Senate Race Report: October 4, 2026

Question: Thirty days from Election Day, after Inside Elections moved seven states in a single October 1 update, the Senate Leadership Fund pulled its money out of North Carolina to defend Kansas, the New York Times found Kansas tied at 45 percent, and Kalshi’s own desk put Democratic Senate control at 63 percent, is the question still whether the Democrats can win the majority, or has it become how many ways they can still lose it?

Answer: The Democrats are the favorites, and this week the professionals said so with numbers rather than adjectives. Inside Elections moved seven states on October 1, six of them toward the Democrats, and projected a Democratic gain of two to four seats against the four they need.¹ Charlie Cook wrote the same day that at least seven Senate races are polling inside the margin of error and called the Senate “the more explosive of the two chambers.”² Kalshi’s own news desk put Democratic control at 63 percent on September 30, up from 49 percent on September 14.³

The week’s most revealing document, though, was not a poll of anything. It was a spending decision. The Senate Leadership Fund, the chief Senate super PAC of the Republican Party, paused its North Carolina advertising and reserved roughly $10 million in Kansas, a state that has not elected a Democratic senator since 1932, on the same day the New York Times and Siena found Kansas tied at 45.⁴⁵ Money moves before ratings do, and this is the clearest signal yet of where the party has decided the line is.

A 63 percent price is not a majority, and the same update that moved six states toward the Democrats moved one away: Minnesota, which Inside Elections cut from likely to lean Democratic.¹ The Democrats still need four net flips and have not banked one. Decision Desk HQ’s model, the most independent column in the table below, puts them at 54 percent rather than 63.⁶ The honest reading is that the Democrats have become favorites for a majority they have not yet assembled, and the thirty days between now and November 3 are the distance between those two things.

What Changed This Week

The Republican Party’s Senate money changed direction, and that was the week. The Senate Leadership Fund, which had already spent about $32 million in North Carolina and had planned $18 million more over the following month, paused its future advertising there and turned to Kansas, reserving nearly $10 million to defend a state the party won by eleven points in 2020.⁴⁷ The exact figure moved across the day’s reporting, from nearly $8.2 million in the Associated Press account to $9.8 million in CBS News’s reading of AdImpact, against the $17.5 million the fund spent in Kansas in 2020.⁷⁴ NBC News noted that Trump carried eight of the ten states where the fund will advertise from October through Election Day.⁸ Semafor was first to report the shift.⁸ The asymmetry in Kansas is starker than the totals: Marshall has reserved $2.3 million of his own money plus $500,000 from other groups, while Hamilton has reserved $590,000 and draws a little more than $556,000 in outside help.⁷ Trump, asked about the Republican nominee in North Carolina, said of Michael Whatley, “I am worried about him,” and then, “I hope he wins.”⁷ Outgoing Senator Thom Tillis told CNN that North Carolina may no longer be worth the investment, saying, “But the reality is, we have to keep a majority.”⁹

The ratings caught up in one update. Inside Elections moved seven states on October 1: Alaska from tilt Republican to toss-up, Georgia and North Carolina from tilt to lean Democratic, Michigan from toss-up to tilt Democratic, Nebraska from likely to lean Republican, Texas from lean to tilt Republican, and Minnesota from likely to lean Democratic, the single move in the set that runs against the Democrats.¹ Its published outlook projects a Democratic net gain of two to four seats, which is worth reading twice: the forecaster that just moved six states toward the Democrats still publishes a range whose midpoint falls short of the four they need.¹ Cook’s own update that day was a column rather than a ratings change, and its argument was arithmetic: at least seven Senate races now sit inside the margin of error, and the Senate, in his phrase, is “the more explosive of the two chambers.”²

