Stanford Political Scientists Weigh the 2026 Midterms: A Wave Election Taking Shape

Robert Grady opened the discussion by thanking the audience and introducing the two guests, whom he described as experts in political science and polling. He said the panel would talk about the upcoming elections and noted that at least eight Senate races were then dead-heat toss-ups.

Grady introduced David Brady as a Stanford political scientist who had written eight books and more than a hundred academic papers, and he promoted Brady's latest book, From Dominance to Parity. Grady summarized its argument: Democrats dominated both houses of Congress from 1932 until the Reagan election of 1980, a realignment that was only partial. Since 1994, when Republicans took over the House under Newt Gingrich and the Contract with America, the country has been in a period of parity in which control has swung back and forth.

Asked what the electorate is feeling, Douglas Rivers said that the country has been on the wrong track for 25 years, but that the numbers have worsened over the last year quite a bit. He said it is a mistake to assume the whole electorate moves together, and he broke the public into groups that respond to events in different ways.

Rivers said the MAGA base, which he put at about 20-plus percent of the electorate, was very negative under the Biden administration and became very positive after Trump took office, at around 95 percent right-track. That figure fell by about 10 points over the last year. Non-MAGA Republicans, in the low teens as a share of the electorate, dropped by about 10 points as well and now sit roughly even at 50-50. Democrats, who were overwhelmingly negative about the country's direction under Trump, are now at 99 percent wrong-track.

Between these groups are independents, whom Rivers said are the reason elections are close. About 25 percent of independents think the country is headed in the right direction, roughly 10 points lower than a year earlier. Rivers said that independent group is the one moving the electorate most.

Grady drew two observations from Rivers's figures. The first was that, even on issues unrelated to politics, people's view of reality now tends to follow their party. He said voters tend to judge the economy as good when a president from their own party is in office and as bad when the other party holds the White House.

The second was that both the non-MAGA moderate Republicans and the independents who backed Trump in 2024 are moving away from him quickly. Grady then turned to Trump's approval, noting Brady's view that the economy is central to elections, and asked how far Trump is underwater on it.

Brady said Trump is in trouble on the economy, with overall economic disapproval at 77 percent, and that his position on inflation is worse. Brady argued that inflation and immigration were the main reasons Trump won in 2024, and said immigration is no longer a strength for him, with his net standing on it only about 15 points negative. When asked to rank issues, Americans put inflation first, the economy second and health care third, and on all three they trust Democrats more to handle them.

Grady returned to the 2024 election polling to show how issue priorities mapped onto the vote. Among voters who put democracy first, Harris won 90 to 6, but they were 9.6 percent of all voters and 7.8 percent of independents. Among voters who named inflation as the top issue, Trump won 64 to 26, a 38-point margin. That was the largest group, at 25 percent of all voters and 22 percent of independents.

Grady then cited YouGov data on Trump's handling of inflation today. Among Democrats, 3 percent approved and 96 percent disapproved. Among Republicans, 67 percent approved and 30 percent disapproved. Among independents, 18 percent approved and 80 percent disapproved, which Grady put at a net of minus 62, a striking reversal given that Trump won independents by roughly 7 points in 2024. Brady added that the shift matters particularly among Hispanic voters, whom he said had moved about 35 to 40 points toward the Democrats compared with 2024.

Rivers said that in 25 years of polling he had never seen anything poll worse than a proposal to tear down the Kennedy Center, with only 6 percent of the public in favor. Grady asked about another issue, having children get rich through questionable means, and Rivers said corruption had not worked for a while, unlike the democracy issues that the Democratic base had embraced. Rivers said Democrats in the midterm campaigns had learned to frame the issues around Trump thinking about himself rather than about voters.

Grady argued that Trump has plenty of popular material to work with. He pointed to GDP growth of 2.2 percent and an unemployment rate of 4.1 percent, while consumer sentiment in the University of Michigan survey had nearly reached its post-World War II low, at about 46, compared with 112 in 2000. Rivers said much of that gloom reflects Democrats turning sharply negative on the economy once Trump took office. He noted that inflation is at 3.5 percent rather than the 9 percent peak, though Grady said prices are about 30 percent higher than in 2020, and Rivers pointed to visible costs such as gas prices and mortgage rates. Rivers's central point was that the problem for Trump is less that the economy is terrible than that he will not acknowledge that any problem exists.

