Some Republicans will argue that 2018 wasn’t a wave for Democrats because they expanded their Senate majority.
That’s wrong.
Given the challenging map, Democrats did about as well as you would expect them to do in the in a wave election.
Democrats look on pace to win about 69% of all Senate races that were up for election this year.
I went back and looked at the overall Senate outcome in midterms since senators began to be popularly elected a little more than 100 years ago. Democrats performance this year ties them for the fifth highest percentage out of the 27 since 1914. It puts 2018 slightly ahead of 1974 (68%).
Now, Republicans can certainly crow about Democrats losing Senate seats even with an unpopular president. On average, the opposition party gains four seats in a midterm. There were certainly points this cycle it looked possible (though difficult) for Democrats to net gain seats.
But Democrats had a lot of exposure. They were defending 26 seats, 10 of which were in unfriendly territory President Donald Trump won in 2016. That reality upends expectations about net loss and net gain. There is, in other words, a bit of a reversion to the mean when one party controls a lot of seats.
If you were to look at the previous 26 midterms and were trying to predict how many seats the Democrats would win based on how many seats they controlled beforehand and the party of the president, you would have estimated that they net lose four seats. They only ended up net losing two seats. In other words, they did two seats better than should have been expected.
But let’s get a little more nitty gritty and get down to the individual seat level. Democrats did pretty well given how many seats were up in red states.
If you were to look back at all Senate elections since 1982 and tried to estimate how well a party would do given the past presidential vote in each state, Democrats did on average about seven points better than expected. This seven-point overperformance, perhaps unsurprisingly, will match or come close to matching what looks to be the Democrats popular vote margin in the House. It also ranks as the Democrats third best over-performance in the last 19 Senate elections.
Of course, Democrats holding so many seats at the beginning of the cycle was to their advantage in a way. Incumbents tend to do better than candidates running in open seats or those taking on so-said incumbents. It’s a little difficult to control for the incumbency factor because it has significantly lessened over the last 15 years ago.
Still, it’s pretty safe to look at midterm elections occurring over the last 10 years. Specifically, let’s examine the last three midterms: 2010, 2014 and 2018.
Republicans net picked up six seats in 2010 and nine seats in 2014. No doubt that’s better than Democrats did this year in the net seat loss/won category.
Democrats though ended up actually controlling a slightly higher percentage (69%) of the seats up for election in 2018 than Republicans did in 2010 (65%) or 2014 (67%).
Looking at the individual seat level and controlling for incumbency and the past presidential vote, Democrats had on-par or slightly better performances than Republicans did in 2010 and 2014. The average Democrat did about 5 points better than expected in 2018. The average Republican did five points better in 2010 and three points better in 2014.
This makes a lot of sense when you look at the national environment. Republicans won the House popular vote in 2010 by about the same as Democrats did in 2018. They did slightly worse in 2014 than this year.
Now, I wouldn’t call the 2018 Senate result a “wave.” I don’t think you can given the fact that Democrats lost seats. Once you control for key factors though, the result was about you’d expect in a wave year.