR, AL, KY, ND, UT, ID, WV, OK, WY

Table 2 shows how EV would hypothetically be distributed across states under scenarios ranging from Trump winning the national popular vote by 7.7 points to the Democratic nominee winning it by 2.2 points. Boldfaced results are expected Democratic wins, with italicized results reflecting uncertainty based on the average 3W-RD miss of +/-5.4 points.

What jumps out are the eight states comprising 78 EV with 3W-RDM between D+2.2 and D-0.4 (MI, CO, NV, MN, VA, WI, NH, PA). This range includes the states that effectively cost Clinton the presidency in 2016: MI, WI, PA. All but Pennsylvania lean slightly toward the Democrats, but even a relatively narrow 2.3-point Trump win in the national popular vote (well within the historic range) could easily put all 78 EV in the Republican column, giving Trump a solid 347-191 EC win. However, because there is a 3.0-point jump from Pennsylvania (truly “The Keystone State”) to Florida in 3W-RDM, a similar 2.2-point national popular vote win for the 2020 Democratic presidential nominee would be expected to yield only a narrow 279-259 EC win. In fact, the 2020 Democratic presidential nominee winning Florida (and the EC 308-230) would seem to require winning the national popular vote by at least 3.5 points. That same nominee winning 347 EV would similarly require winning the national popular vote by at least 6.0 points, adding Ohio, Iowa and North Carolina to her/his column.

In other words, there is clear asymmetry in the distribution of EV that works to the Republicans’ advantage: they can win solid EC victories with smaller national popular vote margins.

Of course, this analysis is purely hypothetical, even before re-emphasizing every state-level Democratic margin in Table 2 is the center of a 10.8-point range of possible outcomes (with roughly 80% certainty). Thus, a Trump win by 2.9 points nationally could result in anything from 259 to 356 EV for Trump, while that same margin for the Democratic nominee could yield anything from 191 to 347 EV.

**********

This broad uncertainty about the outcome in 2020, with perhaps a slight historical edge to Trump (though nothing like being a 2-1 favorite, per Mayhew), is why we now turn to 2020 presidential polling.

Since January 1, 2019, pollsters have tested national match-ups between Trump and 16 announced (or soon-to-announce) candidates for the 2020 Democratic presidential nomination:

  • Former Vice President Joe Biden (15 times)
  • California Senator Kamala Harris (14)
  • Vermont Senator Bernie Sanders (14)
  • Massachusetts Senator Elizabeth Warren (13)
  • Former Texas U.S. Representative Beto O’Rourke (10)
  • New Jersey Senator Cory Booker (7)
  • South Bend, Indiana Mayor Pete Buttigieg (4)
  • New York Senator Kirsten Gillibrand (4)
  • Minnesota Senator Amy Klobuchar (3)
  • Former Secretary of Housing and Urban Development Julián Castro (2)
  • Maryland U.S. Representative John Delaney (2)
  • Hawaii U.S. Representative Tulsi Gabbard (2)
  • Former Colorado Governor John Hickenlooper (2)
  • Washington Governor Jay Inslee (2)
  • Author Marianne Williamson (2) and
  • Entrepreneur Andrew Yang (2)

Five other national polls matched Trump against former New York City Mayor Michael Bloomberg, Ohio Senator Sherrod Brown, New York U.S. Representative Alexandria Ocasio-Cortez, House Speaker Nancy Pelosi (D-CA) and Senate Minority Leader Chuck Schumer (D-NY), none of whom are running for president in 2020; Representative Ocasio-Cortez, who was born in 1989, would actually be too young to serve as president on January 20, 2021.

Not surprisingly, the number of polls conducted with each candidate closely matches her/his current standing for the Democratic nomination, based on NSW-WAPA; Table 3 below shows these values as of April 20, 2019. To calculate how each candidate currently fares against Trump in the 2020 national popular vote, I used the procedure detailed here. Essentially, I weighted “unbiased” polling margins[7] by date-to-election and pollster rating two ways: 1) treating all polls (even from the same pollster) as statistically independent and 2) treating polls from the same pollster as a single “average.” WAPA is the average value from the two methods.

