Editor’s note: This is the last in a four-part series considering the outcome of the November 2018 elections for the United States House of Representatives in the context of our Monopoly Politics projections. Read the first part here, the second here and the third here.
Following all 435 U.S. House of Representative races on Election Day can be a daunting task. Not to worry; we’ve done the work for you, using our Monopoly Politics model to project not only who will win, but by how much.
Compare actual results to our projections for a 50-50 baseline year to determine just how red or blue the much-discussed “wave” taking the House will be.
Pay special attention to these key races; we’ll be right there with you.
This short list, based on analysis by several political commentators and our own Monopoly Politics projections, highlights races that might indicate whether 2018 favors Democrats or Republicans retain control.
Key Races with Women Candidates
A pink wave may be looming thanks to a record number of women candidates on the ballot. According to FairVote’s analysis, as many as 29 new women legislators could join the U.S. House in 2019. Monopoly Politics has not made high-confidence projections for 19 of these contests nor for the seven in which women incumbents are in danger of losing their seats, making these races to watch.
Pennsylvania is a state to watch thanks to the new maps created after the Pennsylvania Supreme Court ruled the state’s 2011 districts were unconstitutional. In the first general election since the lines were redrawn, it’s likely the balance of seats each party holds will shift.
Monopoly Politics projects that Democratic candidates will win five seats, Republicans eight, and the winners of the five remaining seats are not projected. FiveThirtyEight and The Cook Political Report characterize four of these as favoring the Democrat and one as a toss-up, meaning Democrats could pick up an additional three to four seats, dramatically altering the partisan breakdown of the state’s congressional delegation.
A final race to watch in Pennsylvania is the 17th Congressional District, which features two incumbents: Keith Rothfus of the former 12th Congressional District and Conor Lamb of the former 18th Congressional District.
The U.S. House races in Maine will make history no matter who wins because the seats will be determined using ranked choice voting (RCV). In the 1st Congressional District, incumbent U.S. Rep. Democrat Chellie Pingree and Republican challenger Mark Holbrook said they will not rank candidates. Meanwhile, independent candidate Marty Grohman has been an outspoken backer of RCV.
Incumbent U.S. Rep. Republican Bruce Poliquin is facing Democratic challenger Jared Golden and independents Tiffany Bond and Will Hoar in the 2nd Congressional District. The two independents in the race are forecasted to receive a little over 4 percent of the vote share. Under the traditional single-winner plurality system, the two independents could pull enough votes from the other candidates to “spoil” the race. A “spoiler effect” does not occur with RCV however, because votes are counted until one candidate secures a majority.
If Democrats retake the House in 2018, it will likely be by a very slim margin of seats. Monopoly Politics projects the outcomes for 379 of the 435 seats at the highest level of confidence—projections which have proven to be 99.9 percent accurate. Of those, 206 seats are projected to go to Republicans, and 173 to Democrats.
The remaining 56 seats are too competitive to call with a high level of confidence, but a few could make the difference in control of the House.
Could our projections be wrong?
There are some instances in which Monopoly Politics’ projections — though made with our highest level of confidence — might be off. There are two potential reasons.
First, our high-confidence projections may be inaccurate due to a change in voter behavior. Our projections take into account only the last presidential election’s partisan dynamic. Whether the shift from 2012 to 2016 was temporary or indicative of a lasting “Trump Effect” could impact the accuracy of our 2018 projections. Consider, for example, the close race in Iowa’s 3rd Congressional District. The district favored Obama in 2008 and 2012, then swung to Trump in 2016. Polls show the Republican incumbent is trailing, though Monopoly Politics projects a Republican winner.
Following the election, FairVote will analyze how accurate our partisanship metric is compared with those that include both 2016 and 2012 election data. If our metric is more predictive of the results, we’ll know that the Republican base has shifted, with Trump voters forming the new base for Republicans. If it doesn’t, that suggests Trump’s base may be willing to vote for Trump, but not for down-ballot Republicans.
The other reason our projections might be wrong is that partisan waves can increase the chance of a statistical fluke. While our model requires at least a 56 percent partisan preference before making a high-confidence projection, an overall 54 percent Democratic wave would bring those seats which favor Republicans by 56 percent within competitive range.
(Note: Such a wave year would be far outside the norm, which is why our high-confidence projections do not anticipate waves. Our full projections do take this kind of partisan shift into account, and we remain confident our projections will prove reliable when adjusted for 2018’s two-party preference.)
If there is a large “blue wave,” races such as Georgia’s 6th Congressional District may give wins to Democrats even though Monopoly Politics calls the seat “safe” for Republicans.
That control of the House rests on just a few competitive seats highlights how broken the American political system is. Monopoly Politics’ projections are both extremely accurate and extremely simple, without regard to fundraising, scandals or polls. Until the reforms outlined in the Fair Representation Act are in place — including multi-winner districts elected via ranked choice voting — only a fraction of seats will ever be truly competitive.
The upcoming release of Monopoly Politics 2020, the next edition of our series, will project U.S. House race winners nearly two years in advance, underscoring the problematic predictability of our system.
Illustration by Mikhaila Markham