So that your inboxes are not empty, I have pre-written a small batch of articles which I will send out once a week through next month. I’ll also take some time, where I can get it, to write up the June and July polls with Verasight so our monthly trends remain unbroken. Each has a few great questions suggested by subscribers!

My many thanks to everyone who reads SIN, and especially those who support the site with a paying subscriptions. The media economy is in a weird place now where writers are asking subscribers to support them while they take breaks (negotiating even a standard week vacation when 70,000 people rely on you is, to put it mildly, kinda weird?), which I acknowledge is a departure from how employment worked in the past. I understand if you feel the need to pause or cancel your subscriptions, but I hope you’ll stick around for my return in ~8 weeks for the rest of the election cycle — and beyond. I have many cool projects planned for this year!

In the meantime, I’m looking forward to some time away to focus on our kid. I’ve opened up the comments of this post so people can share their parenting tips.

Elliott

The political blogger Matthew Yglesias had a piece out yesterday morning arguing that Texas state Rep. and Democratic candidate for U.S. Senate James Talarico needs to “moderate” his image/brand/positions/etc to defeat Texas Attorney General Ken Paxton in the race, which he says Talarico has a moral obligation to win. Here is the money graf, which commits the Strategist’s Fallacy in prescribing a solution but is otherwise directionally right on the problem: Texas is a red state, so Talarico will have to win independents and some Republicans to carry its U.S. Senate seat.

The article caused a small dust-up on X, with Greg Sargent of The New Republic pointing to his reporting that the Talarico campaign is running on what I generally call “anti-system” politics rather than just relying on one-dimensional moderation to win converts — the smarter play, in my opinion, given the data. I’m not interested in relitigating my ongoing fight with Matt to clarify what “moderate” means and bring some actual data and political science to bear on the conversation, and instead want to focus on the question in the title of this post.

See, underpinning a lot of the Discourse about Talarico’s chances is the claim that Democrats have to win “a lot” of Trump voters to take Texas. Given how red the state is at a presidential level, it’s tempting to conclude that a Democratic candidate can’t win in a state like Texas without appealing to a good chunk of Republicans too. The natural inference is that “good chunk” means 10% to 15% of them — but is that right?

This strikes me like a fitting empirical question for publication that’s all about covering politics with data, not punditry and conventional wisdom. And thanks to a lot of the other data I have crunched here at Strength In Numbers — from Census data to precinct-level election returns to voter-file matched polling data — I have all the tools to actually answer it. So come with me as I actually run the numbers on this very important question: How many Trump voters does James Talarico actually need to win in order to win the Texas U.S. Senate race?

This article is free to read, but it was not free to write. Subscribers to Strength In Numbers get exclusive data-driven analysis of politics and elections, access to premium polling data, and support independent, interactive data journalism. Consider joining as a paying member today:

Upgrade to support Strength In Numbers

If a paid subscription is not in the cards for you, the best thing you can do is share this article with a friend. Word of mouth is the #1 way new people learn about this publication! Thank you.

The partisan baseline in Texas is actually around R+5

To answer the big question today, we first need to figure out three small sets of numbers:

Read more

Perhaps this is why Ken Martin, the Chair of the DNC, didn’t want to release the report in the first place; the most robust answer to the question “Why did Democrats lose in 2024?” (answer: “inflation”) offers few to no action items going forward.

But another point to make about the 2024 DNC autopsy — and a point I had to drop from the original article (it was already quite long) — is that the closeness of the election makes it impossible to determine what really cost Democrats support. For this week’s Chart of the Week (the last one before I take a short paternity leave — don’t worry, I’ve pre-written six weeks’ worth of deep dives to come out in June and July!) I wanted to write another brief but important article in part about the DNC’s 2024 report. In particular, I think we should zoom out and reconsider the autopsy as an analytical product altogether. The reality is that we can’t really learn much from campaign autopsies in close races — and this leads people to adopt a lot of beliefs about party strategy that don’t hold up in the future. That’s genuinely pretty damaging for parties, so they should rethink their approach to studying the efficacy of campaigns.

This article is free to read, but it was not free to write. Subscribers to Strength In Numbers get exclusive data-driven analysis of politics and elections, access to premium polling data, and support independent, interactive data journalism. Consider joining as a paying member today:

Upgrade to support Strength In Numbers

If a paid subscription is not in the cards for you, the best thing you can do is share this article with a friend. Word of mouth is the #1 way new people learn about this publication! Thank you.

The 2024 election was really close, and that limits what we can learn about what worked

Donald Trump won the “tipping-point” state (the one whose electoral votes push the winner past 270, if you order the states by Democratic/Republican vote margin) by just 1.7 percentage points in 2024. Trump defeated Kamala Harris by just 120,000 votes in Pennsylvania. The excess electoral votes he won in Michigan and Wisconsin were decided by 110,000 votes — a fifth of the size of the smallest state.

The problem this presents for election analysts is that every campaign decision becomes potentially decisive for the outcome. There were 152 million votes cast in 2024, but the contest was decided by just 230,000 in those three states above — or 0.15 percent of all voters. When a race is that close, almost anything plausibly mattered. Inflation, as I wrote most about last week, could have flipped it on its own. So could Joe Biden clinging on until July, the administration’s policy on immigration and Israel/Gaza — or even soft turnout in Detroit and a few rainy days in Pennsylvania. Anything could have mattered — which means everything did.

Another way to put this is that the smaller the margin, the more explanations become mathematically plausible. In a 10-point loss, you need a large, systematic failure to explain the result. In a 1.5-point loss, you don’t. Empirically speaking, factors such as presidential popularity, wars, inflation, GDP growth, real disposable income, and candidate favorability all have effect sizes for vote margin that are at least double the size of Trump’s winning margin in 2024.

And that doesn’t even cover those factors we can’t study quantitatively: A slightly different nominee-selection timeline in 2024, a smaller turnout decline among core Democratic voters, a few thousand more persuadable voters breaking left, or a marginally worse Republican ad campaign could each have pushed the election the other way. The problem is not that these explanations are wrong. The problem is that we can’t use the result of the election to order them by influence.

Close elections are what social scientists call overdetermined — that is, the product of too many factors to decipher what dominated. This means they are also underpowered as causal evidence; when everything was enough to cause the loss, nothing is uniquely to blame for it. The election just doesn’t carry enough information to tell you which factor was the factor.

When this happens, people resolve the uncertainty about what matters by writing takes that reaffirm their priors. Several high-profile Democratic Party donor groups and centrist think tanks boosted the autopsy (with its scant new data and research) as additional evidence that their own ideological agenda is the only way forward (despite the new data that says this is overconfident). That’s not analysis so much as projection.

Worse, the results of the election can also be used to argue mutually incompatible conclusions. If you think Democrats are too moderate, you can point to youth and nonwhite erosion and say the party failed to inspire its base. If you think Democrats are too liberal, you can point to losses among working-class voters and say the party alienated the middle. If you think campaigns matter, you can point to the relative Democratic resilience in battleground states as evidence the Harris campaign worked (this is, I think, broadly true). If you think campaigns don’t matter much, you can point to the national anti-incumbent swing as evidence of the relative futility of the campaign as well (paradoxically, that’s also true). You can see how limiting analysis becomes when everything mattered.

The opposite problem also matters: autopsies are bad at noticing what worked. If a party loses, the review is naturally organized around failure. But in a close election, the more revealing question is often why the loss was not worse. That was the question I dug into in part in last week’s post.

Share

There’s also a risk of over-learning the wrong lesson

Our desire for clean, simple stories about the world leads to a related problem with autopsies: a tendency to hold on to single stories that extrapolate poorly or otherwise age badly.

The best example of this is the Republican Party’s autopsy of Mitt Romney’s loss in 2012. The conclusion from that report told Republicans to embrace immigration reform and soften their tone toward Latinos to win over more voters in a diversifying electorate.

The reason this example matters is that the 2012 autopsy was not obviously unserious. It was not a stray consultant memo or a cable-news hot take. The RNC billed it as a sweeping review of what had gone wrong, complete with recommendations on outreach, data, digital, the primary calendar, debates and immigration. Among other things, the report called for a $10 million minority-outreach push, comprehensive immigration reform, fewer primary debates, an earlier convention and a stronger data and digital operation. It would have completely remade the campaign goals and operation of the RNC.

And yet, the 2012 RNC autopsy fell apart almost immediately at the next election cycle. In 2015, Trump rode down his golden escalator and implied immigrants were rapists and criminals, and then pursued the harshest immigration policy in a generation — the opposite of everything the report advised. He won anyway. And then, from 2020 to 2024, the GOP’s share of the Latino vote climbed from 36% to 48%, per Pew.

That doesn’t mean harsh immigration rhetoric was the secret to winning in 2016 and 2024. It almost certainly wasn’t. The point is narrower and more important: strategic narratives built from a single election often age badly. The RNC autopsy treated its most notable failing as the central lesson of 2012, when there were many other reasons Romney struggled. Four years later, the party won the White House with a very harsh message on immigration. Eight years after that, its nominee made large gains among the very voters the autopsy had treated as requiring a softer message to convert.

And, not for nothing, after 2020 there was no (public) GOP autopsy at all. (I guess you can’t analyze a defeat your party leader denies happened at all.)

It’s fair for people to want answers from the parties about what they could have done better. These organizations spend (or organize the spending of) billions of dollars every four years; what are they doing right or wrong with that money? People want certainty about what to do next time.

Unfortunately, autopsies simply can’t provide that certainty. A close loss like the Democrats’ in 2024 doesn’t come with one mono-cause; it comes with a dozen, or even dozens. You might as well just randomly pick one of them to fix for the next campaign than spend even more time and money proving the “most important” ones.

And 2024 wasn’t a one-off. Close elections are now the rule, not the exception. The tipping-point state has been decided by three points or fewer in five of the last seven elections.

Based on the other national elections this century, plus the increasing partisan sorting of the electorate, odds are that the 2026 and 2028 elections will be close, too.

None of this means campaigns shouldn’t review their decisions. They should. They should test ads, audit field programs, evaluate polling methodologies, compare persuasion and mobilization experiments, and figure out which operational failures were preventable.

But this isn’t really what most people are doing when they’re looking back at campaigns. They are trying to sell you a magic solution — their magic solution — to the problems that most animate them. And in all fairness, in a close election, the data gives lazy analysts plenty of ammo to work with.

But if you are to read such reports, you should understand them for what they are: noise. Especially in close elections, party autopsies are mostly a distracting exercise in futility.

If you read this far, I’m confident you would get a lot out of a paid subscription to Strength In Numbers. You’ll get weekly premium posts like this one, access to the full historical archive of posts here on Substack, and early access to new data products — plus the ability to add questions to our monthly surveys. Read more here or subscribe using the button below:

Upgrade to support Strength In Numbers

Related Articles

All the writing and data you see at Strength In Numbers is subscriber-funded. If this work is useful to you, upgrade to a paid subscription to keep our independent data journalism project going — and to get the full polling crosstabs, early access to new data products, and my premium Tuesday Deep Dive columns.

Upgrade to support Strength In Numbers

When Donald Trump took office in January 2025, he inherited an economy that was the envy of much of the developed world. America had falling inflation, a tight labor market, strong sales and manufacturing growth, and consumer confidence was recovering steadily for the better part of a year. While prices still seemed high compared to their historical trend, this was really one of the only serious dark spots on the economy when Trump took office. The post-COVID economic recovery in the U.S. looked like something of a miracle when compared to peer countries.

Sixteen months later, everything is changed. Trump’s policy of sweeping global tariffs has pushed up the price of imported goods from groceries to gasoline. Mass deportations have thinned the workforce in agriculture, construction, and food service, squeezing supply and lifting costs in exactly the sectors households feel most. The One Big Beautiful Bill, signed last summer, extended unpopular tax cuts that were skewed to the top while cutting a trillion dollars in funding for Medicaid, SNAP, and other federal subsidies, and let ACA’s enhanced premium subsidies expire at the end of 2025. Headline inflation has climbed back to 3.8%, and gasoline is up 50% year-over-year.

Americans have noticed. In our new Strength In Numbers/Verasight poll, 58% of Americans say Trump’s policy decisions have made the economy worse since he took office. Just 21% say he’s made it better. And this anxiety hits close to home, too. A plurality of 47% say their own household finances are worse off since Trump took office.

The story in this poll is not just that Americans dislike Trump’s economy, it’s that they are connecting national economic discontent to their own household bills — and increasingly assigning responsibility to the president now in office. By a margin of 41% to 16%, Americans now blame Trump rather than Joe Biden for the high cost of living.

What policy solutions do Americans favor? The poll points to a fairly clear agenda. Voters overwhelmingly want lower grocery prices, lower rent, lower health insurance premiums, and higher wages. When forced to pick policies rather than outcomes, they gravitate towards tax cuts for middle- and working-class households (35%), higher taxes on high earners and corporations (29%), stricter regulation of corporate price-gouging (23%), and reducing the federal deficit. They also want to feel like somebody listens to them.

