Four House Days Left. Cook's Three Bands. Build the Board From Both.

Start with the whip notice, not the mood

On Thursday afternoon I printed three documents and laid them on the same desk. The first was Majority Whip Tom Emmer's notice that the House would not vote the weeks of September 21 and September 28 (Bloomberg Government, September 3, 2026). The second was Charlie Cook's column saying Republicans should brace for outcomes that are bad, really bad, or truly horrible (Charlie Cook Politics, September 3, 2026). The third was the Missouri Supreme Court's unanimous order restoring the state's 2022 congressional map for the November general election (Associated Press, September 3, 2026).

None of those pages is a finished midterm. Each is a dated instrument. The Halberstam habit is to refuse the cable summary and rebuild the week from named sources until the spine holds. That is the job today.

Four days is a number you can audit

After Labor Day the House returns September 14. It leaves September 17. That is four legislative days before Election Day on November 3 (Everett Post, September 3, 2026; Associated Press, September 3, 2026). The canceled fortnights had held eight voting days on Steve Scalise's published calendar (NOTUS, September 3, 2026). Leadership left a narrow callback if the Senate clears a budget resolution for reconciliation (Bloomberg Government, September 3, 2026).

Two days earlier, on Tuesday, Speaker Mike Johnson had told reporters Republicans would work every remaining September day on the calendar (Everett Post, September 3, 2026). By Thursday the continuing resolution through December 11 was already signed law (NBC News, September 2, 2026). The September 30 funding cliff that usually pins a pre election Congress to the Capitol had moved. The schedule change followed the signature. Print that sequence. Do not invent a motive the whip notice does not state.

Cook's bands, brick by brick

Cook's most likely scenario, the one he labels really bad, is a Republican loss of 26 to 35 House seats and four or five Senate seats (Charlie Cook Politics, September 3, 2026). He presents that band as roughly the average postwar midterm penalty for the president's party. Either chamber result, in his telling, ends the governing trifecta.

His merely bad scenario is a House loss of 15 to 25 seats. Still a majority loss on his ledger, but short of the historical average, because so few seats sit in true swing districts after recent maps (Charlie Cook Politics, September 3, 2026). His truly horrible scenario keeps the 26 to 35 House band and lifts the Senate loss to at least six (Charlie Cook Politics, September 3, 2026).

He adds a structural offset: mid decade redistricting has cut Republican exposure by about ten seats relative to the 2024 lines (Charlie Cook Politics, September 3, 2026). That sentence does not cancel the loss bands. It explains why a raw historical average can overstate what this particular map will allow. On the Senate side he notes that only Maine's Susan Collins seat and North Carolina's open seat are Republican held contests outside deep red geography, so larger Democratic gains would have to come from states that usually vote Republican (Charlie Cook Politics, September 3, 2026).

The polling shelf he trusts, and the one he discards

Cook told readers to ignore shoddy state polls and watch eighteen high quality surveys from the last three months: eight New York Times / Siena College polls, seven Fox News polls, and four AARP polls (Charlie Cook Politics, September 3, 2026). His bottom line from that shelf is that Republicans are fighting hard in races that used to look like easy holds.

He cites a Cook Political Report tracker figure of 24 percent approval for President Trump among independents, against 81 percent inside the president's own party (Charlie Cook Politics, September 3, 2026). He cites the University of Michigan Index of Consumer Sentiment among Republicans falling from 96.8 in January to 82 in the latest reading he quotes (Charlie Cook Politics, September 3, 2026). Those are not November ballots. They are pressure gauges on the coalition that has to show up.

Missouri puts one seat back on the 2022 lines

While Washington cut calendar days, Jefferson City cut a map. The Missouri Supreme Court ruled 7 to 0 that the 2025 Trump backed districts cannot be used in November unless voters first approve them at the ballot (Associated Press, September 3, 2026; Reuters, September 3, 2026). The 2022 map stays in force for the general. The new map had been built to move Missouri's House delegation from 6 to 2 Republican toward 7 to 1, principally by redrawing Rep. Emanuel Cleaver's Kansas City seat (UPI, September 4, 2026; POLITICO, September 3, 2026).

Primaries already ran under the 2025 lines. The general will not, unless a federal stay changes the order (POLITICO, September 3, 2026). Attorney General Catherine Hanaway said she will take the case to the U.S. Supreme Court (Reuters, September 3, 2026). That is a live appeal, not a finished federal answer. For the House majority math that US Political Rank tracks on the most competitive House districts of 2026, the practical near term effect is that Missouri's clearest intended Republican pickup from mid decade line drawing is off the November board under current state law.

What belongs on the clipboard for the next sixty days

Four House floor days remain before the election. Cook's published bands run from a 15 seat House loss on the soft end to 35 on the hard end he treats as central, with a six seat Senate wipeout reserved for the worst case (Charlie Cook Politics, September 3, 2026). Missouri's eight House seats will be contested on 2022 lines unless a higher court intervenes (Associated Press, September 3, 2026). New Hampshire still votes for Senate nominees on September 8. The Senate board on our 2026 Senate races ranking still needs a Democratic net of four for majority control because Republicans hold 53 and Vice President JD Vance would break a 50 to 50 tie.

I will not mash those instruments into a single prophecy. I will not ignore any of them. A shortened calendar is a constraint on legislation. A Cook band table is a veteran analyst's published range. A state supreme court order is a map receipt. Stack them. Cite them. Let November 3 own the seats. That is how you inform like crazy without turning the scoreboard into a rooting interest.