Q&A of the Day – What states were the pollsters most wrong about in 2016?
Each day I’ll feature a listener question that’s been submitted by one of these methods.
Email: brianmudd@iheartmedia.com
Twitter: @brianmuddradio
Facebook: Brian Mudd https://www.facebook.com/brian.mudd1
Today’s entry: Brian, I’m so glad we have you to cut though the polling bs! You were the only analyst I saw showing Trump with an Electoral College advantage this (last) week. I know you got it right four years ago and that we can’t just trust the polls as they roll in but what I’m wondering is how far off they were. For example, if Biden is shown up 5 in Michigan or 3 in Florida what does that mean? At what point is a lead for Biden in battleground states a real lead for him? Thanks for all you do.
Bottom Line: This note is on back of last week’s Anatomy of a Swing State story showing Trump with a likely 295-243 Electoral College advantage. As I mention in that weekly series, the big miss by most pollsters is within the sampling. For example, in Florida Democrats now only outpace Republicans in voter registration by 0.9%. Thus, any pollster sampling more than 1% more Democrats than Republicans are oversampling them. So back to your question. I’ll use the final RealClearPolitics average of polls from the 2016 Presidential election to first breakout the states Donald Trump won which showed Hillary Clinton in the lead on Election Day.
They were (rounding off the percentages for the ease of conversation):
- Pennsylvania: Clinton +2
- Michigan: Clinton +3
- Wisconsin: Clinton +7
Notably, the final Election Day polling average in Florida showed Trump with a sub-1% lead. What we saw in our state and across the country was a narrowing of the polls leading up to Election Day four years ago in the final week. If you have been following many of the more recent polls you might be seeing similar signs of that occurring once again. The reason the pollsters botched the 2016 election is specifically because of the misses in Michigan, Pennsylvania and Wisconsin. Three states which will likely to be pivotal once again this year. While there are many potential paths to victory for both candidates, in all likelihood, the winner of at least two of those three states will win the election. Now, to most specifically address your question...
As you can see, an average of polls of up to 7% in favor of Biden on Election Day, may still indicate a potential win for President Trump based on what happened four years ago. But there’s also a takeaway for Trump supporters be to mindful of as well. While almost all pollsters botched the election four years ago, the average of the final polling was successful in identifying the winner in 47 states. Completely discarding the polling if you don’t like what you see has the potential to be a primrose path as well. As of today this election is a true tossup.