We already know who will win the midterms: expert

What happened

AlterNet’s headline overstates the evidence: no one already knows who will win the 2026 midterms. Henry Olsen of the Ethics and Public Policy Center wrote in The Washington Post that his model, based on generic congressional-ballot polling and the previous three midterms, projected Democrats winning between 225 and 232 of the House’s 435 seats. That is a forecast, not a result.

Polling analyst Kyle Kondik told AlterNet that President Donald Trump’s weak approval rating, economic dissatisfaction, foreign-policy concerns, and the governing party’s usual midterm disadvantage could cost Republicans congressional seats. Separately, Common Cause policy director Susannah Goodman said the Trump administration was trying to assemble state voter records in a federal database; she reported that several states withheld unredacted records and that Common Cause and its partners had intervened in roughly 17 cases.

A projection is not foreknowledge

Olsen’s calculation identifies the political environment confronting Republicans. It does not settle individual races, turnout, candidate quality, district-level variation, or events between the polling period and Election Day. His range also depends on the assumption that recent relationships between generic-ballot polling and House outcomes remain predictive.

AlterNet converts that conditional analysis into certainty because certainty attracts attention. The underlying reporting supports a narrower conclusion: Democrats appeared favored to win a House majority under Olsen’s model. Declaring the winner in advance replaces analysis with salesmanship.

Power sits with the administration

The most consequential actor in the story is not a polling expert or an opposition voter. It is the federal administration seeking state voter-roll data, including personal identifying information. States control election administration, but the presidency commands agencies, lawyers, enforcement authority, and a national platform capable of pressuring those states.

Goodman described resistance through refusals and legal intervention. Those defenses matter, but they do not equal the executive branch’s institutional reach. The relevant power relationship is an administration demanding sensitive election data and states and civic groups attempting to limit that demand.

The article outruns its evidence

AlterNet repeatedly describes Trump as trying to “rig,” “steal,” or manipulate the elections by obtaining voter rolls and purging likely Democratic voters. But the supplied reporting does not establish that claimed purpose. Goodman documents an effort to centralize voter data and resistance to sharing unredacted personal information; she does not, in the quoted material, provide evidence of an implemented partisan purge.

That distinction is not a concession to euphemism. It is what separates a documented abuse of power from an allegation about intended abuse. The administration’s demand warrants scrutiny on privacy, legal authority, voter eligibility, and potential disenfranchisement. AlterNet weakens that scrutiny when it presents its inference as an already proven plan.

The opposition is not the governing power

The story’s clearest misdirection is subtler: it casts Democrats’ “primary challenge” as warding off Trump, as though protecting election administration were principally a campaign obligation. Democratic officials, litigants, and advocacy groups can resist. But lawful limits must be enforced by courts, state election officials, legislatures, inspectors general, and career administrators.

When institutions shift that burden onto the opposition party, executive overreach becomes another electoral tactic to counter rather than an exercise of state power to investigate and constrain. The weaker actor is told to defend the system while the stronger actor’s authority remains the background condition.

Forecasts cannot substitute for safeguards

This episode exposes two failures that feed each other. Political media treats a polling model as a verdict, then treats an allegation of election interference as established fact. One manufactures confidence about the outcome; the other blurs precisely what government officials have done, what they may be positioned to do, and what remains unproven.

The systemic issue is not whether one pundit can call the House correctly. It is whether institutions controlling election data are subject to enforceable limits—and whether reporting can describe those power struggles without turning prediction into knowledge or suspicion into fact.

Thanks to Alter Net

Source: We already know who will win the midterms: expert