Brady recalled that in 1988 Republicans tried to capture the middle and the suburbs, holding firm on taxes, defense and crime while moving toward the center on environment, education and child care. He said that since the tie election both parties have focused on maximizing their own turnout rather than broadening the coalition.

Brady argued that Trump has not even broadened the coalition inside his own party. He said that after the 2024 election he had told a group of European LPs that if Trump pursued tariffs, mass deportations and retribution rather than high growth and safe streets, he would misread his mandate and pay a high price in the 2026 midterms.

Brady defined a wave election as one in which 30 or more House seats switch parties. He said the usual measure of competitiveness, the presidential vote in the previous election, is less useful than the incumbent's own previous margin. He examined seats where the incumbent won by 2.5 points or less, which would flip with a 2.5-point swing, and seats at 55 percent and 60 percent, noting that in wave years many seats fall into the 55-plus category and roughly 30 to 35 percent do at the 60-plus level.

Brady said independents are moving sharply toward Democrats, with a gap of more than 30 points that he had not seen before. He also pointed to Republican enthusiasm, saying that Republicans are less likely to say they will vote, by 5 to 8 points. Among non-MAGA Republicans, 9 percent for the House and 11 percent for the Senate said they would be willing to consider a Democratic candidate, while only 2 percent of Democrats said they would be willing to vote for a Republican. Brady concluded that he thinks this will be a wave election, certainly in the House.

Grady cited a Pew poll taken about a week before the discussion, in which 63 percent of Democrats said they were extremely motivated to vote, compared with 46 percent of Republicans. He described that as a sizable gap in voter intensity.

Grady then raised the claim, made by some people in the media, White House staff and Republican operatives, that pollsters have systematically undercounted Trump voters, partly because some respondents will not admit their vote or refuse to speak to media callers. Rivers said pollsters have consistently underestimated the Trump vote, but that the size of the miss is much smaller than people think. In 2024, he said, the average poll was one to two points too low on Trump, which amounted to a two-to-four-point error on the lead, a significant miss in an election Trump won by about one and a half points.

Grady asked about mid-decade redistricting, in which California moved in one direction and several Republican states, including Florida, Ohio and Texas, moved the other. Brady said the Downballot project had assembled presidential vote data for 34 of the redrawn districts, which he analyzed by measuring how much Trump won or lost each one by in the previous election. If the wave is as large as he expects, with 30 or more seats changing, Brady said the best Republicans could hope for from the redraw is two or three seats, and in the worst case they might lose two more. Grady summarized that the blue wave might be bigger than people think.

Rivers disagreed. He said the country is in a different political environment than in 2006, 2014 or 1994, and pointed to a dead zone among districts that are Republican-plus-five to Republican-plus-ten. In his view, a uniform five-point swing would give Democrats about 15 or so seats, while a swing between five and ten points would add only about 15 more seats. Rivers said Democrats would need a swing of about 10 points or more to get one of those large gains, and that his own numbers had bounced around, so he had no conviction on whether the figure was 8 or 11 points.

Grady then showed data on 55-plus seats, noting that in 2006 the party lost 30 seats, 17 of them in districts at 55 percent or higher.

Grady opened the Senate section with the current 53-47 Republican majority and said Republicans were most vulnerable in Maine, where Susan Collins is running. He noted that Trump is badly underwater in Maine, and that the University of Virginia and Cook ratings both list the race as a toss-up. Grady also mocked the Democratic candidate selection in Maine, saying the party had first picked a candidate who had a Nazi tattoo and had abused former partners, and then a candidate who reportedly had two children with a second cousin, asking whether the party had learned anything from the first mistake.

Rivers said Jackson leads Collins by two points, a smaller lead than the two points Sarah Gideon had at this stage. He said Collins had survived five terms and that he did not expect her to survive a sixth.

Grady described Alaska as one of the most unusual races. Republican senator Dan Sullivan, who is also Alaska's attorney general, faces a Democratic-filed candidate with the same name, so both appear on the ballot. Grady said the race might be decided by a point, and Brady noted that Alaska uses ranked-choice voting, so the name confusion could matter. Grady said Democrat Mary Peltola, a former congresswoman, is ahead, but that polling in Alaska is later, less frequent and less reliable than in other states.

Rivers said Alaska, Nebraska and Iowa are all roughly even. Of the three, he thought Alaska was the most likely to go Democratic, though he expected all three to go Republican. Brady agreed, describing Alaska as an interesting place to watch campaigns, recalling a Lisa Murkowski event with about 42 to 50 people.

