I know it’s close to Halloween, but the weird vibes emanating from Democrats about the New Jersey gubernatorial race seem too frightful to be real, as I explained at New York:

Off-year gubernatorial elections are often treated as major bellwethers for the next presidential or midterm elections. So there’s a lot of interest in this year’s contests in New Jersey and Virginia. Both states have leaned Democratic at the presidential level in recent years; Republicans last carried Virginia in 2004 and haven’t won New Jersey since 1988. But both states also have a recent history of swinging against the party controlling the White House in gubernatorial contests. The non–White House party has carried 11 of the past 12 governor’s races in Virginia and eight of the past nine in New Jersey. The exceptional winners were Democrats Terry McAuliffe in Virginia in 2013 and Phil Murphy in New Jersey just four years ago.

These states are different even though they often move in tandem. Virginia is the last state with a one-term limit on governors’ terms, which probably increases the “nationalization” of gubernatorial races since there are never any incumbents. They also have different demographics, though both states have significant non-white voting blocs and large and often political crucial suburbs. Until very recently, Virginia was thought to be a near lock for Democrat Abigail Spanberger, in part because she’s a suburban centrist running a well-oiled campaign against the erratic Republican lieutenant governor Winsome Earle-Sears. But the other factor was Virginia’s heavy federal-employee presence, which hurts Republicans thanks to the unparalleled hostility of the second Trump administration toward the so-called deep state (made vivid first by DOGE’s Elon Musk and then by OMB director Russell Vought, both big-time bureaucrat-haters). Virginia Republican despair has been at least temporarily dispelled by a scandal involving violence-laden texts by Democratic-attorney-general candidate Jay Jones. But the damage probably won’t extend much beyond Jones himself.

There’s a more negative vibe in New Jersey, though, where Democrat Mikie Sherrill is facing Jack Ciattarelli in the governor’s race. Ciattarelli significantly exceeded expectations and polls in his 2021 challenge to Murphy. And the president he is now firmly embracing, Donald Trump, cut his losing margin in the Garden State from about 16 points in 2020 to just under six points in 2024. As Ron Brownstein notes, this is one place (others are Texas and Florida) where you could discern a general broadening of the GOP coalition:

“Trump somewhat improved his showing in 2020 over 2016 among minority voters nationwide, especially those without a four-year college degree. In 2021, Ciattarelli in particular advanced further from those beachheads. He significantly narrowed Murphy’s advantage from 2017 in Passaic and Hudson, the two New Jersey counties with the largest share of Hispanic residents, and improved even in Essex, the county centered on Newark, which has a large Black population. In 2024, Trump ran even better than Ciattarelli in all three of those counties, even becoming the first GOP presidential nominee in the 21st century to win Passaic.”

In 2022, Democrats lost one highly marginal U.S. House seat in New Jersey after redistricting made it “redder.” But they did relatively well otherwise, which complicates the “trending red” narrative, unless you accept the GOP rationalization that this particular election was dominated by an abortion issue that has since receded in significance.

More important, the idea that Ciattarelli will build on his gains in 2021 and Trump’s gains in 2024 to win in November involves ignoring the larger phenomenon of backlash against the party controlling the White House. Are Republicans in New Jersey somehow stronger now than Trump was in November 2024, when he lost New Jersey (albeit by “only” six points)? CNN’s Harry Enten uses favorability numbers to suggest that the always-unpopular Trump isn’t really more unpopular than he has ever been. But given the change of party control of the White House, it’s now Trump’s job-approval numbers that are more relevant, and they are not good, and that’s particularly true among the non-white voters who trended toward Trump in 2024. We don’t have that much publicly released New Jersey data, but nationally Trump’s 2024 gains were concentrated among the “low-propensity” voters least likely to turn out in a non-presidential election. It stands to reason that some of them won’t show up for an off-year gubernatorial contest.

