Your browser does not support JavaScript Facing Hostile Senate Map, Democrats Will Need a Tsunami, Not a Wave, to Reclaim Senate - Judy Ford Wason Center for Public Policy - Christopher Newport University Facing Hostile Senate Map, Democrats Will Need a Tsunami, Not a Wave, to Reclaim Senate
Skip navigation

Judy Ford Wason Center for Public Policy

Wason Center

September 26, 2018

Facing Hostile Senate Map, Democrats Will Need a Tsunami, Not a Wave, to Reclaim Senate

National / Midterm / Elections

Image of a wave.

Here’s an interesting statistic. The reelection record for Senate incumbents is 38 out of 38 in Senate races when their side is on the right side of a midterm effect. Conversely, this means that the party trying to unseat an incumbent senator in a cycle where they are on the receiving end of a midterm effect, let alone one the size of what’s expected in 2018, has never succeeded in the task. The 2018 cycle may be the cycle to finally break this record though, due to the particular Senate seats up for reelection. Democrats find themselves defending seats in  9 states that Trump carried in the 2016 presidential election, 5 of which are particularly fertile ground for Republicans. The 2018 map is quite similar to the 2014 map, which saw Democrats lose incumbents in Arkansas, Louisiana, North Carolina, Colorado, and Alaska, effectively wiping out most of the so-called Blue Dog coalition. Strong Democratic challengers in Kentucky and Georgia also came up empty, even after the party spent millions trying to unseat Senate Majority Leader Mitch McConnell and claim an open Senate seat in Georgia. But there is a key difference for Democrats in 2018, rather than being on the receiving end of the midterm effect, they will now be beneficiaries of what is increasingly looking like a very robust midterm effect.

Given the 7-10 point advantage Democrats have on the generic ballot, and the predictive power that demographic factors and the partisan competitiveness has been found to have on the two-party vote share of Democrats in elections since Trump, it’s feasible that Republicans will come up empty in all 9 senate races. Indeed, several have already slipped out of their grasp, including Sherrod Brown’s seat in Ohio, the “Trumpiest” state of all in the Democrat’s former “Blue Wall” in the Midwest. And in Virginia’s Senate election, the trouncing of Ed Gillespie in what should have been a competitive race in 2017 deterred big-name Republicans from getting into the race, which allowed a controversial Republican with a problematic history of ties to white nationalism to become the party’s nominee. Indeed, it's not infeasible that Democrats will not only hold onto all nine of their seats; they may well pick up Republican-held seats in Arizona, Nevada, Tennessee, and Texas. Due to the uncertainty of turnout of Democratic voters, this cycle could easily end with Democrats controlling both chambers of Congress come January or winning 30+ House seats while still losing 4 or more Senate seats.

I can’t stress enough how much this electoral context is conditioned on the controversial Trump presidency. The in-party controlling the White House always faces hostile midterms but because Trump offends nearly every sensibility governing American political culture, this cycle’s backlash will be especially pronounced. As I lay out in detail in my House forecasts, Democrats’ turnout in 2018 is all but certain to mirror, and in some places even exceed, their high turnout in 2006. And now that we have a much clearer picture of how Independents evaluate Donald Trump, we can consider the implications it will have on the Senate map. Independents are your quintessential “medium voters” and tend to hover somewhere in the 40s on their favorability of the current POTUS unless something like a scandal, controversy, or economic collapse pushes them to the extremes.

When independents abandon moderation, it is a very bad sign for the incumbent party. In the run-up to the 2006 midterms, George W. Bush’s approval among Independents cratered, spending the bulk of the election year in the low 30s. And Independents joined with Democrats to punish Republicans in that midterm, helping Democrats net 31 seats in the House and six Senate seats. Exit polling reveals Independents broke 59% to 41% that cycle. And Independents are sensitive to the midterm effect, even when they aren’t hostile to the incumbent president. In both the 2010 and 2014 midterms  Democrats got trounced by Republicans even while Independents expressed fairly positive views of President Obama.