But this is also a warning for Democrats. As I argued after the DNC’s 2024 autopsy, the anti-incumbent bias that sank Biden in 2024 — and is now sinking Trump in 2026 — does not disappear once power changes hands. Voters will not reward Democrats for being anti-Trump forever. The party will need an affirmative vision for bringing down the cost of groceries, gas, rent, and health care — and the legislative follow-through to actually deliver it. Without that, they will get kicked out the same way Biden was — and the way Trump is being now.

Subscribe now

Trump’s economic ratings are at historic lows

Perhaps the president’s worst numbers are on his handling of the economy, an issue Republicans have dominated for the last 15 years. So let’s start there.

Americans render an overwhelmingly negative verdict on Trump’s stewardship of the economy. By a 58–21 margin, voters say his policy decisions have made the economy worse rather than better since he took office — a net rating of -37, and a deeply negative reading for any president this early in a term. Trump’s straight job approval on the economy is 35%, with 60% disapproving, a net of -25. For context, Bush 43’s overall job approval sank to 25% during the 2008 financial collapse — with the economy his worst-rated issue — and Carter’s overall approval cratered into the high 20s during stagflation. Trump isn’t quite there yet — given polarization, I think he’s unlikely to drop into the 20s, though you never know with this guy — but he’s heading in that direction.

These bad ratings are consistent across questions. Trump’s approval on prices and inflation hit a record low of -47 in May’s Strength In Numbers/Verasight poll. Thirty-nine percent of Americans now name prices as the single most important problem facing the country, the highest share we’ve ever recorded for any issue. And a record 62% include prices in their top three most important problems.

And for what it’s worth, this poll is not an outlier. Quinnipiac (May 15–19) had Trump’s economy approval at 33% — an all-time low for him across both terms (so far). Fox News (May 15–18) had economy disapproval at 71%, also a series high. The New York Times/Siena (May 11–15) had 64% disapproving. CBS/YouGov (May 13–15) found 14% saying Trump’s policies are making them personally better off financially and 57% saying worse — close in shape to our 21/58 split on whether Trump’s policies have helped or hurt the economy, though the questions ask about different scopes. No single poll should be overread, but the pattern across questions — and across pollsters — is unusually consistent here.

We also asked how Americans would grade Trump against their own pre-term expectations. Just 30% of Americans say Trump is doing better on the economy than they expected — but 64% say he’s doing worse, including 45% who say he’s doing “a lot worse” than they expected. Independents say worse by 71% to 19%.

Even among Republicans, 29% — nearly one in three — say Trump is doing worse on the economy than they expected, including 10% who say he is doing “a lot worse.” Pull in independents who lean Republican and the share saying worse climbs to 33%. That is a meaningful crack in his partisan base.

Share

Nearly half of Americans say their personal finances are worse under Trump

It’s not just that Americans feel poorly about “the economy” in the abstract. They also say their own personal financial situation is eroding.

We asked respondents to compare their personal financial situation now to when Trump took office. A 47% plurality say they’re worse off — 24% somewhat worse, 23% much worse. Just 23% say they’re better off. Another 28% say things are about the same.

The intensity gap is worth noting. Nearly a quarter of all Americans (23%) say they are much worse off financially since Trump took office — roughly three times the share who say they are much better off (8%). The share who say they are much worse off is the same as the share who say they are somewhat or much better.

61% of Americans are worried about their household finances

When we ask Americans how worried they are about their household’s financial situation, 61% say they are worried — 24% very worried, 37% somewhat worried.

And the worry is intensifying. 51% of Americans say they are more financially anxious now than they were a year ago, including 25% who are “much more anxious.” Just 15% say they are less anxious.

One common talking point about the economy and the “vibecession” today is that Americans are only worried about the economy in abstract, not their own conditions, which they rate as doing well. This data cuts squarely against that narrative. A majority of Americans say they are somewhat or much more anxious about their own household’s financial future than they were a year ago:

The anxiety is producing visible belt-tightening. More than two-thirds of Americans (69%) say they’ve decided not to make a purchase in the past month because of the cost — 33% “very often” and 36% “somewhat often.”

When we asked respondents to rate how much specific household expenses were contributing to their financial stress, the answers track the news exactly. More than any other category, Americans are upset about groceries, utilities, health care, and housing.

Some things that stick out:

• Groceries and food: 49% say groceries are contributing “a great deal” to their household financial stress; another 33% say “some” — meaning 82% of Americans are stressed about food prices in some way.

• Utilities: 38% great deal, 35% some (73% combined).

• Health care and health insurance: 36% great deal, 32% some (68% combined).

• Emergency or retirement savings: 35% great deal, 30% some (65% combined).

• Housing: 33% great deal, 27% some (60% combined).

We also asked respondents, in their own words, what their single biggest economic worry was right now. I have selected a few examples from across the aisle to embed here, and listed more at the bottom of this article:

“Inflation continuing to climb. Our grocery bills have DOUBLED in the past year. There are things we can do without but food is not one of them!”
— Independent, 45–64

“Inflation and cost of living far outpacing my wage.”
— Republican, 45–64, 2024 Trump voter

“My electric service increased over 300%.”
— Republican, 65+, 2024 Trump voter

Two of the three above are from Trump’s own 2024 voters.

I coded all 1,273 substantive responses (84% of the sample) by topic. The chart below shows the breakdown.

38% of all respondents mentioned inflation, the cost of living, or some general version of “everything is too expensive.” Curiously, that lines up almost perfectly with the share of adults in this survey who say inflation is the country’s number one issue.

When they were more specific about their worries, 18% named gas or fuel prices, 14% named groceries or food, 7% named housing or rent, 13% mentioned jobs, wages, or layoffs, and 6% mentioned healthcare or insurance. Roughly four in ten respondents (36%) named at least one specific household bill by name.

Beyond that core, 4% named Trump or his administration directly, 2% mentioned the war in Iran (almost all tying it to gas prices), and 1% mentioned tariffs specifically. Trump did not need to be named for blame to attach to him — but a non-trivial slice of Americans is putting his name in their own answer, unprompted.

Americans blame Trump for the cost-of-living crisis. They no longer blame Biden.

41% of Americans say President Trump and his administration are most responsible for the cost-of-living problems people face today, more than any other option. That’s more than two-and-a-half times the share who blame Joe Biden and the previous administration (16%). In addition, 11% blame large corporations and CEOs. 8% blame congressional Democrats, 7% say “no one in particular — it’s just the economy,” and 4% blame congressional Republicans. (There are probably many Americans who would blame multiple actors here, but we forced them to pick just one.)

This is one of the biggest political shifts in our poll. For the first three months of Trump’s second term, Americans were still willing to wait and see — in CBS/YouGov’s March 2025 poll, voters still gave Biden slightly more blame than Trump for inflation. By April, the script had flipped: a CBS/YouGov poll that month found 54% saying Trump’s policies were most responsible for the state of the economy versus 21% for Biden’s. By October, the Economist/YouGov poll had it at 57% Trump to 24% Biden. In 2026, the public has settled.

The next question is what Americans want done about this.

What voters actually want

When we asked what outcomes would most help reduce people’s financial anxiety, the runaway winner was simple: lower grocery prices. 64% of Americans said cheaper groceries would help “a lot,” 20% said “some” — a combined 84%. The next-highest item was a raise or higher wages (47% “a lot”), followed by lower rent or housing costs (47%), lower health insurance premiums (44%), and a federal tax cut (42%).

When we asked about outcomes, voters mostly said they wanted relief on household bills. When we asked about policy, they gravitated toward redistribution and corporate accountability.

We asked respondents which two policies below would most reduce their financial anxiety. The top response was a tax cut for middle- and working-class households (35%). The second was raising taxes on high earners and corporations to fund more spending (29%). Stricter regulation of corporations and price-gouging (23%) came in third, ahead of reducing federal spending and the deficit (20%), a higher federal minimum wage (16%), and federal investment in affordable housing (14%).

Notably, that is roughly the inverse of what Republicans passed in the One Big Beautiful Bill. The Congressional Budget Office estimated the bill would increase wealth at the top of the income spectrum and decrease it at the bottom: the top 10% of earners gain 2.7% in after-tax income by 2034, while the bottom 10% lose 3.1%. CBO also projects 10.9 million more Americans uwill be ninsured by the end of the decade compared with current law.

If party elites are trying to find out what policies poll well with average Americans — and thus may ease anti-incumbent bias when they are in power — this list is a good starting point.

Democrats are also winning the “cares about people like you” question

For decades, the question “Which party do you think cares more about people like you?” has been one of the most predictive in American politics. In our May poll, 38% of Americans say the Democratic Party cares more about people like them, compared to just 27% (D+11) who say Republicans. Five percent say both equally, and 23% say neither. That is an indictment, on some level, of both parties — but especially the GOP.

Not coincidentally, the Democratic lead on the House midterm vote in our poll is D+10 among adults.

The Democratic opening — and the Democratic trap

(Now, taking off my pollster hat and putting on my analyst/forecaster hat for a section...)

As I wrote last Friday, Democrats need to have a well-researched and authentic economic plan to save themselves from the political pendulum next cycle:

This, of course, will be hard; defying (midterm political) gravity always is. It requires governing the way Roosevelt and Johnson governed, with the assumption that bold action creates its own political coalition. And it also requires you to look back on your failures with a holistic view of what causes certain outcomes.

I have not yet seen anything close to this from the party, from the think tanks, or from the leading 2028 contenders. There are gestures toward the “abundance” agenda, populism, and “radical centrism,” but these mostly read like positioning exercises, not serious agendas for people who are really hurting. The 2024 autopsy was a chance to start that conversation by being honest about why Harris lost. It failed to do that.

This matters because the fundamentals giveth and the fundamentals taketh away. A Democratic win in 2028 built on Trump backlash, without a generational economic vision behind it, is a Democratic loss in 2032 waiting to happen. The only durable answer to anti-incumbent sentiment is to actually earn re-election. The party has roughly two and a half years to decide whether it wants to think that big — and that creatively.

One thing to do is to emphasize policies that solve these anxieties. Voters want cheaper groceries, higher wages, lower rent, and lower health insurance premiums. They want middle- and working-class tax cuts. They want corporations and high earners to pay more. They want stricter regulation of corporate price-gouging. And they want, above all else, to feel like the American dream is working for them.

Crucially, the policy polled here is nothing radical or hard to defend to swing voters. One message that will work well in 2028 is “Trump promised lower prices. Instead, groceries, gas, rent, and health care are squeezing families while Republicans cut taxes for the wealthy and cut health care for everyone else. We need to decrease costs for the average household, restore federal health care funding, and pass a new federal minimum wage hike so everyone is paid a living wage.”

But the overall point is not about the policy specifics, but about the tenor and focus of the party. If Democrats win the House next November — or the White House in 2028 — they will inherit the same angry electorate that just kicked out the other party. Being not-Trump won’t be enough. Party actors need to put some real energy into developing a believable, affirmative cost-of-living agenda, and the legislative follow-through to deliver it, or they will be the next ones blamed.

(Putting my pollster hat back on.)

The political bottom line

Donald Trump came into office promising to bring prices down. Americans believe he’s done the opposite. Their personal finances have gotten worse on his watch. They are anxious about groceries, utilities, housing, health care, and savings. They are forgoing purchases because of cost. They blame Trump and his administration directly, and they support a broad left-leaning policy agenda — middle-class tax cuts, higher taxes on the rich, stricter corporate regulation — that is essentially the opposite of what Republicans have pursued over the last year.

Republicans went into 2025 betting that the public would forgive any short-term economic pain so long as Trump delivered on immigration, tariffs, and his other signature priorities. They were wrong. There are still several months until the 2026 midterms, and the political environment could change — the economic downturn could ease and prices could fall. But as we saw with Biden, once the public has decided who’s responsible for an economic crisis, it tends not to change its mind.

The May Strength In Numbers/Verasight poll surveyed 1,520 U.S. adults online between May 18 and May 19, 2026. The margin of error for the full sample is ±2.7 percentage points. Full toplines, methodology, and crosstabs are available at gelliottmorris.com/poll.

If you made it this far: Strength In Numbers is subscriber-supported. If this work is useful to you, upgrade to a paid subscription to keep our independent polling project going — and to get the full crosstabs and Tuesday Deep Dives.

Upgrade to support Strength In Numbers

Related Articles

Appendix: What Americans said in their own words

We asked all 1,520 respondents: “In your own words, what is the single biggest economic worry on your mind right now? Please be as specific as you can.” Below is a selection of representative responses, grouped by topic mentioned.