Grady described Republican Mike Rogers as running a television spot that opened with the line Canada is not our enemy, which Grady and Brady presented as a break from Trump on tariffs. Grady also raised criticism of Abdul El-Sayed, the Democratic candidate in Michigan, tying him to the Muslim Brotherhood. Rivers said El-Sayed was running about five points behind the Democratic gubernatorial candidate, and Grady noted that Trump is about minus 20 in Michigan while El-Sayed was up one, though Trump had won the state by about three points last time.

Brady argued that the Democrats would be ahead in Michigan if the other Democratic primary candidate, a woman, had won, and he said that candidate would be up seven or eight points. Grady called it an irony that Republicans' main advantage in some states is the Democratic lurch to the left, and cited Peggy Flanagan, Minnesota's lieutenant governor, who is in a close race with broadcaster Michele Tafoya. Grady also noted that Flanagan had been criticized over fraud cases in Minnesota Medicaid and hospice programs.

Grady said Kansas had not elected a Republican senator in 94 years, referring to the Democrats' chance there. He described Republican incumbent Roger Marshall, an OB-GYN, who had reportedly sued about 700 of his patients over late payments on deductibles and copays, with 81 of them arrested. Rivers said there was a decent chance of a Kansas swing and that he would slightly favor Democrat Adam Hamilton over Marshall because Marshall is a very bad candidate, while noting that races this close are impossible to predict.

Grady then described Montana, where he said Democrats had tried a tactic of nominating a placeholder candidate to get through the primary, with the stronger candidate then running as an independent. In this case the president of the University of Montana was running as an independent, while Democratic nominee Alani Blankhead was reportedly supposed to drop out but refused. Grady said the Republican U.S. attorney was ahead 38 to 30, with Blankhead drawing 16 percent.

Grady said pressure on the Republican Party came from a Freedom Caucus that lost 14 seats, with normal Republicans taking back over, and asked how the party would define itself. On the Democratic side, he pointed to pressure from progressives and the Democratic Socialists of America. He named Peggy Flanagan, Nixon in Florida, and El-Sayed as examples. Grady said both parties appear to be playing to the fringes, which he described as an observable trend over the last 25 years, and he asked whether there is an open space in the middle for a pro-growth Republican who favors research investment, low taxes and regulation, but opposes tariffs and insulting allies.

Brady said money comes from both left and right, and that political activists have long been further left and right, as shown by questionnaires given at conventions in the 1950s. He also said primary turnout is low, so the more extreme voters have an outsized role. Rivers said roughly 55 to 60 percent of Republican primary voters are MAGA, and that anyone who takes an anti-MAGA position is likely to lose in Republican primaries. He said about a third of Democrats roughly think socialism is a better economic system than capitalism, and that DSA has a favorable view among them, with about two-thirds of Democrats saying they would vote for a DSA candidate. Rivers noted that being a DSA candidate does not get a majority among Democrats, but it does get a substantial minority.

Grady noted that current 2028 polling has Kamala Harris leading among Democrats and J.D. Vance leading among Republicans. He said he did not think either would be president in 2028 and doubted that Harris was running. Brady said he did not think either would win, though he expressed no firm conviction. Grady framed the poll leads as largely reflecting name recognition.

An audience member asked whether the independent base had shrunk. Rivers said colleague Morris Fiorina's research suggests people have not moved, with most still in the middle, but the parties have grown more extreme, so the middle voters who switch between parties now follow the party bases. Brady said about 80 percent of Republicans call themselves conservative or very conservative, Democrats are more mixed, and about 55 percent of independents remain in the middle when leaners are sorted out. Grady added that in the presidential states, the independents decide outcomes, citing 2016 and 2020 counties that switched and made the difference.

An audience member asked about prediction markets. Rivers said markets take all polls into account and act like poll aggregators. He described a French-Israeli bettor who believed there were many shy Trump voters in 2020, commissioned polls from Rivers's group, bet heavily on prediction markets and reportedly made around 80 billion dollars. Rivers said the bettor's polls overstated Trump's support and were less accurate than public polls, but that he had been on the right side of the bet.

Grady closed by saying that, in his experience talking with ordinary Americans, most people want the same things: a dependable job, enough money to support parents as they age, good colleges for their children, and safe towns. He argued that the country has moved to the fringes but that a gap in the middle remains open for a more analytic, get-something-done approach.