📊 Where unlikely voters differ the most from likely voters

Marquette Law School poll’s Charles Franklin has published a great summary table breaking down views on various topics between the likely, registered, and unregistered electorates.

You may have heard me harp on this before: polls can look really different depending on whether you’re interviewing all adults or just likely voters. Usually, a poll of all adults will show Trump approval in the 30s while a poll of registered or likely voters is more likely than not to show it in the 40s.

What’s really interesting is not how the overall number shifts based on the inclusion of different respondents, but what these marginal groups themselves believe. So I fed this table to Claude and it had it run an imputation to find what the unlikely voter in Marquette polling thinks, be that registered but not likely voters or unregistered adults. These numbers are just estimates based on the rounded table, and may be noisy, especially for the registered but not likely universe (11% of the public). To wit, here’s the breakdown.

The less likely the voter, generally speaking, the more negative things are for Republicans, in an inversion of the pattern we saw in 2024. 44 percent of likely voters in this poll approve of the job Trump is doing; that number is in the 20s for unlikely or non-voters. We can further generalize that unlikely voters really haven’t liked the incumbents the last couple of go-arounds.

Probably the biggest divergence is on the right direction / wrong track question. Unlikely voters are less than half as likely as likely voters to say things are going in the right direction, and the wrong track number is an estimated 83 percent.

But differences are actually more muted on evaluations of the parties, which don’t differ at all for Democrats in Congress and are more modest for their Republican counterparts. Unlikely voters tend to generally dislike both parties more, but with a sharper gradient for incumbent Republicans.

These differences aren’t as sharp as those for a nonpartisan measure like the right direction / wrong track. Wrong track sentiment doesn’t automatically translate to partisan anti-Republican sentiment — but enough of it does to make a difference in the midterms. Trump approval sits somewhere in the middle — half partisan sentiment, half proxy for wrong track sentiment. Unlikely voters, almost by definition, are much less partisan than likely voters. That makes donning a red or blue jersey a higher hurdle for them than thinking things are off on the wrong track and blaming whoever is in power for it.

Unlikely voters are also quite different in their issue prioritization: less likely to say the war in Iran is their top issue, and more likely to say the economy.

Public opinion has a defined shape and structure. Whatever the direction of public opinion is at any given point, it tends to be more exaggerated among less engaged voters. And those of them who do enter the midterm electorate are disproportionately likely to be independent or swing voters.

📊 New Pew typology just dropped

Pew Research Center’s updated Political Typology quiz categorizes Americans into nine political groups based on their views and values. Here’s their write-up of the results.

Echelon’s 2026 Political Tribes will be coming out in the next few weeks, but Pew’s work aligns with ours in many ways. They find that short of half of each political coalition belongs to an extremely ideological tribe and 15 percent of each coalition is the “wrong” party based on their views on issues — showing an underlying complexity we don’t normally see in most public opinion research. They further divide the most ideological tribes (Loyal Liberals and Leftward Progressives for Democrats, and the No Apologies Right and Family First Conservatives for Republicans) but find a similar shape to less attached partisans. Their Pragmatic and Polite Right maps to our Moderate Right and their Order and Opportunity Left maps to our American Institutionalists.

🗳️ How each state is swinging in 2026

The Argument models how each state would vote if an election were held today. The analysis points to a substantially more Democratic political environment than 2024, with several Republican-leaning states becoming highly competitive.

🗺 Roll your own redistricting

Christopher Kenny has built an Algorithmic Redistricting Atlas that maps how redistricting impacts past and present cycles, letting you tweak the national environment and your preferred redistricting strategy.

🤖 Another leap forward for AI

Claude Fable 5 is out. This is a version of the Mythos model with additional safeguards for cyber and biosecurity. I expect the number of vibe-coded projects I write up in this newsletter will continue to grow at an even faster clip.

AI is replacing human coding, but is also generating lots more code that humans (at least as of now) will need to review. Internally at Anthropic, Mythos is generating 8 times more code than the pre-Claude Code baseline. So, don’t count on there being a jobs apocalypse just yet.

Here are just a few of things people have been cooking up with the new model. Ethan Mollick used it to build an interactive isochronic travel map showing travel times from New York to destinations worldwide, all from a single high-level prompt.

And Nicolas Bustamante asked it to create an interactive map of San Francisco using transit, terrain, and weather data.

😂 Humor

🤠 Do Democrats have a shot in Texas?

The Texas Senate race is set after Donald Trump ousted his second incumbent Republican Senator this month. Nate Cohn makes the argument that a blue Texas blue Texas isn’t just a dream for Democrats anymore — using current national polling, you can get to a place where it’s a better pickup opportunity for Democrats than Ohio or Alaska based on demographics alone:

Or consider another example: Imagine we adjusted the results of the last presidential election by shifting each racial demographic group’s support to match the latest Times/Siena poll results. Based on the poll’s findings, Texas would be expected to tilt Democratic in this national political environment. Interestingly, the other red states Democrats are targeting in their pursuit of Senate control — like Ohio and Alaska — would still remain red in this exercise, even though each voted more Democratic in 2024 than Texas. That’s because Democratic gains among nonwhite voters help Democrats more in Texas, since it is a much more diverse state.

Count me skeptical. Even conceding some Hispanic reversion, you’d not only need to roll back Trump’s gains in 2024, but his and the GOP’s gains in 2020/22 too. Remember that Hispanics swung something like 30-40 points right between both elections combined. And a favorable environment for Democrats gets you back to something like the 2020 pre-election polling, where Biden led by 8 nationally (similar to Dem generic leads today) but Trump was up by just 2 statewide, leading many to take seriously that it could be a part of a 400+ EV plus anti-Trump landslide. The trouble is the partisan baseline then was much softer than it is now — Trump +14 now, but Trump +8 in 2016, Cruz +2-3 in 2018, and ultimately Trump +6 in 2020.

The difference between now and then is not just Hispanic gains — but that the Texas suburbs have revealed themselves to be anything but the “hydrogen bomb of D votes” proclaimed by old-school Election Twitter. Texas has been just as big a magnet for right-leaning migrants from the rest of the country as Florida, bolstering Trump’s position in 2024. While a newly minted Texas exurb might see the R+40 or R+60 margins of what it replaced, R+20 off a larger population base tends to be the norm. And a large share of these exurban voters are largely untouched by the aesthetic considerations that held sway in the old-school Republican inner suburbs in 2016 and 2018.

That said: Ken Paxton is somewhere between Herschel Walker and Roy Moore in candidate quality—and a lot hinges on where he lands on that spectrum. But the campaign to polarize the conservative electorate against James Talarico—with comments heavy on veganism and gender ideology—has only begun. And in a state like Texas, polarization is usually enough.

Looking back on the primary this week, John Cornyn’s 36% in the Texas Republican Senate primary runoff is the lowest two-person share for an incumbent senator in a primary since 1974.

Need an expert for your conference or fly-in to break down the midterms or the latest trends in politics, demographics, and AI? Visit my public speaking page or get in touch directly.

Book Me as a Speaker

🫏 All about that base

Even without independents, your part has an enormous advantage if you’re winning your own party’s voters by Assad margins. And that’s true for Democrats this cycle. Democrats are in array with 96% support among their own voters, to Republicans at 91%. That works out to about a 10 point net advantage for Democrats relative to Republicans. Applied to around 40% of the electorate that might identify with or lean towards the Democrats, that’s good for a 4 point shift in the generic ballot numbers.

Historically, that 91% figure for Republicans isn’t bad—but 96% is astronomical. At this point in the 2018 cycle, both Democrats and Republicans were much less loyal to their parties than they are now, showing how much more it’s become about the base.

And it’s also all about adding safe seats in redistricting. To avoid a midterm wipeout, Republicans have added up to 14 new safe seats, defined as Trump +10 or more, for a total of 204 — but that’s pending redistricting being finalized in Alabama and Louisiana (it’s not moving forward in South Carolina, where Jim Clyburn survives). I talked about the analysis from last week’s newsletter with CNN this week, stating that the biggest consequence of redistricting is not whether Democrats can win the House (they can) but the size of the majority and the number of seats Republicans would have to win back in 2028 (cut by as much as two thirds).

🧑‍🧑‍🧒 18-year-olds are independent — just not always from their parents

VoteHub analyzes voter file data of 18-year olds, finding only 20% of registered voters born in 2008 registered with a major party. Partisanship among the rest varies by state, but it seems to be highly influenced by registration among others in the household. For instance, in North Carolina, very few of these young voters buck their parents of the same party, but there’s a Republican skew in mixed households.

📺 Let’s hear less about PVI and more about DMAs

It’s not just about whether the partisanship of a state makes it competitive — but whether buying ad time is cheap enough to allow you to carpet bomb the state or district with relatively few resources. Crystal Ball reviews which races that a prospective donor can get the biggest bang for their buck, with Alaska taking the top spot.

🌎 A decade since Brexit, the populist right is stronger than ever

In June, it’ll have been a decade since the Brexit vote. Pew Research details the rise of the populist right in Europe since.

🤖 Negative views of AI come mostly from those who don’t use it.

Those who use AI in the workplace are more likely to say it saves time, improves quality of work, and improves their career.

🦊 Where to find FiveThirtyEight data now that it’s been pulled down

Earlier this month, ABC quietly pulled down the archive for FiveThirtyEight. In response, independent archivists have been racing to preserve the site’s work — including a newly expanded database cataloging roughly 13,000 charts and 3,000 illustrations scraped from FiveThirtyEight’s HTML and saved individually for public access. The archive includes a couple of my articles on whether you can buy a presidential election and Black voter turnout.

🎰 Prediction markets are officially big business

Trading volumes on prediction markets have exploded since the middle of 2025.

🗳️ Here’s just how much gerrymandering has wiped out competitive seats

I keep making the point that gerrymandering isn’t just about winning seats for your party. It’s also about taking competitive seats off the board. Now with South Carolina wrapping up its redraw, we’ll soon be able to close the books on the 2026 redistricting wars and know where the dust has settled.

Redistricting alone won’t be enough to protect the Republican House majority, but a new interactive visual from The Intersection shows exactly how valuable the redistricting campaign has been in narrowing the playing field. The tipping point seat in the House (based on 2024 presidential vote) moved from Trump +3.1 AZ-1 to Trump +5.1 PA-10. If the election were being held today, Republicans would be favored to lose either one. But Democrats doing 4 points better nationally than in the tipping point seat would yield them a meager five extra seats. Under this scenario, where a Democratic wave crests in…

Read more

🏭 The new politics of American business

The team at Echelon released a report this week on the voter coalitions supporting and opposing 13 key industries. You can download the full report here.

Applying our political tribes framework to views of American business, we found 7 “tribes” with distinctive belief systems that guide their support or opposition to various industries. We think these are pretty revealing about the dueling impulses within each of the major coalitions, and also about the mindset of conflicted swing groups. Here are what those tribes look like:

Read more

It’s the 400th edition of The Intersection! Here’s what was covered in past centennial editions:

• No. 100 — December 13, 2019: Blue Wave in the UK — Boris Johnson’s landslide victory

• No. 200 — December 17, 2021: Americans prefer suburbs, Democrats’ second post-2020 Latino postmortem

• No. 300 — January 19, 2024: Trump’s Iowa victory

Thank you for your support as we chronicle the evolution of data-driven politics. Here’s to the next 100.

🏛️ VRA Section 2 is effectively a dead letter. Are Dems out of good redistricting moves?

The Supreme Court’s long-awaited decision in Callais v. Louisiana has come down. As expected, it renders the Voting Rights Act’s Section 2 a dead letter — but stops short of eliminating it entirely. But the standard of proof is so high that it will do nothing to stop a bonanza of Southern redraws, creating maps that could look something like this:

And Democrats could well counter with the following in states that they control, entirely eliminating Repub…

Read more

🇺🇲 Trump’s strongest soldiers lag run-of-the-mill Democrats in midterm motivation

We’ve been tracking motivation in the midterms for quite some time, and this month Kristen broke the numbers down by MAGA vs. non-MAGA Republicans.

MAGA Republicans are more motivated to vote, but even they lag all Democrats and leaners by 9 points in being extremely motivated to vote in the midterms.

We asked a similar question to more directly isolate the Trump factor in the midterms — getting an even wider spread. This question specifically focused on their party winning control of Congress to either stop or help Trump’s agenda. All Democrats were 16 points more extremely motivated than MAGA Republicans. Non-MAGA Republicans were 35 points less extremely motivated than Democrats.

Read more

I want to focus less on the margin than the turnout, which was not only well to the right of 2025, but also 2024 and maybe even 2021—when Republicans won.

Republicans obviously didn’t succeed in defeating the referendum, but it showed that the fight was winnable, and more money invested early might have made a difference.

Absent a victory, Virginia provides Republicans with a rare optimistic counterpoint to the building narrative of a Blue Wave. If the fight is framed the right way, Republican voters will turn out the same or better relative to Democrats than 2024, even with an unpopular president in the White House. And if they recreate this performance even to some degree, they are virtually assured control of the U.S. S…

Read more

🤖 The false promise of silicon samples

Following polling analysis cycle after cycle is like being stuck in the movie Groundhog Day. If there’s a polling miss, people will swear off horserace polling, but the breathless reporting of head-to-head polls always returns. Pollsters who publish outlier results or don’t weight on education are mocked, yet those pollsters persistently reappear.

This also applies to more novel approaches. In 2024, there was some buzz around a company called Aaru that promised synthetic AI respondents could be both cheaper and more predictive than real polls. The basic premise: instead of polling a real person, poll an LLM fed a diet of the latest news. Their results ended up tracking closely with the polling averages — at the time, a tell that this new AI form of “polling” was basically a copy-paste of normal polling rather than providing any sort of value-add. It ended up being just as off as the polling averages and forecasts which show…

Read more

😡 America’s new age of discontent

Americans think the economy just plain sucks.

In 2024, you might have thought that electing Donald Trump would reset the public’s view of the economy back to what it was during Trump’s first term. After all, that’s people back then said they voted for. As I’ve noted before, there was reason to think this might work: economic perception rose sharply in the transition from Obama to Trump in 2016 and stayed pretty high throughout his first term—withstanding the shocks of the Covid pandemic.

That hasn’t happened this time.

A Reuters chart of economic approval over the last three presidencies drives the point home. Economic approval did briefly recover on Trump taking office in 2025 and then it quickly reset back to Biden-era levels. In the midst of the Iran war, it’s nosedived to a new low of 29% in their polling.

Even if inflation worries eventually subside, it seems likely that AI job loss anxiety will be next as the main narrative d…

Read more

📊 Democrats no longer outnumber Republicans in the electorate

YouGov’s deep dive into the Cooperative Election Study — which has interviewed 60,000 Americans in every even-numbered year since 2006 — shows a consistent decline in the share of the electorate identifying as Democratic, now down to a near-even break with Republicans.

This is a trend I wrote about right after the 2024 election, except that in exit polls, Republicans actually outnumber Democrats and have since 2020. But it’s good to have a longitudinal dataset with solid subgroup breaks on party ID to do this analysis on. These trends by group confirm that the trend has been consistent across generations, with all groups getting more Republican as they get older.

Read more

⛽ Americans don’t hate the Iran strikes. They hate what comes next at the pump.

Military action against Iran isn’t especially unpopular: in our latest omnibus, approval of U.S. military operations against Iran runs at a net -6 (43% support, 49% oppose) — better than Trump’s overall job approval (net -16) and far better than his foreign policy numbers (net -21). We also found that actual support for the strikes outpaces support for a theoretical Iran strike from January, essentially the same thing we saw before and after the Venezuela operation.

Despite the MAGA base actually being united on Iran, and the not-terrible numbers on the strike itself, Trump’s approval is at a new low. And the reason why seems pretty clear: rising gas prices from the shutdown of the Strait of Hormuz striking an affordability raw nerve. While Venezuela was a long-term net positive for U.S. energy supply, Iran is a short term net negative. Americans can be relatively deferential to a Pre…

Read more

🫏 The Democratic Party’s moderates are hiding in plain sight

A new Manhattan Institute survey of nearly 2,600 Democratic voters finds a coalition that’s far more moderate than its loudest voices suggest. The party breaks into three blocs: Moderates (47%), Progressive Liberals (37%), and a “Woke Fringe” (11%) that takes consistently maximalist positions. Just 22% want the party to move further left while 38% want it to move toward the center. On issue after issue — immigration, trans policy, DEI, crime — the median Democrat looks nothing like the activist left that dominates the discourse.

🗳️ The 26 seats that will decide the House

DecisionDeskHQ has mapped out the 26 Republican-held districts most likely to flip in November. Five are seats Harris carried in 2024, but the main battlegrounds will be the 11 seats that Trump carried by 5 points or less. (And yes, this is all still subject to further redistricting in states like Virginia and Florida.)

Read more

⭐ Lone Star primary takeaways

• White voters are in the drivers’ seat of Democratic primaries. Much was made of a Latino vote surge in the Democratic primary, but in fact, turnout surged more in white counties, especially in white, college-educated suburbs. That was the key to James Talarico’s victory. The education divide is increasing white liberal control of Democratic primary, jeopardizing the historic veto power Black voters have had in many primaries. Black voters came out strong for Jasmine Crockett, but she was unable to break through much beyond that to build a multiracial coalition that also included Hispanic and Asian voters, both groups that backed Talarico.

Read more

After 88 years of tracking how Americans view their president, Gallup has quietly exited the presidential approval polling business. Writing for DDHQ, Geoffrey Skelley breaks down what this means — and why it’s not as surprising as you might think.

The immediate numerical impact on polling averages will be modest, since approval is one of the most-polled questions in American politics. But Gallup’s departure isn’t without consequence. The firm had been a consistent outlier on the low end of Trump’s approval numbers during his second term, so its absence could nudge aggregate averages slightly upward. More broadly, it ends a nine-decade continuous time series that has served as a foundational benchmark for understanding public attitudes toward the presidency.

Gallup’s pullback from political polling has been gradual. After a poor performance in the 2012 presidential race, it stopped surveying elections entirely ahead of 2016. It scaled back f…

Read more

The phrase “silent majority” that derived from the speech is seen as marking a jumping off point for cultural polarization. But whatever version of polarization Nixon was cooking with was extremely popular, boosting his net approval to +47 points.

As a result of examples like this and those in the decades that followed, political reporters have been trained on the idea that a President’s job approval rating is highly responsive to the events of the day. Any fluctuation in these ratings however small must be due to whatever is in the news that week and definitely not random statistical noise.

I raise the Nixon example as evidence of the fact that presidential job approval is much less variable than it once was. There have been more recent examples of presidential approval jumping into the 90s overnight, but these involved the U.S. entering major wars. Nixon was able to engineer a 20 point bump just by giving a speech. The idea of Donald Trump — or Joe Biden or Barack Obama or Bill Clinton — doing this is laughable on its face.

Not only do modern approval ratings not fluctuate nearly as much, but their overall levels throughout a president’s term resemble each other more and more. Nate Silver’s postage stamp plot of approval of various presidents captures this pretty well. Donald Trump is now almost exactly at the approval rating recorded at this time during his first term — and Joe Biden’s only term. The approvals recorded by the last three individuals to hold the office are much more stable than anything that came before.

From this point forward in Trump’s first term, his approval was effectively a flat line, moving up by a few points during the initial Covid emergency, declining during the chaotic months that followed, stabilizing before the 2020 election, and declining after January 6th. The major crises of 2020 and 2021 only merited shifts of a few points in presidential approval.

One other feature of modern approval ratings is that they largely follow the same trendline over a four year period. There’s an initial honeymoon period that fully wears off sometime late in the first year or early in the second year — and the trend is mostly flat from there on out. Only during presidential re-election years do you start to see a slight uptick in ratings, as more voters begin to align their approval with their vote intent. But otherwise, the trend is flat throughout the entirety of a presidential term. And across the Trump and Biden presidencies, approval seems to be anchored in the low 40s.

With stability like this, we need to stop looking for event-driven explanations for presidential job approval and start looking at the forces that shape these numbers across presidencies.

Read more

🤖 If AI is like the Internet, it won’t cost jobs

My early career built on being the first in Republican politics to ride the wave of digital transformation. And “first” is hardly an exaggeration. In the late 1990s, I built political websites and a proto-blog completely for fun and love of the game. This eventually got noticed and led to early jobs at the RNC and the George W. Bush campaign in 2004, where I started out as one of just two staffers running email and the website. Back then, the number of people in the GOP with this mix of skills could be counted by finger on one hand. I didn’t feel any real competition in the field until the late aughts — more than a decade after I coded my first website in 1995. A few years later, employment in the digital field skyrocketed — with hundreds of staffers on presidential campaigns doing the work of a handful of us in 2004.

The current moment with AI feels like the takeoff moment the Internet had in the late 1990s an…

Read more

↔️ Who do voters see as extreme—or in the middle?

The perceived ideology of politicians is always a fascinating question, with Donald Trump seen as more moderate than his Democratic rivals in each of his winning presidential campaigns. But recent YouGov data has Trump viewed as the most polarizing figure in America, with 50 percent calling him far-right, more than the 42 percent who call Bernie Sanders far-left.

Some interesting characteristics of this data beyond just the individual data points: First, party leaders pay an extremism penalty. They tend to be viewed as more extreme than if they were a backbencher with the same political positions. And second, the center is an exceptionally unstable place. Only a handful of politicians are genuinely perceived as centrists. And that’s usually because they’ve fallen dramatically out of favor with the bases of their own party (Liz Cheney, Andrew Cuomo), rendering them unable to win a primary election and thus electe…

Read more

👫 Red state population boom

The latest Census population estimates are out, and they show a slowing down of population growth thanks to a fall in international migration — and a continuing shift of the population into states won by Donald Trump in the 2024 election.

At the state level, nineteen states lost population to domestic migration to other states, with California experiencing the biggest net loss — 229,077 people. With the crackdown at the border and stepped-up immigration enforcement, domestic migration is what’s increasingly deciding whether states will gain or lose House seats after the next decennial census in 2030.

Every state projected to lose House seats was a Biden state in 2020, while every state that would gain seats was a Trump state in 2024. If these reapportionment projections came to pass, Trump would have won by 9 more electoral votes in 2024 and would not have needed to win any of the “Blue Wall” states to win.

Lest we get too far out fro…

Read more

🎂 One year of Trump 2.0

At the one year mark of presidential terms, you tend to get polls. Lots of them. And you’ll definitely get a New York Times/Siena poll.

The headline here is this chart from Nate Cohn (who should be the one doing the write-up for the print edition -ed.) showing most of the groups of the groups that swung hard towards Trump reverting back to their 2020 support levels — or worse.

Has Trump lost his coalition? That depends on how you define what the coalition is.

In 2016 and 2020, Trump realigned a lot of voters — white and nonwhite — who were better aligned with a populist and traditionalist vision of America but who previously voted Democratic. Those gains appear to be durable. If you want to compare 2026 to 2018, let’s compare demographic support levels in that election to this one as measured by the NYT/Siena poll, as Ryan Girdusky does in his excellent National Populist Newsletter:

• Democrats lead 18-29-year-olds by 27 points, down 22 poin…

Read more

📈 Nate Silver’s latest pollster rankings are out

Echelon Insights once again scores in the top-tier of pollsters nationally in Nate Silver’s latest pollster ratings.

Lots of pollsters will tell you that accuracy is only a small piece of the puzzle. After all, if the point of polling is to help clients improve their position, you hope your clients will do better on Election Day than they do in your polls.

Nevertheless, we welcome the opportunity to be judged for whether we ultimately got it right. Even if polls are a snapshot in time, clients are much better off with an accurate snapshot than one that’s too optimistic or pessimistic.

And we’re particularly proud to have next to no bias — meaning we don’t systematically overestimate one side or the other.

Lots of pollsters and forecasters can become hot commodities by shading their numbers towards a certain outcome that then comes to pass in any given election. Statistically, polls and forecasts will tend to miss i…

Read more