Despite a nearly 50% approval rating in 2010. Democrats lost an astounding 63 House seats that cycle, and Independents broke for Republicans by the exact same margin they broke for Democrats in 2006: 59% to 41%. The same is true for the 2014 cycle. Obama’s approval rating was in the low 40’s among Independents and Independents broke for Republican House candidates by 12 points.

So how will Independents react to an incumbent Republican Party headed by Donald Trump, whose approval rating among Independents hovers in the mid to high 30s? History suggests they will break strongly in favor of Democrats, and combined with the big uptick in participation by riled up Democrats, this is almost certain to produce senate race outcomes that would have been unfathomable with a Democrat in the White House.

In order to estimate expected two-party vote share of Democrats running in competitive Senate races, I created state-level PVI scores (partisan voting index) following the Cook Political Report methodology they use to estimate scores for House districts. For those of you that have already taken time to read my forecast for the House races, you will know that PVI scores are found to be extremely powerful predictors of Democratic candidates’ two-party vote share in the Virginia 2017 elections. However, it should be noted that PVI’s are NOT predictors of special election outcomes since Trump’s election. Democrats won races when PVI’s were both small and large. Still, these state-level PVI’s can give us a useful baseline for understanding just how receptive or hostile some of these races are for Democrats and I expect that when I model the results of 2018, they will be statistically significant predictors of Democratic performance. I have included states that have competitive Senate or gubernatorial races this cycle and coded states that have elected at least one statewide officeholder of one party and one or more statewide office holder for a different party (denoted in purple).

In terms of PVIs the outlook would look pretty good for the GOP to knock off some incumbent Democrats. Several of the seats they hope to pick up from the Democrats strongly favor Republicans. One such state is West Virginia, where Republicans hold a nearly 9 point advantage and Donald Trump won the popular vote by 42 points. Even better is North Dakota, where Heidi Heitkamp is running for reelection in a state whose partisan composition  favors her Republican opponent by 14 points. PVIs also show Republicans in strong positions in open races in Tennessee and to a lesser extent, Arizona. In Texas, Ted Cruz is trying to stave of the insurgent candidacy of Democrat Beto O’Rourke, who as expected (by me, at least) has managed to tap into the enthusiasm of Democrats to make Texas a possible pick up target. In Tennessee, Democrats are running an ideal nominee in Phil Bredesen, a former governor of the state known for his bi-partisan appeal. Both of these candidates face an uphill battle based on the partisan composition of their state (Texas is R+8 and Tennessee is R+13). Yet, polling shows both are within reach of winning.

One of the insights garnered from my Virginia analysis is the ability of certain demographic characteristics to offset Republican PVI advantages. In special elections, federal and state legislative districts with high numbers of college educated voters were able to overcome large, structural advantages for Republicans, at least in those open races. States that are more urbanized, diverse, and educated are well situated to produce a surge in the turnout of Democratic voters. And although Trump’s approval ratings in red states are higher than the national average, evidence from the Texas Politics Project at the University of Texas, Austin suggests that Trump may not be that much of an asset to Ted Cruz. In their June 2018 poll, 38% of respondents strongly disapproved of the president, nearly 10pts more than those who indicated they strongly approved (29%).

But the real trouble for Cruz is in the cross tabs. Although 25% of Independents strongly approve of Trump, 39% strongly disapprove. If Independents base their Senate vote on their support of Donald Trump, Ted Cruz could be in for the fight of his life. Because Republicans greatly outnumber Democrats, the only way that O’Rourke could actually win the race is if Democratic Party turnout surges a lot and Independents break in favor of him by at least 5 points. Recent polling from Marist/NBC News indicates this may well be possible. The poll finds Cruz and O’Rourke currently splitting Independents evenly while Greg Abbott, Texas’ popular Republican governor is winning Independents handily.

