2018 Gubernatorial Elections: Where the REAL action is

With the recent—and thoroughly warranted—attention on the excellent Democratic prospects for recapturing control of the United States House of Representatives (“House”) and their improving (though still less than 50%) chance to do the same in the United States Senate (“Senate”) this November 6, there has been insufficient focus on the 36 gubernatorial elections being held simultaneously.

In fact, I would argue that from a long-term perspective (innovative policy making, redistricting following the 2020 United States Census, etc.), this is where the real electoral action is. And currently Democrats only hold 17 governor’s mansions, compared to 32 held by Republicans and Independent Alaska Governor Bill Walker.

I first addressed the importance of governors in June 2017:

“In an age of increasing partisan polarization and Congressional gridlock, governors have emerged as crucial policy leaders far from Washington DC. On the conservative side are recent innovations by Republican Governors such as Sam Brownback of Kansas, Scott Walker of Wisconsin (prompting an unsuccessful 2012 recall election) and Rick Snyder of Michigan. Governors could choose whether or not to accept Medicaid expansion under the Affordable Care Act, as Republican Governor John Kasich of Ohio continues to note.

More recently, Democratic Governors have attempted to block Trump Adminstration actions. Washington’s Jay Inslee was a key leader in blocking iterations of the travel ban. California’s Jerry Brown has emerged as a leader on climate change, especially after President Trump announced the withdrawal of the United States from the Paris Climate Accord.”

In this post, I present an analogous comparison of “fundamentals” (state partisan lean [3W-RDM], expected Democratic “advantage” in 2018 of 8.9 percentage points, incumbency) to the current polling average (WAPA; average Democratic margin of all publicly-available polls conducted in 2018 adjusted for statistical bias and weighted by date and pollster quality) I recently conducted for 2018 Senate races. Unlike those races, however, I calculated a WAPA for all 36 races, as shown in Table 1.

Table 1: Summary of 2018 Polling Data in 2018 Gubernatorial Elections

State # Polls/ Pollsters Raw Margin Bias-Adjusted Margin Average Pollster Rating Adjusted

 Poll Average

Adjusted Pollster Average Final Ave
AL 2/2 D-19.5 D-19.6 2.7 D-18.4 D-18.4 D-18.4
AK 4/3 I-9.3 I-9.5 2.8/2.6 I-9.4 I-9.6 I-9.5
AZ 6/4 D-3.7 D-4.1 2.8/3.1 D-5.9 D-8.2 D-7.1
AR 2/2 D-37.0 D-37.5 3.0 D-37.2 D-37.2 D-37.2
CA 6/6 D+14.3 D+14.5 3.0 D+17.6 D+17.6 D+17.6
CO 2/2 D+6.0 D+5.9 3.0 D+5.8 D+5.8 D+5.8
CT 3/3 D+8.7 D+8.4 2.8 D+9.2 D+9.2 D+9.2
FL 10/7 D+2.9 D+2.8 2.8/2.8 D+3.0 D+3.4 D+3.2
GA 5/5 D+2.4 D+2.2 3.0 D+1.4 D+1.4 D+1.4
HI 2/1 D+25.0 D+25.7 3.3 D+24.8 D+24.8 D+24.8
ID 3/2 D-10.0 D-11.0 2.6/2.5 D-10.7 D-10.9 D-10.8
IL 12/9 D+14.4 D+14.6 2.3/2.4 D+14.6 D+14.7 D+14.7
IA 2/2 D+0.0 D-0.1 3.8 D+3.5 D+3.5 D+3.5
KS 3/2 D-0.3 D-0.9 2.7/2.7 D-1.0 D-0.5 D-0.8
ME 1/1 D+0.0 D-0.6 3.3 D-0.6 D-0.6 D-0.6
MD 6/5 D-14.0 D-14.0 2.8/2.8 D-13.7 D-13.5 D-13.6
MA 3/2 D-36.3 D-36.3 3.6/3.5 D-35.8 D-36.5 D-36.2
MI 10/8 D+8.2 D+8.2 2.8/3.0 D+8.9 D+9.2 D+9.1
MN 3/3 D+6.3 D+6.0 3.5 D+6.1 D+6.1 D+6.1
NE 0  
NV 5/4 D+0.4 D-0.4 2.8/2.6 D+0.2 D+0.2 D+0.2
NH 3/2 D-21.3 D-23.3 3.1/3.2 D-22.1 D-22.1 D-22.1
NM 6/6 D+7.3 D+6.7 2.9 D+6.8 D+6.8 D+6.8
NY 8/5 D+21.4 D+21.4 3.3/3.6 D+20.1 D+19.1 D+19.6
OH 12/9 D-2.3 D-2.5 2.7/2.7 D-0.3 D-0.5 D-0.4
OK 4/2 D+2.0 D+0.6 2.6/2.7 D-0.1 D-0.1 D-0.1
OR 5/5 D+4.2 D+4.5 2.5 D+4.2 D+4.2 D+4.2
PA 7/5 D+14.1 D+14.0 3.1/2.9 D+13.6 D+13.0 D+13.3
RI 3/2 D+1.3 D+1.7 3.1/3.0 D+1.7 D+1.6 D+1.6
SC 2/2 D-7.5 D-7.9 3.2 D-7.8 D-7.8 D-7.8
SD 1/1 D-4.0 D-5.4 2.0 D-5.4 D-5.4 D-5.4
TN 4/4 D-12.0 D-12.2 3.2 D-15.4 D-15.4 D-15.4
TX 11/9 D-14.7 D-14.6 3.0/2.9 D-15.4 D-15.1 D-15.2
VT 0  
WI 8/5 D+3.9 D+3.6 3.5/3.5 D+3.6 D+3.9 D+3.7
WY 0  
AVE 5/4 D-1.5 D-1.8 3.0/3.0 D-1.6 D-1.7 D-1.7

