By Ashok Nilakantan Ayers
NEW YORK: America is heading toward the November 3 midterm elections in an unusually unsettled political mood. Voters are worried about the price of almost everything, the cost of borrowing, gasoline and utility bills, wars abroad, immigration, jobs and the disruptive possibilities of artificial intelligence. And increasingly, they are asking a simple question: Is the country moving in the right direction?
An NBC News reporting trip through some of the most competitive House and Senate territory captured that frustration at street level. In New Jersey, Pennsylvania, Ohio, Iowa and Nebraska, voters interviewed by NBC repeatedly returned to the economy and the cost of living. Some lifelong or repeated Trump voters said they believed the president had failed to deliver on promises or had become disconnected from the pressures confronting ordinary households.
That mood matters because midterm elections traditionally become, at least in part, a referendum on the party occupying the White House.
But the 2026 election is not simply a national referendum. It is 50 state political environments and 435 individual House contests, with turnout, candidate quality, local issues and redistricting capable of producing results very different from the national polling numbers.
And that is where the apparent contradiction in this election begins. Donald Trump insists Republicans can produce a “big surprise” on November 3. He made the prediction as early as February 2025, when he told Republican governors that he expected the midterms to be a major surprise. He has now embarked on an intensive campaign schedule, with Reuters reporting that he plans 32 days of political travel before Election Day.
Yet the national numbers entering October are difficult to ignore. A September New York Times/Siena poll of 1,503 likely voters put the Democratic congressional candidate ahead of the Republican candidate 51 percent to 43 percent, an eight-point advantage. Trump’s approval rating in that survey was just 38 percent, while voters disapproved of his handling of the economy, immigration, tariffs, the Iran war and the cost of living.
An Emerson College poll conducted September 21-22 was even more striking: Democrats led Republicans 53 percent to 42 percent, an 11-point advantage on the generic congressional ballot. Trump had a 39 percent approval rating against 58 percent disapproval. Sixty percent of respondents said Trump’s economic policies were making the economy worse, compared with 28 percent who said they were making it better. Those numbers explain why Republicans are concerned. But they do not automatically mean a Democratic sweep.
The NBC reporting from the road helps explain the numbers better than a national poll can. In Pennsylvania, one three-time Trump voter complained that everything had become more expensive and blamed the administration’s policies and the war for adding to the burden. In Iowa, farmers are dealing with the consequences of trade policy and disrupted commodity markets. In Nebraska, a nurse described colleagues taking second jobs because the cost of living had risen faster than their household budgets.
That is politically significant because voters do not necessarily distinguish between inflation, energy prices, tariffs, interest rates and foreign-policy consequences when they sit at the kitchen table and calculate the monthly bills.
Emerson found that the economy was the single most important issue for 41 percent of voters. Jobs, tariffs, gasoline prices, immigration and interest rates were also high on the list.
There is another complication for both parties: artificial intelligence. The New York Times/Siena survey found voters divided over which party they trusted more on AI, but strongly opposed the rapid expansion of data centres. Sixty-one percent opposed new data-center construction, including 47 percent of Republicans. For politicians in states courting billions of dollars in AI investment, that could become an unexpectedly potent local issue.
The Senate is where the Democrats’ national advantage becomes considerably less straightforward. Republicans enter the election with a 53-47 Senate majority and Democrats need to gain four seats to control the chamber. Several Republican-held seats are now genuinely competitive.
In Ohio, the latest polling has Democrat Sherrod Brown narrowly ahead of Republican Jon Husted, with one October 1 poll showing Brown at 49 percent and Husted at 46 percent.
In Iowa, Republican Ashley Hinson has been running close against Democrat Josh Turek. A September 30 poll put Hinson at 43 percent and Turek at 47 percent, while other polling has produced different results.
Maine is similarly competitive, although the polling picture has varied. One October 1 survey had Democrat Troy Jackson and Republican Susan Collins tied at 50 percent apiece, while other recent surveys have shown narrower margins in either direction.