Table 3: Current WAPA margins for hypothetical 2020 matchups between Donald Trump and selected Democratic presidential nominees

Democratic nominee Nomination polls1 Margin vs. Trump
Joe Biden 27.9% 7.9%
Bernie Sanders 19.9% 3.7%
Kamala Harris 9.8% 1.3%
Elizabeth Warren 8.3% 0.5%
Beto O’Rourke 6.8% 1.6%
Cory Booker 4.3% 1.0%
Pete Buttigieg 4.3% -5.1%
Amy Klobuchar 2.6% -5.2%
Kirsten Gillibrand 1.0% -0.2%
Andrew Yang 0.9% -7.5%
Tulsi Gabbard 0.8% -9.6%
Julián Castro 0.8% -7.5%
John Hickenlooper 0.3% -9.7%
John Delaney 0.3% -9.1%
Jay Inslee 0.2% -11.6%
Marianne Williamson 0.0% -10.1%
Other2 11.9% 4.2%
Margin weighted by nomination polls 3.1%
Margin weighted by nomination polls (w/o Biden) 1.3%

 1 Not listed are Ohio U.S. Representative Tim Ryan (0.1%), California U.S. Representative      Eric Swalwell (0.1%) and Miramar, Florida Mayor Wayne Messam (0.01%)

 2 The weighted average of the five non-candidate matchups

As of April 2019, Biden—with near-universal name recognition—fares best against Trump in 2020, beating him by nearly eight points, which translates to around 347 EV (and perhaps more if he wins Georgia’s and Arizona’s 27 combined EV as well). More surprising, though, is Sanders’ solid performance against Trump—beating him by 3.7 points in the national popular vote (and likely winning Florida for 308 EV).

Moreover, both Biden’s and Sanders’ prospective margins are well outside the historic range computed above. The next tier of candidates—Harris, Warren, O’Rourke, Booker and Buttigieg—essentially tie with Trump, averaging a -0.1% loss in the national popular vote; Gillibrand also would essentially tie Trump (-0.2%). Of this group, Harris, O’Rourke and Booker fare best, beating Trump in the national popular vote by between 1.0-1.6 points and eking out an EC win, more likely than not. The other candidate with at least three public head-to-head matchups, Klobuchar, for now fares as poorly as fellow Midwesterner Buttigieg, losing the national popular by just over five points.

I would not read too much into polls showing Yang, Gabbard, Castro, Hickenlooper, Delaney, Inslee and Williamson losing the 2020 national popular vote to Trump by between 7.5 and 11.6 points. Both sets of polls were conducted by HarrisX, a firm rated “C+” by FiveThirtyEight.com and for whom no “skew” has yet been calculated. They may be affiliated with Harris Interactive, another C+-rated pollster with a significant pro-Republican (mathematical) bias of 1.5 points.

Nonetheless, it is noteworthy (and likely reflecting the strong Democratic desire to defeat Trump in 2020) that the better a candidate is doing in the nomination polls (i.e., NSW-WAPA), the better that candidate currently fares against Trump; the correlation between the two values is 0.84.

But most interesting of all is that a possible proxy for “generic Democrat” (weighted-adjusted average of the five candidates not running for president) would theoretically beat Trump in 2020 by just over four points (putting Florida, Iowa, Ohio and North Carolina in play).

To get a sense of how a typical Democrat nominee would fare against Trump in 2020, I calculated a weighted average of the 17 margins listed in Table 3 using each candidate’s NSW-WAPA (a rough proxy for “likelihood of winning the nomination”). By this measure, the typical 2020 Democratic presidential nominee would beat Trump nationally by 3.1 points, just outside the upper end of the range based upon recent electoral history. However, if you remove Biden’s projected 7.9-point win, the margin drops to 1.3 points…roughly how Harris, Warren, O’Rourke and Booker would theoretically fare against Trump. They would still be slightly favored to win, though arguably 164 EV from 14 states (NM to NC in Table 2) would be in play.

**********

But what about the states?

Since January 1, 2019, head-to-head 2020 matchups between a named Democrat and Trump have been conducted in 12 states (AZ, IA, MA, MI, MN, NV, NH, NC, PA, SC, TX, WI; mean 3W-RDM = D-1.9). Table 4 shows how specific Democrats (as well as “typical Democratic nominee”) currently fare against Trump (minimum 2 matchup polls in state), sorted from most- to least-Democratic by 3W-RDM.

