Congressional Republicans aren’t in a forgive-and-forget type of mood with pollsters who were badly off the mark on Election Day.
A series of high-profile pollsters ahead of Nov. 3 predicted Democrats would win a Senate majority and expand on their control of the House, winning up to 15 new seats. Instead, Republicans did better than expected. The GOP now has 50 Senate seats, to 48 for Democrats, and two Georgia runoff elections are set for Jan. 5. In the House, Democrats are likely to have the smallest majority in that chamber in two decades.
The Maine Senate race was among the biggest polling flubs this year, with Republican Sen. Susan Collins seen trailing her Democratic opponent Sara Gideon, the speaker of the Maine House of Representatives.
Less than a week before Election Day, a poll conducted by Colby College in Waterville, Maine, had Gideon leading Collins 47% to 43% among likely voters. Running for her fifth term, Collins received a polling average from FiveThirtyEight of 41% to Gideon’s 46%. The RealClearPolitics average placed her 5 percentage points behind Gideon, while the Decision Desk HQ gave Collins a 33% chance of reelection.
“The so-called public polls were so far off. It should cause pollsters to take a real look at how they do polling. And I’ll give you some examples of what we saw when I had my pollster take a look at the public polls,” Collins told reporters on Capitol Hill Thursday. “And, for example, one of the public polls published by the Bangor Daily News said I was losing Aroostook County. That is where I’m from. There was no way on earth I was going to be losing Aroostook County. I won it by 68% of the vote, just to give you an idea.”
The public poll numbers were not the only numbers that were against Collins, as she operated her campaign on a $76 million fundraising budget compared to Gideon’s $130 million. But ultimately, Collins pulled through, beating her opponent by a 9-point margin on election night.
“Two things I’ll say about the polling. One is our internal polls were much more accurate than the public polls,” Collins said. “And it was frustrating, because I kept wanting to release our internal polls, but campaigns don’t do that, and it did not show me with 9 points. But the last night that we polled showed me ahead by 4, just to give you the latest one.”
According to the Maine Republican, her campaign did a deep-dive in some of the public polls and noted that the sample sizes were wrong for how many people from each county that the pollsters interviewed.
“We found that they didn’t adjust for age, for gender, for party registration — all which is critical to an accurate poll,” Collins said.
Other prominent Republicans were predicted to lose their races as well.
A late October Morning Consult poll showed South Carolina Republican Sen. Lindsey Graham trailing behind his Democratic opponent Jamie Harrison, a candidate who outraised Graham this cycle, by 2 points. Other polls, including Bloomberg, showed Harrison in a dead heat with Graham or Graham being slightly ahead. Graham won his reelection by 11 points.
A Republican consultant told the Washington Examiner that Graham’s internal polls always had him ahead of Harrison by at least 4 points. Graham chided pollsters on the night of his victory, saying, “To all you pollsters out there, you have no idea what you’re doing.”
The fate of the Senate majority has yet to be decided by the pair of Georgia runoffs, but House Speaker Nancy Pelosi is going into a new Congress with an unexpectedly smaller majority that shocked modelers and political pollsters.
“Democrats are set to have the slimmest Democratic majority since World War II. Pundits doubted us polls were stacked against us. The Cook Report said we would lose 15 to 20 (seats). Majority Leader Steny Hoyer said just days before the election that we would lose 15 seats,” House Minority Leader Kevin McCarthy said on Thursday.
As of Friday, House Republicans flipped 10 seats this cycle, and seven races, the majority of which have the Republican currently leading the count, are still waiting to be called.
