Within the Margin

We’re deep into the endgame now, and with that comes my final full-board ratings update for the House of Representatives. There are a few races where a last-minute data point might push a race into a new tier, but this is just about where I’m going to rest my case in the House. This update comes with nineteen moves, all of them in Democrats’ direction. That might seem like a lot, but it’s actually much less than what many of the prominent election insiders have done lately…because they waited longer than I did to start shifting races toward Dems in the first place. Among those nineteen moves, five seats are moving from Tilt Republican to Tilt Democratic – so the big board now shows Democrats gaining a net total of 25 seats over their current 215 (which includes one vacant but safely blue seat).

Here’s how the board looks:

Remarkably, I’m projecting Democrats to end up with five more seats than they had after the 2018 midterms. Democrats are entering these elections with considerably more seats to their name than in 2018 when they had 194, so this smaller projected net gain would still put them ahead of that year. I do not offer this projection this lightly: I entered 2026 expecting Dems to flip the House, but with fewer than 230 seats. But the year has unfolded dramatically, with Trump and his party now in a much worse position than they faced even in 2018. A notable difference this time around? Trump’s handling of the economy is rated poorly whereas he still had positive approval on the economy in the fall of 2018…when his party suffered a shellacking anyway with a net loss of 41 seats in the House. Households are battling stubborn inflation and a dramatic spike in gas prices, and these things stem in large part from Trump (largely supported by his Congressional allies) implementing heavy and erratic tariffs, then starting a war in Iran without realizing our adversary could close the Strait of Hormuz and severely impact oil shipping. Voters understand these causes and effects and are apportioning blame accordingly.

Polling, of course, influences this rosy projection for Dems. The public polling this cycle has been all too scarce, but what we’ve seen shows a very wide battlefield with Democrats competitive or leading in deep-red territory. We’ve repeatedly heard insiders from both parties say that their private, internal surveys shows similar results. Most importantly, the parties’ money moves reflect what the polling says: both parties are spending in Republican strongholds, and many swing seats have already been triaged – by Dems confident of victory and Reps conceding likely defeat. An across-the-board polling error is always possible – but it can and does run in either direction. While Republicans have beaten their polls with Trump on the ballot, polling in fact slightly underestimated Dems in the last two midterms, and dramatically so in last year’s statewide races in New Jersey and Virginia. There are methodological reasons – acknowledged by quality pollsters – to worry about Republicans being under-sampled because they’re less likely to answer polls, and to worry about whether you can weight your way out of a bad sample. There are also methodological reasons – also acknowledged by quality pollsters – to worry that Democrats are in fact being under-estimated because respondents misrepresent who they voted for in 2024, and pollsters are increasingly weighting to recalled vote in an effort to address the first issue noted above. Generally, for all of the hullabaloo about rigged polling or bad samples, the best pollsters work incredibly hard at their craft and still usually deliver a product reasonably predictive of outcomes.

Something else to keep in mind: the world around you. Tariffs. The Iran War. ICE far exceeding the “secure the border” talk of 2024 and instead bringing chaos to neighborhoods and workplaces. A president more focused on renovating a ballroom and painting a reflecting pool than addressing these other problems. And this unfolds as his party wavers between outright support and feeble acquiescence. You wouldn’t expect that president, or that party, to be terribly well-regarded right now, no matter how inured you are to Trump-era politics after his second victory in 2024.

Alright – with that out of the way, let’s take a look at the 19 races shifting in this (likely final) House update:

I’m going to group these moves as best I can.