Increasingly, it looks like the behavior of Independents will mirror their behavior in the 2006 cycle. Recent polls from NBC/Marist finds that in Arizona, Democrat Kyrsten Sinema leads Republican Martha McSally by 17 points among Independents. In Ohio, incumbent Democrat Sherrod Brown leads his Republican challenger Jim Renacci by 21 points. And in Florida, incumbent Democrat Bill Nelson leads Republican Rick Scott, the state’s current governor, by 9 points. Should indepedents break strongly in favor of Democrats in many of these Trump states, Democrats should be able to avoid what would otherwise be a bruising senate cycle.

To arrive at my senate (and gubernatorial) race ratings I estimate the expected share of the two-party vote for each Democratic Party candidate using my demographics model. This provides a baseline for the expected two-party vote share of each Democratic candidate for each race. The variables included in this model include the state’s PVI score, the % of the state’s population that is college educated, and the population density of each state, and the diversity level of the state. Other factors are then considered that may add or subtract from the Democratic candidate’s expected two-party vote share. These factors include incumbency, fundraising, party support, primary turnout, registration data, outside spending, candidate quality, polling, the presence of competitive house races and/or another competitive statewide race, and campaign strategy. Predictions will be refined as new data becomes available.

Toss Ups

My initial ratings produce 4 pure toss ups: Tennessee, Florida, Texas, and Missouri. I need to make an important point regarding Florida. This race would be ranked as Lean Republican right now if the Democratic nominee for governor was the Democratic Party’s preferred candidate Gwen Graham. Like Nelson, Graham is a centrist (survivable) who lacks charisma and perhaps more importantly, an understanding of how much the electorate has changed over the past decade (not survivable). And with Ron DeSantis as the party’s gubernatorial nominee, Rick Scott is well-positioned to run as a moderate, even though he has an 8 year track record as governor as a solid conservative. Instead, Nelson will be joined by Andrew Gillum, who as an unabashed progressive Democrat and the state’s first African American nominee, will tap right into the energy that permeates the Democratic Party’s base in the Trump Era. If the Nelson campaign has the sense to tether their candidate to Gillum, it may be enough to offset the lack of enthusiasm Florida Democrats have toward Nelson. Demographically, Florida is capable of producing a surge in Democratic turnout, and several House districts currently occupied by Republicans have strong potential to flip to the Democrats. Adding Gillum to this mix adds to the already significant enthusiasm Democrats hold, and may be enough to push this race in Nelson’s favor. That said, given Nelson’s centrism and the pervasive, but erroneous, belief among the Democratic Party’s most established strategists that moderate, boring candidates are an asset, it is possible that the Nelson campaign will decide to keep Gillum at a distance. This would be a fatal miscalculation and one of the most important things I’ll be looking for as we come into the final stretch is how the Nelson campaign positions themselves regarding Gillum. Rick Scott is a formidable candidate, and is running a shrewd, strategic campaign as evidenced by the tweet his campaign posted distancing himself from Trump’s Puerto Rico twitter tantrum during Hurricane Florence. Scott has out-raised Nelson by $21 million and has been spending almost all of it in an ad blitz while Nelson has been largely invisible. Due to this massive spending Nelson reports $14 million dollars cash on hand whereas Scott only has $3 million heading into the final stretch. Although Republican turnout increased in the primary over 2014, turnout on the Democrat’s side increased sharply, up 88% over 2014. More importantly though, Democrats managed to get within striking distance of Republicans in terms of total ballots cast reaching 48% of total ballots. These are all good signs for Nelson. 