Michigan is another closely watched contest. Recent polling has Democrat Abdul El-Sayed ahead of Republican Mike Rogers, with one aggregate putting the Democratic margin around three points.
North Carolina currently presents a larger Democratic lead in polling, with Roy Cooper ahead of Republican Michael Whatley in the latest available polling average.
Then there is Texas. The fact that a Democrat such as James Talarico is polling ahead of Republican Ken Paxton in a state Trump carried comfortably in 2024 illustrates just how far the national political environment has shifted. A recent poll reported Talarico at 51 percent to Paxton’s 45 percent.
Alaska has also entered the conversation, with Democrat Mary Peltola running competitively against Republican Senator Dan Sullivan. None of these numbers should be read as forecasts. Senate races are decided state by state, and individual polls can move considerably during the final month.
The House presents a different mathematical problem. Republicans hold only a narrow majority, and Democrats need relatively few net gains to take control. The latest national generic ballot numbers suggest that the Republican vulnerability is real, but the precise number of seats that could change hands remains uncertain.
A Washington Post analysis based on generic-ballot polling and presidential results estimated a Democratic House range of 225 to 232 seats if current conditions persisted. But that is a model, not an election result, and redistricting makes translating a national vote percentage into a precise House seat count particularly difficult.
Pennsylvania provides a useful snapshot. A September Times/Philadelphia Inquirer/Siena survey found voters preferring Democratic control of the House 56 percent to 42 percent.
Nebraska’s Omaha-based 2nd Congressional District, meanwhile, is regarded as competitive after Republican incumbent Don Bacon’s retirement, with Democrats holding a slight edge in national prognosticator assessments.
New Jersey’s 7th District offers another warning sign for Republicans. Republican Tom Kean won the district by 5.4 points in 2024, while Trump carried it by only 1.2 points. The district’s Hispanic population and concerns over immigration policy add another layer of uncertainty.
So, is Trump really behind? By the national polling indicators, yes. But “behind” does not mean defeated. The evidence entering October shows Democrats leading the generic congressional ballot by anywhere from several points to double digits, depending on the poll. Trump’s approval rating is underwater, and voters are particularly dissatisfied with the cost of living and economic conditions.
At the same time, Republicans retain structural advantages in several Senate contests, incumbents can outperform the president’s approval rating, and the national generic ballot does not determine the Electoral College or a House majority directly.
There is also a historical lesson: polling describes the electorate at the moment respondents are questioned. It does not capture every late campaign development, turnout decision or October event. That may be what Trump is counting on.
His strategy is increasingly to make the election a choice between his administration and the Democrats rather than allowing individual Republican candidates to run entirely on local issues. His campaign appearances are designed to energise the Republican base and turn enthusiasm into turnout. Reuters reported that Republicans are simultaneously trying to determine how closely their candidates should associate themselves with Trump.
The Democrats face their own problem. Their advantage is being driven substantially by dissatisfaction with Trump and the economy, but voters also continue to express reservations about the Democratic Party. The Times/Siena survey found that 50 percent regarded Democrats as too far left, while socialism remained an unpopular label.
That leaves November 3 as something of a national political stress test. The country is frustrated. The Democrats currently have the stronger numbers in the national congressional ballot. Republicans still possess competitive Senate terrain and the ability to turn out Trump’s intensely loyal base.
And Trump, despite the polling deficit, is betting that turnout, late campaigning and the particular political geography of American elections can produce the surprise he has been predicting for nearly two years.
The real question is therefore not whether Americans are angry. The NBC road trip suggests they plainly are. The question is where that anger finally lands on the ballot — and whether it produces a broad national shift or a collection of sharply contrasting state-by-state results.
On November 3, the answer will come not from Washington, the White House or the polling studios, but from voters in Ohio, Pennsylvania, Iowa, Texas, Michigan, Maine, Alaska, New Jersey, Nebraska and dozens of other closely fought districts and states. And that is why, with thirty days to go, the polls may tell us which way the wind is blowing — but not yet how hard it will blow. (IPA Service)
