The day after the election, veteran Colorado pollster Floyd Ciruli and Fritz Mayer, the dean of the Josef Korbel School of International Studies at the University of Denver deconstructed how the polls and the media missed their mark, again.

How did history repeat itself?

“I think that the industry will once again be spending a considerable amount of time considering that,” Ciruli said.

From Sunday to Tuesday, the narrative was emerging that the gap was closing in key states. Polls are always a snapshot in time.

“It was getting tight in Arizona and Florida and Georgia, a lot of places where it had been 7, 8 points up or even higher,” Ciruli said. “They were overly optimistic in Texas, but the numbers in North Carolina had been much bigger.

But the closing race was not the new story anybody was reporting so late in the race.

“I would attribute that there was no talking about it to a couple of things,” he explained. “No. 1, everybody was talking about the huge turnout … and secondly how it was going to processed (after the election).”

The misses were obvious and scattered. The final composite of polls, for example, had Biden up by 1 percentage point in Florida, where Trump won by 2. 

“I think there were some folks who were in denial,” Ciruli surmised of missing the late change in voter intensity in some states. 

He pointed to how the race changed at the end in 2016, after FBI director James Comey releasing a letter than he was reopening an investigation against Clinton, which was shut down just before the election without finding anything new. Because Clinton held steady when the letter was released, the public, press and pollsters missed the late shift.

“We have a tendency to ignore what’s going on at the end,” when the dice of the race has been cast, Ciruli said. “That surge was really toward the end.”

Ciruli said pollsters thought they had dealt with one of their most significant failures in 2016.

“We thought, ‘Well, even if we can’t get to the Trump voters, we can get to voters who are like the Trump voters with the same demographic characteristics – Anglo men, rural exurban areas, working class, etc., etc., and we’ll get it by weighting the data.’

“I think the national polls show that has not worked properly yet.”

Mayer thinks there is an underestimation of the fervor and mobilization in small towns in rural America that pollsters miss that benefits the right. 

He noted, also, that Trump’s win in Florida appeared to ride on the shoulders of Cuban Americans in the southern tip of the state. That somewhat skewed the expectation of Democrats from Latino voters, because Cuban Americans tend to be more conservative.

In Colorado, however, polls predicted consistently and accurately how the top races came out in favor of Democrats.

Former Gov. John Hickenlooper and presidential nominee Joe Biden held comfortable double-digit leads since clinching their nominations.

Hickenlooper dispatched incumbent Sen. Cory Gardner by nearly 10 percentage points, while Biden won Colorado by better than 13.

Four years ago, Hillary Clinton managed only a 5-point spread against Donald Trump.

Wednesday afternoon’s discussion was sponsored by the Crossley Center for Public Opinion Research, the Josef Korbel School and the Scrivner Institute of Public Policy.

Ciruli is the founder and director of the Crossley Center for Public Opinion Research at the University of Denver, which teaches the art of measuring the mood of the electorate at any given moment.