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

TL;DR

I estimate a 46.3% chance that Abdul El-Sayed wins, making Mike Rogers the narrow favorite. The national environment favors Democrats, but the latest pre-primary matchup polling and candidate-specific evidence still lean toward Rogers. The first reputable post-primary polls will determine whether this forecast moves above or farther below 50%.

Context

El-Sayed narrowly won the Democratic primary, leading Haley Stevens by 14,893 votes out of more than 1.5 million counted when the Associated Press called the race. Rogers was unopposed for the Republican nomination and previously came within 19,006 votes of winning Michigan’s other Senate seat in 2024. (apps.npr.org)

Every available El-Sayed–Rogers poll predates the primary. Stevens, Mallory McMorrow, Chuck Schumer and the Democratic Senatorial Campaign Committee quickly backed El-Sayed after his victory, reducing the danger of an organized party split but not proving that moderate primary voters have consolidated behind him. (The Atlantic)

Evidence

Michigan’s historical backbone favors Democrats, but the trend does not. The table shows every Michigan Senate election since the last Republican victory; margins are the difference between the Democratic and Republican nominees’ vote shares.

Democrats have therefore won ten consecutive Michigan Senate elections, but their last four margins fell from 13.28 points to 6.50, 1.68 and 0.34 points. The best analogue is 2024: Slotkin beat the same Republican opponent even as Trump carried Michigan, but she was a better-tested general-election candidate than El-Sayed. (en.wikipedia.org)

The full El-Sayed–Rogers polling series tracked by 270toWin is below. The dates are those listed by the aggregator; LV means likely voters and RV means registered voters.

The simple average of the latest three polls is Rogers +4.4, but it gives the Glengariff R+10 result disproportionate influence. My recency-, quality- and sponsorship-adjusted estimate is closer to Rogers +1.6 before nomination effects. I add 1.5 points for Democratic consolidation, then subtract 0.7 points for a modest prior that Michigan polls may again understate Republicans. That leaves the polling component at approximately Rogers +0.8 with a 4.2-point standard deviation, or a 42.4% El-Sayed win probability.

The polling-bias adjustment is deliberately small. The final 2018 RCP average overstated the Democratic margin by 1.8 points; the 2020 average was almost exactly correct; and the 2024 average overstated Slotkin’s margin by about two points. The record supports some caution, not automatically adding two or three points to Rogers. (realclearpolitics.com)

Same-sample comparisons show a candidate penalty for El-Sayed, but its size is unstable. EPIC-MRA had Stevens two points ahead of Rogers while El-Sayed trailed by three, a five-point difference. The late-July Glengariff survey had an eleven-point difference, but Mitchell found El-Sayed tied while Stevens trailed by three. Taking only the two worst comparisons treats a noisy, contradictory signal as settled fact.

The national environment is El-Sayed’s strongest evidence. The Silver Bulletin average placed the generic congressional ballot at Democrats 48.2%, Republicans 41.7%, or D+6.5 on August 5, while Trump stood at 41.5% approval and 54.7% disapproval on August 6. The Michigan EPIC-MRA poll was harsher, giving Trump a 33% positive and 63% negative job rating. Yet that same survey still had Rogers ahead of El-Sayed, which shows why national fundamentals cannot simply be transferred point for point into this race. (natesilver.net)

Money gives Rogers an edge, though not an overwhelming one. Federal filings covering January 1, 2025 through July 15, 2026 show El-Sayed with $14.53 million in nominal receipts, $11.98 million spent and $2.55 million cash, versus Rogers with $10.90 million received, $6.68 million spent and $4.47 million cash. Republican allies reserved $45 million in Michigan advertising, while Senate Majority PAC announced an initial $20 million Democratic reservation. Reservations can be changed, and Democrats are unlikely to abandon a pivotal seat. (senatemajority.com)

The primary provides mixed information. El-Sayed survived more than $60 million in outside spending and drew record recent participation, proving that he has a real organization and fundraising base. But the final five-poll primary average had him ahead by 7.4 points, while his counted lead when the race was called was just under one point. That miss is a warning against assuming his enthusiasm advantage will translate cleanly into November turnout.

My quantitative synthesis uses three lenses. The polling model gives El-Sayed 42.4%; a fundamentals model based on Michigan’s D+4.5 baseline, the 2024 Rogers result and the current national environment gives 55.9%; and a candidate, coalition and resources stress test gives 36.9%. Weighting those components 50%, 35% and 15% produces 46.34%, rounded in the prose to 46.3%.

I fixed that estimate before checking markets. Polymarket showed El-Sayed at 63 cents and Rogers at 40 cents on $246,404 of displayed volume; normalizing the two prices gives about 61.2% for El-Sayed. PredictIt showed a similar roughly 61% midpoint. I do not move to the market price because the venues are correlated, the contract is not especially deep, and the direct candidate polling points the other way. Cook Political Report and Inside Elections both classify Michigan as a toss-up, which is closer to this forecast than the market’s clear Democratic advantage. (polymarket.com)

What's non-obvious

The strongest anti-El-Sayed argument is not simply that he is progressive. It is that EPIC-MRA found Rogers ahead even in a sample that rated Trump 30 points underwater. That is direct evidence that El-Sayed can lose enough split-ticket or undecided voters to cancel a very favorable Democratic environment. But the claim that he is necessarily five to eleven points weaker than another Democrat goes too far, because Mitchell found the opposite in both June and late July. (June Mitchell poll; July Mitchell poll)

The market and much of the bullish case appear to price the Democratic environment and Michigan’s long Senate streak more heavily than candidate-specific polling. The bearish case often makes the reverse mistake by treating one Glengariff poll as the true race. The evidence supports neither a 60% Democratic favorite nor a decisive Rogers advantage. It supports a close race with Rogers narrowly ahead.

Uncertainties

There is no post-primary general-election poll. That is the largest gap. Two or three high-quality surveys conducted after voters have absorbed the nominations could move the estimate by ten percentage points, especially if they report support among independents, Black voters, people over 45 and Stevens primary voters.

The other gaps are the eventual outside-spending balance, whether anti-El-Sayed groups continue into the general election, and whether the national Democratic advantage persists through November. The current reasonable range is 35.0%–58.0% for an El-Sayed victory; most of that width reflects uncertainty about consolidation and correlated polling error rather than ordinary sampling error.