Tennessee is interesting. Everything I read and hear says the Democrats are bullish on Bredesen picking up this seat. But with a PVI score of R+14 and a quality, Trump Republican nominee with a significant national profile, this race is one in which a lot will have to go right for Democrats to pull off a victory. Yes, Bredeson is a popular, former two-term governor of the state with some bi-partisan appeal who left office fairly recently and whose favorability rating in June was 67%, 17 points higher than his Republican opponent’s rating. And Tennessee is another low-turnout state. Just 28% of eligible voters cast ballot’s in the 2014 midterms. Low turnout decreased the probability of Democrats winning competitive elections during the Obama years but under the Trump Administration Democrats have made up huge partisan disadvantage gaps in special elections with low turnout. Lackluster turnout among Republicans, combined with a surge in Democratic Party turnout, can overcome Republican’s double digit PVI advantage, just as it has done in numerous special elections. And in Tennessee, as in the majority of states, Democratic primary turnout was way up, up 55% over the 2014 midterms. That said, Democrats cast just 32% of the ballots in the primaries, 380,000 compared to the 723,000 cast in the Republican primary and recent polling shows Republicans actually lead the generic ballot question there, at least among registered voters. Bredesen leads among Independents by 7 points. And looking back at Tennessee over the course of the past decade it is clear that the state has turned a lot more Republican since Phil Bredeson won his initial bid for governor in 2002 in a close contest. When Bredesen pulled off his victory, Tennessee’s PVI score was just +5 Republican. It’s now triple that. With a PVI of R +14 and a college education rate of just 32%, my quantitative model spits out an estimated two-party vote share of just 33% for Bredesen. So the qualitative aspects of my ratings are overriding what the model predicts for this race, for now. Its quite possible that some of the white, working class voters who left the Democratic Party over the past decade of the state’s ongoing Republican realignment might have found memories of Bredesen’s tenure as governor. For him to win, this will have to be the case. Bredesen is pulling over some support that is going to the Republican nominee in the governor’s race. He is outperforming the Democrat’s gubernatorial nominee by roughly 10 pts. Bredesen is facing the same level of hostile electorate that Manchin and Tester face, without the benefit of incumbency, but he is a known quantity as a former governor. For now I am manually handicapping this race, which statistically is a “likely Republican” race, into the toss-up category for now.

Can Democrat Beto O’Rourke actually win in deep red Texas? Truthfully, I don’t know. Those who have been following my predictions since they first launched on July 1st know that I predicted Texas’ senate race would be competitive way back when Cruz’s polling average advantage was 9+ points and everyone thought I was crazy. The fact that we’re talking about Texas as toss up is a testament to two things. First, and most important, the fundamental advantage Democrats have in this cycle and second, how important a good candidate is.  I recognized immediately that Beto O’Rourke was going to draw comparisons to 2004 Barack Obama, so I am not surprised to hear others invoking that comparison now. That level of candidate quality is a rare thing, and combined with the enthusiasm Democrats have it may just be enough to pull a W out IF (and you’ll note, that literally a big if) Latinos finally increase their participation rates. It wouldn’t take much, mind you. But political scientists have been waiting in vain for the Latino surge for most of the past decade and it has yet to manifest. It certainly didn’t in 2016 when many thoughts Trump’s hostility to Latino immigrants would drive up participation. The theoretical threat of Trump’s campaign promises is now nearly two years into tangible policies including massive crackdowns by ICE as well as the “zero tolerance” policy that led to the deeply unpopular family separation policy. One of the things I’ll be looking closely for in Texas and Florida both is some actual evidence of backlash among Latino voters. But as special as Beto O’Rourke is, make no mistake about it. If Hillary Clinton was president, there is no way he’d be competitive in this race.

