By Zachary Gorchow
President of Michigan Operations
Posted: January 22, 2024 9:50 AM
The polling-media industrial complex is churning to life again in our state as the presidential election year begins, and a new chapter will soon be written in Michigan's turbulent history when it comes to polling elections.
Two of Michigan's pollsters, EPIC/MRA and Glengariff, are coming off a good 2022 when both found Democratic Governor Gretchen Whitmer running away with the governor's race, which Whitmer won by 10.5 percentage points. Other pollsters, like Mitchell Research & Communications and Trafalgar, had the race a statistical dead heat, which represented a polling error outside the margin of error.
The 2018 governor's race saw Mitchell and Glengariff close to Whitmer's 9.5 percentage point victory margin. EPIC/MRA and Target Insyght, however, missed a bit with final polls showing a five- and four-point spread, respectively. That's not an egregious polling error, however.
The 2016 and 2020 presidential cycles were what really upended Michigan's pollsters, which have struggled mightily to draw a polling sample that models the electorate in the Donald Trump era.
There was the infamous polling debacle in the Democratic presidential primary, where virtually every pollster, local and national, had Hillary Clinton with a yawning lead over Bernie Sanders. Mitchell had Clinton +37 and EPIC/MRA had Clinton +24. Sanders won by a little more than 1 percentage point.
For the general election, Mitchell had Clinton +5 on Trump, EPIC/MRA had Clinton +4. Trafalgar had Trump +2. Glengariff, which did its last poll in mid-October, had Clinton +14. Trump won by less than a percentage point.
The 2020 presidential primary saw some overestimation for now-President Joe Biden's victory margin over Sanders in the primary, which Biden won by 16.5 points. EPIC/MRA and Mitchell had Biden +21 and Target Insyght had Biden +41, a major polling error. Glengariff, however, had Biden +6, off by 10.
For the election between Trump and Biden, Mitchell, EPIC/MRA and Glengariff all had Biden +7. Several national polling outfits also had Biden with a clear lead. Trafalgar was at the other extreme with Trump +2. Biden ultimately won by 2 points.
There's a tendency among news organizations, though some are improving, to write about these polls uncritically and drop any of our usual skepticism. Except all these polls have a margin of error. Just because a firm polled a race at Smith 48, Jones 46 in September and then in October that firm polled again and got Smith 49, Jones 45 does not mean Smith's lead grew. It's all just statistical noise, but too often the framing in news stories seizes on these statistically insignificant changes.
Seldom do pollsters pay a price for major errors. Does anyone remember the poll another news organization trumpeted in 2018 showing a dead heat between now-Sen. Winnie Brinks and Chris Afendoulis (Brinks won by 16 points), Scott Dianda leading now-Sen. Ed McBroom by 6 points (McBroom won by 2) and the late Kelly Rossman-McKinney up by 9 points on Tom Barrett (Barrett won by 10)? I do, and I don't want to hear about how polls are a snapshot in time, though they are. These were just flat wrong.
Did anyone care besides me? Probably not.
That's my long lead-up to the latest round of polls on the anticipated Biden-Trump sequel.
In November, EPIC/MRA had Trump 46, Biden 41 in Michigan. New York Times/Siena College earlier in that month had Trump 48, Biden 43. About two weeks ago, Glengariff had Trump 47, Biden 39.
In 2020, the final margin was Biden 50.63 percent, Trump 47.85 percent.
The biggest takeaways I had from the recent Michigan surveys:
There are a few analysts out there who think Biden has no shot, be it because of his age, his record, the inflation of the past several years, etc. But once both Biden and Trump have their party's nominations in the summer, voters will have to confront a Biden vs. Trump choice, something that while it seems inevitable now is not yet set in stone. Once that happens, I would expect Democrats to come home, and we'll be looking at a third straight very close race for president.
What Biden cannot afford is a bigger third party vote in 2024. In 2020, just 1.52 percent of the vote in Michigan went to third party candidates. In 2016, it was 5.23 percent. That was a huge difference in those two races.
I am once again screaming into the void for everyone to chill on polling. Take it for what it is and not treat it as unassailable.
The chances of that happening are about as good as the Pistons winning 10 games this year.
KEY NEWS STORIES THIS WEEK
Here are some stories that stood out to me this week in state government and politics.
No one covers redistricting like Gongwer's Ben Solis. His granular coverage throughout the week really gives a sense of what the Independent Citizens Redistricting Commission is doing, and I am fairly certain Gongwer was the only news source analyzing what the potential changes would mean for incumbent lawmakers.
Did you know that just because a bill passes both houses of the Legislature in identical form that doesn't mean it goes to the governor's desk for signature? There were two such bills last year that passed both houses and were ordered enrolled but are now in purgatory. Gongwer's Elena Durnbaugh has the latest.
The Michigan Republican Party mess continues. Crain's Detroit Business reported on the party going into default on a $500,000 loan.
Hard feelings among the disbanded Racial Equity Workgroup created by the state's Opioids Task Force, which Governor Gretchen Whitmer created, Bridge Michigan reports. The workgroup was to advise the state on how to address opioid abuse among communities of color.
Will the independent review of the Judicial Tenure Commission's investigation of alleged judicial wrongdoing, a review launched given the disproportionate number of judges investigated who are Black, be fair? MLive looks into the inquiry ahead.
That's all for this week. Don't forget to listen to Episode 2 of the new Gongwer and WDET partnership on the "MichMash" podcast. And GO LIONS.