2018 Senate: How the “Trump Ten” Races Compared to 2016
A Commentary By J. Miles Coleman
KEY POINTS FROM THIS ARTICLE
— Last week, Republicans tried to knock off incumbent Democratic senators in 10 states won by President Trump in 2016. While some of these Republican challengers won, nine of the 10 ran well behind Donald Trump’s showing in 2016, perhaps not surprisingly given that all of the Democrats in these states had the power of incumbency and the political environment was generally pro-Democratic overall. The one exception where Trump’s performance wasn’t that much different from the 2018 GOP Senate showing was Florida, which is in the midst of a recount.
— The following maps compare Trump’s 2016 performance in these 10 states to the performance by the GOP Senate candidates in each state.
— Most of these maps measure the relative county-level showings by GOP Senate candidates compared to Trump two years earlier. While this is explained in much more detail below, keep this in mind: The red counties in the relative performance maps are where the GOP Senate candidates declined less than their statewide underperformance of Trump, while the blue counties are where the Republican Senate candidates declined more.
— The comparisons help to illustrate the places where Trump may be more popular than other Republicans, and where other Republicans may have more strength than Trump.
Tracking GOP performance in Senate Democrats’ “Trump Ten”
Heading into the 2018 cycle, Democrats seemed to have many advantages, as the out-party typically does in midterm years. However, one factor that was decidedly slanted against them was the Senate map. A majority of the Democratic caucus — 26 of 49 members — faced the electorate. Further, 10 Democratic incumbents on the ballot represented states that President Trump carried in 2016. In many cases, to win reelection, these senators had to perform significantly better than Hillary Clinton did two years ago.
Going state by state, we’ll see how these Republican challengers compared to the president. All told, Democrats hung on to at least six of these Senate seats (Michigan, Montana, Ohio, Pennsylvania, West Virginia, and Wisconsin), while Republicans won three (Indiana, Missouri, and North Dakota) and very narrowly lead in a fourth (Florida) pending the completion of a statewide recount.
Keep in mind, too, that vote counts are not necessarily final in many of these states, such as Florida and Ohio, although they shouldn’t change dramatically for the purposes of these comparisons.
From a strictly numerical perspective, West Virginia could have been considered the top pickup opportunity for the GOP. Trump won the Mountain State by 42 points, his biggest win among the 10 states included in the Trump Ten.
Still, Sen. Joe Manchin (D), from his decades in public life, was a known commodity. During his time in the Senate, Manchin cast his votes carefully, which ultimately helped him beat state Attorney General Patrick Morrisey (R). Compared to Trump, Morrisey did significantly worse in every county:
NOTE: For any of the following maps included in this article, click on the map to see a larger version.
Map 1: West Virginia, Trump 2016 vs. Morrisey 2018
Map 1 certainly shows that Morrisey broadly ran behind Trump. In the next map, and in subsequent states, we’ll look at the relative margin by county. Overall, Morrisey did 45.0% worse than Trump statewide (he lost by 3.3% compared to Trump winning by 41.7%). In counties shaded in red, Morrisey’s deficit compared to Trump in 2016 was less than 45.0%, while in blue counties, it was greater than 45.0%.
Just to make sure this is perfectly clear, let’s give a couple of examples. In McDowell County — the southern tip of West Virginia — Trump won by about 51.1 percentage points in 2016. Two years later, Morrisey lost the county by about 0.4 points. So Morrisey ran about 51.5 points behind Trump in that county, a deficit greater than the 45.0 points he ran behind Trump’s margin statewide. So that’s why McDowell is shaded in blue. On the flipside, Trump won Hancock County — the northern tip of the state — by 44.1 points. Morrisey carried it by 2.5 points. So the county is shaded in red because the difference between Trump’s 2016 margin there and Morrisey’s margin (41.6%) is less than the statewide difference (45.0%). As you look at these maps, just remember this: The red counties are where GOP Senate candidates ran ahead of their statewide decline compared to Trump’s showing from two years prior, while the blue counties are where those Republican Senate candidates ran behind their statewide Trump deficit.
The remainder of the maps that follow for all 10 states will show these relative changes between Trump in 2016 and GOP Senate performance in 2018:
Map 2: West Virginia, relative county-level performance, Trump 2016 vs. Morrisey 2018
The relative margin map shows the trends more clearly than the simple margin map, with its ubiquitous blue. Not surprisingly, the region where Morrisey held up best was the panhandle. This region, which is increasingly becoming part of the Washington D.C. suburbs/exurbs, was a key part of Morrisey’s win in his competitive primary victory back in May. Because the area has relatively more transplants than other parts of West Virginia, the name recognition of an incumbent like Manchin could carry less weight. Additionally, the panhandle also has generally been more Republican in some recent statewide Senate and gubernatorial contests than much of the rest of the state. By contrast, and while his margins there weren’t as strong as in past years, the southern coalfields have a certain loyalty to Manchin. Morrisey ran furthest behind Trump in this ancestrally Democratic, heavily unionized area.