In Missouri, Claire McCaskill maintains a modest advantage in the polling average over her Republican challenger Josh Hawley.  Hawley is dogged by the perception that he lacks the focus needed to unseat a formidable incumbent. Another advantage for McCaskill (and other Democrats in red states) are Trump’s trade tariffs, which are not popular in Missouri. Hawley certainly lags far behind McCaskill in terms of fundraising. She has raised $22 million dollars compared to Hawley who has only raised $5 million, a deficit he will need to close. Still, at R+9, this is not a shoe in for McCaskill.  As such, this race is handicapped from Lean Republican to toss up. As in all of the senate races there are three ongoing issues that may affect the final classification of this race including the ongoing effects of Trump’s tariffs, the confirmation vote of Brett Kavanaugh which became even more complicated after a women came forward to accuse him of sexual assault decades ago, and possible new indictments in the Mueller probe connected to the plea deal Paul Manafort has reached pledging his full cooperation in the Russia investigation.  I should also note that Obamacare will take center stage for the 5th cycle in a row. Only this time, the politics have inversed. After the GOP’s “repeal and replace” slogan was revealed to be empty campaign jargon and the public got a good look at Republican replacement options, it is no longer en vogue to run on killing the Affordable Care Act. Now Democrats are running on healthcare and Republicans are running away from healthcare. In the 2017 Virginia elections as well as the special election in Pennsylvania CD 18, health care was the most salient issue, and voters overwhelmingly preferred Democrats’ approach of shoring up Obamacare over the Republicans’ actual “repeal and replace” plans. Top on the list for voters is protection of the preexisting conditions ban, which disallows health insurance providers from excluding people who have had previous health conditions such as cancer or heart disease, or ongoing conditions such as diabetes. I expect Democrats will make protecting Obamacare preexisting conditions rules a centerpiece of their health care messaging, and if they are smart, they will hammer their Republican opponents on healthcare both in their direct mailers and ads and Joe Manchin has already reprised his famous shotgun ad to take  “dead aim” at efforts by Republican attorney generals to gut it. 

Leans Democrat (Democrat Incumbent) 

Speaking of Manchin and his colleague in Montana. With a PVI of R+19 in West Virginia and R+9 in Montana, there is no doubt that both Manchin and Tester are facing hostile electorates.  Both states broke for Donald Trump in 2016 by larger margins than they went for Mitt Romney over Barack Obama in 2012.  In terms of the Montana senate race, either candidate has the edge depending on the polling aggregation you use. Tester is trying to position himself as appropriately supportive of Trump, choosing to highlight legislation he’s sponsored on behalf of veterans signed into law by the president. Tester’s campaign mocked his Republican for pretending to be a rancher in a brilliant “All hat, no cattle” ad and authenticity attacks can be effective. But both Tester and especially Manchin will have to endure an onslaught of negative ads by the NRA. And although Montana is in the top 25 states in terms of the percent of its citizens that hold a bachelors degree or more,  West Virginia is dead last. Either way, with a PVI of R+9 in Montana and R+19 in West Virginia my demographics model does not produce a winning vote share for either of these two incumbents.

Its in situations like Tester’s and Manchin’s where the qualitative handicapping comes in really handy. In Manchin’s case West Virginia has a PVI of R+19 and the smallest percent of college educated adults in a state that is also 94% non-Hispanic white. As such, its no surprise that the model produces a dismal estimate of two-party vote share for Manchin of just 28%. In the 2016 presidential election Donald Trump won West Virginia with 68% of the vote. Yet, the state also elected a Democrat as governor that day (although he switched parties shortly after the election). Jim Justice  was the beneficiary of massive split-ticket balloting. Justice beat his Republican opponent by 7 points, 49% to 42%. This means that somewhere around 19% of Trump voters split their ballots to vote for a Democrat for governor because they liked the man (although some of the difference is due to ballot drop-off as well). Given West Virginia’s recent and still enduring tendency to vote “coal over party,” I think Manchin has a solid chance of pulling out the “W” despite what my model predicts for this race. There is no way that Manchin will pull in less than 30% of the two party vote, even if he does end up losing his seat.  In Tester’s case, Montana’s got a decent share of college educated voters and is far more diverse than West Virginia. Tester has a 50% favorability  and Montana is one of the few states that saw Republican participation actually slip from their 2014 numbers in the primaries.  Another good sign for Tester is polling shows that Montana Democrats are more energized to vote than Montana Republicans. Ultimately, it is Manchin and Tester’s fit for the states they represent and their established ability to survive in hostile electoral environments that give me the feeling that they may be able to capitalize on the environment just enough to hold onto their seats. However, prepare for the possibility that one of both of these races will move into the toss-up category as the cycle continues and more data becomes available.

