Michigan Senate race
Abdul El-Sayed (D) and Mike Rogers (R) are running for the U.S. Senate seat in Michigan. Across 37 polls, the Frothier average has Abdul El-Sayed at 47.7% and Mike Rogers at 44.8%, which is El-Sayed +2.9. After blending the polls with how Michigan usually votes, the estimate is El-Sayed +3.8. That gives Abdul El-Sayed a 76% chance of winning, and the race is rated Lean D. The seat is currently held by Democrats.
Where the race stands
How the polls have moved
The polling average over the last four months. Above the middle line El-Sayed leads; below it Rogers leads.
Recent polls
Newest first. Polls paid for by a party or a campaign are tagged and count for half in the average.
| Pollster | Ended | Sample | El-Sayed (D) | Rogers (R) | Margin |
|---|---|---|---|---|---|
| Quantus Insights | Oct 6 | 744 LV | 48.4 | 47.2 | El-Sayed +1.2 |
| YouGov | Oct 4 | 3,947 LV | 49.0 | 48.0 | El-Sayed +1.0 |
| Mitchell Research & Communications | Oct 1 | 682 LV | 45.6 | 41.4 | El-Sayed +4.2 |
| Trafalgar GroupR poll | Sep 30 | 1,085 LV | 46.9 | 45.1 | El-Sayed +1.8 |
| Beacon Research/Shaw & Company Research | Sep 28 | 1,028 LV | 50.0 | 49.0 | El-Sayed +1.0 |
| Marist University | Sep 27 | 1,235 RV | 51.0 | 44.0 | El-Sayed +7.0 |
| Guidant Polling and Strategy | Sep 27 | 800 LV | 48.0 | 43.0 | El-Sayed +5.0 |
| Big Data Poll | Sep 24 | 678 LV | 46.7 | 42.1 | El-Sayed +4.6 |
| co/efficient | Sep 23 | 843 LV | 45.0 | 45.0 | Even |
| GBAOD poll | Sep 22 | 800 LV | 48.0 | 44.0 | El-Sayed +4.0 |
| The New York Times/Siena University | Sep 22 | 605 LV | 49.5 | 44.5 | El-Sayed +5.0 |
| Cygnal/Beacon Research | Sep 21 | 600 LV | 45.0 | 40.0 | El-Sayed +5.0 |
| Suffolk University | Sep 20 | 500 LV | 46.8 | 39.6 | El-Sayed +7.2 |
| Torchlight StrategiesR poll | Sep 19 | 643 LV | 43.1 | 44.4 | Rogers +1.3 |
| InsiderAdvantage | Sep 17 | 1,200 LV | 46.6 | 44.7 | El-Sayed +1.9 |
| The Washington Post/George Mason University Schar School of Policy and Government | Sep 14 | 803 LV | 48.0 | 45.0 | El-Sayed +3.0 |
| Emerson College | Sep 14 | 1,000 LV | 48.2 | 45.7 | El-Sayed +2.5 |
| Trafalgar GroupR poll | Sep 9 | 1,079 LV | 46.3 | 44.8 | El-Sayed +1.5 |
| CNN/SSRS | Sep 6 | 843 LV | 47.0 | 44.0 | El-Sayed +3.0 |
| Glengariff Group, Inc. | Sep 3 | 600 LV | 44.4 | 45.8 | Rogers +1.4 |
| Abacus Data | Aug 28 | 332 LV | 47.0 | 42.0 | El-Sayed +5.0 |
| EPIC-MRA | Aug 28 | 600 LV | 47.0 | 43.0 | El-Sayed +4.0 |
| Michigan State University Institute for Public Policy and Social Research/YouGov | Aug 20 | 779 LV | 50.0 | 45.0 | El-Sayed +5.0 |
| Susquehanna Polling & Research, Inc. | Aug 17 | 800 RV | 46.0 | 39.0 | El-Sayed +7.0 |
| Fabrizio Ward/Impact Research | Aug 11 | 877 LV | 48.0 | 47.0 | El-Sayed +1.0 |
How this estimate is made
Every general-election poll goes into one weighted average: newer, larger and likely-voter polls count for more, and no single pollster can flood it. The average is then blended with how the state voted for president in 2024 and today's national generic ballot. Races with many recent polls lean on the polls. A win estimate is a probability, not a call: a candidate with a 70 percent chance loses three times in ten. Full method and sources.
Poll data from The New York Times poll database, CC BY 4.0, via PollResults.org. Averages, ratings and projections are Frothier calculations.