Groceries and the cost of everything

“Cost of living (rent and groceries) are increasing at staggering rates and wages can’t keep up with it.”
— Democrat, 30–44

“Grocery prices have gone through the roof here in South Florida! I am extremely worried at the huge increase in basics such as meat!”
— Democrat, 65+

Gas and utilities

“Gas and grocery prices are absolutely INSANE.”
— Independent, 45–64

“Energy expenses (LP gas, electric, fuel: gas & diesel).”
— Republican, 65+

Housing and rent

“Right now I am worried about housing, rent is already high and I’m having issues keeping up — not sure what I will do when it goes up more.”
— Republican, 30–44

Healthcare

“Health care — I can’t afford the health care that I need and I’m dumping my funds into insurance that doesn’t cover enough of my costs.”
— Independent, 30–44

Wages and jobs

“I graduated college a year ago and I’m still unemployed.”
— Democrat, 18–29

“I am worried that my job will end and I will become homeless.”
— Democrat, 45–64

Responses from 2024 Trump voters

A reminder that economic anxiety is not confined to Trump’s opponents:

“Inflation and cost of living far outpacing my wage.”
— Republican, 45–64, Male, Trump 2024

“My electric service increased over 300%.”
— Republican, 65+, Male, Trump 2024

“Rising costs of living, especially housing and healthcare, making it harder to save or plan for the future.”
— Republican, 18–29, Male, Trump 2024

“My health care going up and my raise not even covering it.”
— Republican, 30–44, Male, Trump 2024

“Life savings gone; living on Social Security alone.”
— Republican, 65+, Female, Trump 2024

“Layoffs imminent as executives believe they can replace folks with AI.”
— Republican, 30–44, Male, Trump 2024

Attorney General Ken Paxton defeated incumbent U.S. Senator John Cornyn in the Republican Senate runoff today, giving Democrats their preferred and weaker opponent. That’s not because Paxton is all that much more “conservative” than John Cornyn (whatever that means these days), but because he’s just generally a bad dude who has earned a lot of bad press over the last year. Paxton has what we call in political science a “negative candidate residual.”

The general impression that Paxton is a worse general election candidate than Cornyn would have been is not misguided. Empirically, hypothetical general election polling showed Talarico doing about 1-2 points better against Paxton than Cornyn across many polls. That is not an earth-shattering difference — but it might be enough in a close race.

And this race is looking really, really close.

Texas hasn’t elected a Democrat statewide since 1994. For 32 years, every cycle has come with a fresh round of “is this finally the year?” pieces, and for 32 years the answer has been no. I’ve written some of those pieces myself — including one for The Economist back in 2019 calling Texas the Democrats’ “white whale.” Then 2020 happened, then 2024 happened, and the state seemed to drift further from the party’s reach.

Given that, I expect that people will view this piece skeptically, and I want to be careful here. But the data is the data, and right now the data on Texas’s 2026 Senate race looks genuinely different from past cycles. I think we have a genuine tossup on our hands. Could this year be the Democrats’ year?

This article is free to read, but it was not free to write. Subscribers to Strength In Numbers get exclusive analysis of politics and elections, access to premium polling data, and support independent, interactive data journalism. Consider joining as a paying member today:

Upgrade to support Strength In Numbers

If a paid subscription is not in the cards for you, the best thing you can do is share this article with a friend. Word of mouth is the #1 way new people learn about this publication.

Texas has all the makings of a Democratic victory

Start with the polling. Hypothetical general election numbers at FiftyPlusOne currently show Talarico running a couple of points ahead of Paxton (and there have been some particularly bullish surveys for Talarico recently).

That alone is promising data for the Democrats, even if it is early (and somewhat hard to believe).

But remember, Texas is not all crimson red all of the time. Back in 2018, Beto O’Rourke lost to Ted Cruz by 2.6 points. That was the closest a Democrat had come to a statewide Texas win in decades. Importantly for our benchmarking purposes, it happened in a year with a roughly D+7 to D+8 national environment. In other words: Democrats came within striking distance of Texas in a very good year for their party — about 9-10 points to the right of the national vote.

Now look at where things stand for 2026. The national generic ballot today is roughly D+6 to D+7 — not quite 2018, but close. In fact, Democrats seem to be running even with Democrats at this point in that last “wave” election — the generic ballot at this point in 2018, per my historical average, was also D+6. If Democrats gain another 1-2 points in the next 6 months (which is extremely doable in a midterm), that puts them in line with 2018 or better.

And there’s a real argument that the D+6 national number understates Democrats’ actual position in non-presidential electorates today. I’ve made this case in detail at this blog already: the people who turn out in midterms and specials have been disproportionately Democratic this cycle, and special election results have been running well to the left of 2024 presidential numbers. This alone means we should expect Democrats to do better in 2026 than in 2018, all else being equal.

Zoom in to Texas specifically, and the picture gets even more interesting. My state-level generic ballot estimates now have Democrats at 50.7% of the two-party vote for House candidates in the Lone Star state — essentially even with, or slightly ahead of, Republicans on the generic ballot in a state Trump won by 14 points just 18 months ago.

That’s a remarkable shift, and it lines up with what my MRP model of Trump approval is showing in Texas, too: the president is about 20 points underwater with registered voters in Texas, with most of the state’s metro areas — including the Houston and DFW suburbs — registering net-negative approval. When the leader of the GOP is upside-down in your state, every Republican on the ticket inherits some of that drag.

In a separate piece breaking down the race in March, I laid out six reasons Texas could actually flip: a potentially very weak Republican nominee in Paxton (scandal-ridden, polling 2-3 points behind Cornyn in head-to-heads against Talarico), a Democratic recruit who plays well with independents and suburban voters, eye-popping primary turnout numbers, a diversifying Democratic electorate, a continued Latino swing back toward Democrats in special elections, and Trump’s poor numbers in the state. Generally, the environment has only gotten better for Democrats since then.

Texas also has higher-than-average shares of Hispanic and low-income voters, both of which have trended away from Trump and the Republicans at higher rates than the population as a whole.

The data is hard to argue with

None of these factors is individually decisive on its own. But that’s largely the point — they don’t have to be.

Beto O’Rourke’s loss in the 2018 Texas Senate race shows how genuinely competitive the state can be, with the right candidate and a favorable national environment. In a D+7-8 year, our expectation should be for a close race. If the environment is comparable to 2018 — and the Texas generic ballot, Trump’s sagging approval, and recent special elections all suggest it might be — and Talarico campaigns as well against Paxton as he has so far, the math starts to add up.

The Texas Senate race is looking close. My gut is telling me to be shy here. History is telling me to be shy here. But the data is compelling, so I’m going to be bold… I think it’s a genuine tossup.

And now, back to editing tomorrow’s poll release…

Subscribe now

The kind of original polling analysis and data journalism I’m publishing here at Strength In Numbers doesn’t exist anywhere else in the media today. Paid subscribers make it possible for me to run custom surveys, build models like the MRP estimates cited above, and spend my time on public-interest journalism full-time. If that’s a mission you can get behind, consider supporting Strength In Numbers today, and get access to tons of exclusive data-driven journalism like this.

Upgrade to support Strength In Numbers

President Donald Trump’s approval rating is essentially where it was a month ago — and a month before that, and a month before that. Our latest Strength In Numbers/Verasight poll, conducted May 18–19, finds 37% of U.S. adults approving of his job performance and 60% disapproving. The net rating of -23 is within the margin of error of April’s -26 and is the second-worst reading we’ve recorded since we began tracking the question a year ago.

Yet despite his stable topline rating, the president’s net approval on prices and inflation — the issue voters say they care about most — fell to a new record low of -47, a 1-point drop from April’s -46. Just 25% of Americans now approve of the way he’s handling prices, while 72% disapprove. 57% of all U.S. adults disapprove strongly.

This has been the basic math of the president’s popularity for the last year. He cannot dig himself out of a hole with the average American if he continues to marginalize them on their core priorities. The issue voters care most about — the one they say is the single most important problem facing the country today — is also the one where the president is most unpopular.

Below, I report the results from the monthly political tracking questions in our May Strength In Numbers/Verasight survey. In another article tomorrow, I’ll dive into our new data on how stress over prices and job security affects the average American, and what policies they want leaders to pass to address those challenges.

Headline poll findings

• Job approval: 37% of U.S. adults approve of Trump’s job performance; 60% disapprove (net -23, essentially unchanged from -26 in April and the second-worst reading in our trend)

• Generic ballot: Democrats lead Republicans 51% to 43% among registered voters — an 8-point margin — and 51% to 41% among all adults. Democrats have led the generic ballot in every poll we’ve conducted.

• Direction of the country: 52% say things are going poorly and major, disruptive changes are needed. Just 8% say things are going well. This is a continuation of deepening pessimism from last month.

• Prices: Trump’s net approval on prices and inflation has fallen to -47, a new record low. 39% of Americans now name prices as the single most important problem facing the country — also a record in our polling.

• Issue approval: Trump is underwater on every single issue we tested except border security, where he sits at net 0 (48% approve, 48% disapprove).

• Party trust: Democrats lead Republicans on 8 of the 12 issues we tested, including all the top voter priorities — prices, jobs, health care and, elections and democracy.

Strength In Numbers is able to publish original, independent political polling thanks to the paying subscribers who support this site. To become a paid member and support work like this, plus get at least one exclusive premium analysis article each week, click the button below.

Upgrade to support Strength in Numbers

Democrats have led the generic ballot in every poll we’ve conducted

On the generic congressional ballot, Democrats lead Republicans 51% to 43% among registered voters — an 8-point margin. Among all U.S. adults, the lead widens to D+10 (51% to 41%). 7% of registered voters — and 8% of all adults — say they don’t know who they would vote for.

Democrats have led the generic ballot in every single Strength In Numbers/Verasight poll since we began fielding in May 2025. Across 12 monthly polls (we skipped December 2025), Democrats have never trailed, with margins ranging from +5 to +10 points among registered voters.

The May D+8 among registered voters is essentially in line with April’s D+7 result and within the normal range of poll-to-poll variation. It also matches the May aggregate of public polls (roughly D+6 to D+8 depending on the source); the New York TImes/Siena poll last week had Democrats up 11 among registered voters.

Trump’s rating on prices hits a new record low

And if you thought Trump’s position on inflation and the economy couldn’t get any worse...

Our new polling of Trump’s popularity at the issue level shows a continued slide for the president. Trump is underwater — sometimes deeply so — on every issue we tested except border security, where he sits at exactly net 0 (48% approve, 48% disapprove).

His worst rating, by a wide margin, is on prices and inflation: at -47. Trump’s approval on prices has now gotten worse in every single month of 2026 — -31 in January, -35 in February, -40 in March, -46 in April, and now -47 in May. Just 25% of Americans approve of how he’s handling prices, while 72% disapprove. Worse, 57% strongly disapprove. More than half of all U.S. adults don’t just disapprove of how the president is handling the cost of living, but say they strongly disapprove.

Trump’s other issue numbers are scarcely better. He’s at -28 on health care, -25 on jobs and the economy, -23 on foreign policy, -22 on elections and democracy, -22 on government funding and social programs, -21 on trade, and -19 on education. Even on his strongest non-border issues — deportations, immigration, and crime — he’s at -12 across the board.

As for most of our recent surveys, border security is the lone exception; Trump is tied on the issue. Forty-eight percent approve, 48% disapprove.

Prices keep climbing as the country’s top problem

When we ask Americans to name the single most important problem facing the country, 39% say prices and inflation — the highest share we’ve ever recorded for any issue. That’s a 5-point jump from April (34%), and prices outpace the next-closest issue, elections and democracy, by nearly three to one.

13% name elections and democracy, 12% name jobs and the economy, and health care rounds out the top four at 7%. When we let respondents name their top three problems, instead of forcing them to pick one, the gap widens further: 62% select prices, 40% name jobs and the economy, and 33% pick health care. Elections and democracy comes fourth at 25%.

The tough reality of the math for Republicans is that the issue where he is most unpopular is also the issue Americans care about most. And this isn’t just a Trump problem anymore; our poll shows this is also having severe consequences for the Republican Party, too.

Democrats lead on the issues voters care about most

When we ask voters which party they trust more to handle each issue, Democrats lead on 8 out of 12 policy areas. They hold their biggest advantages on health care (D+20), education (D+15), government funding and social programs (D+15), and elections and democracy (D+10). They also lead on prices and inflation (D+10) and jobs and the economy (D+8) — the two issues voters rate as the country’s biggest problems, and on which Trump and Republicans had a clear lead in the 2024 election.

Republicans lead on just four issues, all in their traditional wheelhouse: border security (R+12), crime and public safety (R+4), deportations (R+3), and immigration (R+1).

And the data has been moving generally in the Democratic direction over the last year. The biggest shift since we began polling is is on inflation, where Democrats have gone from D+1 in May 2025 to D+10 today, and on immigration, where the Republican lead has decreased from R+12 last year to R+1 this month.

When we ask voters which party would do a better job handling their personal most important problem, Democrats lead 49% to 36% (15% are unsure). That 13-point advantage is a 1-point improvement from April’s D+12 and a more significant rise from previous months, with the Democrats at D+8 in February and March 2026 and D+4 in May 2025.

52% say things in the country are going poorly — still near a record high

The share of Americans who say “things are going poorly and major, disruptive changes are needed” sits at 52% in May, down slightly from April’s record high of 55%. Just 8% say things are going well; 36% say things could be going better. Combine the two and 88% of Americans are dissatisfied with the direction of the country.

The trend is (still) not Republicans’ friend

Trump’s overall approval has stabilized at a deeply unpopular level. Since January, his approval has fallen from 40% to 37% and his net rating from -18 to -23. His rating on prices — the issue voters care about most — has fallen every single month this year, and now sits at a record -47. That is significantly worse than rating for Joe Bien at this point in his term in 2022.

