Election Predictions Look Different (and Better) Now that All the Races are Over
January 22, 2021 · 9:09 AM EST
More than two months later, the 2020 elections are still handing out lessons. The latest: Don’t judge election projections until all the results have hatched.
Lost in the aftermath of the siege on the U.S. Capitol was the fact that Democrats won two Senate runoffs in Georgia. Not only does control of the Senate have a profound impact on President Joe Biden’s agenda and appointments and Republicans’ ability to launch investigations, the result lined up with pre-election expectations.
After Nov. 3, the decided narrative was “Here we go, 2016 all over again.” After President Donald Trump lost by less than expected and House Republicans overperformed, handicappers were forced to go on an apology tour.
Now with the benefit of a full slate of results, the preelection projections look different, and better.
The Senate switching party control should not be a surprise. By October 2019, it was clear the chamber was in play.
Here’s what we wrote in Inside Elections:
“With President Donald Trump struggling to recreate his 2016 Electoral College victory, control of the Senate should be regarded as in play. Republicans are still more likely than not to maintain control of the Senate, but Democrats have a legitimate path to control, particularly if they win the White House.”
Not bad for a declaration made more than a year before the elections and before COVID-19 and the economic fallout were imaginable. Winning in Georgia to seal control was not the most likely path to Democratic control, but the party’s cautious optimism for 2020 was fueled by multiple takeover opportunities. If one (or four, or five) races failed, they still had enough to get across the line.
The Democratic trifecta should not be considered a surprise either.
A year before the elections, Democrats winning the White House, taking control of the Senate and maintaining control of the House was the second most likely outcome.
Here’s what I wrote for CQ magazine:
“This result requires a surge in Democratic voters and continued revolt by independent voters against Trump as seen in the 2018 midterm elections. … Trump would be the catalyst to unify and energize the Democratic Party, causing him to lose some of the states he won by the closest margins in 2016, including Michigan, Pennsylvania and Wisconsin, while also turning Arizona, North Carolina, Florida and Georgia blue on Election Night.”
Of course, the president won North Carolina and Florida, but it’s not terrible foresight a year before the elections. It shows that even through multiple seismic and historic events, the overall battlefield didn’t change all that much over the course of the cycle.
By October 2020, the Democratic trifecta was the most likely outcome. I said so in Roll Call on Oct. 7, “In a year of surprises, a win by Joe Biden should not be one” and similarly in Inside Elections on Oct. 17, “Democrats Poised for a Sweep.”
Ticket-splitting Uncommon
Our projections certainly weren’t perfect. I’ve admitted that our House projection was way off, even as we correctly anticipated Democrats maintaining their majority. Yet while our down-ballot projections weren’t as precise as I would have liked, they were rooted in the correct analysis in the lack of ticket-splitting.
One of the common arguments Republicans use to discredit Biden’s victory is to question how House and Senate Republicans could have done so well and yet Trump lost at the top of the ticket. That argument is rightly skeptical of a large population of ticket-splitters.
But the fact is that there weren’t many ticket-splitters. Where Trump won, Republicans won and where Trump lost, Republicans lost, with a few exceptions.
In the 2018 midterms, Democrats crept into Trump territory by winning House seats that the president had carried in 2016. With Trump back on the ballot in 2020, he won more of those new Democratic seats than expected and boosted GOP challengers. Even though the 2020 presidential results have yet to be crunched in many districts, it’s likely that most of the GOP gains were in districts where Trump outpaced Biden.
At the Senate level, ticket-splitting continues to be the exception, rather than the rule. In 79 Senate races over the last two presidential cycles, just one state voted for a different party for president than it did for the U.S. Senate. That makes GOP Sen. Susan Collins’ victory in Maine in November all the more impressive. Even the Georgia runoffs are less surprising considering Biden won the state.
Aside from monitoring the fallout of the violent invasion of the Capitol and the potential for future armed insurrection, it’s also important to listen to the politicians. What politicians think happened in the election will guide their future behavior. If Republicans continue to believe that they had a largely successful election, based on gains in the House, they won’t feel the impetus to change. And that’s quite a takeaway after losing control of two entire branches of government.