Will 2014 Midterms Be a Wave Election?

Will 2014 Midterms Be a Wave Election?

By Sean Trende - August 28, 2014

Analysts are beginning to wonder: Where is the Republican wave? You can find Larry Sabato, Kyle Kondik and Geoffrey Skelley grappling with the question here (with apologies to Kondik and Skelley, I will just reference Dr. Sabato as the author in the future), or Nate Cohn here.

My answer isn’t that dissimilar from Dr. Sabato’s or Cohn’s ultimate conclusion: It is just too early to know for certain. Cohn more or less conclusively demonstrates this in his piece, when he observes that the full extent of the 2006 wave wasn’t apparent until October of that year, the 2008 wave didn’t crest until mid-September, and the 2010 wave didn’t fully appear until mid-August. To that we might add that the size of the 1994 wave wasn’t fully appreciated until after the elections, while in the “mini-wave” year of 2002, smart analysts were suggesting a Democratic wave might appear in August.

In other words, if 2014 is going to be a wave year, we shouldn’t really expect to see it yet. Just remember, at this point in 2010, Ohio was a single-digit Senate race, with Lee Fisher having led in polls are recently as June (Rob Portman won by 18). PPP found Richard Burr leading by two points (he won by 12). Kelly Ayotte’s lead was eight (she won by 24), Marco Rubio was up two on Charlie Crist (he won by 19) and Russ Feingold led Ron Johnson. Republicans lost a special House election in May that many observers thought was a referendum on their ability to take back the House. In 2006, Maria Cantwell looked vulnerable against Mike McGavick, Ben Cardin narrowly led Michael Steele, and Tom Kean Jr., led Bob Menendez (this was true until late September). Given this, the fact that we don’t see a wave in all of its glory just yet is largely uninteresting.

With that said, should we expect to see a wave? Sabato and particularly Cohn do a good job presenting the evidence for why we might not expect a wave. It is a coherent case, and if you’re designing mental probabilities, you would be foolish to assign a zero percent chance to it. In fact, I think you’d be foolish to assign anything below, say, a 30 percent likelihood to it.

My goal here is to present counter-evidence and respond to some of the arguments against a wave that have been presented. My goal is not to make firm predictions (just as neither Sabato nor Cohn makes firm predictions). As I said, I think it is far too early for that. My goal is simply a presentation of counter-evidence. I likewise think you’d be foolish to assign a likelihood of less than 30 percent to this case, but neither should you assign a 100 percent chance to this.

We’ll proceed in three steps: (1) defining a wave; (2) making the case for a wave; and (3) examining the arguments against a wave.

The terms “wave” and “tsunami” are bandied about so often that it is easy to lose sight of the fact that there is no real definition for them. Everyone seems to agree that a “wave” election involves big gains for one party or the other. But what counts as “big”? This is complicated by the fact that most analysts agree that the GOP would not see large House gains this year even in a wave election, so we only have Senate races to analyze.

Sabato and Cohn both suggest that even GOP gains of seven Senate seats would not constitute a wave. Sabato has called a GOP gain of seven seats “Gale Force Whitecaps,” reserving “Tropical Storm Wave,” “Republican Tidal Wave,” and “Full GOP Tsunami” for gains of nine, 11, and 14 seats, respectively. Cohn writes “[t]he Republicans will still have a good chance of picking up the Senate without an anti-Democratic wave. There are so many Democratic-held seats up for grabs in red and purple states this year that the GOP could take the Senate under neutral conditions.” Brandon Finnigan likewise places the bar for the GOP at more than seven seats.

Without a firm definition of a “wave,” it is hard to engage the point. But I think the GOP knocking off three incumbents and picking up three open seats would typically be construed as a wave election. For what it is worth, I’ve held this view for quite some time: Last summer, when we were actually in a neutral environment, I gave the GOP around a 30 percent chance of taking back the Senate, assuming that things might break the GOP’s way. This February, my model -- which I noted was probably too favorable to Republicans -- suggested that a presidential approval rating of 50 percent resulted in the GOP picking up five Senate seats as the most probable outcome.

To better see why I think picking up six seats would constitute a wave, let’s look at wave and non-wave midterm elections since the election of FDR. The canonical wave midterm elections are: 1934, 1938, 1946, 1958, 1966, 1974, 1982, 1994, 2006, and 2010. I tend to believe that 1942 should be on the list (the GOP won the popular vote for the House, and picked up 47 House seats and nine Senate seats), and think 2002 is a borderline case. I also tend to think that 1982 is exaggerated as a wave year. But for our purposes, we will stick with canon.
So let’s look at Table 1:

Table 1: Out-Party Performance in Wave Midterm Elections.

This table shows the canonical wave years, the total number of seats gained by the out-party (e.g., the Republicans when there is a Democratic president, and vice-versa), the number of incumbents the out-party defeated, and the number of incumbents the out-party lost. I also attempted to control for the favorable nature of any map by including the percentage of incumbents defeated in states where the president had performed under his national average in the preceding election (i.e., states with a favorable partisan index for the out-party).
At the bottom, I included the median result for each column. I also included a row containing a scenario for 2014 where the GOP picks up six seats, defeating three of the four Democratic incumbents in Republican-leaning states in the process.

Next, the non-wave years:

Table 2: Out-Party Performance in Non-Wave Midterm Elections.

To my mind, a GOP pickup of six seats looks a lot more like what we’d traditionally call a wave election. A pickup of six seats is just below the median result for wave elections. Knocking off three incumbents is just below the median for wave elections, while knocking off 75 percent of incumbents in favorable states would actually be an above-the-median performance even for a wave year. The only times that mark was matched or exceeded were 1958, 2006, and 2010 (in 2010, the GOP defeated the only Democratic incumbent running in a Republican-leaning state).

Compare this with the non-wave years, where the out party typically gains 0.5 seats, defeats 2.5 incumbents, loses two incumbents, and defeats only 19 percent of the incumbents with favorable partisan indexes. The only years that even look a little bit like our hypothetical 2014 are 1942 -- which I personally think is a wave year -- and 1986, which, unlike 2014, really was a case of withdrawn coattails in states that Reagan turned “red,” but only for presidential years (note how few blue state Republicans lost that year; there were few “accidental” blue state senators, who were therefore less vulnerable to withdrawn coattails).

If we expand to include presidential years, including 1932, 1936, 1948, 1952, 1964, 1980, and 2008 as wave elections, very little changes. The median wave election involves a gain of six seats, the defeat of four incumbents (without losing any), and the defeat of 58 percent of incumbents in friendly states. The median non-wave year involves gains of one seat, defeating two incumbents, and defeating 17 percent of incumbents in friendly states.

1 | 2 | Next Page››

Sean Trende is senior elections analyst for RealClearPolitics. He is a co-author of the 2014 Almanac of American Politics and author of The Lost Majority. He can be reached at Follow him on Twitter @SeanTrende.

A Reply to Rush Limbaugh
Mona Charen · December 9, 2014
Against the Wave: Three Winning House Democrats
David Byler · December 14, 2014
Obama, Congressional Dems Show Cracks in Unity
Jim Kuhnhenn · December 7, 2014
Democratic Louisiana Senator Mary Landrieu Defeated
Melinda Deslatte and Bill Barrow · December 6, 2014

Sean Trende

Author Archive

Follow Real Clear Politics

Latest On Twitter