State Bid San Har War O’Ro Boo Oth Overall Implied NPV
MA1 1 1 1 D+32.6 D+10.5
MI2 5.8 4.5 3.5 3.0 1 1 D+3.6 D+1.4
NV3 1 1 1 1 1 D-1.0 D-3.0
MN4 1 D+15.5 D+14.0
WI5 10.7 5.3 1 1 2.6 D+6.1 D+5.4
NH6 1 10.6 1 5.2 1 D+8.1 D+8.2
PA7 9.3 6.4 1 1 -0.6 D+5.9 D+6.3
IA8 4.8 1.1 -6.8 -3.4 1 1 1 D-1.4 D+3.3
NC9 -3.0 -10.3 -9.8 1 1 -7.3 D-5.6 D+0.4
AZ10 1 1 1 1 D-6.2 D+3.5
TX11 -1.9 1 -7.9 1 -1.9 1 D-3.3 D+12.0
SC12 1 1 1 1 1 1 1 D-7.7 D+8.0

1 Emerson College, 4/4-4/7/2019, 761 registered voters

2 O’Rourke: Firehouse Strategies/Optimus, 3/19-3/21/2019, 540 likely voters; Klobuchar: Emerson College, 3/7-3/10/2019, 743 registered voters

3 Emerson College, 3/28-3/30/2019, 719 registered voters; Oth=Buttigieg

 4 DFM Research, 2/26-3/3/2019, 719 registered voters; Oth=Klobuchar

 5 Emerson College, 3/15-3/17/2019, 775 registered voters; Oth=Klobuchar

  6 Emerson College, 2/21-2/22/2019, 910 registered voters

  7 Emerson College, 3/28-3/30/2019, 719 registered voters

 8 Emerson College, 1/30-2/2/2019, 831 registered voters; Other=Gillibrand; mean of  Brown, Pelosi

  9 PPP, 1/4-1/7/2019, 750 registered voters

  10 OH Predictive Insights, 2/12-2/13/2019, 600 likely voters

  11 Quinnipiac University, 2/20-2/25/2019, 1,222 registered voters; Other=Castro

  12 Emerson College, 2/28-3/2/2019, 755 registered voters; Other=Klobuchar

Most public 2019 state-level polls have been conducted by Emerson College (B+, D+0.1), who occasionally test “trial heats” when they conduct Democratic (and, less frequently, Republican) nomination polls in a state. While Emerson is a high-quality pollster, it would help to see other high-quality pollsters test these matchups. Also bear in mind that these are primarily polls of “registered voters,” not voters deemed likely to vote; the latter polls often tilt slightly Republican. And, of course, the 2020 U.S. presidential election is still more than 18 months away.

Nonetheless, Table 4 reveals some interesting things. One, it is clear pollsters believe that Biden, Sanders, Harris and Warren (≥14 head-to-head matchups)—in that order—are the likeliest 2020 Democratic presidential nominees, followed by O’Rourke (10 matchups); this perfectly matches the top five using NSW-WAPA. Pollsters asked about Booker, Gillibrand and Klobuchar earlier in the year, but have ceased doing so, reflecting their stagnation in the race.

As they do nationally, both Biden and Sanders perform very well in the states—especially in the key swing states of Michigan, Wisconsin, New Hampshire (one poll has Biden up 10 points), Pennsylvania and, most importantly, Iowa. Removing Biden from the analysis, in fact, lowers hypothetical margins a mean 2.2 points across these states, albeit without changing the state “winner.” Harris, Warren and O’Rourke do not perform as well, but neither do they fare poorly.

Overall, Democrats (using the same weighting method as national polls) are collectively ahead by solid margins in Michigan (3.6), Wisconsin (6.1) and Pennsylvania (5.9)—the states that cost Clinton an EC victory in 2016; strong 2018 Democratic performances in these three states may not have been a fluke. They are winning New Hampshire (D+0.1) by 8.1 points. And they are holding their own in Iowa, a state trending sharply Republican—from D+2.0 after 2012 to D-4.7 after 2016; being behind only 1.4 points could be a terrific sign for Democrats in 2020.

The most striking positives for Democrats in Table 1, however, are being down “only” 3.3 points in Texas and 7.7 points in South Carolina (albeit in a single set of late February polls). First, these averages imply an 8-12-point lead in the national popular vote, which has not happened since Clinton won reelection in 1996 by 8.5 points. Moreover, this could confirm the southeast and southwest are trending Democratic…although the averages from Arizona (D-6.2) and North Carolina (D-5.6) suggest otherwise.

Being collectively 5.6 points down in North Carolina, in fact, is not ideal for Democrats, as it implies essentially a tied race nationally. The worst news for Democrats in Table 4, though, is a set of polls conducted by Emerson College in Nevada in late March. Nevada has been trending steadily more Democratic[8], so to be collectively down roughly one point there is worrisome, because it implies a national environment where Trump wins by 3.0 points. Across the other 11 states, on average, the Democratic nominee would beat Trump by 6.6 points!

Still, the overall message from Table 4 is that a hypothetical Democratic nominee

  • is faring well precisely where (s)he needs to (MI, WI, NH, PA—even IA) and
  • leads the 2020 national popular vote by 5.8 points–roughly the margin by which Democrats now lead in the “generic ballot” (e., some variant of “If the election for U.S. House were held today, would you vote for the Democratic or Republican candidate?”)[9].

**********

I cannot emphasize enough the following caveats to this analysis:

  1. The 2020 presidential election is more than 18 months away.
  2. National polls only reveal so much.
  3. State-level polling is limited and mostly conducted by a single (respectable) pollster.
  4. Most polls are of registered voters, not likely voters, introducing some pro-Democratic (mathematical) bias
  5. Hypothetical matchups reflect name recognition as much as considered choice.
  6. I have not controlled for a possible Independent run by Schultz.

That said, signs are good for Democrats right now. The best-known Democratic candidates (and current adjusted polling leaders)—Biden and Sanders—are beating Trump both nationally and in the key states. The next tier of candidates—Harris, Warren and O’Rourke—while less well-known, are also faring relatively well, beating Trump by just over one point on average. As they get better known, that margin could increase (just as the Biden and Sanders margins could shrink).

And while a wider historical frame shows Trump as much as a 2-1 favorite (winning the national popular vote by as much as 7.7 points), recent history favors the Democrats to win the national popular vote by between 1.5 and 2.9 points—though even then their nominee would not be guaranteed an Electoral College victory.

Moreover, current polling suggests Democrats are ahead of even those positive numbers. Using national polls only, a typical Democratic nominee is ahead by 3.3 points in the national popular vote, while state-level polls imply a lead nearly double that: 5.8 points! Removing Biden from the analysis drops these values to 1.8 and 3.6 points, respectively, precisely in line with recent electoral history.

All in all, then, I would rather be the eventual Democratic presidential nominee than Trump in 2020.

In April 2019, anyway.

Until next time…

[1] Weighted-adjusted polling average

[2] National-and-state-weighted WAPA

[3] For now, I ignore a March 8-10, 2019 poll by Change Research which substituted Vice President Mike Pence for Trump. I also ignore (for now) five national polls (and Emerson College polls from Iowa, Michigan, New Hampshire, South Carolina and Wisconsin) that also include former Starbucks CEO Howard Schultz, who may run for president as an Independent.

[4] Democratic percentage of total vote minus Republican percentage of total vote. Weights for most recent 3W-RDM are 2008=1, 2012=2, 2016=3

[5] This drives our eldest daughter—a Democrat like her parents—crazy because she LOVES elephants.

[6] Mayhew, David R. 2008. “Incumbency Advantage in U.S. Presidential Elections: The Historical Record.” Political Science Quarterly 123:2.

[7] Using the FiveThirtyEight.com “Mean-Reverted Bias” listed in their Pollster Ratings.

[8] Starting with the three presidential elections from 1984-1992, Nevada’s 3W-RDM are D-8.5, D-6.9, D-5.0, D-2.5, D+2.0, D+3.2, D+2.0

[9] The margin has varied between D+2 and D+9—and mostly between D+5 and D+7—over the last few weeks, according to polls listed on FiveThirtyEight.com.

A wicked early look at the 2020 Democratic presidential nomination

On March 11, 2019, the Democratic Party announced that its 2020 national convention will be held at the Fiserv Forum in Milwaukee, WI. This is a reasonable choice, given that Wisconsin—which 2016 Democratic presidential nominee Hillary Clinton lost by 0.8 percentage points—is one of the true swing states in presidential elections.

It also means the formal selection of the 2020 Democratic presidential and vice-presidential nominees will not take place for another 15 months (July 13-16, 2020). Nonetheless, nearly two dozen Democrats have either declared their intention to run for president in 2020—or are strongly leaning toward doing so.