Aside from approval, other indicators are also flashing red. The country is very dissatisfied, with the vast majority saying major political and economic changes are needed. The president is underwater on almost every issue, and if you ask voters who they trust more to handle their personal most important issue, they say the opposition party by a growing 13-point margin.

In terms of electoral consequences for this poor performance, Democrats have led the generic ballot in every poll we’ve conducted since May 2025. Their margin has grown in recent months. When a president is this unpopular this close to a midterm, his party usually loses badly. Bush, at this same approval level in May 2006, saw his party lose 30 House seats and the majority later that year.

If our May numbers are anywhere close to where things stand in the fall, Republicans should be preparing for significant losses, up and down the ballot.

Over the next few days, Strength In Numbers will release more data from our May survey with Verasight — including a closer look at how Americans are feeling about their personal finances, who they blame for the cost-of-living crisis, and what they want Congress to do about it. These data will offer a lot of information to readers looking to inform debates about policy agendas that could be used to blunt the power of anti-incumbent sentiment in 2028 and 2032, as I wrote about last Friday. Subscribe today to get those fresh polling dispatches delivered straight to your inbox.

Subscribe now

Methods statement: Verasight collected the data for this survey from May 18–19, 2026. The sample consists of 1,520 United States residents ages 18 and above. The data are weighted to match the April 2026 Current Population Survey on age, race/ethnicity, sex, income, education, region, and metropolitan status, as well as to a running three-year average of partisanship distributions from the Pew Research Center NPORS benchmarking surveys and population benchmarks of 2024 vote. The margin of sampling error is ±2.7%.
G. Elliott Morris prepared the topline document for Strength In Numbers. Strength In Numbers had input on question wording, but all other methodological decisions were made and carried out by Verasight to ensure independence of the data-generating process behind these results. Verasight also reviewed the questionnaire.

You can download a full topline file, key crosstabs, and full methodology statement at the Strength In Numbers/Verasight polling portal. The paywalled section containing crosstabs and more interactive graphics has been updated with these numbers.

If you have any questions about this poll or the release, please email polling[AT]gelliottmorris.com.

Have a suggestion for next month’s poll? Leave it in the comments below.

If you’re a frequent reader of Strength In Numbers, I’m confident you will get a lot of value out of a paid subscription. You’ll get access to all of my analysis, including weekly Tuesday Deep Dives and other exclusive posts, and you’ll be supporting independent data journalism — like this original polling.

Upgrade to support Strength in Numbers

Related Articles

Upgrade to support Strength In Numbers

I had intended to publish a wonky Chart of the Week for today about why I think this article by Charlie Cook overstates the impact of partisan sorting and understates Democrats’ chances for the midterms… but then the news cycle got in the way. The piece got pretty “weedsy” anyway, so it’s probably better as a Deep Dive for premium subscribers. I have scheduled it to publish next month while I’m out on paternity leave.

To take its place, the news gods have given us the Democratic National Committee’s “autopsy” of the party’s defeat in the 2024 election. There is a lot of controversy surrounding this report, mostly stemming from the fact that the current DNC chair promised to release it when he was running for his position in 2025 and then abruptly changed his mind after winning. He apparently flip-flopped back to publishing the report yesterday after CNN got a hold of portions of it.

Unfortunately, there is little to learn from the autopsy. The story is mostly about what’s not in it, which I’ll get to. As for what is included, start with the fact that it contains many factual errors (this is a “draft” report and they they presumably would have been caught in a full copy edit, but still), as evidenced by warnings such as the one below that appear on nearly every page:

The autopsy does not really present a unified theory of why Kamala Harris lost in 2024. The unnamed author writes, among other things, that the party moved too far left and failed to define Trump and go negative effectively. But it arrives at these conclusion with a surprising lack of new research, and with a lot of pundit conventional wisdom and “analysis” you’d expect from someone committing the Strategist’s Fallacy. I don’t know how much the DNC pays for reports like these, but presumably if it really wanted to understand why Harris lost, it would want to invest some resources into fact-finding research like surveying state party chairs and local officials, as well as talking to voters who switched sides. Maybe it could even do some original causal research on its theories.

But the biggest problem is that the autopsy straight up ignores the major reasons Harris lost in 2024. Yes, it’s bad enough that the report doesn’t mention that party bosses failed to coordinate an early exit for Joe Biden, who was too unpopular to win. And there is no mention of Israel/Gaza, low turnout in the cities, and nothing on Harris’s race or gender. But this is a data-driven site, so I want to really focus in on what the numbers can tell us.

When we boot up the data, it’s obvious the main reason Harris lost — and the reason I am going to explore here, at this website, it being a data-driven website — is that 2024 simply had too much inflation-induced anti-incumbent sentiment for the incumbent party to overcome. This is curiously missing from its main diagnosis. The word “inflation” isn’t mentioned in the autopsy a single time (except in the context of inflation-adjusted ad spending).

This week’s Chart of the Week is about the real reason Democrats lost in 2024.

Subscribe now

The fundamental(s) issue

The reality of the 2024 election is that it was going to be hard for a Democrat to win, regardless of who they were or how they campaigned. The broader economic and political conditions were so favorable to Republicans that you would have expected Trump to win about 90% of the time, regardless of campaign or candidate effects.

Political scientists have been pointing out for decades that you can predict presidential elections reasonably well using just two pieces of information: how voters feel about the incumbent president, and how voters feel about the economy. There are many variants of this model — such as the “Bread and Peace” model (Douglas Hibbs), the “Time for Change” forecast (Abramowitz), Ray Fair’s “Fair” model, and Wlezien/Erikson’s work with “Leading Economic Indicators” — but all use a similar set of economic and political “fundamentals” to predict the result of the election.

None of these models is perfect, but they’re a useful baseline for forecasting elections and understanding why certain results happen. One use of the predictions is to distinguish the part of an outcome that’s about candidates and campaigns from the part that’s structural.

The chart below shows a model I built for exactly these explanatory purposes. It uses Gallup’s net approval of the incumbent president measured in June and the trailing two-year average of the University of Michigan’s Consumer Sentiment Index to predict the incumbent party’s share of votes cast for the Democratic and Republican candidates for president — what we call “two-party” vote share. That’s it: two variables, both measured before Labor Day, both knowable months before any ad runs, and in some cases, before candidates are even picked.

The dashed line is a perfect prediction. The shaded line shows the 80% prediction interval for the 2024 election, and the solid blue dot shows the actual result of the race.

In 2024, Kamala Harris received about 49.3% of the two-party vote. The model — fit on data from 1956 through 2020, with 2024 held out — predicted she’d get about 48%, with an 80% prediction interval of 46.6% to 49.9%. Harris’s vote share lands on the upper end of this range, but still squarely inside of it.

In a sense, then, the surprise of the election is that Harris did as well as she did, considering the prevailing factors against her. Given Biden’s approval rating in June (deeply underwater, in the high 30s) and two straight years of the worst consumer sentiment readings outside of a recession, the Democratic nominee was on track to lose the popular vote by 4 points. She lost by 1.5.

Contrary to the autopsy and Strategist thinking, you don’t need a theory about Harris’s past issue positions, or whether the campaign spent too much on broadcast and not enough on connected TV, or whether Future Forward (a Democratic Super PAC) and the campaign had aligned strategies, to explain the 2024 outcome. The fundamentals already explain it.

I have found that people don’t like giving or hearing this explanation for election results. I recently tried at a meeting with congressional staffers, and I think it left them dejected and/or upset, because it left them with few action items besides “run anti-incumbent candidates when the economy sucks.”

Another reason consultants don’t focus on structural factors more often is that they can’t sell you any services to solve that problem, because there’s nothing you can do about them.

But generally speaking, “inflation was high, and Harris was going to lose anyway” is a much better explanation for 2024 than anything else in isolation. Since we have the track record of this inference, and it has been helpful historically, we can also be confident it will continue to be helpful in predicting future elections. A lot of the other explanations people are giving about 2024 don’t extrapolate well to future elections (”we should align the campaign’s interests with its Super PACs” — what? duh!).

Subscribe now

Incumbents did poorly everywhere in 2024

If you only read the DNC report, you’d think 2024 was a uniquely American failure of Democratic messaging, organizing, and candidate definition. And maybe you don’t buy my “fundamentals” regression of election outcomes. Well, to that person I’d emphasize that 2024 wasn’t just bad for Democrats; it was bad for incumbent parties all across the globe.

As data journalist John Burn-Murdoch documented in the Financial Times, every single governing party in a developed democracy that faced voters in 2024 lost vote share. The Tories were obliterated in the UK. Macron’s coalition lost its majority in France. The LDP lost its majority in Japan. And on and on. Incumbents from Portugal to South Korea either lost outright or were badly weakened.

What did all these countries have in common? Not Kamala Harris’s communications strategy. Not the Democrats’ rural organizing budget. Not whether the party had gone “too far left” on social issues. They had inflation. The post-pandemic price shock hit virtually every developed economy, and voters everywhere punished whoever happened to be holding the gavel when prices went up.

Burn-Murdoch said on Twitter/X the day after Trump won:

We’re going to hear lots of stories about which people, policies and rhetoric are to blame for the Democrats’ defeat.

Some of those stories may even be true!

But an underrated factor is that 2024 was an absolutely horrendous year for incumbents around the world.

This should give the structural theorists more ammo in rebutting the autopsy truthers. When incumbents lose across dozens of countries with different parties, different candidates, different media environments, and different campaign tactics, the parsimonious explanation should be the one variable they share. In 2024, the global post-COVID inflation rate was far beyond what was observed in prior decades.

Most people in DC just don’t care about understanding voter psychology

Party autopsies are basically my Strategist’s Fallacy in its purest form. Tactical choices look enormous when you stare at them up close, but largely miss the point. From inside a campaign war room, every ad decision, every rally location, every reporter interview feels like it could swing the race. And in a truly close election, some of these factors might add up.

But the campaign decisions are small factors dominated by a much more influential set of fundamentals that predict roughly where the final result will land, regardless of who’s running the campaign.

The autopsy’s diagnoses — Democrats didn’t define Trump, didn’t go negative enough, didn’t engage male voters, didn’t show up in rural areas, didn’t invest enough in digital ads, didn’t have a “permanent campaign” strategy — could all be simultaneously true and roughly irrelevant to the 2024 outcome. They might matter at the margins in a future election where the fundamentals are neutral. But they probably didn’t matter much in 2024 because the fundamentals weren’t neutral.

Worse, by attributing the loss to strategic failures, the autopsy invites the party to learn the wrong lessons. If you decide Harris lost because she didn’t run negative enough ads against Trump, you’ll spend 2028 running more negative ads. If you decide she lost because the campaign didn’t have a clear definition of the candidate, you’ll spend 2028 obsessing over the candidate definition. Neither will help if the next Democratic nominee inherits another period of high inflation or low presidential approval. And neither will be necessary if they inherit a recovering economy and a popular incumbent.

The deeper problem with the autopsy is that it imagines a voter who doesn’t exist. The kind of voter the report’s recommendations would persuade — someone weighing Harris’s issue positions against Trump’s, watching campaign ads carefully, updating their beliefs in response to messaging frames — is essentially a Washington consultant, not your grandma who can’t afford to pay her bills because gas is up 50% and electricity subsidies just ended. One of the problems with autopsies is that voter psychology takes a lot of work to understand well, but the people who have that skillset largely aren’t the type of person the DNC is hiring to audit their choices.

Start with what we know about the anti-system voter. A large and growing share of the electorate doesn’t think in terms of “left vs right” or even “Democrat vs Republican.” They think in terms of “the people running things are screwing me, so throw the bums out.” They are not persuaded by an ad explaining that the “opportunity economy” is a better plan than the “Trump economy,” because their objection to the status quo isn’t about specific policies — it’s about the status quo in general. A large portion of voters simply cast their ballot against whoever they perceive as being more a part of “the system” or “the establishment.” And this can be a decisive bloc!

This is also why the hidden axis of American politics — the well-being-oriented, “non-ideological” dimension that runs orthogonal to traditional ideology — keeps catching strategists off guard. Most Washington-based political professionals were trained to think about voters along a single left-right spectrum, and to design messages to “move” voters along it. But a huge fraction of the electorate is non-ideological. They hold a mix of progressive and conservative views that are basically uncorrelated with one another, and they choose candidates based on cues that have little to do with the policy debates that animate party staff. They want change. They want competence. They want leaders to “fight” for them. They do not want to hear about left-right identity politics if the price of a gallon of milk is $6.

And then there is the low-information voter, who decided 2024 more than any other group. The voters who broke hardest for Trump in 2024 were the ones who paid the least attention to politics. These are voters who, in our surveys, cannot name the party in control of Congress, don’t follow the news regularly if at all, and make decisions mostly based on vibes and what their social groups are saying. The DNC autopsy spends pages on messaging strategy aimed at engaged voters and almost no time on the people who actually moved.

This is what strategists miss when they treat campaign choices as the dominant drivers of vote choice. But you can’t ad-target your way out of an inflation problem. You can’t define your opponent if your opponent’s main appeal is that he is not the person currently in charge while gas is $5 a gallon. And you certainly can’t run a 107-day campaign aimed at low-information voters and expect them to suddenly start paying attention to your six-point plan on housing. Whether you “defined your opponent early” has no bearing on the dominant force in politics.

