Will Republican Pollsters Try to Skew the Final Averages? They Haven’t Yet 

Just before the 2022 midterm elections, The New York Times’s data guru Nate Cohn flagged that many poll averages had become heavily influenced by “a wave of polls by firms like the Trafalgar Group, Rasmussen Reports, Insider Advantage and others that have tended to produce much more Republican-friendly results than the traditional pollsters. None adhere to industry standards for transparency or data collection.” And as many longtime pollsters had produced fewer surveys, if any, “in some states, nearly all of the recent polls were conducted by Republican-leaning firms.” 

And in October 2024, The New Republic warned of a potential flood of “red wave” polls, reminding readers of a late October 2022 poll from Trafalgar that showed U.S. Senator Patty Murray of Washington state up by a mere 1.7 points, triggering “a raft of ‘Is Patty Murray in trouble?’ stories” before she won by a comfortable 15 points, and catching a new outfit named Quantus Insights bragging on social media how a September poll of theirs flipped the average presidential election margin in North Carolina in Donald Trump’s favor.  

In this year’s midterm election, we’re again seeing poll averages dominated by firms with Republican ties, including the ones mentioned above, as well as co/efficient and National Public Affairs. With one week until Election Day, there’s still time for a late burst of junky polls to push the averages in a Republican direction. But so far, these pollsters haven’t. Their results mostly align with other independent outfits.  

Let’s look at polls of key U.S. Senate races sampled since Sunday, October 4. 

In Michigan, the only nonpartisan pollster in the field has been the Marist Poll, which showed Democratic nominee Abdul El-Sayed ahead by two points. Most of the rest came from Insider Advantage, Trafalgar, and Rasmussen Reports, which all showed El-Sayed leading by 1.2 to 2.3 points. Yesterday a poll commissioned by a progressive political action committee, A Fight Worth Having (which doesn’t appear to be directly involved in the race), has El-Sayed up five.  

The same trio of Republican pollsters has sampled Maine this month, as have Fox News (which uses a bipartisan duo for its polls) and Suffolk University. The Republican polls range from Susan Collins, the six-term Republican incumbent, leading by half a percentage point to the Democratic nominee Troy Jackson leading by two points. Similarly, Fox News found a tie, and Suffolk has Jackson up by two. 

Minnesota has four polls, two from Republican pollsters, but all show the Democratic U.S. Senate nominee, Lieutenant Governor Peggy Flanagan, up three to four points. Georgia is a similar story: four polls, two from Republicans, all between seven- and eight-point leads for Senator Jon Ossoff, the Democratic incumbent.  

In Alaska, Insider Advantage puts the Democrat Mary Peltola, a former U.S. Representative, up in the first round of the ranked-choice voting contest by 5.6 points, and Rasmussen Reports pegs her lead at 2 points. The only other pollster in Alaska since October 4 has been the Democratic-affiliated firm Data for Progress, which found a three-point lead for Peltola over the incumbent Republican, Dan Sullivan. (Alaska Survey Research sampled a poll in early October just before our arbitrary cutoff date with Peltola leading by two.) 

The only public polling in Kansas conducted entirely in October is from Insider Advantage, which has conducted two polls, and one from the A Fight Worth Having PAC. The first Insider Advantage poll put Democratic nominee Adam Hamilton up one; the second put incumbent Republican Roger Marshall up one. If that was intended to suggest momentum for Marshall, it hasn’t worked, as both parties are furiously spending. And the result of the latter poll matches that of the A Fight Worth Having poll. 

Trafalgar and Rasmussen Reports have South Carolina with small leads for the Republican-appointed incumbent Darline Graham of three and four points, respectively, which has fed belief among Democrats that their nominee, Dr. Annie Andrews, a pediatrician, has a real chance. Yesterday a poll published by Targoz Market Research projected an even tighter race with Graham ahead by a single point, but all three are statistical dead heats.

