If it's in the News, it's in our Polls. Public opinion polling since 2003.


Senate 2018: At Least for Now, the Elephant Endures

A Commentary By Kyle Kondik


— In what is a somewhat unusual development, the incumbent party is not a clear underdog in any single Senate race with just two months left to go. But there are at least a couple of cases where the incumbent party is likely behind.

— Republicans remain favored to hold the Senate, but the Democrats do have a path.

— Texas goes from Likely Republican to Leans Republican.

Table 1: Crystal Ball Senate ratings change

Map 1: Crystal Ball Senate ratings

With Labor Day in the rearview mirror, the race to Election Day is on. The national picture remains favorable to Democrats, but because of the Senate playing field, the Republican majority still will be difficult for Democrats to dislodge even in an optimal environment.

One oddity as we reach the final stretch of the campaign is that while there are many very competitive Senate races, we don’t consider the incumbent party to be a clear underdog in any seat they currently hold at this juncture. That’s not to say the incumbent party is leading in every seat they already hold — in at least two states that does not appear to be the case, as we will discuss below — but no seat appears to be a write-off loss. That’s a change from the past several cycles at around Labor Day. In 2016 at this time, Sen. Mark Kirk (R-IL) seemed like an almost sure loser, and he did lose; in 2014, open Democratic seats in Montana, South Dakota, and West Virginia did not appear salvageable for Democrats. In 2012, Angus King (I-ME) was well on his way to winning an open seat in Maine and was widely expected to caucus with Democrats (he still does and is a big favorite for reelection this year), and Republicans were in the driver’s seat to pick up an open seat in Nebraska. We could go back further, but we hope you get the point. So this cycle is different in that we do not have any single race rated worse than a Toss-up for the incumbent party.

That said, the incumbent party is close to being a clear underdog in two races. Let’s start with those. We’ll go seat by seat and then finish up with an overall look at the Senate picture. Remember, Republicans have a 51-49 Senate majority, so Democrats need to net two seats to win a majority.

Despite what is a good environment for Democrats across the nation, the shakiest incumbent in the nation is a Democrat: Sen. Heidi Heitkamp (D-ND). We’re holding her race as a Toss-up, but it could (should?) be rated as Leans Republican. North Dakota is probably the most Republican state where Democrats are playing defense in the Senate this year, and Heitkamp barely won in 2012 against a fairly weak opponent, then-Rep. Rick Berg (R, ND-AL), a first-termer who unsuccessfully tried to make the jump to the Senate.

Rep. Kevin Cramer (R, ND-AL), elected in 2012 to replace Berg, has said publicly that he has consistently led in polls. And we have to say, we are not familiar with any recent polls showing Heitkamp leading. While there’s little recent that’s public, we have heard of several different private surveys from sources on both sides of the aisle. As best we can ascertain, Heitkamp is trailing, and probably not just by a point or two. On the flip side, nobody we’ve spoken to seems to believe Heitkamp is personally unpopular: It’s just that it’s hard to be a Democrat running for Senate in North Dakota these days. Democrats hope that Heitkamp, by the end, can translate her good personal numbers into another razor-thin victory. But it seems like one would rather be Cramer than Heitkamp at this point, even though we’re not quite at the point where we want to go to Leans Republican yet.

Meanwhile, in Arizona, Republicans were pleased that Rep. Martha McSally (R, AZ-2) emerged from a primary last week. She will face Rep. Kyrsten Sinema (D, AZ-9), who had no real primary opposition and has been bolstering her image for months. Republican operatives are upset that Sinema basically got a free pass from conservative outside groups prior to the primary — in other words, Republican outside groups were not aggressively making the case against Sinema while McSally was getting beaten up in her own primary. (Actually, some Republicans believe that Sinema has gotten a free pass for years: After a competitive initial race in 2012, she won easily in 2014 and 2016, leading some to lament that she was allowed to set herself up for an expected statewide run largely unimpeded). Perhaps as a result of the structure of the race, Sinema appears to be leading as the general election begins, and we (and others) have noticed a persistent weakness for President Trump in what is otherwise a Republican state. Many seem to believe that Nevada is a better pickup opportunity for Democrats than Arizona, because Nevada is a state that Hillary Clinton narrowly won while Trump carried Arizona, albeit by just 3.5 points, down from what one would expect. And yet, it may be that Arizona is actually a slightly better Democratic target, in part because of Sinema’s head start and because it’s an open seat. In Nevada, it may ultimately be that the environment is difficult for Sen. Dean Heller (R) to overcome, but he’s still standing in his race against Rep. Jacky Rosen (D, NV-3) and Nevada is a 50-50 kind of state. We look at the race between Sen. Bill Nelson (D-FL) and term-limited Gov. Rick Scott (R-FL) in the Sunshine State in much the same way. In terms of campaign strength, Scott has seemed to run circles around Nelson, who has been caught flat-footed against a very aggressive and well-funded challenger. That said, the environment could still save Nelson. A Quinnipiac University survey released Wednesday showed the Florida Senate race as a dead heat.