Which brings us to Joe Donnelly in Indiana. With a PVI score of R+9 and the diversity and urbanity of Indiana’s metropolitan areas Donnelly is capable of winning his reelection campaign despite the low rates of college education in Indiana. The latest round of polling has been of good quality and seems to suggest that Donnelly has an advantage. Trump’s approval rating among Independents in Indiana is in the low 40s and among likely voters, 43% of Independents report “strongly disapproving” of the job Trump is doing as president.  Democrats enjoy an 8 point generic ballot advantage Democrats Indiana’s Independents and they are breaking in favor of Donnelly by 20 points. Although this will be be Donnelly’s first race in a non-presidential election year, he will be benefit from a large enthusiasm differential in favor of Democrats and from a couple of competitive House races ion Indiana that may drive up turnout in a couple of key areas. Still, Donnelly is largely untested in a competitive statewide election as his first candidacy was against Republican Richard Mourdock, a “Tea Party” Republican who ousted 36-year  incumbent Richard “Dick” Lugar in the Republican primary, but who proved to be a weak candidate in the general election. This time around Donnelly will face a much stronger challenger in the Republican nominee Mike Braun, a former member of the House. Although Braun’s fundraising is keeping pace with Donnelly, it is only because Braun cut his own campaign a $6 million dollar check. Beyond that, his fundraising has been sluggish, and with Republicans forced to defend more than 40 House seats and perhaps Ted Cruz’s senate seat in Texas, it is not clear how much investment from the party Braun will be able to attract. What helps this race into the “Lean Democrat” category is the state’s surprisingly narrow PVI score, just R+1. Indiana is also the 16th most populous state in the Union and with a fairly robust population density. However, Indiana is also heavily reliant on agriculture, and Trump’s trade wars are not popular even after the GOP’s $12 billion bail out to farmers. Although Trump’s net approval in Indiana is still positive (51% approve), it has declined since January of 2017.

Lean Democrat, Republican Held Seat

Republicans find themselves defending open senate seats in both Arizona and Tennessee for the same reason: the current Republican senator was forced to resign because they are “Never Trump” Republicans. Facing dignity-destroying primary elections where they would be forced to abandon their candor about Donald Trump, both Jeff Flake (AZ) and Bob Corker (TN) instead elected to gracefully exit stage left. Now, partially due to the lack of the incumbency cushion Flake and Corker would have provided in these races and the broader appeal both held to the general electorate because of their courage to call out Trump, both these seats lean towards the Democrats. Republicans dogged a bullet when Martha McSally won the party’s nomination, benefiting from the inclusion of not just one, but two, Trump Republicans in the primary. Ironically, it was Donald Trump himself that made this possible via his pardon of Sheriff Joe Arpaio. Quality polling gives a slight advantage to Democrat Krysten Sinema and with a PVI of only R+5, a college education rate of 36%, a highly diverse population, very low midterm participation rates, and the fact that in Arizona, the vast amount of the population of the state is clustered entirely around two urban areas, Arizona is prime territory for a Democratic pick up. Sinema has another advantage, she is, how should I put this, aesthetically pleasing. And as I teach my campaigns students, there are two things that are always going to help or hurt you: the candidate’s appearance and the candidate’s charisma. And although both of these things are arguably malleable, the baseline matters and research consistently show that better looking candidates will competitive elections more often then their more mundane opponents. Don’t shoot the messenger!

Dean Heller in Nevada is facing an upward climb to hold onto his Senate seat in the face of insurgent Democratic voters. With a PVI of D+1 Heller has the unfortunate distinction of being the only Republican senate incumbent defending his seat in a state that favors Democrats, albeit slightly. And the Democratic primary produced more bad news for Heller because Democrats cast the majority of ballots in primary (51%). In 2014, they cast just 31% of them. Thus far, these has been little quality polling in this race but the high level of diversity and the fact that President Trump is under 40% approval in Nevada doesn’t bode well for Heller. And in the new “Year of the Woman” Heller faces an additional hurdle from Jacky Rosen, who is tapping  into the nearly 20pt gender gap on approval of Trump.