Texas is the most heavily polled state, with 13 polls collectively producing an average lead for Democrat James Talarico of 1.9 points as the state representative attempts to become the first in his party to win a U.S. Senate seat in the Lone Star State since 1988. Three Republican outfits—Rasmussen Reports, Quantus, and Insider Advantage—show slightly more favorable results for Republican nominee, Texas Attorney General Ken Paxton, with two putting Paxton ahead by less than a point and one showing a tie. But Trafalgar gave Talarico an above-average 3.5-point lead.

Iowa is one state where we see a more pronounced divergence between Republican and non-Republican pollsters. Insider Advantage, Trafalgar, and co/efficient put Republican Ashley Hinson ahead by 1.5 to 3 points in her efforts to capture an open U.S. Senate seat, while Cook Political Report, YouGov, and Marist put Democrat Josh Turek ahead by 1 to 3 points. Emerson College, however, has Hinson up two.

Rasmussen Reports and co/efficient are more optimistic for Republicans in Ohio than other pollsters, giving incumbent Republican Jon Husted a slight one- or two-point lead. Insider Advantage’s 2.3-point lead for the Democrat, former Senator Sherrod Brown, is closer to where the rest of the polling industry is at, with five other polls giving Brown leads between three and six points.  

In a few red states where independent candidates are trying to pull off upsets, polling is scant and varies wildly by outlet. A Democratic firm tied to the charter school movement, Concord Public Opinion Partners, has the independent candidate leading in each state: Nebraska’s Dan Osborn by seven, Montana’s Seth Bodnar by two, and South Dakota’s Brian Bengs by one. However, Quantus has Montana’s Kurt Alme ahead by 16 and South Dakota’s incumbent Republican Mike Rounds by 22. National Public Affairs, working on behalf of a conservative political action committee, has Pete Ricketts, Nebraska’s incumbent U.S. Senator and a Republican, ahead by a modest three points. Emerson College has South Dakota’s Rounds up eight and Mason-Dixon has him up by 19, making the Concord’s South Dakota number look more suspect than the Quantus number. Otherwise, these states lack nonpartisan polling.  

So overall, we don’t see much evidence of Republican pollsters warping the averages with just a week to go until Election Day. But note that despite the concerns flagged in 2022 and 2024, we didn’t see much warping in those years either.  

In 2022, a lot of Republican pollsters lowballed Democratic U.S. Senate candidates in Arizona, Georgia, Nevada, and New Hampshire but so did nonpartisan pollsters. The one case with marked divergence, besides the aforementioned Washington state, was Pennsylvania, where several Republican-affiliated polls sampling fully or in part in November showed Republican Mehmet Oz winning, while the only other poll in the field late, Marist, had Democrat John Fetterman winning by six. Fetterman wound up winning by five. 

There were even less fishy data in the final round of 2024 presidential election polls. In nearly every swing state, Republican-affiliated pollsters didn’t dramatically alter the final averages, didn’t produce numbers all that different from other pollsters, and didn’t oversell Trump’s actual performance.  

None of the above amounts to me personally vouching for the credibility and accuracy of every Republican-affiliated outlet, let alone every individual poll. Nor am I arguing that poll aggregators should include data from outlets that don’t meet industry standards.  

But what we are seeing from both Republican-affiliated and unaffiliated pollsters in Alaska, Iowa, Kansas, Maine, Minnesota, Michigan, Ohio, South Carolina, and Texas, at least, are close races that could go either way, with Democratic candidates overperforming in many red states and underperforming in some blue states. Any late flurry of polls, whatever ulterior motives there may be, isn’t going to materially change those narratives.  

And even if we see some late news coverage suggesting that Republicans are closing the gap or taking the lead in certain races, Democratic donors and door-knockers are unlikely to give up. Such reportage could just as easily guard against complacency and ensure maximum effort through Election Day. So even if Republican pollsters are planning to manipulate the final averages, they would only be wasting their time.  

The post Will Republican Pollsters Try to Skew the Final Averages? They Haven’t Yet  appeared first on Washington Monthly.