Matt Grossmann

Matt Grossmann

@mattgrossmann.bsky.social

Michigan State political scientist & IPPSR Director; Hooked bookstore/cafe Co-owner; Science of Politics Podcast; New book: Polarized by Degrees

12.5K followers171 following1.8K posts

Turnout rises quickly with age, especially in midterms, & survey respondents overreport their confirmed participation. In 2026, young respondents are both more enthusiastic & more Democratic than usual in many important elections. Hard for pollsters; overperformance key for Dems

Lot of fear of people lying to polls. We don't have much evidence for it & don't need it to explain error. People uncomfortable sharing their choice are more likely to be in the now ~98% of people who don't respond than to take ~20 min survey to lie. Non-response much bigger deal

The 2026 electorate is likely to be older & whiter than 2024 (as is normal in midterms), meaning some Dem gains will be in lower-turnout groups. But 2026 is also expected to be more educated & more Democratic (the education gap is high but not expected to grow or decline)

Generic ballot polls currently imply a gain in the Democrats' 2-party vote share (from the prior election) slightly above the post-Watergate election & not seen since 1948 (& comparable to 1966 & 1994 for Republicans, but not 2010)

In addition to observing what is changing, we should also attend to what we thought was changing that is not. A few candidates: 1) Republican foreign policy was too hawkish & is facing backlash 2) Voters care about inflation, especially gas prices; nothing specific to Biden

3) Voters move in parallel against the party in power; presidents don't keep all their partisans 4) Republicans did not change their economic agenda to reflect downscale constituents (& thus don't get credit for doing so) 5) Parties out of power can be leaderless & fine

6) culture war salience is on long-term upward trend, but voters & candidates can shift back to economics 7) backlash means party out of power both wins back some voter groups & extends their gains with others 8) Party turnout differential & swing voters move in same direction

Polls point to a clear blue wave. The chance of systematic polling error favoring the Democrats is higher than the chance that (a) anything shifts before the election or that (b) key races will move in different unexpected directions. Either the polling is off or a wave is coming

Matt Grossmann reposted

From my just-released article in @pspolisci.bsky.social

Matt Grossmann@mattgrossmann.bsky.social

special elections are predictive of general election outcomes, particularly during midterm election cycles and in competitive districts, a good sign for Democrats in 2026 www.cambridge.org/core/service...

We are starting a Policy Entrepreneurship program (as a grad student working group to start) next semester. See below for the outline. If you have done anything similar and have advice, please share.

There is likely to be a reinventing government plan in preparation for the next Democratic administration; it may entail various efforts to improve bureaucracy in addition to repealing Trump efforts: www.vox.com/politics/505... But this didn't happen much under Biden

vox.comThe radical Democratic plan to change the government for goodThe party is finally mounting an aggressive response to Trump’s slash-and-burn second term. Could their ideas for a rebooted, reimagined Washington actually work?

There is no disagreement among parties or interest groups that El-Sayed is underperforming other Dems or that Cooper & Peltola are overperforming them, just as there is no disagreement that Paxton is underperforming and Collins is overperforming. Trust where the sides spend $ over what pundits say.

Data center expansion & the AI backlash Data centers are built in places with less capacity to object. Will backlash start feasible AI regulation or empower shifting harms to others? New #ScienceOfPolitics podcast with Vasu Kaul & Daniel Schiff www.niskanencenter.org/data-center-...

niskanencenter.orgData center expansion and the AI backlash - Niskanen CenterBetter regulation is possible but does run up against a traditional political constraint: that the powerful tend to offload harms to those without the power to object.

In the unlikely event that Dems won TX, IA, & AK but lost MI & ME, it could hurt further left candidates leading up to 2028 as a consistent electability narrative emerged. But most other scenarios would likely be interpretable by both sides of internal Democratic Party feuds

Some argued US backsliding could be addressed with legal penalties against Trump 1, citing global examples. But targeting Trump may have aided his comeback. Now Brazil & France may add evidence that backsliding parties gain from prosecution. Who has updated this analysis?

argument that AI will spread expert consensus & aid technocracy: www.conspicuouscognition.com/p/how-ai-wil... Reasonable (especially for intra-elite debate). But increasing role of experts & educated in governance is a longstanding trend (that this continues, not reverses); & still provokes backlash

conspicuouscognition.comHow AI Will Reshape Public OpinionSocial media democratised public opinion, shifting influence away from elites and experts to ordinary people. LLMs will partly reverse this trend. They are a powerful, new technocratising force.

All MSU programs also meet the earnings threshold, though there is again wide variation in earnings. The nationwide averages by major show similar rankings, with political science 26/61 widely offered majors.

Matt Grossmann reposted

NEW: In March the White House released an FY27 budget request that zeroed out the National Science Foundation's directorate of Social, Behavioral and Economic Sciences (SBE). For months it was unclear what would happen. Now, with FY26 over, the damage to SBE is coming to light. My reporting:

science.org‘Complete devastation’: How NSF gutted its social sciences researchAgency body that funds sociology, linguistics, and archaeology studies in U.S. gave 90% fewer new grants this year

Like national race forecasters, analysts of state legislative races are relatively timid in seeing opportunities for wave election pickups. But if the generic ballot shifts 12 points, with both differential partisan turnout & a large swing, it suggests many unexpected pickups

Eg a uniform 12 point margin shift would produce a 63–47 D House & 23–15 Senate in Michigan & a D 53-46 Assembly & D 18-15 Senate in Wisconsin. Even in gerrymandered NC, Dems would win House 64-56 & lose Senate 26-24

The shift in the generic ballot against the president's party continues. Although consistent with the historic pace & direction, this should be close to the largest 2-year shift in generic ballot polling on record; see below for the averages since the 30s www.natesilver.net/p/generic-ba...

Matt Grossmann@mattgrossmann.bsky.social

Majority parties of the president lose generic ballot support as midterms approach, especially at the end. It responds to falling prez approval & thermostatic public opinion. Backlash to the Iran War & immigration crackdowns will help Dems, but are consistent with normal cycle

Since 2000, the average college course description has more political content & more liberal slant. Changes are modest relative to persistent cross-field differences; student demand for liberal content rises over the 2010s & peaks in 2020 papers.ssrn.com/sol3/papers....

papers.ssrn.comThe Political Content of College CoursesWe measure two dimensions of politics in college courses-the amount of political content and its partisan orientation-using course descriptions from nearly 1,00

I enjoyed Blockers, the new Michael Lewis book on DOGE. Again, the reporting at the time was mostly correct: it was haphazard & inexperienced efforts that degraded government without achieving savings. But most effects are long-term & harder to see. hookedlansing.com/book/9781324...

hookedlansing.comBlockers: Rebels in the Deep StateAn inside look at the early days of Trump’s Department of Government Efficiency, as told by the public servants it targeted and one of America’s best storytellers. Blockers. It wasn’t a compliment. In...