Put another way, just since January 1, 2019, there have been…

  • 37 national polls (17 by Morning Consult alone)
  • 6 Iowa Caucuses polls
  • 5 New Hampshire Primary polls
  • 1 Nevada Caucuses poll
  • 3 South Carolina Primary polls
  • 1 poll each from the primaries in Alabama, California, Michigan, Florida, Wisconsin, Pennsylvania and Oregon

conducted of the race for the 2020 Democratic presidential nomination, for a total of 59 different polls in three months.

The 37 national polls themselves queried voters about 39 potential nominees; when you include candidates only included in state polls (Massachusetts Representative[1] Joseph P. Kennedy III and billionaire Tom Steyer in New Hampshire; Miramar, FL Mayor Wayne Messam in Alabama and Nevada), the number increases to 42. And not one of these 59 polls listed Ohio Representative Tim Ryan, who declared his presidential candidacy on April 4, 2019—bringing the total to 43.

**********

The point is, despite the Iowa Caucuses—the first contest to choose delegates to the 2020 Democratic National Convention—being nearly 10 months away (tentative date February 3, 2020)—it is not too early to begin to examine public polls assessing voters’ 2020 Democratic presidential nominee preference.

Thus, just as I did with public polls for governor and United States Senate (“Senate”) in 2018, I am collecting and aggregating 2020 Democratic nomination polls released publicly since January 1, 2019. The difference from 2018, though, is that rather than “project” who the 2020 Democratic nominee will be, I am simply taking snapshots of the relative ordering of candidates for that nomination.

Given that convention delegate selection occurs at the state level, I calculated a version of my “weighted-adjusted polling average” (WAPA) for each candidate, both nationally and within each state contest for which I have polling. I then calculated a single WAPA across all of these levels, which I explain below.

I calculated each candidate’s national/state-level WAPA this way:

First, I weighted the raw percentages for each candidate by

  1. The FiveThirtyEight.com pollster rating, converting their letter ratings to a numeric value (A+ = 4.3, A = 4.0, etc.). Thus, each weekly Morning Consult (B-) tracking poll is weighted 2.7/4.3 = 0.628.
  2. A ratio of two values: a) the number of days a poll’s midpoint[2] is since January 1, 2019 and b) the number of days between December 31, 2018 and the contest being assessed (state contest or July 13, 2020, the first day of the 2020 Democratic National Convention). Thus, the University of New Hampshire poll of 240 likely 2020 New Hampshire Primary voters conducted February 18-26, 2019 had a midpoint of February 22, 2019, which is 54 days before the primary. There are 407 days between December 31, 2018 and February 11, 2020, the likely date of the 2020 New Hampshire presidential primary. Dividing 54/407 gives this poll a weight of 0.133.

In essence, recent, higher quality polls count more in WAPA than older, lower quality polls. I also considered weighting each candidate poll percentage by the number of respondents. However, given that the 12 national Morning Consult tracking polls average 13,173 respondents, while the other 25 national polls average 558 respondents, that would give way too much weight to the former.

Second, to account for pollsters releasing multiple polls of the same contest (e.g., the 37 national polls noted earlier were conducted by just 14 unique pollsters[3]), I averaged two versions of the candidate polling average detailed in the first step:

  1. Treating each poll as a statistically-independent event.
  2. Calculating aggregate averages for each pollster then aggregating those.

To complete these steps, I needed to make two decisions regarding each poll.

  1. I assigned a value of 0% to any candidate not listed in a poll who was included in at least one other poll. I realize this may introduce some slight mathematical bias[4], but as FiveThirtyEight.com’s Nathaniel Rakich wrote, “The good news is that, in most polls, the candidates who are rotated in or out are polling poorly, so it doesn’t make a huge difference whether they’re included or excluded.”
  2. If a pollster released two versions of a poll on the same day—with and without one candidate (usually former Vice President Joseph R. Biden, Jr. or Hillary Clinton)—I used the more inclusive poll. While this could potentially introduce significant mathematical bias, this is (hopefully) mitigated by the infrequency of the practice; the bias will also drop considerably over time as the field of 2020 Democratic presidential nomination candidates becomes more set.

Having calculated 12 different candidate WAPA, the final step was to combine them into a single “national-and-state-weighted WAPA” (NSW-WAPA).

