27OCT

Cook moves six seats, all one way

3 min read

13:20UTC

Cook Political Report changed six House ratings on Thursday 3 September and moved every one toward the Democrats. Silver Bulletin's new FLIPR model, launched on Tuesday 11 August, then moved on a court order rather than a poll.

← Back to Senate Democrats borrow $15m, outspend GOP Jump to analysis ↓

PoliticsDeveloping

Key takeaway

Cook shifted six House seats toward Democrats with no move in the other direction.

Cook Political Report published six House rating changes on Thursday 3 September and moved every one of them toward the Democrats.⁠¹ Michigan's 10th and Florida's 14th went to Toss Up. Ohio's 15th and 10th and Colorado's 3rd fell from Solid Republican to Likely Republican, and Michigan's 8th rose from Lean Democratic to Likely Democratic. Nothing moved the other way. Cook rates every House district on a seven-point scale and revises in batches, so a batch with no counter-move is the part worth noticing.

Florida's 14th repays a second look. Kathy Castor is the sitting Democrat there, yet the seat was rated Lean Republican, because a rating describes the district and Florida's 24R-4D map pushed hers rightward. A long-serving Democrat in a Republican-leaning seat now sits in a toss-up. The batch follows a summer in which Sabato's Crystal Ball moved three governor's races in July and the president's approval reached the worst reading of either of his terms.

Silver Bulletin launched its 2026 forecast model on Tuesday 11 August. Forecast with Leading Indicators, Polls and Expert Ratings (FLIPR) runs 40,000 simulations per update and forecasts House, Senate and gubernatorial races jointly, so it can price a Republican Senate alongside a Democratic House rather than treating the two as separate bets.⁠²

Its numbers moved after the Missouri ruling. FLIPR's estimate of Democrats winning House control went from about 81 per cent to about 83, and its average simulated Democratic seat count from 230 to 231. Its estimate of Emanuel Cleaver holding Missouri's 5th District went from 25 per cent to 98.⁠³ A court order rather than an opinion shift produced that jump, so it says nothing about what voters think. Silver Bulletin publishes these as one model's outputs, not as a statement about what happens on 3 November.

Deep Analysis

In plain English

Two different ways of tracking which party is likely to win The House both moved toward the Democrats this week. Cook Political Report, which uses expert judgment, shifted six individual House races in the Democrats' favour and none the other way. Separately, a statistical model called FLIPR, which runs tens of thousands of simulations, jumped its Democratic control estimate from about 81 to 83 per cent, and its estimate that Missouri Congressman Emanuel Cleaver keeps his seat leapt from about 25 to 98 per cent after Missouri's court ruling on the map.

Deep Analysis

Root Causes

The scale of Cleaver's swing traces directly to Missouri's Supreme Court ruling suspending HB 1: the district's competitiveness depends entirely on which map is used, and a simulation model updates that input instantly, while a qualitative service like Cook adjusts ratings on its own review cycle.

That difference in update mechanism, not disagreement over direction, explains why the two outlets' reactions differ so much in size.

What could happen next?

Meaning
Missouri's court ruling had an immediate, quantifiable effect on a single House race's national forecast, showing how much one state's redistricting fight can move overall control estimates.
Risk
If Missouri's map fight reverses again, through the pending SCOTUS stay or the November referendum, FLIPR's Cleaver estimate has no fundamental basis to hold at 98 per cent.

Source Landscape

This story draws on neutral-leaning sources

Primary parallel: FiveThirtyEight's 2018 House forecast produced large single-day swings whenever a court settled a contested map, closest in Pennsylvania's 2018 congressional redraw, because probabilistic models reprice a seat's entire distribution the moment its district lines become fixed, rather than adjusting gradually.

Counter-parallel: Pennsylvania's 2018 redraw was final and litigation-ending; Missouri's map question remains open pending both the SCOTUS stay application and the November Proposition A vote, so FLIPR's current estimate could move again in either direction before the election.

Consensus view: Cook Political Report moving six House ratings toward Democrats with none moving the other way is a directional signal independent of any single model, since Cook's ratings are qualitative analyst judgments rather than simulation output.

Counter-view: Silver Bulletin's FLIPR model, built on 40,000 joint simulations per update, produced a far larger single-event jump, Missouri's 5th District retention estimate moving from about 25 to 98 per cent, than any qualitative rating change could capture, because a model instantly reprices a seat once new district lines become the base case.

Key tension: Whether the size of FLIPR's swing reflects a genuinely large change in the seat's fundamentals, or an artefact of a model recalculating from a low base once a court ruling resolved which map applies.

Sources:Cook Political Report·Silver Bulletin·Silver Bulletin

Mentions:Nate Silver's Silver Bulletin →Missouri →Cook Political Report →Kathy Castor →Florida →Michigan →Pennsylvania →Sabato's Crystal Ball →Colorado →

First Reported In

Update #16 · Missouri suspends its map, and votes on it

Cook Political Report· 7 Sept 2026

Read original →

Causes and effects

Caused by

Missouri map suspended, sent to voters

The Missouri map suspension is what moved FLIPR's House and Missouri-5th estimates.

Occurred 3 Sept 2026

Read story →

Cook moves seven seats, none back

Cook's 3 September moves follow its own 18 June ratings round, extending the same series of Democratic-direction changes.

Occurred 18 Jun 2026

Read story →

This Event

Cook moves six seats, all one way

Six independently rated races shifted in one direction with nothing moving back, which is a stronger signal than any single survey.

Different Perspectives

Missouri voters

Missouri voters rejected Proposition A by 61.83% to 38.17% at 95% reporting, repealing the 2025 map that split Kansas City. Trump carried the state, yet its 2022 lines now stay for future elections too.

Senate Leadership Fund (SLF)

SLF reported about $43.9m of independent expenditures from 27 October to 1 November, almost all against Democrats, including $12.37m against Sherrod Brown. Seven of its ten targets won; Dan Osborn lost in Nebraska and two are uncalled.

Richard Hudson, chair of the National Republican Congressional Committee

Hudson, who runs House Republicans' campaign arm, conceded North Carolina's 9th District to Richard Ojeda before any AP call. Democrats lead 10 of the state's 14 House districts on the unofficial feed.

Department of Justice (DOJ)

DOJ lost before Judge David O. Carter on 28 October and a 9th Circuit panel on 30 October on its demand to set aside 20,000 California ballots, and says it will appeal the Alaska dismissal. No court has set a ballot aside.

Senator Dan S. Sullivan

Sullivan, who trails Mary Peltola 48.2% to 47.9% on first choices, told CNN he would accept the result. DOJ's suit to stop Alaska certifying anyone but him was dismissed on 1 November.

Scott Colom's campaign

Colom's campaign estimates that about 15,000 affidavit ballots remain to be counted in a race he leads by roughly 65 to 100 votes. Those ballots count only if voters confirm their ID or eligibility by 5pm on 10 November.

Written with AI from Lowdown's published reporting. Not individually reviewed by an editor. Editorial standards.