Kansas produced the week’s polling and the week’s strangest endorsements. The New York Times and Siena, in the field September 22 to 30 with 605 voters, found Adam Hamilton and Roger Marshall tied at 45 percent, with the Libertarian David Graham at 3 percent and six percent undecided; the Sunflower State Journal read Graham at 5 percent.¹⁰ Trafalgar, a Republican firm, had Hamilton ahead 44.4 to 43.5 in a survey taken September 29 to October 1 with 1,095 likely voters and a margin of error of 2.9 points.¹¹ Those land on top of a Global Strategy Group poll for the Democratic Attorneys General Association, in the field September 13 to 16, that put Hamilton ahead 47 to 43.¹² Then Mike Hayden, a Republican governor of Kansas from 1987 to 1991, endorsed Hamilton in the Wichita Eagle on October 2, faulting Marshall’s “support for tariffs and the war in Iran, which have increased costs for farmers and other Kansans.” He is the second former Republican governor of the state to do so, after Bill Graves.¹³

The national environment kept widening. Quinnipiac, in the field September 24 to 27 with 1,032 registered voters, put the Democrats ahead 51 to 39 on the generic congressional ballot and the president at 34 percent approval against 61 percent disapproval, which the pollster noted matches the highest disapproval it has recorded since January 2021.¹⁴ The Economist and YouGov found the Democrats ahead by sixteen among likely voters, 53 to 38.¹⁵ The Silver Bulletin’s average, updated October 2, sits at Democrats plus 9.0, up from 6.7 at the start of September.¹⁶

The exchanges moved with the polls. Kalshi’s news desk reported on September 30 that its Senate-control market had reached 63 percent for the Democrats, up from 49 percent on September 14, with the balance-of-power market pricing a Democratic sweep of both chambers at 62 percent and a Republican sweep at 8.³ The day before, its “blue wave” contract, which requires Democrats to hold at least 218 House seats and 49 Senate seats, touched 87 percent, with the higher-threshold “blue tsunami” contract rising from 40.6 to 53.¹⁷ On October 2 it reported El-Sayed at 73 percent in Michigan, an all-time high for him.¹⁸ Ohio went the same way: Sherrod Brown climbed 9.8 points since September 9 to 61 percent, against Jon Husted’s 40.¹⁹ In Texas, Kalshi had Talarico at 60 percent on September 28 and raised him to 62 on September 30 as Ken Paxton fell from 41 to 38 after an audio recording surfaced.²⁰

Michigan tightened to the point where two respected polls disagree about whether it is a lead at all. Fox News, in the field September 24 to 28 with 1,028 likely voters, has El-Sayed ahead by a single point, 50 to 49; Marist, in the field September 24 to 27 with 1,235 registered voters, has him ahead by seven, 51 to 44.²¹ Rogers went up with a seven-figure coordinated buy with the Senate Republicans’ campaign arm, an advertisement titled “Threat” that asserts El-Sayed has ties to “the Muslim Brotherhood and Al-Qaeda.”²² El-Sayed, campaigning in Flint on October 3, said hecklers had been offered $200 to disrupt him.²³

Georgia and North Carolina stopped being contested in public. Karl Rove, speaking at the Texas Tribune Festival, said of both states, “We’re not going to win either one,” and a prominent Georgia Republican told ABC News, “I have never seen an environment this bad for Republicans in Georgia.”⁹ The Republican chair in Peach County, Tim Waters, told the Associated Press that Mike Collins “just doesn’t have what it takes to beat Ossoff.”²⁴ The numbers underneath that judgment: Ossoff has spent $57 million with $13 million more reserved, Collins-aligned outside groups have spent more than $52 million, and Collins’ own campaign has no advertising time booked.²⁴ In North Carolina, a Catawba College and YouGov survey released October 3 put Roy Cooper ahead 47 to 35 among 885 likely voters, widening to 48 to 37 with leaners allocated, with the president’s approval in the state at 41 percent.²⁵

Maine enters its debate phase this week with the race tied in one poll and a three-point Democratic lead in another. AARP’s survey, in the field September 20 to 22 with 982 likely voters, put Troy Jackson ahead 50 to 47, with Collins leading among voters over fifty by 53 to 44 and Jackson ahead by twenty-five among those under fifty.²⁶ CBS News and YouGov, in the field September 23 to October 1 with 1,144 registered voters, described the race as tied.²⁷ The first of four debates is October 6, the day after early voting opens.²⁸