Share

One thing the party could have controlled: a real primary

There is one thing the party could have controlled that the autopsy almost entirely sidesteps: the decision to stick with Biden and not have a primary at all in 2023-2024. Biden’s approval rating had been underwater for nearly three years by the time he dropped out. The forecasting models were warning early that Biden’s odds of winning in November were perilous and trending lower. As we have seen, based on the fundamentals, they may have been even worse.

By letting the party nominate Biden without a primary or convention, party bosses closed themselves off to those paths to November that ended in victory. The party had years, not weeks, to coordinate a graceful handoff to a successor who could have built a real campaign around economic reform, distanced her/himself from the Biden era, and had a fair shot at competing against an incredibly flawed opponent.

The autopsy mentions the truncated timeline for Harris’s campaign in passing, but does not engage with the actual decision-making chain that produced the truncated timeline at all. That omission tells us a lot about the Democratic Party as an institution — about who has power, who exercises it, and who is ultimately (not) being held accountable for 2024.

You can argue that even a hypothetical Democratic nominee who entered the race in 2023 would have lost given the fundamentals. I think that’s probably true! But “we would have lost by less and the party wouldn’t be in this crisis of legitimacy” is a meaningful counterfactual, and it’s the one piece of self-criticism the institution most needed to do. (I might argue that the soul searching about 2024 — the notion of an “autopsy” in general — to explain a 1.5-point defeat is all a bit dramatic anyway, but your mileage may vary.)

Subscribe now

Looking to 2028 — and 2032

If you are the type of person who buys into the autopsy’s Strategist Brain theories of political outcomes, then you are going to dramatically misread the moment in 2026 and 2028.

If Democrats lost in 2022 and 2024 because economic conditions were bad and they were in charge, they should expect to win in 2026 and 2028 if conditions are bad. That’s because they are the out-party now. But if you buy the explanation for 2024 presented in the DNC autopsy, and ignore the structural factors at play, you are less likely to see this as the dominant messaging strategy for your campaign.

Consider that Trump’s approval is already underwater enough to give Democrats the House and maybe the Senate in 2026, and presidency in 2028. Consumer sentiment, after a brief post-election bump, has cratered as tariffs have inflated costs of goods and the war against Iran has caused gas prices to spike. If the trends we’ve seen in general and special elections in 2025 and 2026 hold, we are looking at a substantial Republican defeat in the midterms and a genuinely favorable environment in 2028.

But here’s the catch, and it’s a big one. Those very same fundamentals that will create positive conditions for Democratic wins in 2026 and 2028 will also start working against them the moment they take power. Anti-incumbency in the post-pandemic era has been unusually durable — voters across the developed world have been throwing out whoever is in charge, regardless of ideology, for five straight years now. If Democrats win the House in 2026 and the White House in 2028 on the back of Trump’s economic mismanagement, they will inherit the same trap that swallowed Harris: a public that is structurally pissed off about the cost of living and willing to punish whoever happens to be in charge when the next election rolls around.

We should not expect economic anxiety to simply fall away. A new New York Times/Siena poll fielded May 11–15 finds that 88% of potential Democratic supporters say the American economic system is unfair to most Americans, and 83% say the political and economic system needs major changes (63%) or to be torn down completely (20%). Only 1% want no changes at all.

The only way to escape the gravity of anti-incumbent sentiment is to do something big enough that voters actually feel it, or signal that you’re trying. You don’t need a 12-point plan and to rattle off macroeconomic indicators from the Oval Office; Biden tried that. The next Democratic candidate will need something on the scale of the New Deal or the Great Society — a generational project that reorders the relationship between the American economy and the American worker, and that voters can point to and say “My life is better because of this party.”

Democrats need a president who, in 2029 — like Zohran Mamdani in New York City — says the cost of housing, healthcare, childcare, and energy is the central political question of our time, and who proposes to do something on the scale of that problem. There are lots of options on the table for a party that wants to go big. Build millions of new homes. Cap the price of insulin and expand negotiation to every essential drug. Subsidize childcare or make it a public good. Build out clean energy infrastructure fast enough to combat rising costs from AI infrastructure and climate change. Tax the people who have captured most of the economic gains over the last 20 years and use the money to fund the policies ordinary people actually buy. Pick fights with the industries that profit from scarcity and dysfunction — and win them.

This, of course, will be hard; defying gravity always is. It requires governing the way Roosevelt and Johnson governed, with the assumption that bold action creates its own political coalition. And it also requires you to look back on your failures with a holistic view of what causes certain outcomes.

I have not yet seen anything close to this from the party, from the think tanks, or from the leading 2028 contenders. There are gestures toward the “abundance” agenda, populism, and “radical centrism,” but these mostly read like positioning exercises, not serious agendas for a people that are really hurting. The 2024 autopsy was a chance to start that conversation by being honest about why Harris lost. It failed to do that.

This matters because the fundamentals giveth and the fundamentals taketh away. A Democratic win in 2028 built on Trump backlash, without a generational economic vision behind it, is a Democratic loss in 2032 waiting to happen. The only durable answer to anti-incumbent sentiment is to actually earn re-election. The party has roughly two and a half years to decide whether it wants to think that big — and that creatively.

Next week’s Chart of the Week will dig into a new Strength In Numbers survey with a lot of new data on how voters are actually thinking about the economy right now — what they blame for high prices, who they trust to fix it, and how the two parties’ economic brands have shifted since Trump took office. (Subscribe if you haven’t already!)

Share

I would have liked to see a DNC autopsy with more data, more research, and more attention to the real hard work on voter psychology, economic anxiety, and the path forward. The 2024 autopsy was the party’s chance to confront the thing it actually got wrong. Harris lost because prices were high, Biden was unpopular, and voters across the world were in a mood to punish whoever was in charge. It is surprising to read a hundred pages about messaging discipline and rural organizing and still miss that fundamental fact. And yet.

Democrats are about to be handed a win in 2026–28 by voters who are angry at the party currently in power. If they mistake that for vindication of a new strategy that fiddles around the margins — instead of the same structural anti-incumbency that buried Harris — they will spend the 2030s out of power and wishing they had thought bigger when time was still on their side.

If you got something out of this piece, consider a paid subscription to help fund future articles like this one.

Upgrade to support Strength In Numbers

Related Articles

So this is a short preview post of this month’s Strength In Numbers/Verasight poll release. This full poll will be released on Monday.

Strength In Numbers is a reader-supported newsletter on polling, elections, and democracy. If you want more data-driven independent political journalism, join the community as a paying subscriber for $10/month or $90/year, and get access to weekly exclusive posts and support independent polls like these ones.

Upgrade to support Strength In Numbers

Here is the big finding: Americans overwhelmingly reject the idea of spending taxpayer money on President Trump’s new White House ballroom. Overall, 68% of adults said they opposed spending taxpayer money on security upgrades for the ballroom. A striking 57% oppose it strongly. Just 21% of adults said they support it.

In our May survey — fielded May 18-19, 2026, among 1,520 U.S. adults — we asked respondents about a Republican proposal to put $1 billion in taxpayer funding toward security upgrades for Trump’s new ballroom project, which is being built next door to the White House. Trump originally promised the ballroom would be paid for entirely by private donors.

The more striking result is in the partisan breakdown. Even Republicans don’t like the plan, with 42% in support of the funding and 44% oppose to it.

Among political independents, 61% oppose the funding, including 52% who oppose it strongly. Just 14% are in favor, with the remaining 24% saying they “don’t know.” Unsurprisingly, Democrats are all but unanimous, with 86% strongly opposed.

Share

In our new poll, 51% of Americans say prices or the economy are their number one issue, and President Trump’s net approval rating across both issues is -36, with 66% opposed and 30% in favor.

Americans report high levels of anxiety about both their personal finances and the nation’s economy as a whole. At a moment when voters are already skeptical of Trump’s priorities, asking taxpayers to spend $1 billion on a ballroom project that Trump said private donors would fund is exactly the kind of thing that reinforces voters’ anxieties. Thus, an unusually bipartisan majority of Americans do not want to pay for it.

The May 2026 Strength In Numbers/Verasight poll surveyed 1,520 U.S. adults from May 18-19, 2026. The margin of sampling error is ±2.7%.

Here are a couple thoughts about this poll, and then a big story to dig into with you all:

Here at Strength In Numbers we are, of course, delighted with coverage of polling that serves the unique small-”d” democratic purpose of offering a check on the party in power between elections. As the data below show, this is not a presidency for working-class people, and covering public opinion polls is one way to advocate for the people.

And PS, the NYT poll also happened to land basically exactly on top of our current average at FiftyPlusOne.news. It was obvious to some of us some time ago that Trump was historically unpopular in a way that would hurt his party — even while pundits have been arguing otherwise:

The Democrats also posted an 11-point margin among registered voters in the NYT poll’s “generic ballot” question, which measures support for U.S. House candidates across the country. That matches our January–February Strength In Numbers/Verasight polls. Our May poll comes out next week.

But the result that should really cause you to update your beliefs about this year’s midterms is not in the topline or the Times’ official reporting, but hidden deep in the poll’s crosstabs.

A big story of the 2024 election was that Donald Trump made inroads with non-voters who traditionally leaned Democratic, and even got some of them to turn out for him. But now it’s looking like he has squandered those gains and put his party on the path to midterm defeat in the House — and possibly the Senate.

In this week’s Deep Dive, we look at how the least politically engaged Americans are feeling about Trump, the midterms, and the broader political and economic environment of the United States right now. Are the Americans who put Trump in the White House now set to give his party a historical thumping in November? Let’s boot up the data.

Independents and non-voters have turned on Trump

Read more

In case you haven’t had enough of me in podcast form this week…

1. I chatted with Paul Krugman about the latest redistricting news and Democrats’ still-solid chance of taking the House, as well as my work (in part with David at The Downballot) on the impacts of what I guess is now getting called “price anxiety” on consumer sentiment. Paul and I put out some hypotheticals for what may be causing the “Vibecession 2.0” — the period after January 2025 when our predictions of consumer sentiment, even accounting for excess prices, are higher than they should be (still a lot lower than if you ignored excess prices). That interview, and the transcript, are now up at Paul’s site.

2. I was also interviewed on WNYC’s podcast On The Media about my article on the “fundamental statistical error at the heart of the Supreme Court’s Callais decision” that effectively killed Section 2 of the Voting Rights Act. Link here.

Thanks as always for supporting this Substack, which gives me the institutional freedom to do interviews like these and affect public discourse on polling, voter psychology, and democracy — in all the spare time I have between crunching the numbers and writing these articles! (It’s not a lot of time.)

More next week.

Leave a comment

Upgrade to support Strength In Numbers

Asked how much Americans’ financial situations were motivating him to make a deal with Iran, President Donald Trump told reporters on May 12, 2026: “Not even a little bit. I don’t think about Americans’ financial situation. I don’t think about anybody.” Stopping Iran from getting a nuclear weapon, he added, was “the only thing that matters.”

Statements like these exacerbate the White House’s problems getting the average person to view Trump’s presidency favorably. And they are especially damaging when it comes to prices.

Now think about the broader political context under which this episode happened.

Republicans across half a dozen states spent the last year passing new congressional gerrymanders to give themselves an unfair advantage over Democrats in this November’s midterm elections. After the Supreme Court blew up Section 2 of the Voting Rights Act in Louisiana v. Callais two weeks ago, GOP legislatures in Alabama, Louisiana, Tennessee, and elsewhere have moved to cut even more Democratic seats from the map. Democrats could end up needing to win the popular vote in the House by 4 points or more in order to win the majority of seats.

This is a huge boon to a party that, just three weeks ago, was staring down the barrel of a virtually assured loss of the House and likely loss of the Senate, too.

And yet, it doesn’t change the big picture of how Americans feel about the president today. As I repeated in my podcast episode on Thursday, very few Americans actually like what Trump is doing.

This week’s Chart of the Week is: An update on Trump’s approval rating. Nobody really likes this guy.

Share

How low can he go?

In the average of polls at SIN sister site 50+1, Trump has a roughly 37% approval rating today and a 60% disapproval rating.

Looking at the approval data for the past 18 years, this 37% rating makes Trump worse off than any president at any point since the end of George W. Bush’s second term:

Note from this graph that not only is Trump’s rating at the all-time low for a president since 2009, and lower than that of any president at this point in their term, but it also shows no sign of slowing down. Zoom in and you can see how he is predictably losing support with the American public week after week after week:

Measured in terms of the difference between his approval and disapproval — what we in the biz call a “net approval” rating — Trump is also lower than any modern president at this stage of a term. The previous floor on the chart belonged to Biden, who sat near -21 in mid-2022 — a number which likely cost Democrats the House that year.

The chart below shows the same data as above, but recalculates approval into one number: net approval. The line is green when a president is “above water” (has a positive rating) and turns pink the moment more people disapprove than approve.

It’s the fairest way to stack presidents against each other — and it’s brutal for Trump.