Gubernatorial elections are woefully under-polled: only 163 polls have been conducted for 36 races in 2018, or just 4.5 per election. These polls were conducted by an average of 3.6 pollsters, meaning most pollsters have only polled these races a single time. Only five of these elections—Florida, Illinois, Michigan, Ohio and Texas—have been polled as many as 10 times, with Illinois and Ohio being polled 12 times each. At the same time, Nebraska, Vermont and Wyoming have not been (publicly) polled at all, Maine and South Dakota have only been polled once, and 13 states have only been polled two or three times. Of these 163 polls, just over half (52.1%) were conducted since July 1. Overall, the quality of the polling is marginally better than in the Senate races I analyzed recently: the average poll was conducted by a B-rated pollster. And their skew (according to the FiveThirtyEight.com pollster ratings) has only been slightly pro-Democratic (0.3 percentage points, on average—note, however, that 24 polls were conducted by non-rated pollsters).


This analysis is divided into five parts:

  1. Safe seats
  2. States with Republican governors most likely to elect a Democrat
  3. States with Democratic governors that are (not very) vulnerable.
  4. Popular Republican governors in Democratic states
  5. Alaska

Just as a reminder, “expected” margin is each state’s 3W-RDM plus 8.9 plus party-specific incumbency advantage; I described how I calculate incumbency advantage in my updated Senate race post. The only difference with gubernatorial races is that I used data from the 2014, 2010 and 2006 elections—the last three times these 36 states (excluding New Hampshire and Vermont which hold gubernatorial elections every two years). Also, Republican incumbency advantage, for unknown reasons, dropped from a bonus of 16.2 percentage points (“points”) in 2006 to just 0.5 points in 2010 to a loss of 8.4 points in 2014. Averaging these values yields a Republican gubernatorial incumbency “advantage” of just 2.6 points; Democrats, by contrast, had gubernatorial incumbent bonuese of 24.0, 1.4 and 7.1 point, respectively, for an average of 10.8 points.

