Usually, when political hacks talk about the “undervote” we are talking about how many fewer votes a particular race had than votes for President or Governor in a district or state. However, the 2016 election exposed a new phenomenon: the presidential undervote.
In Arizona, there were 2,554,240 votes cast for President. However, there were a total of 2,661,497 ballots cast. That means that 107,257 people voted in the election but did not cast a vote for President. That is an astounding number when you compare it with the undervote in previous presidential elections. In 2012 it was 34,777, in 2008 it was 28,771, and in 2004 it was 28,734.
And the presidential undervote this year was not unique to Arizona. A Washington Post analysis shows that in the 33 state where they were able to collect the data, 1.7 million people cast a ballot but did not vote for President. Arizona ranked 2nd highest as a percentage of people who did not vote for President (behind Montana).
Obviously, the record number of people abstaining in the Presidential election is a reflection on the candidates. Donald Trump had some seriously unfavorable ratings heading into Election Day, as did Hillary Clinton. And Clinton didn’t inspire the Democratic base the way Obama did in 2008 and 2012.
Clinton was also a terrible candidate, and while the media fawned over the “amazing” operation coming out of Brooklyn, we now know that they were too smart for their own good. They made some assumptions about the election, created a model, and then never tested the model with inputs throughout the campaign.
On the morning of Election Day, internal Clinton campaign numbers had her winning Michigan by 5 points. By 1 p.m., an aide on the ground called headquarters; the voter turnout tracking system they’d built themselves in defiance of orders — Brooklyn had told operatives in the state they didn’t care about those numbers, and specifically told them not to use any resources to get them — showed urban precincts down 25 percent. Maybe they should get worried, the Michigan operatives said.
Nope, they were told. She was going to win by 5. All Brooklyn’s data said so.
President-elect Donald Trump’s campaign on the other hand was hopeful, but not confident, that he would pull off the win. They also had a data operation, but the difference was they were continuing to get inputs and making changes to strategy up to the weekend before the election. It was not just on a whim that Trump went to Pennsylvania, Michigan, and Wisconsin in the last days of the campaign.
The question is how will future Presidential elections look? Is the 2016 presidential undervote an anomaly, or is this the new normal? We won’t know for four more years.