To be clear, we still call Arizona, Florida, North Dakota, and Nevada Toss-ups. In all likelihood, Democrats need to win at least three of four to avoid losses this year, and a sweep is likely necessary for them to win the majority.

Back to 2018. The two other closest races are in Indiana and Missouri, where Sens. Joe Donnelly (D-IN) and Claire McCaskill (D-MO) are also in Toss-up races against former state Rep. Mike Braun (R) and Missouri Attorney General Josh Hawley (R), respectively. Missouri seems like a total Toss-up, as a head-to-head matchup by NBC News/Marist showed on Tuesday morning (47%-47%), although the inclusion of third-party candidates gave McCaskill a 44%-40% edge. Our understanding is that internal campaign polls on both sides also show a very tight race. Indiana may be a slightly easier though still challenging hold for Democrats. For one thing, Donnelly is closer to the political center than McCaskill (he is not pro-choice on abortion), and he may have a small polling lead, depending on who one believes (we have heard some vague details about internal polls on both sides). NBC News/Marist has Donnelly up: That poll found Donnelly leading 49%-43% in a two-way race, but with a narrower three-point lead when a Libertarian was included in the poll. Again, these are Toss-up races. Trump’s precise approval rating may be meaningful in both of these states; while the president won them each by close to 20 points, his approval rating may be only around 50% or so in each. Ipsos, with whom we’ve been working, estimates Trump’s approval at around 53%-54% in both states (as noted in our new Political Atlas), while, in Indiana and Missouri, Marist had Trump’s approval at a more modest 48% and 45%, respectively.

Sens. Joe Manchin (D-WV) and Jon Tester (D-MT) remain modest favorites in their states, each of which Trump won in landslides of different proportion (he carried West Virginia by 42 points but Montana by “just” 20). After a recent Trump visit to the Mountain State, Manchin remains ahead even in Republican polling — a recent poll from a group supporting state Attorney General Patrick Morrisey (R) pegged Manchin’s lead at six points. Still, Trump’s visits do seem to move the numbers for Morrisey to at least some positive degree although perhaps not durably, but Manchin has to be concerned that the president could try to will Morrisey to victory by hitting the state several times at the end of the campaign. It’s one of the few places where Trump is very clearly an asset to a GOP candidate. Additionally, Manchin’s approval numbers are not that great, yet he is a state institution who is clearly leading at the moment. Meanwhile, Tester did not win a majority of the vote in his two prior victories, although he doesn’t have to this time, either, because a Libertarian is on the ballot. Republicans released a recent poll showing state Auditor Matt Rosendale (R) up 47%-45%, but it seems as though that was an outlier, at least for now. Both states are in the Leans Democratic column; if the Republicans end up beating one or both Democratic incumbents, it probably means they would be making at least a small net gain in the Senate.

To win a Senate majority, Democrats have to win every race we’ve mentioned so far — unless they can pull off an upset in the next two states we’re discussing.

There’s considerable disagreement between the parties about Tennessee, where Republicans argue Rep. Marsha Blackburn (R, TN-7) is favored and now leading but where Democrats say ex-Gov. Phil Bredesen (D) is up (most public polls have shown the latter, although some of those were polls for Democratic groups and are dated). We continue to see Tennessee as a heavy lift for Democrats just because the state has trended so clearly Republican in recent years; if this was a more 50-50 state, we’d probably call this race a Toss-up.

Probably 2018’s most covered Senate race comes in another red state, Texas, where Sen. Ted Cruz (R-TX) is being pushed by Rep. Beto O’Rourke (D, TX-16), who is raising gobs of money and is, at the very least, running a very spirited campaign. One way to track that is through the news O’Rourke has generated on social media, such as a recent viral video in which he defended the right of NFL players to protest during the Star-Spangled Banner. O’Rourke is dominating the race in terms of social media — an index from Ipsos that tracks social media indicators actually sees O’Rourke as a small favorite in the race — while traditional handicappers and polling has Cruz as the favorite. We are sticking with the latter view, but we also are moving the race from Likely Republican to Leans Republican. Here’s our logic: While Hillary Clinton lost Texas, she got within single digits of winning (Trump’s winning margin was nine points). Given the environment and O’Rourke’s resources, shouldn’t he be able to get closer than nine points? We’re inclined to think so, which means the margin in this race could be something like the mid-single digits or even closer. To be clear, we’d still rather be Cruz in big, red Texas, but Cruz himself is ringing the alarm bell about how close the race may be, and a major outside conservative group, the economically-focused Club for Growth, has entered the race. The Club probably wouldn’t be doing so if it thought Cruz was a 100% lock to win. This is all a long way of saying we still like Cruz’s chances, just not quite as much as before. Before we leave Texas, one observation, passed along by a Lone Star State Democrat and friend of the Crystal Ball: O’Rourke’s message about the NFL players seems like a better message for the nation as opposed to Texas, which is more culturally conservative than the nation as a whole. In other words, a viral video with that content might not necessarily be an asset in the state. But it does further raise O’Rourke’s profile amongst Democrats, and it seems possible that O’Rourke could be a 2020 Democratic presidential contender, win or lose.

That covers the 10 races — six held by Democrats, four held by Republicans — that we feel are the most competitive. Assuming no seat flips in the next group of races — the 25 more we rate as Likely or Safe for the incumbent party — Democrats need to win eight of the 10 Toss-ups/Leaners to win a majority. Again, that is possible, we just think there’s a less than 50-50 chance of that happening, at least at this juncture.

Of the remaining seats, perhaps the two most interesting right now are New Jersey and Wisconsin. In the Garden State, Sen. Bob Menendez (D-NJ), who’s never been particularly inspiring, is dealing with the fallout from his near-conviction on corruption-related charges (he was saved by a hung jury). Menendez now has to work to put away his opponent, wealthy former pharmaceutical executive Bob Hugin (R), who has already spent millions of his own money on the race. A recent Quinnipiac poll had Menendez up by just a so-so 43%-37%, although in a Democratic state in a Democratic environment, the incumbent has the easier path to 50%. New Jersey probably will grit its teeth and send Menendez back to the Senate, but the Democratic Senatorial Campaign Committee has already had to spend at least a small amount of money on the race in the form of a coordinated ad buy with Menendez. Republicans say that this is the Democratic version of Tennessee — a state where the incumbent party is favored to win but still has to spend money to win, diverting resources from more competitive states. One Democrat compared this race to Connecticut’s open-seat race in 2012, where now-Sen. Chris Murphy (D-CT) had to deal with an aggressive self-funder in the form of Linda McMahon (R), who is now the federal Small Business Administrator (McMahon, who ran in 2010 as well, actually led some polls in that race as late as early October). Murphy ended up winning by about a dozen points. We’re holding at Likely Democratic for now in New Jersey.

We’re doing the same in the Badger State, where Sen. Tammy Baldwin (D-WI) is running for a second term against state Sen. Leah Vukmir (R), who is close to Gov. Scott Walker’s (R) political organization. Baldwin has never struck us as particularly strong, although she seems to hold a lead of somewhere in the mid-to-high single digits (the respected Marquette University Law School’s poll had Vukmir within two points of Baldwin, but other polls have not shown it as close). After former Sen. Russ Feingold’s (D) lead against Sen. Ron Johnson (R) surprisingly evaporated at the end of the 2016 cycle, allowing Johnson to win a second term, Democrats can’t take anything for granted in Wisconsin, although at this point we expect Walker will have a closer race than Baldwin. But we also think that Baldwin’s position may be a little shakier than that of Sens. Debbie Stabenow (D-MI), Tina Smith (D-MN), and Sherrod Brown (D-OH), fellow Midwest incumbent Democrats in states that Trump won or came close to winning that we also rate in our Likely Democratic column (Smith, an appointee, is running in a special election).