There is some talk that Sherrill is either running a bad campaign or is unexciting to Democratic-base voters — or both. But as Brownstein points out, Sherrill, like Virginia’s Spanberger, is a strongly positioned candidate this year:

“Spanberger and Sherrill are both centrist Democrats with national security backgrounds who were elected to the House of Representatives as part of the 2018 blue wave. Both have run careful, disciplined campaigns that have drawn criticism from some Democrats for failing to ignite much enthusiasm or passion, but also praise from analysts in both parties for avoiding missteps …

“Sherrill and Spanberger have each maintained a predominant focus on the cost of living. That contrasts with their Republican opponents, who have attempted to revive several of the wedge issues that have benefited the GOP in recent years.”

So why is there such pessimism, bordering on incipient panic, about Sherrill among Democrats? She has been comfortably leading in polls other than those taken by the notoriously pro-GOP Trafalgar-InsiderAdvantage and Quantus Insights outfits, and she’s leading even in those. But some Democrats fear the polls never entirely capture Trump or pro-Trump votes. Beyond that, there is something I can describe only as a sort of superstitious belief that Trump has changed everything, repealing all the rules we used to understand about how elections are won and lost. Check out this take on Sherrill by Politico’s Michael Kruse:

“What at the outset of her political ascent was a formula that looked like a model — former Navy helicopter pilot, former prosecutor, suburban mother moderate in ideology and mien — at this messier, storm-the-gates moment marked by economic populism and TikTok histrionics, often can feel as if it falls flat. ‘She’s playing the politics,’ as twentysomething Jersey City mayoral candidate Mussab Ali put it, ‘of 2018.” And it’s 2025.'”

Some of this angst about Sherrill could simply reflect intraparty tension that has little to do specifically with New Jersey, as younger and more outspokenly progressive Democrats yearn for the candidate nominations they feel centrists have squandered (though Sherrill dispatched a couple of leftward rivals in the June primary). But you could also argue that Democrats are still in trauma over what happened nationally in 2024. Since they can’t believe a convicted felon campaigning on a pledge to seize authoritarian powers to pursue vengeance against his enemies actually won the presidency, they fear anything’s possible now, particularly if their candidates aren’t as loud and vicious as the president their opponents adore.

Like Virginia, New Jersey is competitive enough that upsets are possible on November 4. But its Democrats should stop spooking themselves once Halloween has passed.

All our debates over Democratic strategy are strictly dependent on protection of our right to oppose the current administration at all, as I explained at New York:

When looking at judicial review of Trump 2.0’s many audacious power grabs, it’s easy to get bogged down and tangled up in legalisms. Constitutional law is complicated. Federal court procedures are not designed to cope with unprecedented assertions of presidential power advanced almost hourly in places all over the country.

But now a three-judge panel of the Seventh Circuit Court of Appeals, in a ruling that halted a National Guard deployment in Chicago, wrote some sentences that cut through the fog like a powerful search light and reached the real point of contention:

“Political opposition is not rebellion. A protest does not become a rebellion merely because the protestors advocate for myriad legal or policy changes, are well organized, call for significant changes to the structure of the U.S. government, use civil disobedience as a form of protest, or exercise their Second Amendment right to carry firearms as the law currently allows. Nor does a protest become a rebellion merely because of sporadic and isolated incidents of unlawful activity or even violence committed by rogue participants in the protest. Such conduct exceeds the scope of the First Amendment, of course, and law enforcement has apprehended the perpetrators accordingly. But because rebellions at least use deliberate, organized violence to resist governmental authority, the problematic incidents in this record clearly fall within the considerable day-light between protected speech and rebellion.”

In other words, the judges (one of whom was appointed by Trump, another by George H.W. Bush) slapped down as absurd the administration’s claim that protests against ICE’s activities in Chicago constitute a “rebellion” that warrants otherwise illegal deployments of military force in a U.S. city. And neither Donald Trump nor Pete Hegseth nor Kristi Noem nor Tom Homan nor Pam Bondi can turn these protests into the equivalent of the Whiskey Rebellion, the Civil War, or a foreign invasion. Nor can Texas governor Greg Abbott, who is eager to send his own National Guard units to Democrat-governed Illinois in what amounts to a war between the states.