Alaska produced the week’s one forecaster reversal. Decision Desk HQ, which on September 25 became the first forecaster to put a flip on the state, withdrew it: its forecast page updated October 2 rates Alaska a toss-up with Peltola at 60 percent.²⁹ The money there is enormous and tilting Republican. PACs have spent more than $35 million on the Senate race since mid-July, with groups backing Dan Sullivan outspending those backing Peltola by about $4.5 million, roughly $20 million to $15.4 million.³⁰ Peltola answered with an advertisement that distances her from her own party: “If it’s not good for Alaska, it’s not getting my vote.”³¹ Reporting from Alaska Public Media suggests the September New York Times investigation into her treatment of staff has not moved voters, with correspondent Liz Ruskin writing, “I don’t recall hearing from any voters who said this changed their vote, even if it did change their feelings about Peltola.”³²

Iowa’s polls split and its averages barely moved. Fox News had Turek ahead 49 to 47 among likely voters; Rasmussen had him ahead 47 to 43; the New York Times and Siena had Hinson ahead 48 to 47.³³ The Senate Leadership Fund added $20 million to its Iowa buy, taking its total there to $53 million, in a state where AdImpact counts $90 million spent, the sixth most expensive Senate race in the country.³⁴ The one debate both candidates have accepted is October 7 on Iowa PBS.³⁵ Nebraska got its first poll in three weeks, a co/efficient survey in the field September 28 to 30 with 1,113 likely voters that put Ricketts ahead 45 to 40, and the Republican money advantage there is as wide as anywhere: $7 million on air since September 1 with $8 million more booked, against $3.5 million spent and $1.5 million reserved on the Osborn side.³⁶

The war kept setting the price of everything else. Brent settled the week between $102.25 and $105.28, WTI between $89.38 and $92.87, and diesel, which Kalshi’s desk noted had hit an all-time high of $6.53 a gallon, remains the cost that reaches deepest into a farm state: reporting on the Kansas race put diesel up 79 percent from a year earlier, with ranchers facing losses of up to $600 a head.³⁷ Over the weekend at least two more tankers were struck near Oman and Iran, and Iran’s parliament speaker restated that “The Strait of Hormuz will not open until Iran’s seven conditions based on the Islamabad Memorandum are met.”³⁸

The Defense

Michigan is still the most competitive Democratic seat in the country, and for the first time this cycle the two best-known polls disagree about whether the Democrat leads at all. Fox News has the race at a single point; Marist has it at seven.²¹ The market has moved the other way, with El-Sayed at 73 percent on Kalshi, an all-time high.¹⁸ The caveat is unchanged and worth repeating: El-Sayed runs ahead of his party’s national numbers in a state that is even on the partisan index, and a one-point reading thirty days out is not a lead so much as a coin flip with better fundraising.

Georgia is the second seat, and the reporting has stopped treating it as a contest. No new Georgia poll was fielded this window; the September surveys this series reported had Ossoff between eight and eleven points ahead, Cook has moved the race to likely Democratic, and the Republican nominee is being outspent badly enough that his own campaign has booked no advertising.²⁴ A runoff is the only plausible Republican path, and a Republican county chair saying on the record that the nominee “just doesn’t have what it takes” is not the sound of a party planning for one.²⁴

Minnesota is the defense seat that will not settle, and this week it was the one place the forecasters moved against the Democrats. Big Data had Flanagan ahead 40.0 to 37.8, narrowing to 45.8 to 43.6 with leaners allocated; InsiderAdvantage had her ahead by one, 46 to 45, and its pollster noted the result “goes against many of the trends we are seeing in other Senate contests.”³⁹ Cook’s own reasoning is the counterweight, and it is unusually blunt: the race, the report said, was moved to likely Democratic because, in its words, “We’re skeptical of some public polling that has shown the race close.”⁴⁰ One of these is right, and the margin between them is the difference between a comfortable night and a five-flip requirement.