It is well-established that candidate performance in Iowa affects candidate performance in New Hampshire, in turn affecting candidate performance in every contest that follows; I still think this is the definitive work on the subject:

Bartels Presidential Primaries.JPG

Therefore, I weigh the earliest primaries and caucuses—Iowa, New Hampshire (February 11[5]), Nevada (February 22) and South Carolina (February 29)—higher than subsequent contests, which in turn I weigh higher than national polls:

  • Iowa, New Hampshire = 5
  • Nevada, South Carolina = 4
  • The weighted-adjusted average of subsequent state contests = 2
  • National polls = 1

The weighting formula to combine 12 WAPA into a single NSW-WAPA is thus:

(IA*5 + NH*5 + NV*4 + SC*4 + OtherStateAverage*2 +National)/21

Yes, the results from Iowa and New Hampshire will change the subsequent national and state polling, but in ways that are currently unknowable. This formula represents a compromise between underweighting the early contests on one hand and overcomplicating the formula on the other hand.

Also, I chose to use an average of the WAPA from the post-South-Carolina contests because, as of this writing, I only have a single poll from each of seven such contests. I may revisit this decision as more polls are conducted of 2020 Democratic presidential nomination preferences in these states.

**********

As of March 31, 2019, here is the relative position of 2020 Democratic presidential nomination candidates, regardless of announcement status.

Table 1: Weighted-adjusted polling averages for selected 2020 Democratic presidential nomination possibilities

Candidate Running? National Iowa NH NSW-WAPA
Biden ??? 29.9 26.6 24.5 28.5
Sanders Yes 20.6 18.5 23.5 20.8
Harris Yes 9.9 9.0 11.9 10.1
Warren Yes 6.5 8.4 8.4 8.2
O’Rourke Yes 7.2 5.1 5.0 6.2
Booker Yes 4.0 4.5 3.8 4.2
Klobuchar Yes 2.0 3.4 4.4 2.7
Buttigieg Yes 1.1 4.2 0.6 2.4
Gillibrand Yes 0.8 0.3 1.4 1.1
Gabbard Yes 0.5 0.4 0.3 0.9
Yang Yes 0.5 0.1 0.0 0.8
Castro Yes 1.0 0.9 0.2 0.6
Bloomberg No 1.3 0.0 1.5 0.4
Brown No 0.6 0.5 0.7 0.3
Delaney Yes 0.4 0.7 0.3 0.3
Hickenlooper Yes 0.7 0.3 0.0 0.2
Inslee Yes 0.4 0.6 0.0 0.2
H. Clinton No 1.6 0.0 0.0 0.09
Swalwell ??? 0.0 0.1 0.0 0.09
Bullock ??? 0.2 0.2 0.0 0.06
Bennet ??? 0.0 0.2 0.0 0.05
M. Obama No 0.7 0.0 0.0 0.03
J. Kennedy III No 0.0 0.0 0.1 0.03
McAuliffe ??? 0.1 0.0 0.0 0.02
Holder No 0.2 0.0 0.0 0.02
Kerry No 0.3 0.0 0.0 0.02
de Blasio ??? 0.1 0.0 0.0 0.01
Moulton ??? 0.0 0.0 0.0 0.007
Winfrey No 0.1 0.0 0.0 0.005
Abrams ??? 0.1 0.0 0.0 0.004
Williamson Yes 0.1 0.0 0.0 0.004
Steyer No 0.0 0.0 0.01 0.003
Cuomo No 0.02 0.0 0.0 0.001
Messam Yes 0.0 0.0 0.0 0.0
Ojeda No 0.0 0.0 0.0 0.0
Ryan Yes 0.0 0.0 0.0 0.0
7 Others No 0.0 0.0 0.0 0.0
DK/Other 9.1 16.0 13.3 11.7

There is a glorious absurdity to listing candidates averaging well under 1.0 percentage points, but 32 candidates garnered at least 1% support in one or more of the 59 polls assessed. However, given that one Democratic National Committee criterion for participating in the first 2020 Democratic presidential debate (hosted by NBC, MSNBC and Telemundo June 26-27, 2019[6]) is to register at 1% in at least three designated national polls released after January 1, 2019, even that low level of support matters.