It’s increasingly clear that treating political opposition as a rebellion is at the heart of the administration’s legal case for the militarization of political conflict that goes well beyond protests against ICE raids. In MAGA-speak circa 2025, the “Democrat Party” is now the “Radical Left,” and everything it does is presumptively illegitimate and probably illegal. Just yesterday White House press secretary Karoline Leavitt bluntly asserted that “the Democrat Party’s main constituency are made up of Hamas terrorists, illegal aliens, and violent criminals.” Earlier this week House Speaker Mike Johnson said the peaceful No Kings rally in Washington planned for October 18, which will feature massive displays of Old Glory and countless patriotic gestures, is insurrectionary: “This ‘Hate America’ rally that they have coming up for October 18, the antifa crowd and the pro-Hamas crowd and the Marxists, they’re all going to gather on the Mall.”

This follows onto the threats of repression broadcast by the president and by his top domestic policy adviser, Stephen Miller, after the assassination of Charlie Kirk. Both men blamed this crime by a deranged individual on Trump opponents writ large, with Miller going so far as to suggest that calling his boss “authoritarian” was an illegal incitement to the kind of violence that murdered Kirk, and an act of “terrorism.” Trump’s subsequent executive order called for a literal war on “antifa,” the shadowy and scattered network of protesters, that is useful in the ongoing clampdown precisely because it’s nowhere and everywhere. Meanwhile, his so-called secretary of War called in the entire leadership of the U.S. armed forces to mobilize them for duty against the “enemy within.” This entire escalation of rhetoric, to be clear, is the logical culmination of the president’s relentless campaign of demonization throughout the 2024 campaign that treated opponents as anti-American, anti-Christian crooks who were deliberately destroying the country and importing millions of criminals to steal elections.

Suffusing this militant attitude is the pervasive belief in MAGA circles that Trump’s narrow 2024 victory represents a mandate to do whatever he wants. It’s unlikely, in fact, that the swing voters who pulled the lever for Trump because they wanted lower gasoline or grocery prices or better border control bought into the full Trump 2.0 agenda, which is why his job-approval numbers are well underwater. But even if they did buy the whole enchilada, the 49.8 percent of voters who backed Trump do not have the right to revoke the constitutional rights of the remaining 50.2 percent. That would be true, moreover, had the 47th president actually won the “historic landslide” he keeps mendaciously claiming.

The words of the Seventh Circuit judges really do need to become a rallying cry against the administration’s efforts to use every bit of power it can amass to silence and intimidate opponents and critics. Political opposition is not a rebellion and doesn’t justify a repression that turns half the country into suspected terrorists. This president has more than enough power to pursue his policies without ruling like a king. Enough is enough.

There’s a lot of strange impatience among Democrats about receiving an official DNC “autopsy report” about 2024, but at New York, I made an argument that it won’t be that helpful until the next presidential cycle arrives.

Look, I get it: There are many reasons Democrats feel the need to look back at the electoral calamity of 2024. The Democratic presidential nominee, Kamala Harris, has books to sell. Joe Biden loyalists feel they must rehabilitate his tarnished image. Operatives and donors who were knee-deep in the Biden or Harris campaigns naturally have scores to settle and grudges to air. And above all, the ideological warriors of the Democratic left and center want to blame each other for the debacle, just as they’ve blamed every Democratic defeat large or small on each other since about 1968.

In wallowing in the 2024 defeat, Democrats are avidly assisted by Republicans experiencing intense Schadenfreude at their misery. The GOP is deeply invested in spinning the close 2024 results into an irreversible realignment that will make Donald Trump and his heirs masters of the universe until the end of time.

So Democrats will not be able to eschew 2024 reminiscences altogether. But they should give it a try. The Washington Post reported earlier this week that the Democratic National Committee was slow-walking its official “autopsy report” on 2024 until 2025 elections are over, out of concern that negative discussion of the party (and, for that matter, of the accuracy or inaccuracy of the “autopsy” itself) might affect organizers’ morale or even voter turnout. Here’s a better idea: Democrats should put off any official 2024 “autopsy” until late November 2026, when the midterms are done.

This recommendation does not stem from a preoccupation with vibes or a belief that Democrats can’t handle bad news or division over what happened in 2024. The more basic truth is that much of what happened in 2024 is probably irrelevant to what will happen in 2026, and revisiting it all is just a big, fat waste of time, at least until the next presidential election cycle arrives. Here’s why.

Different elections, different electorates

Midterm elections are fundamentally different than presidential elections in multiple ways. Basically, different electorates show up for each. Presidential election turnout is invariably higher (it was 67 percent in 2020 and 64 percent in 2024). Voters who participate in presidential but not midterm elections are often referred to as “low-propensity voters.”

Until very recently, Republicans had an advantage among the “high-propensity voters” most likely to show up for midterms. But in the Trump era, that advantage has shifted to Democrats. So a lot of the endless debate over Trump’s gains among low-propensity voters in 2024 might not even be relevant to the 2026 electorate.

Comparative versus referendum elections

Presidential elections are mostly comparative, i.e., a choice between two candidates representing the two major parties (although perceptions of the party controlling the White House have a significant effect on that choice). Midterm elections are mostly referenda on the party in power, particularly when that party has trifecta control in D.C., as Republicans do today. So polls showing that voters favor one party or the other on certain issues can be a bit misleading; their perceptions of the president’s performance on those issues is more germane.

This is why at least some of the fretting about the supposed weakness of the “Democratic brand” coming out of 2024 is probably excessive. In a first-past-the-post system dominated by two major parties, the “out” party will benefit from any and all misgivings about the “in” party. Trump’s persistently underwater job-approval numbers help explain why he’s trying to rig the midterms through gerrymandering and voter suppression.

There is also a tendency, which is real but hard to quantify, for voters who are aligned with or even support the agenda of the president’s party to vote against it once it’s in office as a “check against presidential power.” This helps explain why the party controlling the White House almost always loses congressional seats (and often governorships and state legislatures) in midterms. And no president has tried to grab more power than Donald Trump.

The 2024 election was extremely unusual

The situation facing voters next year isn’t going to resemble the one that existed in the very strange 2024 election. Whether their “brand” is weak or strong, Democrats are not going to be led by 81-year-old Joe Biden and then by a relatively untested Kamala Harris. Yes, some Democrats believe they have too many old politicians in office or running for office, but it’s a different problem from a historically old man being the accepted head of the party and the most powerful person in the world.

Similarly, it makes a world of difference that Democrats will not control the White House and Congress in 2026. There is an ineradicable group of voters (growing larger with younger cohorts) who are profoundly unhappy with the status quo and will swing between the two parties based on who controls the country. This “I hate everything” vote was a millstone for Democrats in 2024. It won’t be in 2026.

The midterm battlegrounds are different

The 2024 election was fought over seven battleground states that were seriously contested by both parties: Arizona, Georgia, Michigan, Nevada, North Carolina, Pennsylvania, and Wisconsin. Trump carried all of them, which created the mirage of a landslide (as though all those 75 million Democratic votes didn’t actually count). In the 2026 midterms, the big battle will be over competitive Senate and especially House races. Of the nine Senate races deemed competitive by Cook Political Report, just three are in 2024 battleground states. Thirty-nine House races are rated as competitive by Cook. Eleven are in 2024 battleground states. Different strokes (and messages) may be appropriate for different folks.

Some political lessons are timeless

Without a deep dive into the particulars of 2024, Democrats clearly made some mistakes that you don’t need an “autopsy” to identify. It’s been obvious at least since the swiftboating of John Kerry in 2004 that falling silent in the face of relentless opposition attacks is almost always a very bad idea — see the Harris-Walz campaign’s decision to look the other way or change the subject as the Trump-Vance campaign relentlessly pounded her using clips from the bizarre 2019 interview in which Harris appeared enthusiastic about spending taxpayer dollars on gender-assignment surgery for prisoners who were also illegal immigrants. I’m reasonably sure future candidates won’t make that mistake.

The threat posed by Trump is no longer hypothetical

The single biggest reason 2024 is relatively useless as a model for 2026 is that Trump won in no small part because a significant slice of voters simply did not buy Democratic claims that he was dangerously authoritarian, cruel, and indifferent to the suffering he wanted to inflict on noncriminal immigrants and people dependent on government help to make ends meet. Some remembered his first term as relatively benign (aside from a pandemic for which he was not blamed), while others, particularly younger voters, thought all politicians were pretty much the same.

We’ve now had more than nine months of dramatic proof that Democratic warnings about Trump 2.0 were, if anything, understated. That won’t matter to Trump’s MAGA base; indeed, their own anger and hostility to democracy seem stronger than ever. But it will matter to many of the same swing voters who opened the door to Trump’s return to power.

Donald Trump lies so much that it’s easy to get lulled into complacency about him. But it’s important to call him out on some made-up “facts,” as I argued at New York:

President Trump says all sorts of wild things on his infamous Truth Social account. But this week in his fury at Fox News for not becoming even more lavishly sycophantic toward him than it already is, he said something so hallucinatory that it really must be called out (emphasis mine):

“Fox should either get on board, or get off board, NOW, but at this point, it just doesn’t make any difference to me. They suck up the Ratings because of us, and then spin them in the Democrats’ direction. They refuse to put up Polls that correctly show me at 65% in Popularity, a Republican RECORD, but if I have a fake bad “Poll,” many of which are done by Fox (One of the worst “Pollsters” out there is the FoxNews Poll), they put them up immediately, and with gusto. Republicans are so tired of this fight with Fox always trying to be so ‘politically correct!’ Thank you for your attention to this matter.”

I have no idea where Trump came up with that 65 percent number, but it wasn’t from any publicly available poll measuring his job-approval rating, past or present. The highest second-term job-approval number I can find in RealClearPolitics’s vast database is from the extremely Trump-friendly outlet Insider Advantage, which showed him at 56 percent approval on the day of his second inauguration (January 20). His all-time-highest job-approval rating in either term was from the equally Trump-friendly Rasmussen Reports, which put him at 57 percent a couple of days after his first inauguration. Neither of those peak numbers, of course, is anything like 65 percent. For context, his net job approval has been underwater every day since March 13 this year according to RCP’s averages, which are generally better for Trump than others. And during his first term, according to the same outlet’s averages, Trump’s net approval rating went underwater, too, a week after he took office and stayed that way until he finally vacated the White House.

As for that “Republican RECORD” Trump cited … well, not really. George W. Bush hit 90 percent and his father, Poppy Bush, hit 89 percent job approval per Gallup (the pollster with the most historical data) immediately after their respective wars with Iraq. Ronald Reagan hit 68 percent job approval twice (in 1981 and in 1986). Gerald Ford managed a 71 percent job-approval rating from Gallup in 1974. Richard Nixon got to 67 percent job approval in 1969 and again in 1973. And Dwight D. Eisenhower got to 79 percent just after his reelection in 1956. Trump is actually the only post–World War II Republican president never to have reached the 65 percent he claims, and he hasn’t even gotten close, not once, ever.

Now it’s entirely possible Trump pays somebody to tell him what he wants to hear about his popularity, much as he apparently hired a failed historian to tell him his narrow 2024 presidential victory was a historic landslide that gave him a mandate to do as he wishes. But if he’s mad a Fox News — or Insider Advantage, or Raz — for not showing him to be the most popular president of them all, he should probably keep it to himself. Better yet, someone he respects should let him in on the not-so-secret reality that even if he believes he’s saved or is saving America, Americans don’t much like him.

