The U.S. general election is on Tuesday.1
The Surprisingly Low Stakes
In the short term, there isn’t very much at stake.
This is not too surprising. In 2024, I ranked the 2024 election only the fifth most important of our lifetimes, and, even after the past two years, I stand by that. However, even if you continue to view the past two years as the greatest disaster in living history, and the 2024 election therefore the single most important election of all time, the 2026 election isn’t going to do much to change the Trump White House. Whatever happens on Tuesday, Mr. Trump will still be in office for two more years.2
The entire U.S. House of Representatives is up for replacement.
If Republicans win, they will continue to lie absolutely supine to the will of Mr. Trump.
If Democrats win, they will launch vigorous investigations into the Trump White House. However, the trend since the Obama Administration has been that the White House increasingly just ignores House subpoenas, so, although quite welcome, it’s unclear what this will actually accomplish.
It is likely, though not yet certain, that House Democrats will eventually impeach Mr. Trump. However, because conviction on impeachment requires the agreement of two-thirds of the Senate, there is essentially no chance that impeachment would have any effect. After all, Trump’s already been impeached as many times as every other U.S. president in history combined—and, each time, acquitted by a minority, despite his obvious guilt of the charges brought against him.3
If House Democrats pass Democratic legislation, Trump will veto it. This is probably for the best, since most partisan Democratic legislation (much like most partisan Republican legislation) is absolutely terrible. However, it also allows Trump to continue to grow executive authority with no effective check in Congress.
One-third of the U.S. Senate is up for re-election. This is Senate Class II, the group of senators last elected in 2020.
If Republicans win, they will continue to lie absolutely supine to the will of Mr. Trump.
It is possible that, with a Republican Senate, Justices Alito and Thomas would retire in 2027 or 2028, in order to be replaced by a Republican president. However, both justices seem to be following Ruth Bader Ginsburg’s arrogant path into dying on the Court and destroying the legal legacy of the movement that put them there. Both should have retired in 2025-2026, and I’m furious that they didn’t.
…or, I would be furious, but then I remember that Donald Trump is functionally the President. Trump is furious at the Supreme Court, which he sees as a traitor court, because it keeps ruling against him when he obviously breaks the law (on tariffs, birthright citizenship, and so forth). He considers Barrett, Gorsuch, and Kavanaugh special failures (n.b.: they were by far the greatest successes of his reasonably successful first term) and has mostly shaken off the influence of the Federalist Society that brought their names forward. I tremble at the thought of who Trump might appoint to replace Thomas now, and I can’t help wondering whether perhaps Alito and Thomas are staying on the Court because they’ve had that thought, too. That, at least, is the charitable interpretation of their decision to stay on.
If Democrats win the Senate, they will stop Trump from appointing any more officials or judges for the remaining years of his term. This would be unfortunate, since Trump’s lower-court judges have been the bright spot of his grim administration. Despite Trump’s best efforts, his judges have not been “loyal” to him, but have consistently been loyal to the rule of law. (I assume he simply can’t be bothered to pay much attention to lower-court nominations, and continues to be dazzled by Ivy League credentials.) In fact, because Trump’s judges are, by and large, originalists, they’ve routinely been more loyal to the letter of the law than the Democratic opposition that spends so much time fretting about the rule of law!
The Senate will not pass any Democratic legislation, either. The Republicans would filibuster it. Democrats may have the votes to break the filibuster,4 but they have no reason to do so until they hold a trifecta. If they did break the filibuster now, Trump would veto their legislation anyway.
If Democrats take either house (or both), they can demand concessions on spending, but, since they cannot trust the President to obey laws and have few means to force his compliance, this will mostly be limited to defunding. In particular, ICE is likely to be defunded, as is the Iran War. If they hold at least one house, Democrats’ leverage will be significant, but not unlimited; holding out for too much for too long in a government shutdown could trigger a premature U.S. debt default, for which they might take much of the blame. The extent to which their power over the purse actually constrains the executive branch remains to to be seen, but it could be substantial. (Under the Constitution, it is supposed to be plenary!)
At the national level, that’s pretty much it.
Lots of state governments are having important elections that will determine control.
Minnesota, for example, must decide whether to hurl itself back into the abyss of a DFL trifecta, or maintain the reasonably effective divided government we’ve had for two years.
States pass most of the laws that make the largest differences in the lives of normal people, so we probably should pay more attention to state races. The actual policy stakes are all here. Laws will be passed, or fall short, based on who controls state houses. The gubernatorial race in Minnesota will have a substantially larger impact on more people than the MN U.S. Senate race, even though Tafoya vs. Flanagan gets far more national attention. However, it’s very hard to track hundreds of races in fifty different states with fifty different cultures and top issues, so our political culture mostly just doesn’t. The main thing politicos notice is that, if Democrats win as many governors’ mansions as expected, they are going to engage in some truly ugly retaliatory gerrymandering, they will win the gerrymandering war for 2028, and Republicans will have no one but themselves to blame for it.
Besides, as far as I’ve been able to tell, nobody at the state level has much of a clue what to do about [gestures] all this, either. There are lots of grand plans on file that amount to “we’ll nibble at the edges of our great crises, subsidize you with your own taxpayer money in the meantime, and hope for the best.” Alas, no state governor can end the immoral war with Iran.
The outcome of the 2026 midterm matters, then, much more for what happens after Trump. In 2029, God willing, Donald Trump will peacefully leave office for the last time.5 It is the nature of public office that poll numbers tend to fall over time, and Trump’s are already very low. This will make it difficult for a Republican—even a “normal” Republican—to win the presidency. It will likely be even more difficult for the Trumpified GOP to win back the House. Senate Class III will also be up for election that year, presenting Democrats with a relatively favorable Senate map.6
In other words: all else being equal, we would expect Democrats to have a fairly good shot at winning a trifecta in 2028, with enough votes to end the filibuster, pack the courts, abolish human rights for the unborn, ban many guns, and “fortify” the democratic process in their favor in future elections. Why wouldn’t they do this? Scruple? Trump has waged utterly unscrupulous political warfare and lawfare for a decade (albeit usually incompetently), and rank-and-file Democrats have made the (likely correct) strategic calculation that they can only restore equilibrium by playing tit-for-tat. Expect no lingering small-r republican scruples in the Democratic Party come 2028 (to the nation’s great loss).
For fifteen years, the Republican Party’s best defense against a filibuster-busting Democratic trifecta has been the GOP’s current “natural” edge in the Senate. All else being equal, thanks to Republican strength among rural voters and the Senate’s bias toward rural land, Republicans should win both Senate seats in 27 states in just about every election. That’s enough to secure a durable, 54-seat majority. Their “natural” Senate majority might occasionally swell up to 57 or 60 seats… or down to 49 or 50 seats in a bad year. After all, even after a decade of some of the worst political malpractice in the nation’s history, where Republicans threw away completely winnable seats in red years for Trumpian manias over and over again, Republicans still have 53 Senate seats. Imagine if they were competent! Or even, hard as it is to imagine, likable!
However, if Democrats win the Senate outright in 2026—especially if they win it by a healthy margin—Republicans are likely to have a difficult time retaking it on a tricky map in 2028, especially if Trump’s support continues to decline. A Democratic trifecta at this impasse would be devastating for the country’s fraying unity and apocalyptic for conservatives.
The most consequential national-level outcome of the 2026 midterm, then, is what it could set up for Democrats in 2028.7
It is also possible that a sufficiently devastating rebuke to Mr. Trump and his movement might set in motion the final collapse of his political support, culminating in an impeachment hearing where even Republicans finally feel free to convict. Millions of disillusioned MAGAns could come, blinking, into the sun, where they would forswear their errors. A mass apology-letter campaign would eventually mollify Paul Ryan and persuade him to enter the 2028 race, and—oh, no, wait, sorry, that’s all from a fanfic I’ve been writing. Got mixed up there for a minute. Smart money says, no matter how shattering the electoral blow to Trump, he will blame it on Indians or Somalians or John Cornyn, his followers will be mollified no matter how obvious the lie, and the movement will continue, as hardy as ever.
It is also worth noting that, even though nobody’s really talking about it yet, the United States has a major fiscal crisis scheduled for the early 2030s, when the saved Social Security surplus (“trust fund”) runs out and the U.S. must finally face a series of impossible policy choices it has deferred for several generations. The senators elected this fall are the first class of elected officials who are likely to be in office at the onset of that crisis. So, who do you want in office: the Democrats who honestly (but very stupidly) believe we don’t need to worry about the budget, because we can solve all our problems by borrowing more, soaking the rich, and cutting the military? Or the Republicans, who correctly insist that budget discipline is urgent, but are lying?8
Oh, I’m sorry! I’m blackpilling again! I gotta cut that out. A friend of mine pointed out recently that my writing has become “bleak,” and I was forced to agree, while insisting that it’s just my writing. In real life, I’m very happy!9
So don’t worry: Jesus Christ has already written the end of history. It all works out for the good guys! God loves you and has a wonderful plan for your life!
Anyway, you already had some vague idea of the stakes. Now that we’ve established just how utterly bleak those stakes are, let’s get to the question everyone really wants answered: which of these dooms will America choose on Tuesday?
Who Will Win?
This is so much easier than it was in 2024!
The Democrats are going to win this election. There’s no question of that. To the best of my knowledge, every single person agrees that the Democrats are going to have a larger popular vote total in the national House elections. The only outstanding question is: how many more votes?
The Democrats “should” gain control of the House just for winning more votes. Republicans are barely holding on to control there anyway, having barely won it in 2022 and 2024 against a slightly gerrymandered, somewhat unlucky map. However, Republicans launched a campaign of blatant gerrymandering of their own last year, and they—narrowly, thanks to some lucky court orders—won this round. The House is gerrymandered to hell (by both parties), and Republicans have somewhat better gerrymanders. (We should ban this.) As a result, Democrats must win the House national popular vote by 2 to 4 points to actually gain enough seats to take control.
The Senate is a much harder push for the donkeys. They have to win by a lot to take the Senate. Since this is Senate Class II, which last faced election in 2020 (a slightly blue year), all the easy pickup opportunities are off the table for them; they already won those states in 2020, and now have to defend them. Not only do they need to hold their seats in purple states like Michigan, Minnesota, and New Hampshire, but they need to unseat the resilient Susan Collins in Maine, then win a reddish-purple-state seat of their own (North Carolina), then win at least two new seats in true-blue red states that backed Trump by more than 5 points, many of them against entrenched incumbents. Miss any one of these tricks, and they lose the Senate.10 This, in an era when presidential vote tracks nearly in lockstep with senate vote.
On this tough map, for Democrats to even reach parity with the GOP in the Senate (50-50, with the vice president as tiebreaker), they would likely need to win by 3 to 5 points. Winning outright would require a rare Blue Wave (D+6.5)—or, at least, some really good luck.
Fortunately for the Democrats, a Blue Wave is almost (but not quite) in the bag. In fact, the polls show that they are at around D+9, on the cusp of a Blue Tsunami. Moreover, in his very generous effort to help Democrats win as many races as possible, Mr. Donald Trump worked hard this year to eliminate two highly re-electable Republican incumbents in must-win races, Thom Tillis (R-NC) and John Cornyn (R-TX). He has gotten them replaced with nincompoops.
As a result, the Democrats are probably going to win the Senate. For me, then, the most interesting question remaining is, “By how much?” A 51-52 seat Democratic Senate majority is very different from a 55-56 seat Democratic Senate majority. (The latter makes it possible they could reach 60 seats in 2028 and overcome the filibuster without abolishing it!)
However, Democrats have done themselves a disservice: they have set their expectations very, very high. Now that they have dreamed of winning even red Nebraska, they may find themselves abjectly disappointed with a “mere” Blue Tsunami, or even with a Blue Wave that falls short of delivering the Senate. Republicans, too, are primed to mock Democrats for anything that falls short of a sweep. They’ll call it a “Blue Ripple” if the Democrats only get 51 Senate seats.
They shouldn’t. For Democrats to take the Senate at all on a map this red would require them to notch popular votes on par with their tremendous victory in 2018. The fact that we’re even talking about bigger possibilities than that shows that the Democrats have already won this election; we’re just cashing out their winnings.
Let’s set realistic expectations. Polls are never exactly right, so let’s consider some different ways the night could go. If polls aren’t exactly right, everyone always screams, “Oh, my, what a huge shock! We never expected this result! This completely changes how I think about politics!” but you, the faithful De Civ reader, will not be shocked. You will have already considered the many ways this race could possibly go, starting with:
Polls Get It Exactly Right
Okay, I just said this never happens, but it does sometimes come pretty close to happening. The last two midterms, in fact, had almost zero average polling error, so the final results turned out very close to what the data had projected. I even got a few lucky hits and managed to predict the exact number of House seats the GOP ended up with, largely going off that polling. The “Red Wave” was always strictly a pundit phenomenon; we never saw it in the data.
If polling in this election has zero average national error, we’re likely to see:
Democrats win nationally by about 9 points, a solid Blue Wave that is right on the cusp of becoming a Tsunami.
Democrats pick up a respectable 20-35 seats in the House, giving them 236-251 seats and comfortable control.
In the Senate, Democrats win 52 or 53 seats. It’s hard to say which ones they are, because Democrats have so many roads to unlikely seats right now.
They could simply win all the races where they are currently ahead in the polls (Ohio, Alaska, Maine, Texas, Iowa, Kansas).
Or they could lose Maine and Texas, but pick up Nebraska and South Carolina to make up for it.
Or Abdul El-Sayed could lose in Michigan, while Dems pull off an upset in Montana.
Or Abdul El-Sayed could just lose in Michigan, full stop, and Democrats would end up with 52 seats instead of 53.
My point is: with Republicans so unpopular and playing defense in so many places, Democrats don’t need a whole lot to go right for them to land on 52-53 seats—that is, if the polls are accurate.
Look, Nate Silver even made a visualization for this very scenario!
This general outcome (NOT this exact map) is about 40% likely to happen. It’s not actually likely to happen, but it’s more likely than either of the other possible outcomes.
If it does happen, Republicans are pretty much toast in the Senate. They could get lucky and eke out a bunch of close races, because Senate races really do involve independent local factors… but it’s pretty improbable that they could block all the Democratic paths to victory in this world.
Republicans really need a polling error in their favor to have a decent shot at holding the Senate. So let’s take a look at that possibility:
Polls Miss Red
This has happened in some pretty high-profile races in the past decade, namely each of the three presidential elections in which Donald Trump was running (2016, 2020 big-time, 2024 a little). It also happened in 2014, so it’s not a Trump-only phenomenon. It could happen again this year.
In fact, there’s a common phenomenon that is highly relevant to this particular set of races: polling error tends to underestimate blue in blue states, and underestimate red in red states. Where are the crucial battlegrounds of this election? Red states! That gives us reason to suspect that Republicans will beat their polls in those states.
What happens if they do? We’ll assume a “normal polling error” of 2-3 points that underestimates Republicans.
Democrats win the House national popular vote by about 6-7 points.
Note that this is still in Blue Wave territory! A Blue Wave is very likely this year, and that’s all there is to it.
Every seat in the gerrymandered House is a battle, but the Democrats eventually fight it out, picking up 5-15 new seats for a total of 220-230.
The Senate becomes a true tossup.
Instead of enjoying clear leads in Ohio, Alaska, Michigan, and Texas, Democrats now discover that those races were much closer than polling indicated. They’re in for a tough fight.
Worse, the polling error likely puts Nebraska and South Carolina out of reach, giving Democrats fewer backup plans if they lose one or more of these races.
Worst of all (for Democrats), the smaller polling lead means Minnesota is suddenly in play.
However, even after accounting for the error, Democratic candidates are still favored to win all four of the pivotal races I just named and Minnesota. They need to run the table in this scenario to capture the Senate, but they’re in a very good position to do that. Republicans probably need a slightly larger polling error to actually be favored in the Senate.
Sometimes, instead of a small national polling error, there’s a larger polling error, but it’s concentrated in a single region of the country—say, a 6-point error in the Midwest. If that happens, it mostly plays out the same way: the Democrats still win the House (with fewer Midwestern seats and more from elsewhere).
However, this scenario could lead to some surprising outcomes. Imagine you wake up Wednesday and Republicans have won Minnesota, Ohio, Michigan, and Iowa… but Democrats still took Maine, Texas, Kansas, Alaska, and (narrowly) Nebraska to secure 51 seats and control of the Senate. That’s a thing that could happen! It’s not very likely, but it’s well within the realm of the plausible!
A larger-than-“normal” polling error could benefit Republicans even more, but it would have to be a doozy indeed for Republicans to hold the House.
Broadly speaking, Polls Miss Red is about 30% likely. Even if it occurs, it doesn’t guarantee Republicans the Senate, but it gives them a fighting chance. I think this is what most people in the MAGA pundit class expect is coming up on Tuesday. It is a real possibility, but they’re wrong to expect it, because it’s still unlikely.
I think they are expecting this partly because they aren’t seriously engaging with the data in the first place and don’t realize how bad it’s gotten out there, but partly because they reject the data outright. This is a common symptom of partisanship (Democrats did it too, in 2010! I was there!), but it sets them up for profound disappointment when the possibilities they were shutting out suddenly come crashing in.
Let’s discuss some of those possibilities:
Polls Miss Blue
People really overlook this possibility, and they really shouldn’t. There’s no iron law that says polling errors must always underestimate Republicans. We’ve had 14 presidential+midterm elections since 1998. Polls got 7 of those races correct (within two points). Polls missed Red 5 times (2024, 2020, 2016, 2014, 2002). But they also missed Blue twice (1998 and 2012). With only 14 trials in our sample, that looks like random noise to me.
Moreover, since pollsters don’t want to live with the shame of underestimating Republicans again, they’ve been working hard to fix their problems. They might overcorrect… or, they might even intentionally err on the side of underestimating Democrats just to be sure they don’t underestimate Republicans again.
Finally, some egregiously bad polling errors occurred in the New Jersey and Virginia gubernatorial races of 2025… and those errors underestimated Democrats. By a lot. That’s our most recent data point: polls underestimated Democrats the last time we polled major races!
So what happens if polls underestimate Democrats in this race? Let’s say it’s another “normal” polling error, a 2 to 3 point miss.
The only good news for Republicans is that the result isn’t quite as bad as 2010 was for the Democrats:
Democrats win the House national popular vote by 11-12 points. This is a Blue Tsunami. It’s bigger than the Republican Revolution of 1994.
Democrats begin to break into the firewall of tough gerrymandered GOP seats. They gain 35-45 seats, landing somewhere north of 250 total seats.
The Senate devolves into a bloodbath.
Texas, Ohio, Alaska, Kansas, and Iowa all fall to the Democrats. They probably aren’t even particularly close. The single bright spot in the night might come when Susan Collins miraculously clings to her battered seat in Maine.
Nebraska and South Carolina are real tossups on this map.
Even Florida, Mississippi, and Montana become risks, although the GOP probably holds them all in the end (and would be very unlucky to lose more than one).
When the dust settles, the Democrats have 56 Senate seats. The GOP has essentially no chance to retake the Senate in 2028, and must depend on winning the presidency or the House if they hope to avoid a Democratic trifecta.
Alternatively, consider a scenario where the polls are mostly fine, but there’s a large, 6-point regional polling error underestimating the Democrats… this time in the South:
Democrats take all the non-Southern states where they currently have a polling lead: Ohio, Maine, Kansas, Alaska, and Iowa. Republicans just barely scrape by in Nebraska in their sole bright spot for the night.
In the South, Talarico wins Texas by almost 10 points, and Darline Graham loses in South Carolina. In a shocker (that isn’t shocking if you’ve followed the polling), the surprisingly close Mississippi Senate race goes to the Democrat. Only Florida offers a hope for Republicans to salvage something from the South, and it’s too close to call.
Democrats wind up with 56 or even 57 Senate seats.
There are many different maps that could result from a Polls Miss Blue night. As with Polls Miss Red, much depends on size and location. I have only attempted to provide some illustrative examples of what a fairly normal Polls Miss Blue night would look like. To wit: all of them are nightmares for the Right.
Blues Miss Blue is about as likely as Polls Miss Red: 30%.
Wow, James, That Sure Is A Lot of Words To Hedge Your Bets and Give Some Vague Probabilistic Pabulum! WHAT’S GOING TO HAPPEN?
My true prediction will always be the probabilistic pabulum. I am always startled by the breathtaking confidence of people who just flatly say, “Democrats are gonna win the House, but they only get 50 seats in the Senate.” How can you predict a specific seat count?!
These people are, obviously, wrong about 90% of the time, which reassures me that they can’t, in fact, predict a specific seat count. However—by the law of averages—they are also right about 10% of the time, which makes the handful of correct guessers insufferable at the next election.
Overall, I think the Democrats have a three-in-four chance of winning the Senate. If Polls Are Right or Polls Miss Blue, that chance becomes 90% or better.
The House is not an absolutely sure thing, but it’s as close to a sure thing as there can be in this era of highly polarized elections. Republicans would need extraordinary luck and a historic polling error underestimating them to hold the House.
James, Seriously, Tell Me What’s Going to Happen. Make No Mistakes. I Have a Gun
Gun to my head, forced to guess at exact House and Senate seat totals… I’m very probably going to die.
However, I would still make my best guess and cross my fingers.
I have noticed something interesting in the past twelve years: polls and conventional wisdom tend to diverge a little bit. For example, in 2012, polls showed Barack Obama winning comfortably, but conventional wisdom held that Romney was closing the gap in the final days and it would be a close race. In 2022, polls showed Republicans narrowly winning the House, with a photo finish in the Senate. However, prediction markets (which are a good proxy for conventional wisdom11) suggested a Red Wave was on the horizon. Republicans thought they had a chance in Oregon!
When there’s a gap between polls and conventional wisdom, any polling error that results is usually in the opposite direction. Romney not only lost worse than he expected; he lost worse than the polls expected. The Red Wave didn’t just fail to materialize; the Red Ripple was very slightly smaller than polls indicated. This happens again and again. I don’t think the pattern is strong enough to count as a rule, but call it my personal rule of thumb: when the vibe is pointing away from the polls in one direction, bet in the opposite direction.
The 2022 experience seems to have shaped the reaction to the 2026 polling. The vibe right now has been, and remains, simple disbelief, on both sides of the aisle, that the Republicans have screwed the pooch so badly that they could end up with 47 Senate seats, even though that’s the modal outcome in Nate Silver’s model. Most people, most prediction markets, and even most pollsters seem to expect some of the polls to be wrong, and for Republicans to end up with 48, 49, or even 50 Senate seats, because they just can’t quite believe the Blue Wave in the polling is actually going to materialize. Kalshi thinks Republicans have a 10% chance of holding the House, and a one-in-three chance of keeping the Senate! Kalshi is wrong!12
What finally convinced me to apply this rule in 2026 was two polls: the very strong NYT/Siena poll showing D+11 on the generic ballot, and a recent Marist poll that managed to show a close race in Maine… by applying a likely voter screen that somehow changed D+9 to D+2 (the opposite of what you’d expect). The consensus is that Maine is close, and it sure looks like Marist has gotten a really, really blue poll result, then done some weird math stuff to try to get their result to fit into that consensus. If they’re doing it, others are doing it, but being sneaky about it. That makes me think that, maybe, the polls could, in truth, be a little more Democratic-leaning than we’ve seen in the polling averages.13
Gun to my head, then, my prediction is that the Democrats do slightly better than the polling suggests. They win 247 House seats and 55 Senate seats.
I will not enjoy this, because the Democrats are evil. I will not vote for any of them, and I strongly encourage you not to collaborate with them, either.
On the other hand, I will enjoy the schadenfreude. While there are a small number of Republicans far enough from the evils of the Trump Administration that I can bear the stench of voting for them, the party is a wretched hive. If anyone ever deserved to lose to the Democrats, it is Donald Trump’s supine Republican Party, face down in its (gilded) pillow as it tries desperately to think of England. Bimbofied as it has been, the GOP may have forgotten everything it once believed about the debt, the unborn, the markets, or the U.S. Constitution—but I haven’t.
Do You Want to Join in the “Fun”? How to Watch Intelligently
The day of the Brexit vote, the very first constituency to report results, Newcastle-upon-Tyne, reported a defeat for Brexit, 51% Leave to 49% Remain. The defeat for Brexit there was expected. However, the margin was wrong: experts had expected Brexit to fail in Newcastle by 12 points, not 2 points. Newcastle had been a “safe state” for Brexit opponents that turned into a very narrow win. I was over at my parents’ for dinner, just buzzing by the computer for a quick results check, didn’t even bother sitting down—and I remember seeing that result, sitting down heavily, and thinking, “Welp. The U.K. just left the E.U.”
The rest of the night went the same. Areas that were expected to vote for Brexit by a narrow margin voted for it by a large margin; areas that were expected to oppose it by narrow margins ended up supporting it by narrow margins. There were a few places where the anti-Brexit “Bremain” vote did better than expected, as there always are… but not many. Networks finally called the race for Brexit hours after the result was clear.
This happened again in the 2016 presidential election. I told my family at around 8:15 PM (Central Time) that Trump was probably going to be the President-Elect, because I could see that all the states Trump needed were getting called for him (or were on the cusp of it), while all the states Clinton needed were tossups… and leaning red.
It happened again in 2020, when De Civ readers were tracking the results of a random House race in suburban Kentucky and realized, whoa boy, the polls missed again this year, bigly, and we’d better buckle up for a long, long night. As it turned out, it was a long day, too; it wasn’t clear to me until Wednesday night that Biden had won the election.14
In 2024, the polls performed reasonably well. The race was the purest of tossups, so even a tiny polling error in either direction would likely determine the outcome. They had a normal polling error (red flavor), and the exact map I’d posted that morning for that scenario (which I’d considered the second-most likely map) ended up being the winner by the end of the night. It was clear to De Civ readers early.
Basically, if you’re paying attention, it’s usually pretty obvious pretty quickly if Polls Missed Red or Polls Missed Blue. Once you have that crucial piece of missing information, the rest of the election is very easy to predict.
I used to explain how to quickly recognize a polling miss in great detail, so you could follow the race at home. I’d show you benchmarks and give you county-by-county tools to see the results as they come in and compare them to expectations.
However, we live in a Bold New Future, and this whole process is pretty much automated now. The New York Times invented the Needle in 2016, and people have had nightmares about it ever since. It’s very good at collecting precinct by precinct data, figuring out in real time where the polling errors were, and using that data to figure out who’s gonna win. I don’t expect it to take long zeroing in on this election!
That said, the Needle is imperfect. It crashed in 2018 due to technical problems and never recovered. In 2020, covid complications made the Needle too challenging to run at scale. In 2024, it was impaired by a strike at the Times, although, as I recall, it worked fine anyway. We can’t necessarily count on the Needle in 2026. Plus, not all of you have Times subscriptions.
(I’ll try to update this article with a gift link to The Needle once the Times has posted the 2026 version, so watch this space!)
There’s also a bunch of Twitter election nerds who are just as excited about election results as I am, but who also have a professional interest in them, and therefore have plenty of time to draw up benchmarks of their own, so they can spend Election Night (the Super Bowl for polling nerds) merrily doing data entry and tweeting out their results. God bless them. You can follow along with my current roster here:
Twitter List: electiondata (they lean pretty left as a whole, because “polling nerd” is not very respectable on the right, but they care about data most of all)
My current problem is that, with all these Internet applications and tweets, I don’t have a good TV program lined up. I never watch any news television and have no idea who’s going to have good election night coverage—and, since I don’t have cable, I’m basically limited to channels that will stream on YouTube. Nevertheless, I want my 12-year-old to watch the returns with me, so I need to find a good, entertaining, nerdy election night show to watch. I welcome your suggestions, readers, by comment or by email!
Will Trump Accept the Result?
The unique evil of Donald Trump is that he forces us to ask this of him.
If the election turns out the way I expect, Trump will certainly never accept the result, in the sense of admitting in public that his side lost. Moreover, if he sees an opportunity to prevent the election from taking full effect, he’ll take it.
The difficulty for Trump is that the available attack surface here is fairly low. I’ve been thinking about doing a full-length article on this, but the trouble for Trump is that there simply isn’t any single national ceremony for him to target the way he targeted January 6—and, even there, his attack was stupid and doomed to fail from the get-go.
Elections are conducted by states. States count the ballots, states certify results, state courts adjudicate (most) challenges. Trump can and will order Republican officials in states to fight to the limit of the law and slightly beyond it, as he did in Georgia in 2020, but here he has several more difficulties:
Some of the key contested states are blue states, where Republican officials just don’t have much power. If Peggy Flanagan beats Tafoya in the Minnesota U.S. Senate race,15 what’s Speaker of the Minnesota House Lisa Demuth going to do about it? She’s our only thing resembling a statewide official.
Many Republican elected officials are willing to bend the law, but few are willing to directly violate a court order for Donald Trump. Trump is routinely disloyal to people who break the law on his behalf, so, even if the Republicans lack integrity, the risk is too great.
The courts are still quite good. They did a great job adjudicating the 2020 election. This includes Trump’s own judges. The federal government will have only very narrow windows of opportunity to disrupt election certifications without a court stepping in.
The Democrats are likely to win in lots of places, and Trump will have to fight and win on many fronts in order to deliver himself a victory.
It wasn’t Trump’s name on the ballot, so what does he really care? He can blame the losers and maybe Mitch McConnell and call it a day. That’s what he did in 2018, more or less.
Mr. Trump could try to keep Congress from accepting certain new Democratic members, but here he has a real big problem: when the Senate first convenes in 2027, before it swears in new and re-elected members, it will be composed of only Class I + Class III senators, because the Class II members have to be sworn in. There will therefore be only 67 Senators at the time Trump wants to act… and 34 of them are Democrats, which is an absolute majority. I need to look into the procedural rules around swearing in senators a little more closely, but I don’t think there’s anything Trump can do there against a temporary (but functional) Democratic majority.
Overall, my assessment is:
Likelihood Mr. Trump verbally rejects the results of the election: very high
Likelihood Mr. Trump actively hinders the certification of the true election results: low-to-medium, depending on the opportunities available to him when the dust settles
Likelihood Mr. Trump successfully disrupts an election certification, causing genuine doubt and confusion that courts have a difficult time resolving: low
Likelihood Mr. Trump successfully changes any election results: extremely low, approaching zero
This is subject to revision once I’ve done more reading, but, with the election upon us, I’m not going to get that reading done until whatever’s going to happen actually happens.
By the time I post this, it may actually be Tuesday.
Mr. Donald Trump, who is technically not the president, will nevertheless hold the office until January 2029.
The presidency has become the most important legislative institution in the country, in addition to its vast executive powers. The president has not assumed the full powers of the legislature, and he still has many limitations (to the endless fury of the base of whichever party happens to be in power), but he has figured out workarounds to allow him to exercise the legislature’s most important powers. Mr. Trump has been a lead innovator in this department, but he builds on a legacy that stretches back to President Obama.
More importantly, Congress gave the presidency truly monumental quasi-legislative powers over the course of the twentieth century, believing that they could hold him in check with the legislative veto. However, the legislative veto was struck down in the early 1980s, and the presidency has been exercising those powers more and more boldly ever since. The vastly overpowered presidential veto ensures that Congress can’t claw back powers it’s granted the president, even if it did so by accident.
In short: in present-day American politics, the presidency is the locus of action and the font of change. In years when the presidency isn’t up, not much is going to change.
About par for the course. Clinton was acquitted on the obstruction-of-justice count by an evenly divided Senate.
On the perjury charge, he was acquitted 45 yea-55 nay, which is an even better result for Bill! That’s insane when you consider how flagrant the perjury was. It should actually be embarrassing to Democrats that Clinton was the only impeached president ever acquitted by a majority of the voting members.
One of my outré theological theories is that Donald Trump is God’s punishment to Democrats for the acquittal of Bill Clinton.
The only Democratic senator on record still supporting the filibuster is Sen. Fetterman (D-PA), so Democrats would need 52 seats for sufficient control to break the filibuster. (This assumes no stealth filibuster supporters.)
I am not saying I’m opposed to Mr. Trump leaving office before 2029, due to voluntary resignation, conviction on impeachment, or even by (natural) death or incapacity. However, I’m planning on 2029.
Senate Class III was last up in 2022, a red year, presenting Democrats with a few soft targets. However, Republicans bungled the 2022 Senate elections so badly, by nominating such Trumpish idiots, that there really aren’t as many vulnerable Republicans as their might otherwise have been.
A 2028 Republican trifecta is much less interesting / terrifying, because Republicans are much less likely to blow up the filibuster and unleash their unhinged will on the nation. We know this, because we just lived through the last two years.
Mr. Trump hates the filibuster, but Republicans—for once, resisting Trump on an important issue—refused to end it. They did this in the face of widespread misinformation—and, frankly, idiocy—within the GOP about how the GOP could end the filibuster to pass the SAVE Act without really ending the filibuster. (That’s the kind of cockamamie conspiracy theory Harry Reid talked himself into in 2013 when he nuked the judicial filibuster, kids. Harry Reid’s arrogance is how we got Justices Gorsuch, Kavanaugh, and Barrett. Don’t be like Harry Reid!)
The anti-filibuster pressure became so intense that a perfectly cromulent, highly electable, reasonably popular Republican senator, John Cornyn, lost his primary to a widely despised troll named Ken Paxton. Now Republicans are likely to lose Cornyn’s seat to Democrats.
Yet the filibuster held.
Perhaps something similar would happen on the Democratic side, and the filibuster would be saved, even with a Democratic trifecta in power. I hope so. Not because I love the filibuster—I very much do not—but because I am mortally afraid of the Democratic legislative agenda.
If this seems familiar, it’s because it’s a paraphrase from the one article I read in The Dispatch this summer.
And, praise God, this is the last election ever where Donald Trump will be president or scheming to become president! Next election, I will have hope!
(I put this in a footnote because it felt like tempting fate to put it in body text.)
…perhaps not much hope, to be sure, but surely, this, this must be the nadir. I was sentenced to twelve years of almost unrelenting political misery, and I’ve survived to year ten. The light is at the end of the tunnel. I’ll emerge into the blasted wreckage of a political landscape wrecked by a Zack Snyder-level apocalypse, as was foretold, with years of rebuilding to do, but emerge I shall.
That was the road I laid out in late 2024, when I was one of the few people saying it was even possible for Democrats to take the Senate.
This may be changing, as prediction markets are increasingly taken over by AI bettors, but there was a lot of dumb money in prediction markets 2016-2025.
DISCLOSURE: having been reassured that it is still legal for Minnesotans to play Kalshi despite our recent law against Kalshi operating within our state, I bet 64 cents that the Democrats would sweep Congress.
I also put 88 cents on Cindy Hyde-Smith (R-MS) to win her race as a hedge, and 44 cents on Samuel Alito to fail to announce his retirement by July 1, 2027. (If he drops dead—very probably the only way he’s leaving the court if the Dems take the Senate—I get my money back.)
I hope fans of my old Playing PredictIt feature enjoyed this footnote! We’re back! In footnote form!
However, there’s a case to be made in the opposite direction: if Marist, a trustworthy pollster, applied its likely voter screen and got this very strange result, maybe there’s something deeply funky with the underlying data.
The Washington Post went so far as to publish a (very blue) poll the other day with a disclaimer that said, basically, “We think this poll may have serious problems with Republicans refusing to answer the phone. We weighted it accordingly, but the numbers still look weird. We’re publishing it out of transparency, but we aren’t sure it’s true.” If this is widely true, then it could mean that the polls are just wrong this year, and that partisan non-response bias (the technical term for “Democrats are much more eager to answer polls than Republicans”) has finally infected a midterm election.
On the other hand, the Washington Post’s doubts of its own honestly-collected polling data, simply because the data was bluer than they expected, might also be interpreted as evidence that the vibes are so strong they’re messing up the objective measurements of pollsters. That would send us back to the first hypothesis, and it strikes me as slightly more likely in this case.
Biden really did win that election. Republicans’ refusal to accept that, yes, 81 million people really did vote for Joe Biden—because the alternative was more Donald Trump—has been the pre-eminent albatross around Republicans’ necks ever since, and continues to deal damage even here in 2026.
Note: Tafoya is not one of the Republicans I can bear voting for. If we have to wholesale sacrifice the cause of unborn rights on the altar of Trump in order to win a Senate seat, the seat’s not worth winning. If she wins, which she could, she’ll do it without me.
Unfortunately, the MAGA crowd will not learn anything, even from catastrophic defeat. They're already spinning a Paxton loss as the result of traitorous RINOS/Never Trumpers who are upset that a True Conservative beat a swamp creature in the primary, rather than the possibility that the squishy middle just doesn't like Paxton's vibe in a very bad year for Republicans. (And I say this as someone who thinks that Talarico is a smarmy heretic. Dude thinks the gospel of Thomas is authoritative, and misreads it to boot.)