MSU Poll Shows Democrats Lead, Republicans Coalesce Behind Rogers

Author: MSU IPPSR / Office for Survey Research | Published: 2026-09-01 | Source: report


Summary

The university’s long-running State of the State Survey (since 1994) puts Abdul El-Sayed ahead of Mike Rogers for Michigan’s open U.S. Senate seat, 48–44 among registered voters and 50–45 among likely voters, with the governor’s race wider (Benson 49–39 RV, 52–43 LV). Fieldwork Aug 10–20, 2026; 1,000 Michigan adults via a YouGov online panel matched to a Michigan adult sampling frame; unweighted bases 1,000 adults, 913 registered, 779 likely voters. No margin of error is stated in the release and the question wording is not published; the full round is embargoed until late February 2027.

The release’s analytic point is generational. El-Sayed runs ahead of the more conventional Democrat Benson among likely voters under 45 but behind her among older voters; Rogers runs ahead of the Republican gubernatorial nominee James, with more success among independents. Director Matt Grossmann’s reading is that El-Sayed is “not really underperforming Benson; he’s just relying on a younger coalition.” Two other findings bear on the race: 37% of respondents say they are less likely to support a Muslim candidate, and Trump’s endorsement is a net negative (55% less likely, 27% more likely).

For the wiki this is the third independent post-primary poll with El-Sayed ahead (after AARP’s 51–50 and Susquehanna’s 46–39), and the first from an academic house. It is also the first to quantify a religion penalty for El-Sayed specifically. Because it is an opt-in online panel, the progressive-overpolling caveat from Wisconsin surprise raises big questions about polls and Democratic voters and I reanalyzed the raw data from Wisconsin's primary polls. Here's what actually went wrong. applies with more force than to the live-interviewer EPIC-MRA survey fielded a week later (EPIC-MRA Statewide Poll of Likely November 2026 Voters (Michigan, Aug 22–28, 2026): frequency report).


Key Claims

  • Senate: El-Sayed 48 / Rogers 44 (RV, n=913); 50 / 45 (LV, n=779). Field Aug 10–20, 2026. YouGov online panel, 1,000 adults.
  • Governor: Benson 49 / James 39 (RV); 52 / 43 (LV).
  • El-Sayed leads Benson among likely voters under 45; trails her among older voters.
  • Rogers outperforms James, especially among independents.
  • 37% say they are less likely to support a candidate who is Muslim.
  • Trump endorsement: 55% less likely, 27% more likely.
  • Crosstabs and wording withheld until end of February 2027.

Direct quotes worth preserving

“Rogers is running ahead of James, with more success among independents. But El-Sayed is not really underperforming Benson; he’s just relying on a younger coalition.” — Matt Grossmann, IPPSR Director

Why citable: the academic pollster naming the coalition structure of a socialist-framed statewide nominee: youth-heavy, not broad, which is exactly the shape that overpolled in August primaries.

“In terms of support, El-Sayed’s Muslim faith may be an issue for many voters, with 37% saying they are less likely to support a candidate who is Muslim.” — IPPSR release

Why citable: the first measured non-ideological penalty on the candidate, separable from the socialism frame.


Connections

Entities mentioned: Abdul El-Sayed, Democratic Party Events: 2026 Michigan Democratic Senate Primary Debates: Does socialism win elections in America?, Why is socialism popular with young Americans?


Contradictions / Tensions

  • Wider El-Sayed lead than AARP (51–50) and EPIC-MRA (47–43 with leaners), narrower than Susquehanna (46–39). Mode differs (online panel vs. live phone); treat as one point in a band of El-Sayed +1 to +7.
  • Grossmann’s “not underperforming Benson” cuts against the Republican frame that El-Sayed is uniquely weak; the numbers do show him 4–5 points behind her on margin.

Notes

Retrieved as the first item of the Sept 4 gap-fill. The corpus still lacks any post-primary poll showing Rogers ahead; the last Rogers lead in the corpus is the pre-primary Fox poll (+4).