With the government shutdown now underway, Democrats need a strategy to fight and win. But first they need to define what victory looks like, as I explained at New York:

2025 has been excruciating for Democrats as they have been playing defense perpetually against a rapacious Trump administration and its lapdog Congress. So now they’re using the one legislative weapon at their disposal by denying Republicans the Senate votes to keep the federal government operating. The move is partially a reaction to grassroots Democratic fury over their unwillingness to take just this step in March when a previous stopgap spending bill expired. Polls currently show a plurality of rank-and-file Democrats favoring a shutdown. And that doesn’t take into account the particular zest for combativeness that is so evident among the donors and activists who tend to dominate intraparty politics, which is mirrored in the incessantly pugilistic language of Democratic politicians these days. I would be rich if I had a dollar for every Democratic public figure who has deplored “bringing a knife/pen/words to a gunfight” recently or has accused the party as a whole of lacking spine.

But now that they have taken on a high-stakes fight with a GOP that controls everything, led by a president who loves conflict and destruction, congressional Democrats and their supporters need to develop a clear understanding of what they can reasonably hope to accomplish. There are three basic outcomes on the table:

Democrats get everything they want

Democrats’ demands for keeping the government open — or, presumably, for reopening it — are clear. They united behind a counterproposal to the GOP’s “clean CR” measure to extend current spending authority until November 21. Democrats called for three major concessions: (1) extension of the enhanced Obamacare premium subsidies enacted in 2021, which are due to expire at year’s end, affecting approximately 22 million policyholders who will otherwise face very large premium spikes; (2) repeal of major Medicaid cuts in the recently enacted One Big Beautiful Act; and (3) reversal of and future curbs on Office of Management and Budget director Russell Vought’s efforts to rescind or cancel previously appropriated funds and to abolish congressionally authorized programs. Presumably they will also demand that Vought’s new threat to lay off huge numbers of furloughed federal employees be canceled or reversed.

To be clear, there is almost no chance Republicans will bend on the Medicaid cuts or Vought’s spending clawbacks. So total victory for Democrats is a chimera, and political leaders should find ways to let “the base” know that.

Democrats score a partial win, brokered by Trump

The policy concession Democrats might realistically hope to win, in part if not in whole, is the Obamacare-subsidy extension. The GOP left this extension out of the One Big Beautiful Act because hardcore conservatives hate Obamacare. However, many Republicans, including some in Donald Trump’s circle, fear the midterm repercussions of a premium spike that would hit a lot of middle-class voters in their own coalition. Owing to GOP divisions on the subject, this concession would almost certainly have to be imposed by Trump himself — and he won’t admit it’s a concession at all but simply a reflection of his own policy preferences. On the other hand, even if Trump goes in this direction, conservatives will fight hard to restrict eligibility for the premium subsidy, forcing Democrats to choose between half a loaf and none. Either way, Democrats could rightly claim to be doing whatever they could to protect the popular health-care benefits that Republicans continuously seek to deny or cut.

Since Vought’s threatened mass layoffs are contingent on a closed federal government, it’s also possible Democrats could secure their reversal as part of a deal to reopen the government along with the traditional provision that furloughed employees receive back pay. Altogether, such an outcome really would represent a rare Democratic legislative victory.

Democrats walk away with only a messaging victory

If Trump and Republicans simply refuse to make concessions, pleasing the government-haters in their ranks and risking economic repercussions, Democratic resistance will likely melt away with time. All it would take is eight Senate Democrats to reopen the government. What the party could claim as a prize in any event is the opportunity to preview its midterm messaging with a major focus on health care and Trump’s contempt for laws and norms. Depending on the length of the shutdown and how it plays with the public, Democrats might also satisfy some of the incessant craving for “fighting Trump” while educating the rank and file on the limits of their powers of obstruction.

Ultimately, the most effective weapon Democrats possess is the opportunity to flip control of one or both congressional chambers in 2026. Having done what they could in 2025, they could essentially tell their base it’s its turn to take up the fight by mobilizing and persuading every vote possible. But if the rifts between Democratic politicians and their supporters are to be healed in anticipation of the midterms, leaders would be wise to keep rank-and-file expectations of what “winning the shutdown” looks like quite low. It’s one battle in a long war, and not an ounce of Democratic passion should be wasted in recriminations over how much “fight” and “spine” House and Senate Democrats exhibited.