New Hampshire had no new public poll this week, which is its own kind of stability. The most recent, a Trafalgar Group survey in the field September 25 to 27, put Chris Pappas ahead 47 to 41, and Cook still rates the race a toss-up alongside Maine.⁴¹ The Republican, John Sununu, is not, as New Hampshire Public Radio put it, “a Trump Republican from central casting,” which is the reason the state has not moved with the national tide as hard as its lean suggests.⁴¹ Pappas declined a New England College debate in late September, which is a choice a leading candidate makes when the alternative is exposure.⁴²

The other nine Democratic seats, Colorado, Delaware, Illinois, Massachusetts, New Jersey, New Mexico, Oregon, Rhode Island, and Virginia, are not in play.

The Offense, Tiered

North Carolina remains the anchor of every path, and this week the money left it. The Senate Leadership Fund paused its future spending there after more than $32 million, and its own planned $18 million for October is now available for Kansas instead.⁴⁷ Cooper led the most recent survey by twelve points, and the Republican Party’s most prominent strategists are now saying out loud that the state is gone.⁹²⁵ The one Republican argument left is that the party is choosing where to lose, and Tillis made it himself: “we have to keep a majority.”⁹

Ohio and Texas are the two coin flips, and both now lean left of even. InsiderAdvantage put Brown ahead 44 to 43 in the field September 28 to 29, with the Libertarian drawing 4 percent; the New York Times and Siena had it 49 to 46.⁴³ Kalshi’s own desk has Brown at 61 percent, up nearly ten points in three weeks.¹⁹ Trump went to Vandalia on October 3 to rally for Husted, telling reporters of Brown, “They threw him out of the Senate once before. He’s horrible,” and the NRSC chair, Tim Scott, described the race as “as tight as ticks, as we say in South Carolina.”⁴⁴ In Texas, Fox News had Talarico ahead 51 to 49 among likely voters and by seven among registered voters, with Paxton underwater on favorability at 40 to 58.⁴⁵ Karl Rove’s contribution was to make the debate question the whole race: “If he doesn’t debate, I think he loses,” he said, and then, “The question is, does he have the guts to do it?”⁴⁶

Maine, Alaska, Iowa and Nebraska are the bench, and this week three of the four got harder to read. Maine has a tied poll, a three-point poll, and four debates beginning October 6, after early voting opens on October 5, which is the arrangement Collins has preferred all cycle.²⁶²⁷²⁸ Alaska has a withdrawn model, a $35 million spending war in which the Republican side leads by $4.5 million, and a ranked-choice ballot that still contains two candidates named Dan Sullivan, which is the fact most likely to decide a close race.²⁹³⁰⁴⁷ Iowa’s polls split in both directions, with Turek ahead by two to four points in two surveys and behind by one in a third, and the Senate Leadership Fund’s $20 million addition is the clearest sign that the party considers the race live.³³³⁴ Nebraska remains the wildcard, three to five points behind in the only survey of the week, in a state the president carried by about twenty points, and still the cheapest fourth seat on the board because an independent needs no party apparatus at all.³⁶

Kansas is the margin seat, and this week it stopped being a curiosity and became an expense. Two more polls put the Democrat at or above the Republican, one of them from the New York Times; two former Republican governors of the state have endorsed him; and the Senate Leadership Fund has reserved roughly $10 million to hold a state the party won by eleven points six years ago.^1011134 The honest caution has not changed and deserves stating again: Kansas produced competitive September polls in 2014 and 2020, and the Republicans won both Novembers by double digits.⁴⁸ What is different this time is not the September poll but the price. A party does not spend ten million dollars in a state it expects to win comfortably, and the money is the tell.

Florida is not a race again, but it is no longer a formality either, at least in the incumbent’s own telling. Ashley Moody declined to debate Angie Nixon and said of the polling, “we are seeing way-too-close-for-comfort polls in many Republican Senate races,” adding, “The polls have been all over the place. Some have us up. Two have us tied.”⁴⁹ Louisiana and Mississippi remain long shots with real activity: Bill Cassidy, the senator whose seat is open, said of Julia Letlow, “I think Julia is going to win the election in Louisiana,” while conceding “it’s just a bad atmosphere for Republicans,” and Letlow dismissed the poll showing a four-point race as “a bogus poll.”⁵⁰ In Mississippi, Scott Colom told reporters that Hyde-Smith is “too chicken” to debate him, and her campaign manager answered that “losing candidates always want a debate about debates.”⁵¹ Montana and South Carolina are Republican-favored and no longer comfortable: Kurt Alme led a three-way field with 45.1 percent against the independent Seth Bodnar’s 26.1 and the Democrat Alani Bankhead’s 18.9, and Alme skipped the Whitefish debate on October 2, leaving the other three on stage without him.⁵² In South Carolina, InsiderAdvantage had Darline Graham ahead of Annie Andrews by a single point, 46 to 45, and the Post and Courier reported Andrews outraising and outspending Graham while noting, “She’s still the underdog.”⁵³

