How Our Predictions Held Up

After the surprises of Election Night 2016, it was nice to have a first Tuesday after the first Monday in November in which things mostly went as expected, and for me personally, as predicted.

In the end, we expected House Democrats to score a net gain of between 30 and 40 seats and control of the chamber; Democrats look on track to pick up between 37 and 40. In fact, it was an outcome that would not have been surprising this past summer, before the Brett Kavanaugh Supreme Court fight, the caravan, or the synagogue shooting in Pittsburgh.

In the Senate, there was an ebb and flow between party fortunes over the last year: Last summer, for example, one of the most plausible outcomes would have been either no net change or one party picking up a single seat. But in the end, after the Kavanaugh fight galvanized the GOP and conservative vote in many of the red states with Senate elections, those Democrats who needed defections from Republicans and/or conservatives found the incline significantly steeper. In the end, it looked like Republicans would most likely score a net gain of a seat or two, and depending upon the outcome in Florida, the GOP will either break even or gain one seat.

We figured a Democratic net gain of about a half-dozen governorships and state legislative chambers. Democrats gained seven governors, while Republicans lost six (the GOP picked up the Alaskan governorship, previously held by an independent); and Democrats gained six state legislative chambers (including Connecticut's, which had been tied, but not counting New York state, where they won complete control of the Senate). The disappointment for Democrats is that in raw numbers, it looks like their net gains in state legislative seats in the low 300s will be little more than their net losses were in the 2014 midterm election, and well less than half what they lost in their disastrous 2010 midterm.

We wrote often of a “tale of two elections”—one for the Senate and one for everything else—and that is exactly what happened. President Trump was an asset for Republicans in many of these Senate races; he is to be credited with going to the right places in the closing weeks of the campaign. But he was also a liability to many Republican House members in suburban districts, where women in general, and specifically college-educated women, couldn’t find Trump’s name on the ballot but voted against anyone wearing a red jersey instead.

Another dynamic that we wrote about a lot was the wave and the wall—the wave was the political environment, and the wall was made up of structural barriers that helped protect Republicans. In the Senate, it was the most favorable map for the GOP that either party has had in modern history. Republicans had the benefit of GOP-drawn congressional district boundaries in many states and natural population patterns that tend to concentrate Democratic voters. So Republicans’ House and state legislative losses could have been much worse; the walls protecting their majorities mitigated much, but not all, of the wave. It wasn’t a tsunami, but it was more than a ripple.

The biggest upset of the night was GOP Rep. Steve Russell’s loss to attorney Kendra Horn in Oklahoma’s 5th District, a diverse, largely urban and suburban Oklahoma City seat—a result that was unanticipated but in retrospect not entirely surprising. The runner-ups for upsets were Rep. Dan Donovan’s loss in New York’s 11th District to Afghanistan veteran Max Rose, and the open seat in South Carolina’s 1st District, where Rep. Mark Sanford had lost his primary to a very weak Republican, Katie Arrington, who went on to lose the seat to Democratic attorney Joe Cunningham.

The much-maligned polls actually had a very good night. There were no upsets in the Senate, and few in the House and governor races. The races that were expected to be close largely were. The average of polls on the generic ballot test was 7 points, and as of now, the national popular vote for the House shows Democrats ahead by 7 points. The reality is that the campaign- and party-sponsored polls were largely on target and the public polling on the state level wasn’t too bad—nothing like the epic 2016 misses in the Wisconsin, Michigan, and Pennsylvania presidential-race outcomes.

Over the coming weeks, we will all be poring over election results and exit polls, slicing and dicing the data to determine what exactly happened, how and why. The longer you look at results, the more texture comes out.

The big news will be the increased diversity in Congress in terms of gender, race, and religion. It is a pretty safe bet that Congress will never look the same again, and that is a good thing no matter which side of the aisle you occupy.

This story was originally published on nationaljournal.com on November 13, 2018