No telling who’ll win Harder-Denham race, polls and experts say

Jeff Denham, left, and Josh Harder, are running for the 2018 10th Congressional District seat.
Jeff Denham, left, and Josh Harder, are running for the 2018 10th Congressional District seat.

A new independent survey of likely voters indicates challenger Josh Harder has a slight edge over incumbent Rep. Jeff Denham in one of the United States’ most competitive House contests.

But the difference is small enough that most experts continue saying the 10th District race is impossible to call.

Harder, a Democrat, was favored by 50 percent of people in the recent poll by the University of California Berkeley Institute for Governmental Studies, while 45 percent preferred the Republican incumbent and 5 percent were undecided.

Factoring in a 5 percent margin of sampling error, “Statistically, that’s a tie. The best we can say is, `They’re tied,’” said Keith Smith, a political science associate professor with Stockton’s University of the Pacific.

The UC Berkeley poll was conducted Sept. 16 to 23, before Senate confirmation hearings for Brett Kavanaugh. Some analysts say Republicans defending the Supreme Court nominee now are more motivated, and that could influence turnout for the Nov. 6 election.

An earlier poll was done in late June by Garin-Hart-Yang Research Group, hired by Democrats. It found Harder and Denham locked in a dead heat at 48 percent each.

That bump of the needle has Inside Elections joining the Cook Political Report and Sabato’s Crystal Ball from the University of Virginia, with all three declining to predict a winner in the Denham-Harder race. Experts looking at both polls and other factors say the race is neck and neck — the same thing they’ve been saying for months, with one exception: Inside Elections, which had been giving Denham a sliver of an advantage, recently changed its prediction to “toss up” status.

Nate Silver’s FiveThirtyEight, which uses a dizzying formula to compute voters’ mood, stands almost alone in saying Harder is ahead by a 2.6-percent whisker, 51.3 percent to Denham’s 48.7 percent.

How important are polls?

“Polls are one piece of information people use to understand the world,” Smith said. “The more the press fixates on polls, the more they matter in people’s understanding.”

But such surveys can be wrong. Remember that almost no polls or experts predicted Donald Trump would prevail over Hillary Clinton in 2016.

“Polls, I think, are good for candidates and bad for democracy,” said David Colnic. He is chairman of the political science department at California State University, Stanislaus in Turlock, the city claimed by both Denham and Harder as their home base.

Candidates use polling to gauge how people are responding to their messages, and campaigns adjust accordingly. But poll results, drummed by media, lead people “to focus on the horse race rather than the issues,” Colnic said. “We’re captivated by who’s in front and we lose sight of the larger issues.”

Local issues usually are discussed when people respond to polls. For example, the Berkeley survey found that health care was mentioned most often as an important subject facing the 10th District. Immigration and gun laws ranked a little lower, followed by the economy and water policies.

Berkeley pollsters used email to survey 726 people — a good sample size, Colnic said — registered to vote in the 10th District, which covers Stanislaus County as well as Ripon, Escalon, Manteca and Tracy. All respondents said they were absolutely certain they would vote in November.

Of the 10th District’s 339,000 registered voters, 37.6 percent are Democrat, 34.4 percent are Republican and 22.6 percent belong to no party.

Notes of interest from the Berkeley poll:

88 percent said they have a high interest in the Denham-Harder race, and 62 percent said this congressional contest is more important to them now than in past years.

30 percent of people who support Trump said it’s important that they vote for Denham in order to support the president, while 47 percent of people who don’t like Trump said their vote for Harder is a way to oppose the president.

49 percent have an unfavorable opinion of Denham, compared to 35 percent for Harder.

61 percent oppose Nancy Pelosi regaining speakership of the House.

30 percent would vote to repeal the gas tax increase in Proposition 6, while 41 percent would not.

FivethirtyEight’s credibility soared for having given Trump, before the 2016 election, more of a chance to win than other polling services. In the current “deluxe” version of its forecast giving Harder the edge, FiveThirtyEight relied on both 10th District polls plus surveys in similar districts; predictions by Cook, Sabato and Inside Elections; trends from other midterm elections; presidential approval ratings; and a bunch of “fundamental” data.

The last category mashes factors such as Denham’s success in four House elections, tempered by the fact that Congress has a dismal 17.9 percent approval rating; the District has a conservative bent, but favored Clinton in 2016 and Barack Obama in 2012; Harder is raising more money than expected; Harder has never held office; and results from the June Primary.

The Los Angeles Times commissioned the Berkeley poll, and reported additional numbers from the 10th District not publicly available. Notably, The Times said 53 percent of white college graduates favor Harder, as do 65 percent of Latinos, while 55 percent of whites without college degrees favor Denham.

Smith said polls can’t capture intangibles that might decide the outcome.

“My expectation is Denham wins because he’s weathered a couple of other elections where people thought he would lose and he still won by a healthy margin,” Smith said. “People like him. He’s got a good personal brand within the district, and those are hard things for challengers to overcome.”

Garth Stapley: 209-578-2390

Polling Service/ExpertDenhamHarder

Nate Silver’s FiveThirtyEight deluxe (current)


UC Berkeley Inst. of Gov. Studies (Sept. 16-23)


Garin-Hart-Yang Research (June 27-July 1)

Inside Elections, Cook, Sabato (current)