In 2016, the election analysis community was wrong. They had overestimated Hillary Clinton thanks to her national lead, underpolled states that Trump ended up winning, and led a nation to believe that one outcome was basically certain. Of course, it did not end up that way. So it’s understandable to see that same community try and hedge its bets more when it comes to the 2020 election. The problem starts to be, though, that these two races aren’t the same race.
National Polling
To be completely fair to the polling industry, when it came to the national polls in 2016 they were within the margin of error right before Election Day.
Basically every pollster except for Monmouth was within the margin of error when it comes to the final tally for the national vote. Most actually had it at Clinton +4. This would not have been far off if 100,000 votes went in another direction. The national polling wasn’t really that far off. In fact, it was pretty spot on, if only overestimating Gary Johnson’s final vote tally. National polling errors were never really the problem.
2020’s national polling
What about this year’s polling then? Well, a major difference are the quality of the third party candidates. Instead of having Gary Johnson and Jill Stein lined up as their candidates, the Libertarians and Greens instead chose Jo Jorgensen and Howie Hawkins. These aren’t exactly high profile candidates. Remember, Johnson was considered to be doing so well at this point in 2016, that CNN gave him and his VP nominee Bill Weld two primetime town halls. Not only does this put us in a two-party election, it also puts the spotlight more directly on both candidates for all Americans.
The numbers don’t show it to be good for Trump at all in this period and time when it comes to national voting. Most solid pollsters show Trump down double digits or very close to being down double digits nationally. Some have countered though with the belief that the polls showed Hillary Clinton with a similar lead at this time in 2016. This claim is false or at the very least cuts out most polls at the time.
As you can see above, polling nationally at this time in 2016 was actually eerily similar to the final polling numbers from 2016. Except for a couple polls that showed Clinton in the double digits range, most showed her at the more typical 3-5 point national vote lead. It was never a consistent near or at double-digit lead for her like we’ve seen for Joe Biden.
Polling the Swing States
The real issue in polling came from the swing states. Pollsters had big misses in Michigan, Wisconsin, Pennsylvania and North Carolina going into election night. All four states were either in the lean D or in the case of Wisconsin, likely D on that night. All four ended up going for Trump. There was especially an underpolling of both Michigan and Wisconsin compared to other states.
Pollsters have at least tried to correct that mistake from last time. Michigan and Wisconsin are being polled more frequently compared to 2016. They have also adjusted their polling to fit the education levels in these states, something that was very off in 2016. Recent results in those polls show good news for Biden too. The last Marquette poll in Wisconsin showed Biden up eight points on the President, while in Michigan the New York Times/Sienna poll showed Biden up by eleven points. A recent poll in Pennsylvania from The Washington Post/ABC also showed Biden up ten points over Trump.
These results and data points matter. While Hillary Clinton had one off leads like this in 2016, we are again seeing this more consistently with Biden this year. A consistent, strong lead in states like Michigan and Pennsylvania. That creates a huge difference for how we should be talking about the election and the chance Biden or Trump ends up the victor.
It is No Longer the Same Atmosphere
The biggest thing the analysts and pollsters are overlooking is something that isn’t data or polling related. It’s the atmosphere surrounding the election. The atmosphere and the culture behind the two opposing campaigns are just so different from the ones we saw in 2016. Clinton was the old school in a time where America wanted to throw the book away.
Trump, who at that point was untested, unknown and unproven in any sort of governmental leadership. He was the trash can to throw the book away into. I can tell you this because I was certainly part of that group who just wanted to throw our governmental norms away. That’s all changed though. People have now seen Trump in action as a leader, especially during the pandemic and have turned heavily against him, more so than they were ever against Hillary in 2016.
The Pandemic changed everything
Before the pandemic, I don’t know if I would’ve disagreed with you that the 2020 election was looking different than 2016. Biden was starting out with a 5-6 point advantage over Trump, but the economy was moving right along as expected and things seemed generally ok. It seemed like people would once again willingly vote for Trump because of the very good economy America was in. The COVID-19 pandemic changed all that and its effects on our election should not be forgotten.
Trump’s failure to show strong, or at the very lest decisive leadership, over the pandemic has effected his approval ratings everywhere, even his own party. The most damning polling number for Trump so far, has to be from Reuters a week and a half ago that showed him at only 80% approval with Republicans. This was the first data point showing the atmosphere has changed. Gone has the energy and the want for a Trump campaign. Even among those who approve of him, there is a tiredness emerging of the tweets, the comments, and the poor interviews. People may still like Trump’s idea’s but they are getting tired of the man. That doesn’t happen without the pandemic.
This is the 2020 Election – Not 2016
It’s time to put to bed the notion that the 2020 election is anything like 2016. With the pandemic, the polling and the changing atmosphere around Trump, the 2020 election is making the case for its own type. It has its own storylines, its own pathways, and will have its own final result. We must talk about it like that as such.
