Sept. 4, 2024, 6:51 p.m.
There’s a polling crisis in America, and it has nothing to do with weighted averages, sampling errors, or the "shy Trump voter" phenomenon. It’s the lack of reliable polling on House races this cycle.
Ask any campaign strategist worth their salt, and they will grumble that, while presidential polling is a dime a dozen, there’s a serious lack of public data on the races that will decide control of an entire branch of government, especially in the House.
Two years ago The New York Times regularly polled several key congressional contests that ultimately decided legislative control in the midterms. But now the New York Times/Siena College poll, rated the best in the country, has shifted its focus to presidential battleground polling.
“House polling in a presidential year is like the puppy bowl as the undercard for Ali v. Frazier,” veteran Democratic strategist Cole Leiter said. “There is a small, devoted core audience standing next to the marquee contest of a generation.”
Other news organizations have also pared back their polling this cycle as media companies reckon with the soaring costs of high-quality polls. If there’s a competitive Senate or gubernatorial race in a state—as in Michigan and North Carolina—pollsters are happy to add in a question, but rarely in states that won’t decide control of the White House.
“The lack of public polling in the House has made it harder to get a clear understanding of what’s beyond that top line,” Noble Predictive Insights Chief of Research David Byler told National Journal.
An analysis of FiveThirtyEight’s polling archive found a severe dropoff in House polls over the last three cycles.
Between Jan. 1 and Sept. 2, 2020, 150 polls were conducted of general elections for the House—113 with a partisan sponsor, 37 without. Between Jan. 1 and Sept. 7, 2022, 141 polls were conducted of general elections for the House, 79 with a partisan sponsor, 62 without. Then this year, between Jan. 1 and Sept. 4, 91 polls were conducted of general elections for the House, 60 with a partisan sponsor and 31 without.
The black hole of House polling makes it more difficult for voters, donors, and pundits to discern which races to follow and potentially support financially. Several campaign strategists and pollsters told National Journal the lack of polling could lead to more surprise results when the dust finally settles after Nov. 5. It also makes holding campaign officials accountable a bigger challenge for the media.
At the Democratic National Convention last month, Democratic Congressional Campaign Committee Chair Suzan DelBene told reporters that Democrats’ margins in House races remained largely unchanged despite the turmoil at the top of the ticket. She might be right, but without public polling there’s no way to fact-check that claim.
It’s not just election nerds. It’s local news and local reporters who are trying to figure out what’s going on in these races. ”
—Jacob Rubashkin of Inside Elections
To be sure, the campaigns, committees, and PACs that will play a big hand in deciding control of Congress are still out in the field with their own polling. Whether they decide to publicize their results is another question entirely.
Strategists point to several trends to explain the lack of reliable horse-race polling in the House. First, most of the races that will decide the chamber, such as those in California and New York, do not overlap the presidential battleground. This makes polling individual races a little more cumbersome, and at times, makes it more expensive to get a representative sample. Second, there is a larger audience for topline and national polling than for individual House surveys, or even for the Senate race in Montana, which could determine control of the upper chamber.
The drama at the top of the ticket hasn’t helped spur interest in downballot races. President Biden stepping aside, and former President Trump surviving an assassination attempt, have renewed interest in presidential toplines when attention could have been diverted down the ballot if the previously sleepy White House race had dragged on as it was. Now pollsters—and readers—cannot get enough of the presidential ticket.
“Sometimes it can feel like one of those kids' soccer games where everyone swarms the ball and it’s this giant cluster,” Byler said.
Enter the election handicappers at The Cook Political Report with Amy Walter and Inside Elections with Nathan L. Gonzales. Traditionally, the forecasters have relied on a mix of private polling and candidate interviews to make their race ratings. But this cycle, they decided to get in the game themselves.
In May The Cook Political Report launched the Swing State Project, a collaboration between the forecasters, GS Strategy Group (a Republican polling firm), and BSG (a Democratic polling firm). The partnership’s focus is mostly on the seven swing states that will determine control of the White House, with other statewide and issue polls sprinkled in.
One of the few reliable sources of House polling is the new partnership between Inside Elections and Byler and his team at Noble Predictive Insights. So far this cycle, the election handicappers have released three polls of House battleground races and plan to release at least two more.
On Tuesday, they released their latest survey, this time from Oregon’s 5th District. The survey found a result within the margin of error in the race between Republican Rep. Lori Chavez-DeRemer and Democrat Janelle Bynum, in a district Biden carried by 9 points four years ago. The Inside Elections poll is one of only five publicly released surveys in the last two years of this district that could determine control of the House.
“We saw an unmet demand for this type of work that we were happy to meet,” Inside Elections Deputy Editor Jacob Rubashkin told National Journal. Rubashkin said the outlet briefly dabbled in polling during the 2018 cycle, but went full-bore this cycle after The New York Times stopped its regular House polling
“There is a demand for this that isn’t being met,” Rubashkin said. “It’s not just election nerds. It’s local news and local reporters who are trying to figure out what’s going on in these races.”
There is some value in the lack of public horse-race polling in the House races. Some strategists said that absence is a welcome change from the anxiety of constant head-to-head matchups. Instead of focusing on the toplines of a race, pollsters can focus on the issue and messaging polling.
“It’s bad for the hobbyists, but it doesn’t impact the ability to run races,” Leiter said about the lack of horse-race numbers.
But the lack of public polls threatens to produce a less-informed electorate. There are hundreds of contests all over the country that will be consequential, and there is a shocking dearth of information. In fact, polling about candidate quality might be even more important in the downballot races that are not getting the wall-to-wall coverage of the presidential race.
“Ultimately, polling is the best tool we have for understanding the thoughts and behaviors of large groups of people,” Rubashkin said.