After West Virginia, North Dakota voted for the president by the largest margin of the Trump Ten, nearly 36%. Sen. Heidi Heitkamp (D) won a very close race in 2012, but this time she lost by 11% to Rep. Kevin Cramer (R, ND-AL). Looking at the relative margin map of Trump vs Cramer, a clear east against west divide emerges:
Map 3: North Dakota, relative county-level performance, Trump 2016 vs. Cramer 2018
Heitkamp was raised in Richland County, the bluest county in the southeast, so she had a considerable personal vote there. Cramer likewise underperformed Trump in heavily Native American counties, such as Sioux and Rolette.
Since the mid-2000s, the western part of the state has seen explosive growth from the oil industry; politically, this has driven the state rightward. McKenzie County, on the western border, is a great example of this. It was at the center of the oil boom; it cast 781 more votes in 2018 than 2012, and Heitkamp’s deficit there went from 28% to 44%.
In Ohio, Sen. Sherrod Brown (D) won reelection to a third term. After initially running for governor, Rep. Jim Renacci (R, OH-16) switched to this race. Brown was reelected, and Renacci did 14.5% worse than Trump. While Appalachia and the Mahoning Valley in the eastern part of the state have generally moved away from Democrats, Trump’s 2016 numbers in those regions are difficult for state Republicans to replicate, especially in competitive races.
Something that stands out on the map is Renacci’s contiguous line of relative strength. It runs through the state’s three major urban centers. In the Cincinnati and Columbus areas, Trump ran notably behind the typical Republican, so Renacci had room for improvement. Up north in Cleveland (Cuyahoga County), Renacci’s relative strength may have more to do with a drop-off in minority voters.
Map 4: Ohio, relative county-level performance, Trump 2016 vs. Renacci 2018
Pennsylvania saw a notable east/west divide. President Trump narrowly flipped the Keystone State in 2016; this year, one of his strongest supporters in Congress, Rep. Lou Barletta (R, PA-11), ran for this seat. After initially looking competitive, Barletta’s campaign faded; he ultimately lost by 13% to Sen. Bob Casey (D).
Casey’s family has been popular in Pennsylvania, especially the west, since his father’s days as governor in the 1980s and 1990s. The western part of the state, especially south of Pittsburgh, was where Barletta fared the worst in relation to Trump.
Barletta did better in many urban centers in the east, where he had room to improve over Trump’s relatively weak showing in 2016. Philadelphia, as well as the state capital of Harrisburg (Dauphin County), gave Barletta some of his best relative showings.
Map 5: Pennsylvania, relative county-level performance, Trump 2016 vs. Barletta 2018
At the beginning of the cycle, Wisconsin seemed like it had the potential to be a top-tier contest, but Sen. Tammy Baldwin (D) won a second term by double digits in a state that the president narrowly carried. State Sen. Leah Vukmir (R) had a base in strongly Republican suburban counties surrounding Milwaukee, which helped her win a competitive primary. As a result, she held up best in the southeastern part of the state. Waukesha County is a great illustration of Vukmir’s relative strength. In 2016, Trump was the first Republican presidential nominee this century to take less than 60% in this reliably GOP county. This left Vukmir with ample room for improvement. However, Vukmir underperformed Trump in the swingier area of the state — basically everything outside of the Milwaukee to Madison stretch in the south.
Map 6: Wisconsin, relative county-level performance, Trump 2016 vs. Vukmir 2018
In Indiana, Sen. Joe Donnelly (D) was defeated in his bid for a second term by former state Rep. Mike Braun (R). Before winning his Senate seat, Donnelly served six years in the House, representing the South Bend area. Donnelly’s old district is essentially the bright blue in the northern part of the state; Braun had the hardest time running ahead of Trump here. Southwestern Indiana was once swingy, but in 2016, Trump posted atypically strong numbers there. Throughout much of the rest of the state, Braun held a narrow but consistent relative advantage over Trump.