The data in Table 1 suggest the following as of March 31, 2019:

  1. The leading candidate, Biden (28.5%), has not yet formally announced whether he will run for president in 2020.
  2. On average, more than half (50.7%) of poll respondents choose someone other than Biden or Vermont Senator Bernie Sanders (20.8%) to be the 2020 Democratic presidential nominee.
  3. Competing for 3rd place is California Senator Kamala Harris (10.1%) and Massachusetts Senator Elizabeth Warren (8.2%).
  4. The three candidates whose support in national polls is understated relative to Iowa and New Hampshire polls are Warren, Minnesota Senator Amy Klobuchar (2.7%) and, especially, South Bend, IN Mayor Pete Buttigieg (2.4%; 4.2% WAPA in Iowa). Hawaii Representative Tulsi Gabbard (0.9%) is also buoyed by WAPA of 2.0% in Nevada and 1.4% in South Carolina.
  5. Despite ranking #5 overall, however, former Texas Representative Beto O’Rourke (6.2%) currently does not poll as well in the early states as he does nationally.
  6. Rounding out the top 10 are New Jersey Senator Cory Booker (4.2%) and New York Senator Kirsten Gillibrand (1.1%).
  7. Half of the current top 10 choices for the 2020 Democratic presidential nomination are women—including four Senators!
  8. Just outside the top 10 are entrepreneur Andrew Yang (0.8%) and former Secretary of Housing and Urban Development Julián Castro (0.6%).
  9. Other than Hillary Clinton (0.09%), the top-listed candidates who have declared they are NOT running for president in 2020 are former New York City Mayor Michael Bloomberg (0.4%) and Ohio Senator Sherrod Brown (0.3%).
  10. Both Bloomberg and Brown continue, on average, to poll just ahead of three other announced candidates for the 2020 Democratic presidential nomination: Maryland Representative John Delaney (0.3%), Colorado Governor John Hickenlooper (0.2%) and Washington Governor Jay Inslee (0.2%)
  11. Seven potential 2020 Democratic presidential nomination candidates—California Representative Eric Swalwell (who may announce on April 8), Montana Governor Steve Bullock, Colorado Senator Michael Bennet, former Virginia Governor Terry McAuliffe, New York City Mayor Bill de Blasio, Massachusetts Representative Seth Moulton, former Georgia House Minority Leader Stacey Abrams—all have NSW-WAPA less than 1.0%.
  12. As for the remaining declared 2020 Democratic presidential nomination candidates: author Marianne Williamson (0.004%) is languishing at 15th among declared candidates (and #30 overall), while Messam registered 0% in both polls listing him; Ryan has yet to be included in a poll.
  13. One candidate, former West Virginia State Senator Richard Ojeda (0.0%), withdrew his candidacy on January 25, 2019.
  14. The other potential candidates (all of whom have announced they are not running for president in 2020) to register at least 1% in any of the 59 polls assessed here are former First Lady Michelle Obama (0.03%), Kennedy (0.03%), former Attorney General Eric Holder (0.02%), former Secretary of State John Kerry (0.02%), businesswoman Oprah Winfrey (0.005%), Steyer (0.003%) and New York Governor Andrew Cuomo (0.001%).
  15. The other seven names to be included in at least one national poll (all registering 0% support) are attorney Michael Avenatti, Los Angeles Mayor Eric Garcetti, Virginia Senator Tim Kaine, California Governor Gavin Newson, former Massachusetts Governor Deval Patrick, former Starbucks CEO Howard Schultz and Facebook CEO Mark Zuckerberg.
  16. Despite being asked about 42 different candidates, roughly 1 in 8 respondents did not state a preference or wanted an unlisted candidate.

In sum, there is currently a clear pecking order among potential 2020 Democratic presidential nominees: Biden and Sanders are the leaders, splitting just under half of the overall vote between them, followed by Harris, Warren, O’Rourke and Booker (29% total). Just behind these six candidates are Klobuchar and Buttigieg (5% total), meaning just eight candidates currently have 83% of the vote between them.

The only other candidate to top 1.0% is Gillibrand, though Gabbard—who rounds out the top 10—rounds up to 1.0%, as do Yang and Castro. However, with 16 declared candidates for the 2020 Democratic presidential nomination appearing in at least one poll (plus the just-announced Ryan) and another eight people contemplating running as well, the race is very much in flux with more than 15 months until Democrats convene in Milwaukee on July 13, 2020.

Until next time…

[1] That is, member of the United States House of Representatives.

[2] If the midpoint falls between days, I use the later day.

[3] I treated Morning Consult tracking polls and the Morning Consult/Politico polls as coming from distinct pollsters.

[4] The arithmetic difference between an estimate and the “true” (unmeasured) value.

[5] Post-Iowa primary/caucuses dates obtained here.

[6] It will be held over two nights to accommodate up to 20 candidates, with no more than 10 appearing each night.