New flips (KY-6, ME-2, MI-7, MT-AL, TX-35): In Central Kentucky, Governor Andy Beshear is working hard to get his ally Zach Dembo elected over Ralph Alvarado – who was the running mate of unpopular former governor Matt Bevin in 2019 and then worked in Tennessee for a while. Probably wasn’t the best nominee. Same goes for bombastic former governor Paul LePage in Maine, who barely carried this district in his unsuccessful race to reclaim the governorship four years ago. He’s 78, lives in Florida and says he’ll continue to do so, and isn’t campaigning hard…in a much redder environment than 2022. I’m finally persuaded Matt Dunlap can pull this off. I held off similarly on MI-7 because of concerns over William Lawrence being too far left to capture this marginal seat, but he’s been skilled enough to ride the wave. Montana’s Sam Forstag combines a cool job – smokejumping – with support from Bernie Sanders and some of his affiliates in a seat that former Senator Jon Tester still carried even while losing in 2024. I think profile and a friendlier political landscape carry the day for the one of the cycle’s more interesting candidates. And finally, I’ve got Johnny Garcia completing a Dem sweep in newly-gerrymandered (by the GOP) South Texas, as a great candidate and truly remarkable Hispanic turnout swamps the GOP despite a redder map.

Dems pulling away (CA-6, OH-7, TX-15): Local Republicans are finally conceding that disgraced abuser Max Miller will not only lose, but probably by a decent margin in Ohio. Republican-turned-independent Kevin Kiley seemed to have a glimmer of hope in his new seat (CA-6) over the summer, but that’s extinguished – safe Dem now. In South Texas, local Tejano music star Bobby Pulido is one of the best candidates of the cycle, and I’m even more confident than I was two weeks ago that he’s flipping this seat and helping statewide Dems along the way by drawing younger Hispanic voters to the polls.

Problematic Republicans present opportunities (non-Max Miller edition – IN-5, NV-2) I thought Victoria Spartz was beatable in the Indianapolis suburbs in 2020 but the region didn’t shift as much as expected. But she’s a odd duck, her opponent has some electoral experience, the environment is bluer, and the lack of a gubernatorial or senate race hear is a recipe for depressed turnout for Republicans who are already seemingly demoralized. And in NV-2, I wrote last month about the GOP’s candidate problems here, making that race more competitive than it should be. Coupled with turnout dynamics in a state where the early vote actually tells us something, I think Flippo’s race is on a knife’s edge.

Mea culpa – I moved too slow (SC-1, VA-5): Events have moved quickly in South Carolina, where the Senate race has Republicans panicking. That’s creating a bit of fright for the GOP downballot, too. I probably should have already had SC-1 in the Lean column. With one of the MAGA super PACs moving money out of SC-1 and into the Senate race, the net effect is probably to address turnout problems caused by the Senate race and in turn avoid downballot disaster in SC-1. Yet even having to do that means the situation is more tenuous there than they’d like, and I feel like Tilt Rep is the right play as a result. And while we try not to read too much into early voting, Virginia ball-knowers increasingly think the strong showing so far in Charlottesville and from Southside’s Black voters point to this race being on the cusp as well. I think that was the case on October 22, and I should have moved it earlier.

Takes a tsunami, but let’s account for the possibility (FL-2, FL-4, TX-2): Dems are running an underfunded perennial candidate in FL-4, but Angie Nixon’s turning out Jacksonville-area voters adeptly in her Senate campaign. If the wave reaches tsunami proportions, this is a place where an outlandish pickup becomes possible. Demographics might do the same in FL-2. And in TX-2, the Dem challenger actually has resources to compete and we’d expect that to pay dividends if we’re truly into uncharted territory.

I don’t really have a category for these, but there’s a candidate quality disparity in AZ-2 between Jonathan Nez and GOP incumbent Eli Crane in their rematch of a 2024 race. One might argue the same in FL-13, where incumbent Anna Paulina Luna might be just a tad too MAGA especially against a decently-funded fellow veteran in Leela Gray. In TN-9, Justin Pearson would seem to be too far left in ideology and demeanor for a newly-gerrymandered and very Republican district, but outside folks are campaigning and spending here. And out on the Western Slope in CO-3, the biggest issue is the top of the ticket, where Republicans nominated a disastrous candidate in Victor Marx for governor and may see even more depressed turnout in their ranks as a result.

Hopefully back soon with a Senate update!