Safe seats. Three heavily Democratic states (average 3W-RDM=D+26.4) will remain in Democratic hands. New York Governor Andrew Cuomo is ahead of Duchess County Executive (and four third-party candidates, including former Sex and the City star Cynthia Nixon on the Working Families line) by 19.6 percentage points (“points”) though Cuomo “should” be ahead by 41.3 points; weighting polls 3-1 over fundamentals puts Cuomo ahead about 27 points. Similarly, Hawaii Governor David Ige leads Republican State House Minority Leader Andria Tupola by 24.8 points; an expected lead of 54.0 points works out to Ige being ahead by 34.6 points. Finally, California Lieutenant Governor Gavin Newson leads Republican businessman John Cox by 17.6 percentage points; an expected lead of 32.1 points works out to Newson being ahead by 22.4 percentage points.

Next January, nine solidly Republican states (Alabama, Arkansas, Idaho, Nebraska, South Carolina, South Dakota, Tennessee, Texas, Wyoming; average 3W-RDM=D-27.2) will still have Republican governors (Table 2); incumbents are bold-faced. Because only one poll has been released of the South Dakota governor’s race, WTD is the simple average of Expect and WAPA. The one remotely-possible upset here is South Carolina if the Democratic wave crests high enough.

Table 2: Safe Republican governorships in Republican states

State Democrat Republican Expect WAPA Diff WTD
AL Maddox Ivey D-20.8 D-18.4 D+2.4 D-19.0
AR Henderson Hutchinson D-22.0 D-37.2 D-15.2 D-33.4
ID Jordan Little D-25.3 D-11.0 D+14.3 D-14.6
NE Krist Ricketts D-19.6 n/a n/a n/a
SC Smith McMaster D-8.1 D-7.8 D+0.3 D-7.9
SD Sutton Noem D-16.9 D-5.4 D+11.5 D-11.2
TN Dean Lee D-16.9 D-15.4 D+1.5 D-15.8
TX Valdez Abbott D-9.1 D-15.2 D-6.2 D-13.7
WY Throne Gordon D-36.8 n/a n/a n/a
AVE     D-19.5 D-15.8 D+1.2 D-16.5

Here are the nominees in each election (Democrat listed first):

  • Alabama: Tuscaloosa Mayor Walt Maddox vs. Governor Kay Ivey
  • Arkansas: Former Arkansas Executive Director of Teach for America Jared Henderson vs. Governor Asa Hutchinson
  • Idaho: Former State Representative Paulette Jordan vs. Lieutenant Governor Brad Little
  • Nebraska: State Senator Bob Krist vs. Governor Pete Ricketts
  • South Carolina: State Representative James Smith vs. Governor Henry McMaster
  • South Dakota: State Senate Minority Leader Billie Sutton vs. U.S. Representative Kristi Noem
  • Tennessee: Former Nashville Mayor Karl Dean vs. businessman Bill Lee
  • Texas: Former Dallas County Sheriff Lupe Valdez vs. Governor Greg Abbott
  • Wyoming: Former State House Minority Leader Mary Throne vs. State Treasurer Mark Gordon

States with Republican governors most likely to elect a Democrat. Thirteen states that currently have Republican governors represent the best opportunities for Democrats to pick up governor’s mansions (Table 3); on average, these states lean Republican (average 3W-RDM=D-4.8). However, term limits mean nine states have no incumbent running, creating an opening for strong Democratic challengers. And Iowa’s Kim Reynolds only became governor when Governor Terry Branstad became Ambassador to China in May 2017). Note that for Iowa (2 polls) and Maine (1 poll), WTD is the simple average of WAPA and Expect.

Table 3: States with Republican governors most likely to elect a Democrat

State Democrat Republican Expect WAPA Diff WTD
IL Pritzker Rauner D+21.0 D+14.7 D-6.3 D+16.2
MI Whitmer Schuette D+11.1 D+9.1 D-2.0 D+9.6
NM Lujan-Grisham Pearce D+15.4 D+6.8 D-8.6 D+8.9
ME Mills Moody D+14.8 D-0.6 D-15.4 D+7.1
WI Evers Walker D+7.0 D+3.7 D-3.2 D+4.5
FL Gillum DeSantis D+5.5 D+3.2 D-2.3 D+3.8
IA Hubbell Reynolds D+2.9 D+3.5 D+0.6 D+3.2
NV Sisolak Laxalt D+10.9 D+0.2 D-10.7 D+2.9
GA Abrams Kemp D-0.7 D+1.4 D+2.1 D+0.8
OH Cordray DeWine D+3.1 D-0.3 D-3.4 D+0.5
KS Kelly Kobach D-14.5 D-0.8 D+13.7 D-4.2
AZ Garcia Ducey D-3.5 D-7.1 D-3.6 D-6.2
OK Edmondson Stitt D-29.2 D-0.1 D+29.1 D-7.4
AVE     D+3.4 D+2.6 D-0.8 D+3.1