Now, maybe you are thinking that since Trump 2.0 is serving a non-consecutive second term, the graph should reflect that instead of treating him as having a second first term. If you do that, you see it is indeed true that Trump isn’t as low as Richard Nixon or George W. Bush were at the nadir of their presidencies. As I wrote in early April, Trump’s are essentially Watergate-level numbers.

But that is both cold comfort, and it’s worth taking into account that those troughs came deep into their presidencies, after years of accumulated damage and (in Nixon’s case) history-changing presidential scandal. On net approval this early, no one has been lower.

Trump’s problems are only partly about prices

Probably the biggest drag on President Trump’s approval right now is prices/inflation.

Voters blame him for the high cost of living and say (correctly) that his policies have been inflationary and reduced growth. A CNN/SSRS poll fielded April 30–May 4 found 77% say Trump’s policies have raised the cost of living in their own community. This includes a majority of Republicans.

The president’s net approval on inflation has also recently cratered to around -40 — and shows little sign of recovering. A new Economist/YouGov poll put it at 25% approve, 69% disapprove — a net of -44:

Worth noting: These numbers are lower than at any point in the Biden presidency:

And things are probably only going to get worse for Trump as time — and the war in Iran — go on.

The April Consumer Price Index report, released May 12, showed prices up 0.6% on the month, putting annual inflation at 3.8% — the highest since May 2023 and up half a point from March. Energy prices jumped 3.8% in April alone — after rising 10% in March — and accounted for more than 40% of the headline gain. The CPI release says real average hourly wages fell 0.5% on the month, while the price of gasoline was up 28.4% over 12 months

And due to some quirks of how the CPI gasoline data is collected, that is a severe underestimate. According to EIA.gov, the retail price of a gallon of gas was flat through much of 2025 — the national average sat near $3 a gallon all year. Then the Iran war began on February 28, and by early March the average had risen to about $3.15; by mid-March, $3.85; by the end of March, about $4.13. And by mid-May, when I’m typing this, it hit roughly $4.63 — a climb of about $1.50 a gallon or roughly 50% from the late-February low, most of it in the four weeks right after the war started.

The chart below shows Trump’s approval on inflation over time (in blue) and the change in the price of a gallon of gas (orange, inverted) since the start of Trump’s second term (I inverted the gas price chart on David’s recommendation, such that the two lines tell the same directional story). When the blue and orange lines move together and downward, it means gas is getting more expensive and Trump’s approval is getting worse. For most of 2025 the orange line is flat while there is a steady chipping away in the blue line. Then, at the end of February, both lines fall off the table together.

Trump’s inflation approval started falling within a week or two of the price spike. Or at least, that’s when pollsters started picking up the shift.

This is textbook “retrospective economic voting.” Voters punish the incumbent for the economic conditions they’re living through, whether that punishment is fair or not.

Redistricting can’t erase the midterm math

But the price of gas isn’t Trump’s only problem; he was at -25 for his handling of prices and -20 overall even before he started his war in Iran. If the November elections had been held on February 27, the day before the bombing in Iran started, his party would have gotten shellacked.

And here is a good reminder that even via gerrymandering and voter suppression efforts, redistricting and registration purges, you cannot artifically alter your approval rating. People either like you or they don’t. And for most of his political career — certaintly from the vantage point of today, May 15, 2026 — voters simply do not like Donald Trump.

This matters for small-”d” democratic reasons and for electoral reasons. On the democratic side, a president governing with the consent of barely a third of the country has a legitimacy problem that even the most biased map cannot paper over. The whole point of representative government is that those in power answer to the people they represent. When 60% of Americans disapprove of the job Trump is doing, and his party responds not by changing course but by redrawing districts to insulate itself from that disapproval, the feedback loop between voters and their government breaks down. One response to this is to remind ourselves and our leaders that, hey, very few people actually want this!

And if legislators won’t listen now, they will listen later; The electoral arithmetic is unforgiving. Democrats lead the generic ballot by roughly 5 percentage points today, which (1) would very likely be enough for them to win control of the House of Representatives; and (b) will likely grow over the next 6 months (the opposition party tends to gain in the polls). Presidents below 50% approval have presided over an average House loss of 36 to 37 seats. You can run the numbers yourself with our Midterm Seat Loss Calculator.

And that is exactly the context in which the GOP’s gerrymandering should be understood. A 5-to-7-seat redistricting advantage is real, and in a neutral environment, it would matter a lot. Via partisan gerrymandering — legalized cheating — Trump has tried to change how big a wave has to be to flip the House.

But even these serious GOP gains would not be enough to overpower the current caused by a pro-Democratic wave election this November. And on this front, Trump’s catastrophic approval rating is a strong positive indicator for the Democrats.

And it’s likely he will soon be haunted, in television advertisements and internal polling memos, by the replay of those words I excerpted at the beginning of this article.

Donald Trump may not be paying attention to the cost of living for the average American. That much was, frankly, already evident.

But you know who is thinking about their cost of living? The average American! And by and large, they are saying — across polls, questions, geographic questions, and the like — that they do not support what Trump is doing as president. That will cost him, regardless of all the rigging.

Share

Related Articles

Subscribe now

Here are the big takeaways:

• Nobody likes Trump, and gas prices are only part of the story. Trump is sitting at roughly 36.8% approval and 59.7% disapproval in the 50+1 polling average — a net rating of about -23, making him the most unpopular president at this point in his term in modern history, worse even than Joe Biden during the 2022 inflation panic. He’s underwater in roughly 40 states, including big red ones like Georgia, Florida, and Texas, and all the swing states. On prices, his net approval is a brutal -40.
  Trump’s recent collapse tracks almost perfectly with the war in Iran: from late February to now, gas prices are up about 54% — and Trump’s approval on inflation is down by about the same percentage. And don’t count on a bounce-back — even if gas prices fall, Biden’s presidency suggests voters don’t forgive you for price shocks on your watch — and logistically it can take months to ship and refine oil imported from Iran, even if Trump does resolve the crisis.

• Voters are losing confidence in Trump’s mental fitness for the job. Worse for the president, 61% of adults (including 30% of Republicans) say Trump has become erratic with age, and only 45% call him mentally sharp enough for the job (down from 54% pre-2024), and just 32% are extremely or very confident in his mental fitness per the Pew Research Center. Among independents, the “mentally sharp” percentage has cratered from 53% to 36%. Those numbers are still about ten points better than Biden’s in April 2024 — but the trend is not Trump’s friend.

• Voters support term and age limits for federal officials. Roughly 80% of American adults — including 78% of Democrats, 83% of Republicans, and 79% of independents — back a maximum age for House and Senate candidates, a level of cross-partisan agreement that’s almost unheard of on a high-salience issue. Additionally, 61% of independents and 73% of Republicans back a presidential age limit, with 54% picking a cap of 79 or younger; 65% support an 18-year Supreme Court term limit, and 74% favor a maximum age for justices.
  The catch: the 1995 U.S. Term Limits, Inc. v. Thornton decision held that states can’t add to the constitutional qualifications for Congress, likely dooming most age-limit laws — though North Dakota recently passed an age-limit law the Court could use to update the precedent, if challenged. In the meantime, primary voters are increasingly opting for candidates

Since this is the last live show for a while, Elliott has pre-recorded a couple deep dives that will trickle into your feed over the next 4-6 weeks.

Share

If you value this work and want to help keep it going, please consider becoming a premium subscriber to Strength In Numbers. Paid subscriptions support this podcast, the newsletter, and the time it takes to do this kind of data-driven analysis of politics and elections. Paying subscribers also gain the ability to send in questions during our live streams, so you can directly shape the conversations we’re having on the podcast.

Upgrade to support Strength In Numbers

If you missed our video livestream, you can watch it by clicking play on the web version of this post at gelliottmorris.com.

You can also subscribe to us on your favorite podcast app to listen on your own time. And if you do listen via one of those apps, please drop us a five-star rating and review if you feel we’ve earned it — it really helps people discover the show!

A reminder that paid subscribers to Strength In Numbers get to participate in our live Q&A!

You can also read the transcript of our conversation by clicking the headline of this article to take you to the web version of the podcast, then clicking the button just below the byline that looks either like a piece of paper or is labeled “Transcript,” like so:

If you’re a reader of Strength in Numbers and haven’t yet subscribed to David’s newsletter, head to the-downballot.com.

Subscribe to The Downballot

And if you’re coming from David’s site, subscribe here to get the numbers behind the news.

Subscribe to Strength In Numbers

We cover the history of how the U.S. wound up with single-member districts in the first place, why proportional representation would render gerrymandering irrelevant and make multiple parties possible, what PR would actually look like from the voters’ perspective, and how Congress could move to a multi-party system without amending the Constitution.

Here are the big takeaways:

• Single-member districts are the problem, not how we draw them. Drutman argues that no “fair map” standard can resolve the underlying contradictions of the single-winner district, which force an impossible trade-off between partisan proportionality, competitiveness, compactness, communities of interest, and minority representation.

• Proportional representation would make every vote count and give voters real choices. Under an open-list PR system, states divide voters into multi-member districts where parties win seats roughly in proportion to their vote share, making third and fourth parties viable and every district competitive. Lee that PR systems also produce more diverse representation, not less, because party leaders want to put forward lists that appeal to broad constituencies.

• It could happen with one act of Congress — and the moment may be closer than people think. PR for House elections doesn’t require a constitutional amendment; Congress can amend the 1967 Uniform Congressional District Act under its Article I, Section 4 powers, and pair it with fusion voting (already legal in New York and Connecticut) for inherently single-winner offices like the presidency. Drutman thinks 2029 could open a window for a broader “democracy reconstruction” with PR as one piece.

Thanks again to Lee for joining me for this special Deep Dive episode of the show.

Share

If you value this work and want to help keep it going, please consider becoming a premium subscriber to Strength In Numbers. Paid subscriptions support the podcast, the newsletter, and the time it takes to do this kind of analysis well. Paying subscribers also gain the ability to send in questions during our live streams, so you can directly shape the conversations we’re having on the podcast.

Upgrade to support Strength In Numbers

A reminder: Elliott and David Nir record the usual Strength In Numbers podcast live every Thursday at 2:00 PM Eastern. We always discuss a few pre-planned topics and then answer questions submitted live by viewers.

You can also subscribe to us on your favorite podcast app to listen on your own time. And if you do listen via one of those apps, please drop us a five-star rating and review if you feel we’ve earned it — it really helps people discover the show!

A reminder that paid subscribers to Strength In Numbers get to participate in our live Q&A!

You can also read the transcript of our conversation by clicking the headline of this article to take you to the web version of the podcast, then clicking the button just below the byline that looks either like a piece of paper or is labeled “Transcript,” like so:

Upgrade to support Strength In Numbers

One programming/framing note: The numbers in this piece are depressing, and frankly, they’re likely to keep getting worse before they get better. I understand that this can be demoralizing to people, but there’s no way around the data here; putting this down on paper is the only way to be specific about what’s happening, which is required for identifying a fix for the underlying problem. So stick with me.

In 1976, the U.S. House of Representatives had 101 “structurally competitive” congressional districts. By “structurally competitive,” I mean seats that either party had a reasonable chance of winning in an electoral cycle that was perfectly tied nationally — those seats where the Democratic Party’s vote margin was within 5 points of the vote margin for the Demoratic nominee for president in the most recent election.

Last November, in contrast, the number of competitive seats was just 42. Under the new partisan gerrymanders that Republicans and Democrats (but mostly Republicans) have passed for the 2026 midterms, the number falls to a new all-time-low of 33.

That’s right: Just 33 out of 435 — less than 8% — of districts were decided by less than 5 points, in terms of partisan lean, last year.

Put another way, sixty-seven percent of the country’s structurally competitive House seats have disappeared in the last fifty years.

This first chart shows the long slide toward “safe-seat democracy.” Competitive seats were common through the 1970s, 1980s, and early 1990s. They started to vanish after the 1992 election, with another sharp decline after Barack Obama’s election in 2008 and the increasing relationship between white racial identity and voting Republican across the U.S.

The new maps — the enacted mid-decade redraws in California, Florida, Missouri, North Carolina, Ohio, Tennessee, Texas, and Utah, plus the likely Callais-driven changes in Alabama and Louisiana — would cut the count from 42 to 33 if we hold the 2024 presidential terrain constant. Because presidential partisanship was not as intense before the 1970s, when data for this series ends, it is reasonable to assert that the House is the lest competitive it has ever been in the history of the United States.

And Republicans aren’t finished. More on that below.

The unfortunate reality of American democracy today is that legislators are less responsive to the concerns of regular citizens than they perhaps have ever been.

1. The half-century march toward “safe-seat democracy”

Read more

As of today, May 11, 2026, it’s been one year since I launched Strength In Numbers to be the new home for my data-driven writing, political analysis, election forecasting, and independent polling.

When I started this publication last year, my email list numbered just 6,000 readers (though many of those addresses were collected in an earlier, then-dormant version of this newsletter, and I was not hopeful they’d stick around). Along with the personal strife caused by the fallout over Disney shutting down FiveThirtyEight, I was pretty pessimistic about things, with little hope that this would be enough to pay my family’s bills and work as a new career for me.

