Longtime Michigan Pollsters Weigh In On What They’ve Learned Since 2016

Presidential polling is a contentious topic and if there’s one thing we can all agree upon, it seems to be that it’s anyone’s race to win — or lose.

While many headlines appear hopeful that Biden can pull off a victory, there’s also a lot of doubt. A cursory Google search of “presidential polls” yields the results “Don’t Believe The Polls, Trump Is winning,” “Biden Leads Trump in Polls As Election Reaches Final Stage,” and “Biden Leads Trump in 2020 Polls But Expect Election Day To Be A Repeat of 2016.”

Who and what are we supposed to believe, especially since 2016 went so differently than what the pollsters had predicted? Detroit Today’s Stephen Henderson talks with two longtime Michigan pollsters about 2016, what they’ve learned since then and who they think will come out on top in the 2020 race.  

Listen: Two veteran Michigan pollsters on what went wrong in 2016 and their thoughts on what the polls indicate for 2020

Guests:

Bernie Porn is a veteran pollster based in Lansing who is a partner and President of EPIC-MRA polling agency.

In looking at this year, Porn says, “I think it’s much different than 2016 in many ways,” adding that “Joe Biden is not Hillary Clinton. Hillary Clinton had higher unfavorables than favorables.” He also notes a much larger number of undecided voters in 2016. 

In looking at the various sources of polling information, Porn says that, to a large extent, he is dismissive of “Trafalgar polls… They correctly called the election in 2016. But of course a broken clock is right twice a day.” Trafalgar polls have predicted a Trump victory from the start of the 2016 race.

Tim Kiska is a University of Michigan-Dearborn Associate Journalism Professor who does tracking on sample counties across the state on Election Night. He will help direct election night coverage for the Detroit Free Press.

In recalling the 2016 election and calling the results for the Free Press, Kiska recalls “we sent people out to 80 precincts across the state on election night and they call us with the numbers form the precincts we had 65 in, and that 65 looked to me like a pretty good sampling of the state in general. In other words, if we had used that same 65 four years earlier we would have been about [less than one percentage point off],” says Kiska.

“I think that a bad call this time around — I don’t want to sound overly dramatic — could damage democracy… I think the pitchfork and the torches come out for me, for good reason” says pollster Tim Kiska on the stakes of announcing the outcome this year. He also talks about the pressure to be first that many media organizations face on elections night as everyone races to make announcements about the results.