The one final race worth mentioning is a special election in Mississippi, where three major candidates are competing in a jungle primary this November, followed by a likely runoff on Nov. 27. It may be that the single major Democrat in the race, former congressman and agriculture secretary Mike Espy, is favored to finish first in the all-party primary, followed by appointed Sen. Cindy Hyde-Smith (R) and state Sen. Chris McDaniel (R), the insurgent who nearly beat Sen. Thad Cochran (R-MS) in a primary runoff the last time this Senate seat was on the ballot in 2014 (Cochran resigned earlier this year for health reasons). Assuming Hyde-Smith and Espy advance to the runoff, the Republican should have an advantage in a racially-polarized electorate where an almost-uniformly Republican white electorate making up somewhere around 60% of the electorate can outvote a uniformly Democratic black electorate making up around 35% or so of the electorate. The race should be competitive but it’s hard to see how Espy could get to a majority in such a state, so long as Hyde-Smith advances to the runoff too. But depending on how the other states go, the Mississippi Senate runoff could extend the Senate battle to a form of overtime.

Every other race not yet mentioned is safe for the incumbent party. Sen. Martin Heinrich’s (D-NM) bid for a second term got a bit more intriguing when Gary Johnson, a former two-term Republican governor who won 3% of the vote nationally as the Libertarian Party presidential nominee in 2016, entered the race late. But the Republican nominee, businessman Mick Rich, elected not to exit the race, and we don’t think Heinrich is in much trouble in what has become a more Democratic state against a split field dividing what very well could be a minority of the electorate. Had Rich gotten out, we might’ve moved New Mexico out of the Safe Democratic column, but Heinrich seems fine, at least for now. Meanwhile, Sen. Tom Carper (D-DE) is facing a challenger, veteran and activist Kerri Evelyn Harris (D), in a primary tonight. The race has been getting some buzz and while Carper should be OK, the recent upsets of Reps. Michael Capuano (D, MA-7) and Joe Crowley (D, NY-14) in primaries have rightly put him on guard.


Keeping the Republicans as favorites in the Senate effectively comes down to this calculation. In order for Democrats to win the Senate, they need to do two of three things: 1.) Win both Republican-held Toss-up seats in Arizona and Nevada; 2.) Hold all 26 seats they currently hold, several of which are in states that Trump won in landslides; or 3.) Win at least one Senate seat in a dark red state where Republicans are currently favored, be it Mississippi, Tennessee, or Texas. At this point, we might peg Democrats as slightly better than 50-50 to accomplish No. 1, but we’d put Republicans as a bit better than 50-50 to prevent Democrats from accomplishing No. 2 and even better to prevent them from accomplishing No. 3. So that’s why Republicans continue to be favored to hold the Senate, in our view. That said, the Democratic path to a Senate majority does not involve them doing something radically out of the ordinary to win: The presidential out party did not lose a single incumbent-held seat in any of the last three midterms in the Senate, for instance, and both Arizona and Nevada (if not the redder Republican-held states) certainly fit the profile of Senate battlegrounds the out party could win in a year like this one. In other words, if Democrats swept the closest races and captured a small majority, it would be surprising, but not shocking.

Nor would it be shocking if Republicans started next year with a larger majority than their current one, thanks to victories in dark red states against some Democratic incumbents.

One thing we do feel confident in saying, though: Barring some major change in the national environment, we would not expect much net change in the Senate. In other words, a net change of something akin to 1998 (no change) or 2002 (two seats in favor of the GOP) is likelier than what we’ve seen in the last three midterms, when the non-presidential party netted at least half a dozen seats in each election.

Kyle Kondik is a Political Analyst at the Center for Politics at the University of Virginia.

See Other Political Commentary by Kyle Kondik.

See Other Political Commentary. 

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.