 Of these elections, the one for which you can most clearly say “stick a fork in it, it’s over” is Illinois. Billionaire businessman Bruce Rauner defeated unpopular incumbent governor Pat Quinn in the Republican 2014 wave by less than four points. Four years later, an equally-unpopular Rauner appears headed for a 16.2-point loss to billionaire venture capitalist J.B. Pritzker.

Two additional races also have the Democrat heavily favored. The specter of Flint’s water crisis hangs over the race in Michigan, where Governor Rick Snyder—who championed the emergency manager law that precipitated the crisis—is term-limited. Former State House Minority Leader Gretchen Whitmer could crack double-digits against state Attorney General Bill Schuette; her adjusted lead in six polls since mid-July is 10.5 points, nearly the “expected” margin. New Mexico, meanwhile, is merely reverting to partisan form (D+6.5) after Governor Susana Martinez won two elections in Republican wave years. U.S. Representative Michelle Lujan Grisham would be the nation’s first Latina governor; she seems headed for a high single-digit win over U.S. Representative Steve Pearce.

Maine’s penchant for supporting Independent candidacies (as in Senator Angus King) likely cost Democrats the governor’s mansion in 2010 and 2014. In 2010, Independent Eliot Cutler won 35.9% of the vote, while Democrat Libby Mitchell only won 18.8%, allowing Republican Paul LePage to win with 37.6% of the vote. LePage proved…controversial…though he still won reelection in 2014 with by just 4.8 points over Democrat Mike Michaud; Cutler’s 8.0% could have made the difference. In 2018, however, state Attorney General Cheryl Mills is in a strong position to defeat businessman Shawn Moody (who won 5.0% as an Independent in 2010). An early August poll showed the race tied (with 22% other/undecided), but Mills “should” be ahead around 15 points; a mid-single-digits win for Mills seems highly plausible.

In four other states with a Republican governor, the Democratic gubernatorial nominee is less-heavily favored. Wisconsin Governor Scott Walker was first elected in the 2010 Republican wave, surviving a recall attempt in 2012 before winning reelection in 2014. This year, state Superintendent of Public Instruction Tony Evers is likely to defeat Walker, whose ill-fated run for president in 2015 did not help him. Evers, who “should” be ahead by 7.0 points, leads by 3.7 points (though only by 2.2 points since mid-August), suggesting a mid-single-digits victory. Florida Governor Rick Scott is term-limited (and running for the Senate). To replace him, Democrats nominated Tallahassee Mayor Andrew Gillum and Republicans nominated U.S. Representative Ron DeSantis. Gillum, who would be the first African-American governor of Florida, rode a progressive insurgency to an upset primary victory while DeSantis decisively embraced President Donald Trump. Gillum, who would be the first Democratic governor of Florida since 1999, leads by 3.2 points (and in all seven polls released since the August 28 primary), slightly lower than the expected 5.5 points, but enough to anticipate a low-single-digits win. Businessman Fred Hubbell looks similarly headed for about a 3-point win in Iowa over Governor Reynolds as both the fundamentals and the polls (only one since January) converge. Finally, Nevada has not elected a Democratic governor since 1990. With popular Governor Brian Sandoval term-limited, however, Clark County Commission Chair Steve Sisolak has an excellent chance to change that. While Sisolak is effectively tied in the polls with state Attorney General Adam Laxalt (D+0.2), he “should” be ahead by 10.9 points. The fact that Laxalt is the grandson of the late Senator Paul Laxalt (and the son of former New Mexico Senator Pete Domenici) may explain the discrepancy. Still, a 2-3 point win for Sisolak appears plausible.