Lean Republican, Democratic Seats

That brings us to Heidi Heitkamp, who faces an uphill battle to retain her senate seat in North Dakota. Heitkamp faces the worst structural disadvantage of any incumbent Democratic senator, North Dakota has PVI of R+14. Like Tester, she faces the state’s only member of the House of Representatives due to the state’s low population which only produces one, at-large district and gives that sole member of the House statewide name recognition and infrastructure. North Dakota is 85% white and although it has some urban areas, North Dakota is the 3rd least populated state in the country. Another issue Heidtkamp has is the state’s relatively robust midterm turnout rates which were nearly 44% of eligible voters in 2014. Heidtkamp has to overcome a lot of hurdles to hold onto this seat but its not impossible based on some results of special elections. In contests where Republicans were able to spend big and field strong candidates like Ohio 12 and Pennsylvania 18, turnout in the special reflected closely turnout in regular general elections and in both cases, Democrats made up huge partisan disadvantages, even though they only won one of the races. This race bears a lot of similarity to these two House races in terms of demographics but in this case, Heitkamp will have the advantages of incumbency. At 44% approval, Heitkamp is right on the cusp of where she needs to be to win reelection. Much will depend on the strategy she pursues here in the final stretch. Her opponent outperforms her on economic issues, but she outperforms him regarding the #1 issue for North Dakotans: health care. North Dakota is a Medicaid expansion state under Obamacare and if Heitkamp effectively harnesses this issue as her senate colleague Joe Manchin is, she may be able to generate enough cross-over support to offset the Republicans’s advantage if she can combine those votes with the expected surge of Democrats galvanized due to Trump. Unlike West Virginia, North Dakota has a lot of college educated voters that may break for Democrats at higher rates than normal as they did in Virginia. However, like other incumbent Democrats serving in Trump states, the confirmation vote for Brett Kavanaugh is a big problem for Heitkamp. Vote for him and she will risk demobilizing the Democratic base voters she absolutely needs to show up in big numbers for her to win. But I have no doubt her strategists are telling her that a no vote will alienate potential swing voters. Either  approach comes with risk and once these decisions are made each race with an endangered Democrat incumbent will need to be reassessed.

Will the perfect record of senate incumbents retaining their seats when on the right side of the midterm effect survive the 2018 cycle? Like Trump’s improbable win in the Electoral College in 2016, it will require a series of things to break perfectly in Democrats’ favor due to the senate map’s structural advantage for the GOP. As Harry Enten points out via twitter, polling has consistently underestimated the two-party vote share of candidates on the side positive side of the midterm effect, especially in so-called “wave years.” And Democrats have already picked up 24 legislative seats they structurally had no business winning. History suggests that despite this tough map, Democrats have a route to control of the senate.

A Note on Nationalized Elections 

As I’ve gotten a few questions regarding what “nationalized elections” means I wanted to provide a working definition. A nationalized election means that a candidate or candidates in a race tie a vote for them to their respective party’s national leader. Republicans have been nationalizing their elections, even down to local judicial elections, since the late 1990s. This is where such well-known mantras such as “activist judge” or “Nancy Pelosi liberal” come from. Indeed, in the GA 6th special election Republicans hammered Jon Ossoff with flyers. Meanwhile, Jon Ossoff did his utmost best not to mention Trump by name and instead based his strategy off a Tip O’Neill’s beyond antiquated belief that “all politics is local.”  The problem is that it is now completely, inarguable wrong. In the polarized era, all politics is national and the candidate that nationalized their race (Handel) beat the candidate that didn’t (Ossoff). O’Neil’s diagnosis is still repeated daily by democratic strategists and candidates, as well as pundits and journalists. Recent interviews of their party officials suggests that Democrats still don’t get this. Only a Democrat would look at a president as scandal-stricken and controversial as Donald Trump and think they should not make the election a referendum on the president. In 2010 and 2014 Republicans structured their entire electoral message as a referendum on Obama and Obama was popular among non-Republicans. It is already clear that Republicans’ messaging strategy for the House is to frame the election entirely around the “specter” of another Nancy Pelosi speakership and the specter of Democrats impeaching Trump. That’s a nationalized message folks, and it works!

Report a problem
Version 3.4