How 2020 will be different from 2016
Could Trump lose the popular vote by 5 million ballots in 2020 and still win re-election?
If you’re like me, you spent the weekend depressed after reading Nate Cohn’s column in The New York Times or David Wasserman’s post on the NBC website.
In each, the two data science-focused reporters looked at the Presidential election polling. Rather than simply looking at the national results, they considered the state by state results and analyzed what these results mean for the election.
The results of their analysis was shocking. While Trump’s negative ratings on a national level seem to make his re-election unlikely, their state-by-state analysis indicated that there was a path for Trump to repeat his narrow electoral college win of 2016, this time while losing the popular vote by as many as five million ballots.
The reason a state-by-state analysis is valuable is because we elect our Presidents on a state-by-state basis.
As a result, if the Democrat wins California by 1 vote or 3 million, he or she still gets all 55 electoral votes. Similarly, if Trump wins Texas by 1.5 million votes or by 1 vote, he gets all 38 electoral votes.
Texas and California are relevant since they are the nation’s two most populous states, and they are unlikely to vote for the Republican, in California’s case, or the Democrat, in Texas.
That’s where the idea of “swing states” comes in. In 2016, Trump won 3 midwestern states by less than one percentage point: Pennsylvania, Michigan and Wisconsin. Indeed, Michigan’s 0.23 percentage point victory was so close, just 10,704 out of a total 4,799,284 cast, that it took days to determine the final winner. Indeed, Jill Stein’s 51,463 votes might have made the difference.
Despite such a small margin, Trump won all 16 of Michigan’s electoral votes. The same can be said for Pennsylvania’s 20 electoral votes and Wisconsin’s ten.
Remember that Trump won the electoral college by a margin of 306 to 232, a difference of 74 electors. If those three states had voted Democratic, a change of less than 39,000 votes out of nearly 137 million cast, Clinton would have won the election.
The problem was that Clinton and her campaign considered these three states as part of the “blue wall,” what they considered an electoral college advantage favoring the Democrats. As a result, they neglected these three states in favor of other states they thought they could win, but didn’t anyway: Florida, North Carolina, and Ohio.
Indeed, throughout the campaign, the candidate’s husband, someone who knows something about winning elections, desperately urged Clinton’s campaign to focus more on these midwestern states. To his consternation, Clinton and her campaign manager Robby Mook pooh-poohed his advice, leading him to actually do some campaigning of his own in these states.
This strategic decision has been characterized as “malpractice,” according to Democratic pollster Paul Maslin.
This is a mistake the Democrats will not make twice. Indeed, in 2018, their efforts in those key midwestern swing states bore fruit. In Michigan, Democratic women won all three of the statewide constitutional offices, and Democrat Debbie Stabenow was re-elected to the U.S. Senate.
In Wisconsin, Democrat Tammy Baldwin, an unapologetic progressive who was the nation’s first LGBTQ Senator, who was originally targeted by Republicans, won a smashing eleven point victory, defeating a female Trump clone. And Democrat Tony Evers unseated the conservative firebrand Scott Walker.
And in Pennsylvania, Democrats won big victories for both Governor and U.S. Senator.
Currently, in all three of these states, Trump faces strongly negative approval ratings. In Michigan, his net approval rating (approval minus disapproval) is -15, in Wisconsin, -14, and in Pennsylvania, -9.
Clearly, when Democrats work hard in these states, even progressive, female candidates, they can win statewide. It’s when they ignore the states, as Clinton did in 2016, that the Democrats lose.
I would add one additional state to the list of targets Democrats should pay attention to: Arizona. In 2018, liberal LGBTQ candidate Kyrsten Sinema, who ran a mediocre campaign at best, defeated veteran Martha McSally. Currently, Trump has a net -7 approval rating there. Should Democrats fail to win Wisconsin, as Nate Cohn, Dave Wasserman and the Economist’s G. Elliott Morris suggested was possible, Arizona could make up for that loss.
Looking at the 2020 race relative to the 2016 race from a purely data-driven perspective misses an important point: campaigns matter.
The 2016 Democratic Presidential campaign almost couldn’t have been worse, and Clinton still almost pulled it off. In 2018, Democrats showed that we know how to win in these key midwestern states, even with progressive female candidates.
My guess and my hope is that in 2020, Democrats will take the lessons of 2016 and 2018 to the bank. That is the factor these analyses failed to consider. But if anything, it reminds us that we cannot take this race for granted, no matter how unpopular Trump looks nationally.
If you liked this post, you might also like:
