Will Abdul El-Sayed win the 2026 US Senate election in Michigan?

TL;DR

I put Abdul El-Sayed at 46.8% to win, which makes Mike Rogers the narrow favorite. Direct polling leans Republican, while the national climate and Michigan’s long Democratic Senate streak lean Democratic. The first credible post-primary polls will move this forecast more than any other single fact.

Context

El-Sayed won the Democratic primary 48.5% to 47.5%. When the race was called, he led Haley Stevens by 14,893 votes out of more than 1.5 million counted. Stevens conceded and endorsed him, while Mallory McMorrow and Democratic leaders quickly backed the nominee. Every available general-election poll was conducted before the primary, so none measures that consolidation directly (Associated Press; The Atlantic).

Rogers is not a generic challenger. He was the Republican nominee two years ago and lost to Elissa Slotkin by only 0.34 percentage points, even as Donald Trump carried Michigan by 1.4 points. That makes 2024 the best reference case: Rogers can come within a fraction of winning, but he also failed in a more Republican national election (2024 Senate result; certified Michigan results). (en.wikipedia.org)

Evidence

Michigan’s historical base rate is Democratic, but the margins have collapsed. Democrats have won every Michigan Senate election since 1996, across both Senate classes. The full streak is:

The 10–0 record is a real Republican hurdle. The last three results—D+6.5, D+1.7 and D+0.3—also show why older landslides deserve little weight. Michigan is now a close state, not a reliably blue one.

The complete public 2026 head-to-head record contains 11 surveys. Results are listed by release date; all used likely-voter samples according to the 270toWin compilation.

The latest-three average is Rogers +4.4, but it gives full weight to Glengariff’s R+10 result. My recency-, sample- and quality-weighted calculation shrinks that outlier rather than deleting it and discounts the pro-El-Sayed Zenith survey. It produces a pre-consolidation center near Rogers +1.5. I add about one point for Democratic consolidation, then subtract a few tenths for Rogers’s money advantage and El-Sayed’s exposure to a new general-election definition campaign. The polling path centers near Rogers +1.0. (en.wikipedia.org)

The fundamentals path is better for El-Sayed. Silver Bulletin’s August 4 national average had Democrats ahead 48.2% to 41.7%, or D+6.5, while Trump stood at 41.5% approval and 54.7% disapproval (generic ballot; Trump approval). Michigan’s June unemployment rate was a seasonally adjusted 5.0%, versus 4.2% nationally, and the state labor force declined during the month (Michigan labor data, published July 16). These conditions favor the party opposing the White House.

El-Sayed gives much of that advantage back. In the latest EPIC-MRA poll, Stevens led Rogers by two while El-Sayed trailed by three—a five-point relative penalty. Glengariff showed an eleven-point penalty. Mitchell showed the opposite, with El-Sayed running three points better than Stevens. I use an expected El-Sayed penalty of roughly 4.7 points relative to Slotkin or a conventional Democrat, with wide uncertainty (EPIC-MRA; Glengariff; Mitchell). Starting from Slotkin’s 2024 result, applying only part of the national swing, then subtracting that candidate penalty gives a fundamentals center near El-Sayed +0.5. (en.wikipedia.org)

Campaign finance gives Rogers a modest edge, not a decisive one. El-Sayed raised $4.57 million during the second quarter and had about $2.7 million on hand on June 30. Rogers raised $2.9 million and had $5.66 million on hand (Detroit News reporting on FEC filings). The Senate Leadership Fund reserved $45 million for Michigan. AIPAC-linked groups spent more than $30 million in the primary and signaled continued opposition to El-Sayed (Associated Press). El-Sayed’s ability to survive more than $60 million in hostile or pro-Stevens primary spending shows resilience, but many general-election voters have not yet seen the attacks.

I do not apply a large mechanical Republican polling correction. The 2020 RCP average showed Peters ahead by 1.7 points, almost exactly his 1.68-point victory. The 2024 polling average showed Slotkin ahead by 3.2, overstating her final 0.34-point margin by about 2.9. The evidence supports extra uncertainty and a small Republican adjustment, not an automatic two- or three-point correction.

My independent model puts 65% weight on the polling path and 35% on fundamentals. It initially gives El-Sayed 45.7%. I model the final Democratic-minus-Republican margin with a calibrated mean of −0.35 points and a standard deviation of 4.4 points, which gives P(El-Sayed wins) = Φ(−0.35 / 4.4) = 46.8%. The central expected result is therefore Rogers by about 0.4 points, with a 90% interval running from roughly Rogers +7.6 to El-Sayed +6.9.

Only after producing the independent estimate did I check markets and public models. The Economist had El-Sayed at about 48%. Cook, Inside Elections and Sabato’s Crystal Ball rated the race a toss-up. Kalshi’s top-two candidate prices normalized to about 57.8% for El-Sayed on roughly $741,000 in volume (Kalshi), while Polymarket showed about 72% on roughly $85,000 in volume (Polymarket). The liquid Kalshi signal moved me upward slightly. The thinner and much more bullish Polymarket price did not.

What's non-obvious

The D+6.5 national environment cannot simply be added to El-Sayed’s state polling. Those Michigan polls were taken in the same political climate. EPIC-MRA sampled an electorate that identified 43% Democratic or Democratic-leaning and 40% Republican or Republican-leaning; it also found Trump at 33% positive and 63% negative. Rogers still led El-Sayed 46% to 43% (EPIC-MRA). Mitchell found a D+6 Michigan congressional generic ballot but a tied Senate race (Mitchell). Two different polling methods therefore find the same thing: El-Sayed runs behind the broader Democratic ticket.

The opposite correction also matters. Mitchell found El-Sayed winning only 85% of Democrats while Rogers won 91% of Republicans. Moving El-Sayed from 85% to 90% among Democrats would add roughly two statewide points in that sample. Stevens’s immediate endorsement makes some recovery likely. El-Sayed also survived a huge spending disadvantage, which shows his coalition is more durable than the label progressive insurgent suggests. The close primary result still cuts against claims that youth and irregular-voter turnout will automatically overwhelm Rogers.

Uncertainties

• No poll has measured the confirmed nominees. The latest fieldwork ended on July 31, before the August 4 primary, and the newest polls still had about 10%–12% undecided (EPIC-MRA; Mitchell).
• El-Sayed’s support among Stevens voters, Black Democrats, older voters and independents remains weakly measured. His primary coalition performed better in college-oriented and affluent communities than in some working-class and Black areas (The Atlantic).
• The scale of Democratic outside spending is not yet clear. Republican groups have a large reservation, and AIPAC has said its opposition will continue, but the fall spending balance is unknown (Associated Press).
• Republican messaging is already using El-Sayed’s full Arabic name and insinuations about extremist ties. The effect of ideological attacks, anti-Muslim prejudice and backlash against those attacks is real but hard to estimate from existing polling (The Atlantic).

The evidence supports a toss-up with Rogers slightly ahead: 46.8% probability that Abdul El-Sayed is elected.