Michigan Senate Voting Record: What 2026 Candidates’ Public Roll Calls Reveal

H2: The Michigan Senate Field: A Historical Pattern of Roll-Call Scrutiny

In the last three cycles, Michigan Senate races have been defined by the public voting records of incumbents and challengers alike. The 2018 contest between Debbie Stabenow and John James saw extensive use of Senate roll-call votes in advertising, particularly around health-care and tax legislation. By 2020, the open-seat race between Gary Peters and John James again turned on recorded votes, with Peters’ support for the Affordable Care Act and James’s business record dominating the airwaves. In 2024, the pattern continued: incumbents’ floor votes on infrastructure and inflation measures became central attack lines in both primary and general-election messaging. For the 2026 cycle, Michigan’s Senate candidates enter a field where every public vote is a potential data point for opponents and outside groups. OppIntell currently tracks 342 candidates across four race categories in Michigan, with 110 Republicans, 220 Democrats, and 12 others. Of these, 320 candidates have source-backed profile signals—meaning their public legislative records, campaign statements, or official biographies have been cross-referenced against verified sources. This creates a rich dataset for understanding what roll-call votes may reveal about the 2026 field.

H2: Party Breakdown and the Voting-Record Divide

Historically, Michigan Senate races have featured sharp partisan divides on roll-call votes. In 2018, Stabenow voted with her party on 95% of partisan votes, while James had no legislative record to scrutinize—a gap that became a campaign theme. By 2020, Peters’ voting record showed 93% party-line alignment, while James again lacked a public voting record, forcing his campaign to rely on business and donor disclosures. For 2026, the party mix is unusually broad: 110 Republican candidates, 220 Democrats, and 12 others. Among Republicans, only a handful have held elected office with recorded votes; most are first-time candidates. Democratic candidates include several state legislators and local officials with extensive voting records in Lansing. This asymmetry means that Democratic candidates’ roll-call votes on labor, environmental, and education policy are likely to be amplified by Republican opposition researchers. Conversely, Republican candidates without legislative histories may face attacks on their lack of public service or on positions stated in questionnaires and interviews. OppIntell’s data shows that across Michigan, the average candidate has 1.51 source-backed claims, but the range is wide: well-sourced candidates have five or more claims, while 259 candidates across the national cycle are thinly sourced with zero claims. In Michigan, the top three most-researched candidates—Gary Peters, Mary Waters, and John Paul Torres—each have extensive public records that researchers would examine for voting patterns.

H2: What Public Senate Roll-Call Votes Show About Candidate Priorities

In prior cycles, Senate roll-call analysis focused on a handful of high-profile votes: the Affordable Care Act repeal attempts, the Tax Cuts and Jobs Act, and COVID-19 relief packages. For 2026, the same methodology applies but with updated legislation. Candidates who served in the Michigan Senate or House have recorded votes on issues like the 2023 clean-energy package, the 2024 school-aid budget, and the 2025 auto-insurance reform. These votes are public and searchable through the Michigan Legislative Website and third-party trackers. For candidates who have never held office, their public record may consist of testimony at hearings, letters to agencies, or statements published on campaign websites. OppIntell’s research methodology cross-references these sources to build a profile of each candidate’s stated and recorded positions. Researchers would examine whether a candidate’s voting record aligns with their campaign rhetoric—a classic vulnerability in debates and ads. For example, a Democrat who voted for a tax increase on corporations but now campaigns as a fiscal moderate would face scrutiny. Similarly, a Republican who supported a school-voucher bill but now courts teachers’ unions would have a record gap. The 2026 field in Michigan includes candidates from both parties with significant legislative histories, making roll-call analysis a critical component of any campaign’s opposition research.

H2: Source-Backed Profile Signals: The Foundation for Voting-Record Research

OppIntell’s platform categorizes candidates by the number of source-backed claims in their profiles. In Michigan, 320 of 342 tracked candidates have at least one source-backed claim, meaning a researcher has linked a public document, news article, or official biography to a specific claim. This is above the national average for states with large candidate fields. The remaining 22 candidates have no source-backed claims, which does not mean they are not credible—only that their public profiles are not yet enriched. For journalists and campaigns, this gap represents an opportunity: candidates with thin public records are harder to attack on specific votes, but they also lack a record to defend. In the 2026 Michigan Senate race, the most-researched candidate, Gary Peters, has a long Senate voting record that researchers would analyze for consistency on issues like labor rights, trade, and health care. Mary Waters, a former state representative and current Detroit-based candidate, has a record of votes on city and state legislation. John Paul Torres, a Republican candidate, has a shorter public record but may have local government or business background documents. The source-backed profile signal is not a judgment of quality—it is a measure of how much verifiable information exists. Campaigns that want to preempt attacks can use OppIntell to identify which of their own votes or statements are most likely to be used by opponents.

H2: Competitive Research Methodology: How OppIntell Analyzes Voting Records

OppIntell’s approach to voting-record analysis begins with the same question a campaign researcher would ask: what public votes has this candidate cast, and how do those votes compare to their current platform? The process involves scraping official legislative databases, cross-referencing with news coverage, and flagging votes that deviate from party majority or from the candidate’s own stated positions. For the 2026 Michigan Senate race, researchers would examine votes on key bills from the past four years: the 2021 infrastructure bill, the 2022 CHIPS Act, the 2023 debt-ceiling deal, and the 2024 farm bill. For state legislators, the focus would be on Michigan-specific votes like the 2023 repeal of the right-to-work law, the 2024 expansion of the earned income tax credit, and the 2025 clean-energy standards. OppIntell’s data shows that across the national cycle, 1,526 candidates are cross-platform-verified across FEC, Wikidata, and Ballotpedia—meaning their identities and basic biographical data are confirmed from multiple sources. In Michigan, 27 candidates are cross-platform-verified, providing a reliable baseline for voting-record research. The platform also tracks which candidates have FEC registrations (111 in Michigan) versus state-SoS-only filings (the remainder), which affects how quickly their donor and spending records become available. For voting-record analysis, the key is that public votes are permanent and searchable; OppIntell’s methodology ensures that no candidate’s record is missed due to incomplete scraping or outdated sources.

H2: What the 2026 Field’s Voting-Record Gaps Mean for Campaigns

In past cycles, candidates with thin voting records faced a different kind of scrutiny: opponents would argue that the lack of a record indicated a lack of engagement or a desire to hide positions. In 2018, John James faced repeated questions about his stance on health care because he had no legislative votes to point to. In 2020, the same dynamic played out, with James’s campaign relying on policy papers and endorsements rather than a voting record. For 2026, the field includes many candidates with no elected experience—particularly on the Republican side, where 110 candidates are competing for the nomination. This creates a research gap that OppIntell’s platform is designed to address: by tracking source-backed claims, the platform can show which candidates have any public record at all, and which are effectively blank slates. For Democratic candidates, the risk is that a long voting record provides ammunition for opponents to pick out unpopular votes. For example, a vote against a popular local project or for a controversial tax bill could be used in a primary or general-election ad. OppIntell’s source-backed profile signals allow campaigns to see, at a glance, what an opponent’s research team would find. The 259 thinly-sourced candidates nationally—and the 22 in Michigan with zero source-backed claims—are not immune to research; they simply require more digging into local news, business records, and social media. The voting-record analysis for 2026 is not just about what votes exist, but about what the absence of votes may signal to voters.