2018 is looking like an important off-cycle election in American politics. In the Senate, Democrats/Democrat-leaning Independents are defending a mammoth 26 seats (includes a special election for Al Franken’s old seat), and ten of those seats are in 2016 Presidential Red states. Of those, seats in Indiana (Donnelly), Missouri (McCaskill), North Dakota (Heitkamp, and West Virginia (Manchin) look particularly susceptible. On the other side, the Republicans are defending just eight seats, and only one is in a 2016 Blue State (Heller in Nevada). In addition to Nevada being a tossup, Arizona (to replace Flake) is also susceptible.
But what does the 2018 House prospect look like? The chart illustrates 80 years of House of Representative elections (1938 is the first year were the House popular vote is straightforwardly available). The location of the Red (Republican) and Blue (Democrats) bubbles in the chart indicate the percentage of seats each party held in each year. Anything above the dashed 50% line indicates a majority. The size of each bubble is proportional to the popular vote; the popular vote is normalized by the total US population for each election year. Bright red and bright blue circles indicate occasions where the popular vote did not match the seat majority. This includes recent examples in 1996 and 2012. A few things to note:
- The US is now redder than people think: Since 1992, the Democrats have had the majority of House seats on just two occasions (2006 and 2008), though they won the popular vote on two further occasions (1996 and 2012).
- The country used to be very blue: The Republicans were in the House minority every year from 1954 to 1994 (they lost the 1954 election, but won in 1994).
- Democrats and 2018: The Democrats have won a single off-cycle election in the House (2006) since 1990. Will 2018 be any better?
- Election turnouts: In Presidential years since 1938, 37% of the US population (not eligible voters) votes in the House elections. This drops to 25% in off-cycle elections.
Data and software: Data on the Senate 2018 race and historic House elections is taken from Wikipedia1,2. The data was compiled and visualized using Microsoft Excel3.
- https://en.wikipedia.org/wiki/United_States_Senate_elections,_2018
- https://en.wikipedia.org/wiki/United_States_House_of_Representatives_elections,_2016
- https://products.office.com/en-us/excel
David Whiteley
If I’m reading this correctly, the Dems prospects aren’t looking good. Can we hope for a grass roots reaction vote against the President and party? Or have we entered a new era where past political examples no longer apply? Thanks for statistical reality check and potential future nightmare.
Richard Middleton
I don’t have the answers of course! I’ve seen a reasonable amount of pessimism from the Republicans and optimism from the Democrats… I’m not sure it’s entirely well placed. Based on current projections (pretty darned uncertain at this point!), the Senate could go either way by one or two seats. And that’s actually pretty good news for the Democrats since, in 2020, they’re defending fewer seats and the Republicans, in a Presidential year, will have a tough time potentially.
The main driver that prompted me to look at the House was the fact that, even though Trump lost the popular vote heavily, the Republicans in the House won the popular vote by ~1.5 million. And they’ve done very well in the house since 1990 (lost the House only twice, and two further times lost the popular vote). That’s why I think the country is Redder than either party knows or admits! I realized that very few people, including the most knowledgeable people I know, didn’t know this!
Grass roots: here’s my thesis to Democrats. Take a look at the seats that Democrats had in 2006 (https://tinyurl.com/y77rczvh) but not in 2016 (https://tinyurl.com/y92lgxta). Almost every single one is a suburban or Blue Dog district. The democrats have the liberal seats pretty much sown up. But what is the message that could win those 2006 seats back? If I were a Democratic strategist, I might argue that the policies ae OK but the framing is poor. For example, why not campaign on the fact that food stamps (SNAP) has a positive multiplier effect for the economy (layman terms: it’s the nest investment America could make economically) or that government/universal healthcare (which the majority of Americans wants) is great for small businesses and exports?