Map 7: Indiana, relative county-level performance, Trump 2016 vs. Braun 2018
Sen. Debbie Stabenow (D) had a closer race than some expected against businessman and veteran John James (R). While Trump carried Michigan in 2016 by about two-tenths of a percentage point, James lost by 6.5%. Trump’s strongest regions were the “thumb” as well as the Upper Peninsula. These areas are heavily white, rural, and have a fading union presence — in other words, exactly the types of places that swung hard to Trump in other parts of the country. James, by contrast, held up better in the Detroit and Grand Rapids areas. The president ran significantly behind the partisan baseline with suburban voters there.
Map 8: Michigan, relative county-level performance, Trump 2016 vs. James 2018
In Missouri, Sen. Claire McCaskill (D) was denied a third term by state Attorney General Josh Hawley (R). Hawley did 12.5% worse than the president, winning by 6% instead of 18.5%. On the relative margin map, Trump’s best counties create something of a crescent around the St. Louis area. In fact, four of the counties that gave Trump his best relative margin — Jefferson, Washington, Iron, and Ste. Genevieve — all supported President Obama in 2008. While McCaskill still lost them, Hawley’s weaker showing suggests that state Democrats still have some residual support there.
Map 9: Missouri, relative county-level performance, Trump 2016 vs. Hawley 2018
Sen. Jon Tester of Montana (D) has been a top target for Senate Republicans over the past few cycles but won a third term last week. State Auditor Matt Rosendale (R) came up about 3.6% short against Tester in a state the president carried by over 20%. Before he was elected auditor in 2016, Rosendale represented a state Senate seat in the eastern part of the state. Not surprisingly, he did relatively well in this area. By contrast, in the counties that Tester was born and raised in, Chouteau and Hill, Rosendale struggled the most.
Map 10: Montana, relative county-level performance, Trump 2016 vs. Rosendale 2018
Florida is perhaps the state where our “relative margin” metric is least useful, as the 2018 Senate result is only about 1% different than the 2016 presidential result. While small, that 1% is certainly significant — while votes are still being counted in some counties, as of this writing Gov. Rick Scott (R) is leading three-term Sen. Bill Nelson (D) by just .15%.
Compared to Trump, Scott was able to make gains in Miami-Dade County, likely due to the fact that Trump underperformed in 2016 with typically Republican Cuban Americans, some of whom then must have backed Scott. Scott likewise performed well in his home county, Collier (Naples), as well as Osceola; the Scott campaign touted its outreach to Hispanics during the campaign, and this result in heavily Puerto Rican Osceola may be some vindication of that. In 2016, Trump posted massive margins in the panhandle, which has been traditionally friendlier to local Democrats. While Nelson did lose support there from his past races, Scott nonetheless was not able to put up margins as crushing as Trump’s in 2016.
Map 11: Florida, relative county-level performance, Trump 2016 vs. Scott 2018
Overall, Republican Senate candidates ran well behind Trump in nine of these 10 states, although that didn’t prevent at least three of those candidates, Sens.-elect Mike Braun (R-IN), Josh Hawley (R-MO) and Kevin Cramer (R-ND), from winning. The one exception, Florida, has yet to be officially decided, although Rick Scott (R) seems to have a leg up on Bill Nelson (D). Meanwhile, as we look ahead to 2020, Trump will need to replicate his 2016 performance — and not the performance of GOP Senate candidates — to carry key states that voted for him in 2016 such as Michigan, Ohio, Pennsylvania, and Wisconsin.
With presidential and downballot coalitions becoming increasingly similar, relative margin offers a way to measure a candidate’s unique geographic strengths and weaknesses.
See Other Political Commentary.
Views expressed in this column are those of the author, not those of Rasmussen Reports. Comments about this content should be directed to the author or syndicate.
Rasmussen Reports is a media company specializing in the collection, publication and distribution of public opinion information.
We conduct public opinion polls on a variety of topics to inform our audience on events in the news and other topics of interest. To ensure editorial control and independence, we pay for the polls ourselves and generate revenue through the sale of subscriptions, sponsorships, and advertising. Nightly polling on politics, business and lifestyle topics provides the content to update the Rasmussen Reports web site many times each day. If it's in the news, it's in our polls. Additionally, the data drives a daily update newsletter and various media outlets across the country.
Some information, including the Rasmussen Reports daily Presidential Tracking Poll and commentaries are available for free to the general public. Subscriptions are available for $4.95 a month or 34.95 a year that provide subscribers with exclusive access to more than 20 stories per week on upcoming elections, consumer confidence, and issues that affect us all. For those who are really into the numbers, Platinum Members can review demographic crosstabs and a full history of our data.
To learn more about our methodology, click here.