One of the many ways midterm elections are different from presidential elections is that there’s an element of subjectivity to determining which party won. In a presidential election year, whichever party captures the White House is seen as the winner, but in a midterm election cycle, it isn’t as clear — a lot depends on the context and on expectations.
In 1998, for instance, Republicans maintained control of both chambers of Congress, losing just five seats in the House to Democrats. If that were to happen this year, it would be considered an incredible triumph for Republicans. But 1998 was the sixth year of Bill Clinton’s presidency, an election cycle typically associated with major gains for the opposition party — and that year, Clinton was enduring the fallout from the revelation of his lying about an affair with White House intern Monica Lewinsky. The GOP performance was seen as so dismal it helped force Newt Gingrich to resign as House Speaker.
As we look toward Tuesday’s elections, it’s worth setting down some markers, and exploring the question: How do we determine whether Republicans or Democrats won?
Tomorrow’s election promises to be unique a number of reasons. One is the phenomenon of President Trump, a figure whose style has disrupted standard political analysis. Typically, factors such as state of the economy and presidential approval are related, but never before in the history of modern polling and economic record keeping has presidential approval been as low as Trump’s with unemployment as low as it is. Furthermore, we could see a rare case of the House and Senate moving in significantly opposite directions given the fact that the strong year for Democrats, with a lot of Republicans holding House seats carried by Hillary Clinton, is coming at the same time as a Senate map heavily tilted toward red states.
In rare cases of midterms in which the House and Senate move in opposite directions, typically the gains in the House have been relatively modest. For instance, in 1970, Democrats gained 12 House seats but Republicans gained two Senate seats. In 1962, Democrats gained four Senate seats and lost four House seats. One would have to go back over a century, to the 1914 midterms, to find a lopsided scenario in which Democrats lost 61House seats (at a time when they had an insane 291-vote majority going in) yet gained three senate seats (in the first election in which Senators were elected by popular vote instead of state legislatures).
Going into 2018, there are some scenarios in which it’s pretty obvious to say who won. If Democrats gain control of both the House and Senate, it would be a crushing defeat of Trump-era Republicans. If they gain 30-40 seats, and keep the Senate around where it is, they should still be considered the victors. But there are other perfectly statistically plausible scenarios where there might be more of a debate, where at least Democrats won’t have a clean win.
For instance, let’s say Democrats gain around 30 seats in the House, providing them with control of the chamber by a relatively narrow margin, but that Republicans add three seats in the Senate, to get to 54. Suddenly, that becomes a disappointing result for Democrats.
On the one hand, there would be an argument that Democrats should be seen as winners given that they gained control of one chamber, and only lost a few seats in the Senate as a result of all the races they had to run in hostile territory. But on the other hand, one has to look at the Senate race in terms of how it sets the parties up for 2020.
As I noted before and as is demonstrated by the interactive graphic below, the 2020 Senate map isn’t necessarily easy for Democrats. There are a few pickup opportunities (such as Colorado and Maine), but they’re also near certain to lose the Alabama senate seat assuming Republicans don’t renominate Roy Moore. So if the dust clears with Republicans at 54 seats, even if Trump is defeated in a landslide in two years, Republicans could still keep the Senate and block the next Democratic president’s agenda.
So, my rough idea is that if Democrats gain 35-40 seats in the House and limit Republicans to no more than 52 seats in the Senate, it will be a pretty unambiguous victory. Once Republicans start getting to 53 of 54 Senate seats, it will become a mixed night for Democrats, even if they win the House. If Republicans gain a few seats in the Senate and miraculously preserve the House, we can start to talk about whether the night was a win for the GOP, even if Democrats gain 20 House seats. But I am open to persuasion.
