We have about six months before the nation faces one of the most consequential elections since the Great Depression. With a country politically polarized, an pandemic killing thousands of Americans, and an economy on the brink of collapse, this election will decide how the nation will face this crisis over the next four years.
“All States are equal, but some States are more equal than others.” (Apologies to George Orwell)
Political scientists and political commentators seem to have settled on thirteen states that they classify as a swing states: Wisconsin, Pennsylvania, New Hampshire, Minnesota, Arizona, Georgia, Virginia, Florida, Michigan, Nevada, Colorado, North Carolina, and Maine.
This post is the first of a series on key presidential swing states. In this post, I will look at Michigan and follow up with analysis on Pennsylvania, Wisconsin and Florida.
These four states combined account for 75 Electoral votes and have a history of close elections and a tendency for voting both Democratic and Republican Presidential candidates.
In 2016, Donald Trump won Michigan’s 16 Electoral votes with 47.25% to Hillary Clinton’s 47.03%. In the past six election cycles the state has chosen two Republicans and six Democrats. It’s clearly an equal opportunity state for both Biden and Trump.
I have collected 14 independent Michigan polls from January through April of this year that included the presidential match up between Biden and Trump. On average of all the polls, Biden leads Trump 47.3% to 42.5%, with an average lead of 4.8%.
In the graph above, two lines represent the percent for each candidate over this 14 poll period. The blue line is Biden’s percent and the red line Trump. Notice that in the beginning of the year, Biden started out ahead, but his lead dropped and eventually recovered. All poll results are listed below.
Only one survey during this period gave Trump a plurality and it was only by 2% (italicized). Using multiple surveys allows for averaging out the statistical error. Almost as important, is having multiple surveys from different pollsters which helps eliminate bias. If you want a good idea of who is leading, in polling more is always better.
Another way to measure Biden’s advantage in Michigan is subtracting one poll question from another. This is called the lead difference. In the chart below, Biden’s net percent average lead was 4.8%.
In the first poll (1.07), Biden started out with a 7% advantage, but after that his lead started to deteriorate until the March 7th survey shows Trump had was leading by 2%.
But surprisingly, after this point Biden’s poll numbers started to improve in almost every other survey there after. So what happened? The only explanation I have (with no empirical evidence) is Covid-19.
Notice that beginning on March 16th, Trump’s begins to loose ground and Biden has bigger gains. It was in the beginning of March that the Covid cases began to explode. And Trump’s percent of the survey vote, deteriorated quickly. Unfortunately, I can’t confirm this hypothesis at this point because I don’t have Michigan Covid case data as of yet. But I will soon.
Politic’s Doctor Verdict: Biden likely win. Trump won the state in 2016 by the slimmest of margins but this years polling shows he has consistently fallen behind Biden. With the campaigns frozen in time, he will need a lot of help from Biden to win the state again.