The nine safe Republican seats, Alabama, Arkansas, Idaho, Kentucky, Oklahoma, South Dakota, Tennessee, West Virginia, and Wyoming, are not in play.

The Math

The forecasters and the markets converged this week, and the convergence has a shape worth naming: the ratings are more cautious than the exchanges, and Inside Elections’ own projection is more cautious still. Inside Elections shows a Democratic gain of two to four seats; Kalshi prices Democratic Senate control at 63 percent; Decision Desk HQ’s model, which uses neither expert ratings nor Cook’s PVI, puts it at 54.¹³⁶ Two of those three describe a coin flip or better for the Democrats; one describes a range whose midpoint loses. The disagreement is not about direction anymore. It is about whether the four seats the Democrats must flip are individually likelier than not, and the honest answer is that three of them are and the fourth is not yet.

The paths have not changed in their order, only in their cost. The market path is North Carolina, Ohio and Texas, plus one of Maine, Alaska, Iowa or Nebraska. That is four flips. The consensus path is North Carolina, Ohio, Maine, and Alaska or Iowa, which requires winning the top four and assumes the defense holds everywhere. The nightmare is any loss among Michigan, Minnesota or New Hampshire, because every defensive loss raises the requirement to five flips, and five flips means winning Kansas or Nebraska or Florida on top of everything else.⁵⁴

The Kansas connection is now the whole of the argument. A month ago Kansas was a curiosity, a week ago it earned a row in the table below, and this week the Republican Party’s principal Senate fund decided it was worth ten million dollars. If the Democrats sweep North Carolina, Ohio and Texas, Kansas is not needed. If any one of those slips, Kansas becomes the margin seat, and this week it acquired the two things a margin seat needs: independent polling that shows it tied, and a party on the other side spending like it believes the polling.¹⁰⁴

The ratings table

The columns are refreshed on different schedules and the differences are real, so the dates are given per column rather than averaged: Cook September 23; Inside Elections October 1; Sabato September 22; RealClearPolitics September 24; Decision Desk HQ October 2; Silver Bulletin October 3. The table gained two rows this week, Minnesota and Nebraska, both of which Inside Elections moved on October 1, one toward the Democrats and one toward the Republicans. It is worth noting where the model columns disagree most sharply: in Kansas, RealClearPolitics and Silver call the race a toss-up while Decision Desk HQ and Inside Elections call it likely Republican.

A word on what the columns mean. PVI is the Cook Partisan Voting Index, and it is the only backward-looking number here. It measures how far a state leans from the nation as a whole, computed from the 2020 and 2024 presidential two-party vote shares weighted toward 2024, and Cook says of it that the race ratings reflect an outlook for which party will win the next election, while the PVI takes a longer view. It earns its place as contrast, which is what makes “Tossup in an R+6 state” legible, and not as a forecast. Lean is a strong prior across the whole map: Pew found 158 of 176 Senate elections since 2012, 89.8 percent, went the way of the state’s presidential carry. But applied as the sole predictor to the nine competitive races this series tracks, lean calls five correctly and misses four, Georgia, North Carolina, Ohio and Texas, every one of which it would call Republican and every one of which the forecasters now favor for the Democrat. That is not a contradiction but a selection effect: a race earns its way onto this table precisely by being a case where the baseline does not settle it. The rule that follows is simple, and it is the reason the PVI column sits first and means least: never argue from PVI against the forecasters, because where the two disagree, the forecasters are the point of the exercise.