Oh, how I was wrong. In this piece, I recap the first year of what you all have helped build here at Strength In Numbers, and share a few updates on what’s coming in the next one.

Subscribe now

Reflections on the first year

First, consider the numbers. Thanks to the generous support and engagement of this community, the last year exceeded even my most optimistic projections for this publication. In terms of metrics, SIN today reaches 68,000 inboxes on a semi-daily basis, and even more via news aggregators and social media traffic (which is dwindling, but still pretty strong on Reddit and Bluesky). Growth is a steady 3-6% each month. Since last May, SIN has not had a negative week in terms of total audience, paying readers, or revenue.

Even better than the quant metrics, I think qualitatively what we’re working on is some really cool and necessary political data journalism. Our monthly poll with Verasight is now cited frequently alongside other “traditional” media surveys (actually, newsletters are more “traditional” than cable television!), and I’m building a suite of interactive data projects that help readers understand polls, election forecasts, demographics, and so on.

SIN has also been early and right on the big stories. I caught Trump’s slide into deep unpopularity earlier than most outlets, documented how latent opinion on immigration and ICE turned hard against the administration months before the conventional wisdom caught up, and overall, we made a serious empirical case that the public wasn’t on board with the current administration’s most aggressive moves. Remember that early in Trump’s term, the media was still writing about how Trump had a “mandate,” and his election was evidence of a permanent “realignment” in American politics away from the Democratic Party.

And from a data products standpoint, this publication is heavily differentiated from what you will find elsewhere in political journalism right now (in business speak, we say SIN has a “good moat”). The MRP model for public opinion on tariffs from last August is still some of the most technically ambitious work I’ve published, and the kind of thing that was also hard to get buy-in for in a newsroom environment. The geographic opinion modeling is now a regular feature of coverage, and I’ll be expanding it soon (more on that below). I’ve also been doing this job for so long that I have data and code for pretty much every story hoarded away somewhere on my computer, which means SIN often reacts faster than other outlets to the big stories on elections and democracy.

Finally, the podcast with David has also genuinely surprised me in two ways. First, more people than I expected listen weekly — about 2,000. Second, I expected it to be a chore. It’s become one of the most fun parts of my week. We did our first live show in April, and nearly 100 people signed up for that too. I’m biased, but in terms of content quality, quantity, and diversity, I think what we’re doing here at SIN is pretty close to the Pareto-optimal for an independent political media startup today.

All of this is to say — across multiple metrics, Strength In Numbers is now a healthy business that (again, I’m biased, but IMO) punches far above its weight. I am grateful to all the members of this community, who make it possible for me to focus my attention on this journalism and push the envelope of what’s possible in the media today (you may have heard, it’s pretty ugly out there).

And I owe a particular debt to the paid subscribers who have turned this corner of the internet into a sustainable career for me. Every paid sub makes a big difference to the bottom line and makes more work possible. And that’s not a marketing line, that’s the actual math of how SIN works: for example, data storage and computer server costs are sustained by a couple of dozen paying subs monthly; a bespoke poll costs about 30 annual subscriptions; a week of my time “costs” about 30, etc.

If you’ve ever clicked the upgrade button, gifted a subscription to a colleague, or forwarded a post to a friend: truly, thank you. You directly helped make this possible, and own a portion of its success!

Upgrade to support Strength In Numbers

New products and programming notes

So with that look back, here’s a look forward. There’s a lot of business news to announce, and I didn’t want to send three different posts to your inbox in one day.

Below are two important flags about new and upcoming data projects I’ve been working on, and an important note about my schedule this summer.

First, I’m launching a new data product called Strength In Numbers Pro

I’ve been working on a new polling data platform for a few months that I’m happy to share publicly today. It’s a self-serve data portal for political professionals that distills polling data to the state, congressional district, and local geographic level using a dynamic MRP model, and it’s available starting today at pro.gelliottmorris.com.

SIN Pro is targeted at politics professionals working in data services, campaigns, independent expenditure groups, foundations, corporate political risk teams, advocacy organizations, and really anyone else who needs public opinion data for a wide variety of questions and geographies on demand. If you need up-to-date polling data in a certain state, county, or congressional district, this should be your first stop.

I built SIN Pro with the same methodological rigor and transparency that I bring to everything else here at SIN. The modeling pipeline is fully detailed for subscribers and inspired by both academic research and a decade of my practical experience in polling and forecasting, and built using high-quality data (polls of real people, not AI). There are no black boxes, no vibes, and no proprietary “secret sauce” that covers up errors other pollsters and analysis shops run into. SIN Pro uses the data methods I’d want to see if I were on the other side of the table, paying for the numbers. (Currently, SIN Pro is powered by the same raw data behind our monthly collaboration with Verasight, but for redundancy and diversification of sample, I’m adding more providers to that pipeline soon.)

Pro is a natural extension of the editorial work I do here every day. Strength In Numbers the newsletter is for any reader who wants to better understand U.S. politics using data and hard math. Strength In Numbers Pro is for the people who also want access to that data or deeper consultant-level insights about the state of politics right now. If you work in politics professionally, take a look. We have trend lines of stuff like Trump approval, the generic ballot, party ID, party trust to handle voters’ most important problem, and more at the state and demographic level, and distill it all down to the county level. You can download all these estimates for whatever you need.

The hard pitch to clients is the following: Due to the sophistication of the modeling effort behind the scenes, this is both cheaper and more accurate than trying to do a traditional poll of a small geographic area. The national and state-level trendlines are also more accurate than for most other pollsters, too (other firms jump around a lot more due to weighting problems and non-sampling error). Again, the idea is if you are working on a project about public opinion at the state, CD, or local level, SIN Pro should be your first (and maybe last!) stop; either we can answer your question directly, or give you data to help improve your project at the start (say, identifying the cities that are most anxious about the impacts of Trump’s tariffs).

Initial pricing is $50/month or $500/year, which is what some other political analysis publications charge for delivering a lot less. If you would like to try it out, head on over to pro.gelliottmorris.com. Prorated refunds are available for your first 30 days.

Second, the timeline on my midterm elections forecast

Over at FiftyPlusOne.news, we’re probably on track to publish our 2026 House and Senate forecasting models in early June. The model itself is actually mostly done; the harder problems are (1) solving all the edge cases that appear in the infrastructure that runs things automatically with new inputs, and (2) that the House maps won’t stop moving! Mid-decade redistricting fights in several states mean we’re essentially forecasting a moving target. If it gets to July and maps are still not set, we’ll publish anyway, with appropriate caveats about which districts are in flux. Election forecasters are the real victims of mid-decade gerrymandering.

As with other projects that are a team effort with Mary and Katie, the forecast will formally live at FiftyPlusOne.news and will be free to read — we believe that communicating uncertainty about public polling data is a public service, not something to hide for only readers who are well-off enough to afford yet another media subscription. 50+1 is funded primarily by data licenses, not subscriptions, so we don’t have to put the forecast behind a paywall to make the business math work. (The forecast data feeds will also be available on our API.)

Methodologically, this is a Bayesian “fundamentals plus polls plus ratings” model that “stacks” various predictions on top of each other based on how reliable they have been historically at predicting midterm outcomes. It’s the forecast I wish I’d had the time to build at The Economist and FiveThirtyEight, and I’m confident it’ll hold wel up against (what’s left of) the competition. (My model for 538 beat Polymarket in 2024.)

Third, SIN will be on a reduced schedule next month to accommodate my paternity leave

Burying the lede, I know. My wife and I are expecting our first child in the next three weeks, so I’m taking leave starting the first week of June (or earlier, if the baby has other ideas). This means SIN will be on a reduced schedule starting next month.

Such that paid subscribers aren’t left without the content they’re paying for, I’ve pre-recorded several deep dive podcasts to run next month, plus scheduled a few fun poll articles on perennial topics and current party debates. Because of some other scheduling conflicts, David and I will record our last live podcast for a while this Thursday, May 14. The one exception is my write-up of the monthly Strength In Numbers/Verasight poll, which, as always, will come out in the third week of the month.

I’ll probably still post a chart or two when something genuinely important happens (I’m not great at being offline), but the chart of the week and Sunday roundup will be on break.

2026: More strength and more numbers!

It has been a great first year of running this solo business, especially given the relatively low probability of success at the start. I have a lot of plans for making Strength In Numbers even more valuable to readers over the next six months, and then ahead of 2028 after that.

A year ago, I was nervous that an independent, evidence-first political newsletter wouldn’t find an audience large enough to sustain a second career. You all proved me wrong.

Yet now is no time to rest on our laurels. Today we are living through a confluence of three major threats to polling, political journalism, and democracy: (1) serious public opinion research is getting harder and more expensive; (2) the empirically minded political press is shrinking rapidly, with few outlets providing regular novel coverage; and (3) the demand for clear-eyed, numerate analysis of politics and elections that doesn’t both-sides democracy has never been higher. In taking on these three threats to honest political data journalism, I think what we’re building here really matters.

Together, this publication is showing that, in a political and media environment that rewards noise, partisanship, and lies, there’s still strength in numbers.

Thank you for an extraordinary first year!

Elliott

Upgrade to support Strength In Numbers

Upgrade to support Strength In Numbers

The Democratic Party’s path to a majority in the U.S. House this November just got significantly narrower. On Friday morning, May 8, 2026, the Supreme Court of Virginia invalidated the redistricting referendum voters passed in April that would have given Democrats up to 4 additional seats in the U.S. House of Representatives (though they will likely pick up two anyway in a “blue wave” midterm).

Writing for the majority, Justice D. Arthur Kelsey declared the referendum "tainted" by a technical constitutional violation about when a general election actually happens — specifically, whether it's a single day or the whole multi-day voting process that includes early voting. Virginia's constitution requires the General Assembly to pass a proposed amendment, hold an intervening House of Delegates election, then pass it again. Clear enough, right?

Well, not according to Kelsey. The Commonwealth argued the relevant election was just Election Day (Nov. 4, 2025), so the legislature's Oct. 31 vote cleared the bar by four days. The Court rejected that, holding that an "election" encompasses the combined act of casting and receiving early, absentee, and in-person ballots across the entire voting window. Because early voting had begun Sept. 19, the Court ruled the Oct. 31 vote came in the middle of the intervening election rather than before it, thus invalidating the amendment and the new map.

So now we are left with the following map of the number of seats targeted by Democrats and Republicans (drawn by The Downballot’s Stephen Wolf), in eight states that have new maps in place for 2026.

Judging by these figures, the Democrats are set to lose a net of 8 seats from redistricting this cycle (14R - 6D).

Except, that estimate of R+8 seats is not quite accurate. Despite how much I’ve seen these figures floating around online as the pseudo-official estimate of the number of seats Democrats are losing to redistricting, there are two complications: First, whether a seat was “targeted” does not mean it will necessarily “flip” — several Dem and GOP targets in CA and FL are still unlikely gains for the other party, which I’ll explain momentarily. And second, Democrats are likely to lose at least two more seats due to gerrymandering in Alabama and Louisiana.

So I think it’s time to do some math. How many seats are Democrats out now that their VA gerrymander is off the table? How bad could it get? And, accounting for that, how large a margin do they need in the House popular vote this November to still win the House majority? In other words, can a blue wave “save” them? In this article, I boot up mapmaking software and a House election simulator to find out.

That and more below the paywall. The upshot is that in a short two-week period, the functional end of the Voting Rights Act at the hands of Samuel Alito in his Callais decision and this invalidating of Democrats’ own partisan gerrymander in Virginia leaves the party on the brink of what once seemed like a distant doomsday scenario ahead of the midterms. They will have to make up a deficit of at least half a dozen seats due to aggressive GOP gerrymandering.

How many House seats has GOP gerrymandering cost the Democrats?

Read more

Upgrade to support Strength In Numbers

The six Republican-appointed justices on the United States Supreme Court have found a magical solution to political polarization. All you have to do is take a partisan election result and subtract out the effects of party loyalty on the result.

That, more or less, is what the Court wrote when it invalidated the Voting Rights Act last week. In Louisiana v. Callais, decided 6-3 on April 29, 2026, the conservative majority told voting-rights plaintiffs they must now “control for party affiliation” before their evidence of racial bloc voting will count under Section 2.

That sounds like a neutral statistical fix, but in reality, it’s a bad control — an error called “conditioning on a mediator variable“ that would get your paper sent back to you with lots of red ink in statistics 101. The problem is that in modern America, party isn’t a variable that operates independently of race. Rather, political party is largely downstream of one’s race. If you subtract the effects of political party from the analysis of polarization, you are subtracting away the very evidence of polarization you are trying to study!

This is important (not just a piece for nerds) because Republican legislatures are already moving ahead with new partisan and racial gerrymanders based on SCOTUS’s new theory. Tennessee passed a 9-0 GOP map this week that splits Memphis’s majority-Black and solidly Democratic 9th District into three majority-white, Republican-leaning seats. Mississippi’s governor has called a special session for May 20. Louisiana is losing at least one of its majority-Black districts. And Alabama, Georgia, and South Carolina could be next. (On this week’s podcast, David and I recap these new gerrymandering efforts that are unfolding with unprecedented haste.)