Tallahassee Mayor Andrew Gillum, a Democrat, (left) leads U.S. Representative Ron DeSantis to be the next governor of Florida

Two Republican-leaning states with term-limited Republican governors are pure toss-ups. First, former State House Minority Leader Stacey Abrams is vying to become the first African-American—and first female—governor of Georgia, and its first Democratic governor since 1995. She “should” be 0.7 points down to state Secretary of State Brian Kemp (who also ran a Trump-like ad), but polls show her ahead 1.4 points, which works out to an anticipated margin of D+0.8. And in Ohio, Democrat Richard Cordray, the first director of the Consumer Financial Protection Bureau, is tied in the polls (D-0.3) with state Attorney General (and former Senator) Mike DeWine. Cordray “should” be ahead by 3.1 points, which works out to an anticipated margin of D+0.5.

Finally, there are three Republican states (average 3W-RDM=D-23.7) where polls and/or weak Republican candidates give Democrats hope, merited or not. Arizona Governor Doug Ducey “should” be ahead of Arizona State University Professor David Garcia by 3.5 points, though he actually leads by 7.1 points, which works out to an anticipated margin of 6.2 points. Former Kansas Governor Sam Brownback had approval ratings in the mid-20s when he became Ambassador-at-Large for International Religious Freedom in January 2018. Then, state Secretary of State Kris Kobach—controversial in his own rightnarrowly edged Brownback’s replacement Jeff Colyer in the Republican gubernatorial primary runoff. This has given Democrats hope that State Senator Laura Kelly could be Kansas’ next governor: while she “should” be trailing by 14.5 points, polls show her trailing by less than one point (which still works out to a 4.2-point loss). And in Oklahoma, Democrats chose former state Attorney General Drew Edmondson, and Republicans chose businessman Kevin Stitt, to replace unpopular term-limited governor Mary Fallin. This race should not be remotely close (D-29.2), but polls have this race essentially tied (D-0.1), likely because of Stitt’s position on teacher pay raises. The most plausible outcome, however, remains a high-single-digits Stitt victory.

Bottom line: Democrats are heavily favored to win the governorships of Illinois, Michigan, New Mexico and (more tentatively) Maine, and they are at least modest favorites in Wisconsin, Florida, Iowa and Nevada. They are even-money in Georgia and Ohio, while Arizona, Kansas and Oklahoma appear just out of reach. Still, Democrats could easily net as many as 10 governor’s mansions from this group of 13 election.

States with Democratic governors that are (not very) vulnerable. Six Democratic states (average 3W-RDM=D+7.1) with Democratic governors are the only chance Republicans have to flip a governor’s mansion—at least based on polling (Table 4). And while it is true that Democrats are “underperforming” expectations by an average of 14.7 points in these six states, they still lead by an average of 6.7 points.

Table 4: States with Democratic governors that are (not particularly) vulnerable

State Democrat Republican Expect WAPA Diff WTD
RI Raimondo Fung D+37.7 D+1.6 D-36.1 D+10.6
OR Brown Buehler D+28.4 D+4.2 D-24.3 D+10.2
CO Polis Stapleton D+11.1 D+5.8 D-5.3 D+7.1
MN Walz Johnson D+10.4 D+6.1 D-4.3 D+7.2
CT Lamont Stefanowski D+21.7 D+9.2 D-12.5 D+12.3
PA Wolf Wagner D+19.3 D+13.3 D+6.0 D+14.8

Rhode Island Governor Gina Raimondo only beat Cranston Mayor Allan Fung in 2014 by 4.5 points; Moderate candidate Robert Healey won 21.4% of the vote. Raimondo still has only middling approval, which could explain why she barely leads Fung in a rematch (D+1.6), fully 36.1 points below where she “should” be. Despite appearing headed for high-single-digits win, this is the governor’s race that should most worry Democrats. Less vulnerable is Oregon Governor Kate Brown, the nation’s first openly bisexual governor, though she “only” leads State Representative Knute Buehler by 4.2 points, fully 24.3 points below expectations. Nonetheless, I expect her to win by around 10 points. Three other states with retiring Democratic governors look solid for Democrats:

  • In Colorado, U.S Representative Jared Polis looks like a 7.1-point winner over State Treasure Walker Stapleton
  • In Minnesota, U.S. Representative Tim Walz looks like a 7.2-point winner over Hennepin County Commissioner Jeff Johnson (who beat former two-term governor Tim Pawlenty by almost nine points)
  • In Connecticut, businessman (and 2006 Senate nominee) Ned Lamont looks like a 12.3-point winner over businessman Bob Stefanowski

As for Pennsylvania Governor Tom Wolf, he could easily be on the “Safe” list, as he appears headed for double-digit win over State Senator Scott Wagner.

Bottom line: In these six states with Democratic governors, only Raimondo in Rhode Island seems remotely vulnerable, and even she is somewhat likely to win.

Popular Republican governors in Democratic states. In 2014, Maryland and Massachusetts voters narrowly elected centrist Republicans Larry Hogan and Charlie Baker, respectively, governor. In 2016, New Hampshire and Vermont voters narrowly elected Chris Sununu (son of former governor John Sununu) and Phil Scott, respectively, governor. The common theme seems to be normally Democratic voters (average 3W-RDM=D+18.1, with New Hampshire D+0.1), selecting a moderate “check” on overwhelmingly Democratic legislatures (less so in New Hampshire). Thus, these governors should sail to reelection (Table 5) over former State Senator Molly Kelly (NH), former NAACP CEO and President Ben Jealous (MD), former state Secretary of Administration and Finance Jay Gonzalez (MA) and Vermont Electric Cooperative CEO Christine Hallquist (who would be the first transgendered governor). In fact, in the three states with polling, these Republican governors are over-performing expectations by an average of 45.1 points!

Table 5: Popular Republican governors in Democratic states

State Democrat Republican Expect WAPA Diff WTD
NH Kelly Sununu D+6.4 D-22.1 D-28.4 D-15.0
MD Jealous Hogan D+28.9 D-13.6 D-42.5 D-3.0
MA Gonzalez Baker D+28.4 D-36.2 D-64.5 D-20.0
VT Hallquist Scott D+34.0 n/a n/a n/a
AVE     D+24.4 D-24.0 D-45.1 D-12.7

Bottom line: While Jealous could make a race of it in Maryland, suffice it to say that this lifelong liberal Democrat is voting for Charlie Baker.

Alaska. This is the one state where Republicans are likely to pick up a governorship—by defeating the Independent Walker, who defeated Republican Governor Sean Parnell in 2014 with a Democratic Lieutenant Governor (Brian Mallott). In 2018, Walker will face former Democratic Senator Mark Begich and Republican former State Senator Mike Dunleavy. Dunleavy currently leads both Walker and Begich by between nine and 10 points in what “should” be a toss-up (I+0.5). Multi-candidate races are notoriously tricky to gauge, but the likelihood is that Walker and Begich split the non-Republican vote, giving Dunleavy a high-single-digits win.

Conclusion. Democrats need to net eight governor’s mansions to have a 25-25 split nationally. As of September 16, 2018, they appear well on their way to doing just that. They are clear favorites in Illinois, Michigan and New Mexico (and probably Maine), and they are likely also to prevail in Florida, Iowa, Nevada and Wisconsin. Georgia and Ohio are toss-ups, while Arizona, Kansas and Oklahoma may be just out of reach. Only in Rhode Island do Republicans have even a remote chance of netting a governorship (and Raimondo is still favored), while they will almost certainly flip Alaska from Independent to Republican.

Overall (and with all necessary warnings about polling accuracy, unforeseen events and margins of error), the (very unlikely) worst-case scenario is Democrats net two governor’s mansions, while the (very unlikely) best-case scenario is they net 13 (or more) governor’s mansions. The likeliest outcome is a net of between six and 10 governor’s mansions, with my money on the higher end of that range.

Until next time…

Leave a Reply

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

WordPress.com Logo

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

Google+ photo

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

Twitter picture

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

Facebook photo

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

Connecting to %s