There are two traps in reading that column, and both make it look inconsistent with numbers quoted elsewhere. The first is that PVI is denominated in vote share for one party, not margin, which makes it roughly half the size of a margin-style lean: Cook calls Texas R+6 while Silver’s partisan lean calls the same state R+11, and both are right, because the denominators differ. Never place a PVI figure beside a margin figure without saying which is which. The second is that the scale flattens elasticity, how hard a state swings when the nation swings, and elasticity is orthogonal to lean, correlating with it at only r = +0.12 across these races. New Hampshire, at 1.15, and Maine, at 1.13, are among the most elastic states in the country, with many cross-pressured moderates who swing hard; Georgia, at 0.90, is among the most inelastic, with a bifurcated electorate and few persuadable voters, which is why it can be rated comfortably Democratic while leaning Republican on the baseline. Texas, at 1.03, is dead average. A state can be strongly partisan and highly elastic at once, which is why a lean column can never stand in for a competitiveness column.

IE is Inside Elections, Nathan Gonzales’s nonpartisan newsletter, one forecaster among several. The rating scale runs, strongest to weakest, solid, likely, lean, tilt, and toss-up, and Tilt is Inside Elections’ own rating: Cook, Sabato and Decision Desk HQ do not use it, so a Tilt appearing only in the IE column is not a disagreement but a finer scale. Decision Desk HQ publishes win probabilities that map onto the same ladder: above 95 percent is solid, 80 to 95 likely, 65 to 80 lean, and 35 to 65 a toss-up for either party. (flip) marks a rating that forecasts a change of party from the current holder.

A note on the sources. One column in that table is arithmetic and six are not. PVI alone can be reproduced from certified FEC totals, which we have done, matching Cook’s published figures; everything else is somebody’s assessment. Cook, Inside Elections, Sabato and RealClearPolitics are analyst judgment, which means their ratings cannot be independently reproduced from any underlying data, because the rating is the data. That is expert opinion in a table rather than a measurement, and it is not a scandal: election analysis is not a solved problem. RealClearPolitics is worse than unaudited, because it blocks automated reading, so we cannot say whether that column is independent judgment or a composite of the others; treat it as unlisted rather than as a seventh opinion. Decision Desk HQ and Silver are published models, and Decision Desk HQ is the most independent column here, using neither expert ratings nor Cook’s PVI and deriving its prior from the most recent presidential margin adjusted for the national swing. Silver’s FLIPR is not independent in the same way, because in its default build it feeds the Cook, Inside Elections and Sabato ratings in as an input at roughly one-sixth weight, so where his column agrees with theirs the agreement is partly built in and is not corroboration. Silver has acknowledged the recursion risk himself, observing that if expert raters calibrate against his forecast, the whole process becomes somewhat recursive. All six read the same polls and talk to the same operatives, which means six agreeing columns are not six confirmations; the signal is where they disagree, and this week that is Kansas, Iowa, Ohio and Texas. Freshness differs per column and is stated above rather than averaged, and this week that matters: Inside Elections moved seven races in a single update, and no other column has caught up with it yet.

The Close

The report lands thirty days from Election Day, and the interesting thing about this week is how little of it was about the candidates. The Senate Leadership Fund did not stop spending in North Carolina because Roy Cooper said something; it stopped because the race stopped being worth the money. Inside Elections did not move seven states because of a debate; it moved them because the polling underneath had shifted. And Kansas is not competitive because Adam Hamilton is a phenomenon, though two Republican former governors evidently think well of him; it is competitive because the price of fuel and fertilizer in a farm state has become a political fact for which the incumbent has to answer.

That is what makes the final month more legible than the last one. The Democrats are favored, the forecasters have conceded the direction, and the one thing standing between them and the majority is arithmetic they have not yet performed: four flips, no losses, and a fourth seat that is either North Carolina, Ohio and Texas sweeping cleanly, or Kansas at ten million dollars and climbing. The party that spent the week moving money out of a purple state and into a red one has told us which of those it expects. Whether it is right will be settled in the same places it always is, by turnout in a handful of states, thirty days from now.

PRH | huffmanwrites.org | © Philip Huffman

Notes