This week’s Chart of the Week is: a simple table (and one causal diagram) that shows how the Court’s new test makes racial polarization vanish on paper, while it is very much still alive in real life.

Share

What the Court decided in Callais

To win a Section 2 vote-dilution case involving single-member districts, minority-group plaintiffs have traditionally had to clear three “preconditions” set by the Court in Thornburg v. Gingles. First, they must show that the minority community is large and compact enough to form a majority in a reasonably configured district. Second, they must show that minority voters are politically cohesive. Third, they must show that the white majority votes as a bloc often enough to defeat the minority group’s preferred candidates. Only then does the court move to the broader “totality of circumstances” inquiry.

Callais rewrites that framework in two places. On the second and third preconditions, Justice Alito says plaintiffs “must control for party affiliation” when proving minority political cohesion and white bloc voting. Failure to do so, the majority says, risks confusing partisan effects for racial ones.

Second, the Court makes it harder for plaintiffs to clear the first precondition — the illustrative-map hurdle. Plaintiffs do not merely have to show that an additional majority-minority district can be drawn in a reasonably configured way. Their proposed map must also satisfy the state’s “legitimate districting objectives,” including the state’s “specified political goals.” If the state’s goals include a target partisan distribution, a specific margin of victory for incumbents, or another constitutionally permissible goal, plaintiffs’ maps must achieve those goals “just as well.”

In other words, where the state asserts partisan advantage as one of its objectives, plaintiffs may now have to produce their own map that preserves the same partisan advantage as the map they are challenging. A state can defend a racially dilutive map as a partisan map, then require plaintiffs to draw an alternative that leaves the partisan outcome intact. As Adam Serwer wrote in The Atlantic last week, Alito and his allies have now given politicians permission to discriminate against voters, as long as they say it’s for partisan purposes. (What other major purpose would there be?)

In Callais, the Supreme Court concluded Louisiana’s second majority-Black district could not survive that test. The plaintiffs that brought the lawsuit over the state’s original racial gerrymander failed at every stage: their illustrative maps did not meet Louisiana’s nonracial goals, including political goals; their bloc-voting evidence did not control for partisan preference; and their totality-of-circumstances evidence did not show an objective likelihood of intentional discrimination.

Samuel Alito should take a stats class

In her dissent, Justice Kagan points out the nonsensical nature of this “updated” VRA. If minority citizens vote mainly for one party and majority citizens mainly for another, she writes, then under the majority’s rule “none of that difference can count” in determining whether minority voters were diluted. Plaintiffs must remove from the equation the very thing they are trying to prove: “polarized voting preferences.”

The majority’s logic assumes that race and party are two separate things that happen to be correlated — and that party can therefore be treated as a confounder. In statistics, a confounder is an outside variable that creates a relationship between two other variables. If party were truly a confounder, then controlling for it might help isolate the independent effect of race on voting behavior.

But that is not the causal story here. One’s political party is not some outside variable exerting pressure on voting in isolation of race. In modern American politics, it is often a mediator — one of the mechanisms through which race is politically consequential. Race helps shape party identification, and then party identification helps shape vote choice. So when the Court tells plaintiffs to “control for party,” it is not asking them to remove statistical noise. Instead, it is asking them to remove one of the main pathways by which racial polarization operates and is able to be measured.

That is the big error at the heart of Callais. The Court treats party as though it explains racial polarization away, when in fact party today is largely how racial polarization shows up in elections — which are, y’know, contests between parties.

This is such a basic error that the natural assumption is that the justices are engaging in bad-faith reasoning wilfully. Party identification is not merely a preference that sits next to race in a regression. And we know that party ID is largely downstream (though in some weird cases, upstream) of voters’ racial identities.

If race shapes party, and party shapes vote, then “controlling for party” is really just “ignoring the effects of race as moderated through party identity.”

The diagram below shows party as a mediating variable for vote — a variable that sits on the causal path between A and C, where A causes B and B causes C. You can see with this graph that “controlling” for political party doesn’t “isolate” the effect of race, it deletes it:

Read the diagram from left to right. In English, this chart shows that race shapes which party people identify with. Party ID then shapes their vote choice. There’s also a small direct effect of race on vote, but it’s swamped by the path that runs through party.

When the Court tells plaintiffs to “control for party affiliation,” it’s telling them to block that big middle arrow. What’s left is the small, not-party-related, direct effect of race on voting.

Strength In Numbers is made possible by paying subscribers. Become a member today.

Read more

Share

Here are the big takeaways:

• The economic vibes may not be “normal” again until 2029 at the earliest — and that’s the optimistic scenario. Consumer sentiment is at its lowest level in the 60+ year history of the University of Michigan’s survey. Our model says the main (but not only) culprit is “excess prices”: Household goods cost roughly 15% more than they would have under the pre-2020 trend of 2% inflation, and people haven’t yet forgotten. If inflation returns to ~2.7%, sentiment recovers to its historical median around April 2029. If inflation stays at 3.5% or higher (as Trump’s tariffs, deportations, and the Iran war suggest it might), sentiment may take decades to recover to pre-COVID levels. An anecdote from a reader in Austria suggests it took residents about eight years to stop complaining about euro prices after the country switched its currency from the schilling, which lines up eerily well with our 2029–2030 projection.

• Republican states are dismantling Black voting districts at breakneck speed after Callais. One week after the Supreme Court gutted the Voting Rights Act, the dominoes are already falling. Louisiana’s governor canceled the state’s primaries by invoking a statute normally deployed after hurricanes; Alabama is passing a law to allow do-over primaries so it can erase the Black-majority district created in 2024; and Tennessee just released a map that surgically splits Memphis into three almost equal pieces, turning one majority-Black district into three Trump+20 white districts. Combined with potential moves elsewhere throughout the South, Republicans could net 13–15 seats from racial gerrymandering alone—and in time for the 2026 elections.

• We’re in a redistricting doom loop, and the only way out is structural electoral reform. America has a severe racial and partisan gerrymandering problem. Once one party abandons fairness, the other has to respond — and it’s a race to the bottom that shafts us, the voters. Elliott coded a computer redistricting simulator to show just how easily a 55-45 state can be turned into 80% one-party representation when partisan maximization replaces fair drawing. The Roberts Court has now ruled four times that partisan gerrymandering isn’t justiciable, and Republicans in Congress have derailed Democratic attempts at a remedy.
  Elliott vouches for a system of proportional representation, arguing that America’s district-based system was built for an era without parties and is no longer fit for purpose.

If you value this work and want to help keep it going, please consider becoming a premium subscriber to Strength In Numbers. Paid subscriptions support this podcast, the newsletter, and the time it takes to do this kind of data-driven analysis of politics and elections. Paying subscribers also gain the ability to send in questions during our live streams, so you can directly shape the conversations we’re having on the podcast.

Upgrade to support Strength In Numbers

If you missed our video livestream, you can watch it by clicking play on the web version of this post at gelliottmorris.com. We record the podcast live every Thursday at 2:00 PM Eastern. We always discuss a few pre-planned topics and then answer questions submitted live by viewers.

You can also subscribe to us on your favorite podcast app to listen on your own time. And if you do listen via one of those apps, please drop us a five-star rating and review if you feel we’ve earned it — it really helps people discover the show!

A reminder that paid subscribers to Strength In Numbers get to participate in our live Q&A!

You can also read the transcript of our conversation by clicking the headline of this article to take you to the web version of the podcast, then clicking the button just below the byline that looks either like a piece of paper or is labeled “Transcript,” like so:

If you’re a reader of Strength in Numbers and haven’t yet subscribed to David’s newsletter, head to the-downballot.com.

Subscribe to The Downballot

And if you’re coming from David’s site, subscribe here to get the numbers behind the news.

Subscribe to Strength In Numbers

Here are the big takeaways:

• What political scientists got right about Trump in 2020 and 2024. Masket readily admits the discipline underestimated Trump’s ability to win the 2016 Republican primary, expecting his celebrity-without-insider-support campaign to flame out the way most do. But once Trump was in office, political scientists were largely correct in warning about election denial, attacks on the press, threats of political violence, and the refusal to commit to a peaceful transfer of power. And after 2020, plenty of pundits wrote that Trump would simply “fade away” — despite primary elections and media coverage showing his iron grip on the GOP.

• The Callais decision is supercharging a redistricting arms race. With the Supreme Court effectively gutting the protections of minority-majority districts, Alabama, Louisiana, Tennessee, and other Republican-controlled states are moving to redraw maps mid-decade. Masket is optimistic and speculates that now that both parties are going to war over new maps, they may try to find some truce in new redistricting guidelines so they aren’t drawing new gerrymanders every two years.
  
  Both Elliott and Masket are increasingly drawn to proportional representation as a structural fix to these problems. A system where politicians can’t draw their own district lines has its own advantages, and having third, fourth, and even regional parties could force coalition-building and blunt the worst effects of partisan sorting. Some sort of system that acknowledges the primacy of parties, instead of denying them wholesale, is probably way healthier for democracy in the long run.

• Media pundits and DC analysts learned the wrong lesson from 2024. Masket surveyed hundreds of Democratic county chairs after the election and found the single most common explanation for Harris’s loss was inflation and anti-incumbent backlash. He thinks that’s roughly right — and notes there is little the party could have done to avoid it. Joe Biden’s stimulus, labor support, and broadly effective economic policy yielded him “roughly nothing in terms of politics.” Elliott’s working theory for what actually moves voters is that highly visible, durable empathy with working-class people can buy an incumbent party that is presiding over economic stress some clout with voters, such as the case of Zohran Mamdani in NYC and Taylor Rehmet in Texas’ 9th state seat district. This is an are both agree the parties and policymakers are having trouble figuring out.

Thanks again to Seth for joining this second Deep Dive version of the show. It was a lot of fun, and I (Elliott) personally learned a lot!

Share

If you value this work and want to help keep it going, please consider becoming a premium subscriber to Strength In Numbers. Paid subscriptions support the podcast, the newsletter, and the time it takes to do this kind of analysis well. Paying subscribers also gain the ability to send in questions during our live streams, so you can directly shape the conversations we’re having on the podcast.

Upgrade to support Strength In Numbers

A reminder: Elliott and David Nir record the usual Strength In Numbers podcast live every Thursday at 2:00 PM Eastern. We always discuss a few pre-planned topics and then answer questions submitted live by viewers.

You can also subscribe to us on your favorite podcast app to listen on your own time. And if you do listen via one of those apps, please drop us a five-star rating and review if you feel we’ve earned it — it really helps people discover the show!

A reminder that paid subscribers to Strength In Numbers get to participate in our live Q&A!

You can also read the transcript of our conversation by clicking the headline of this article to take you to the web version of the podcast, then clicking the button just below the byline that looks either like a piece of paper or is labeled “Transcript,” like so:

But if things are so dire for Trump and the Republicans, why are GOP members of Congress still voting with Trump the vast majority of the time? Wouldn’t we expect them to be more responsive to public opinion in the face of very probable electoral defeat in November? Why are they passing $1 billion funding bills for his White House ballroom, giving Immigration and Customs Enforcement more money, and letting him make war against Iran without authorization?

The simple answer may be that most members fear losing their individual seats in primary elections more than they fear losing power overall — namely, via swing-seat members losing general elections. And we see why this is a legit fear via the primary elections in Ohio and Indiana that were held on Tuesday, May 5, 2026.

First, some context: Last December, 21 Republican state senators in Indiana voted against a mid-decade redistricting plan requested by President Trump that would have redrawn the state’s congressional map to create two additional GOP House seatsahead of the 2026 midterms. The vote was one of the first significant intra-party defeats of Trump’s second term — and in such a red state, the loss came as quite a shock.

Then on Tuesday night in Indiana (as of writing this paragraph at 1:12 AM ET on Wed., May 6), six out of eight of those Republican state senators who voted against the gerrymandered maps lost their primaries — five of them to challengers the president had endorsed, as The Downballot’s Stephen Wolf noted on Bluesky.

Since most state legislative incumbents seldom lose their primaries — the incumbent re-election rate is typically 95% or higher — these losses are notable. They were a direct test of how much sway Donald Trump still holds over the Republican base, and every other elected Republican in the country was watching. In the end, the results were not close.

Republicans are likely to control the Indiana state senate with roughly 40 seats to Democrats’ 10 after November’s elections. With these new six pro-redistricting votes, they will likely pass Trump’s new gerrymander before 2028.

I’ll have more to say in a column later this or next week about the new districting doom-loop playing out across the country. But for now, I want to explain how I think this episode illustrates a fundamental electoral tension for the GOP — one that is hard to impossible for members to escape, and which could lead to big losses in 2026 and 2028.

Safe seats, unstable majorities

I try to keep most of Strength In Numbers free because the news — what the polls say, what just happened in an election, etc — should be widely accessible; public data deserves public commentary. But premium analysis takes a lot of time, so I try to paywall a healthy number of posts for premium subscribers.

The rest of this piece connects the results in Indiana to the broader coalitional math heading into 2026 — including some thinking about how safe-seat Republicans are tanking their party’s future. To read it, become a paying subscriber for $10/ month, or $90/year if you’d rather pay once.

Upgrade to support Strength In Numbers

Read more