Tools chosen: polymarket_direct polymarket_related kalshi_related claude_news gdelt_news wikipedia code_execution

Sub-questions (Fermi decomposition)

Planner reasoning

This is a bucketed seat-count question (GOP 220-224) whose probability depends on the shape of the seat distribution centered on a likely Democratic-leaning midterm environment plus 2025-26 redistricting changes. The Polymarket price for this bucket and its sibling buckets is the primary anchor; Kalshi's control-of-House market and generic-ballot data provide cross-checks. I'll pull market prices, news on redistricting/forecast models, and use code to convert a mean/SD seat distribution into bucket probabilities.

# Current state The House is currently 220R–213D with 2 vacancies. This market asks whether the FINAL 2026 midterm seat count lands specifically in the 220–224 range — a narrow band requiring Republicans to roughly hold their current position despite a Democratic-leaning environment. Polymarket prices this exact bucket at just 8.2%, reflecting that most forecasters see the plausible outcome distribution centered either below 218 (Dem takeover) or, less likely, comfortably above 225. # Timeline of key events - 2025-07: Texas passes GOP-favored congressional map at Trump's request, launching the mid-decade redistricting wave (confirmed, Wikipedia). - 2025-late/2026-early: Missouri and North Carolina pass new GOP-favorable maps; California passes Prop 50 (Dem-favorable map) in response; Ohio required to redraw (no bipartisan support under state constitution); Utah Supreme Court strikes down existing map as unlawful gerrymander (confirmed, Wikipedia). - 2026 (date unspecified): Virginia passes a Dem-favorable redistricting amendment, subsequently invalidated by Virginia Supreme Court (confirmed, Wikipedia). - 2026: Louisiana v. Callais Supreme Court decision limits VRA-based map challenges, prompting more Southern states (Florida) to consider redistricting; Florida calls special session (confirmed, Wikipedia). - 2026-03: Everything Policy seat-by-seat analysis projects Democrats gaining ~10 seats, calling the race "very close" (reported). - 2026-04 to 06-30: Cook, Sabato's Crystal Ball, and Inside Elections all move ratings toward Democrats with no reversals back to Republicans (reported, Crystal Ball). - 2026-07-07/08: Maryland lawmakers announce August special session on redistricting targeting the state's lone GOP seat (confirmed, multiple outlets). - 2026-07-16 to 08-12: Generic ballot polling shows consistent Democratic lead (RCP: D+6.8; Silver Bulletin: D+8.1 likely-voter adjusted); Trump approval among independents and overall near record lows (reported, multiple polls). - 2026-07-27: Kondik/Rasmussen analysis argues 2026 map bias is comparable to 2018 (Dem wave year), meaning redistricting gains alone likely insufficient to save GOP majority (reported). - 2026-08-04: Maryland Democrats pass redistricting amendment that could flip the state's remaining GOP seat if approved by voters (confirmed, NY Post). - 2026-08-13: Cook Political Report moves CO-03, OH-10, OH-15 from Safe to Likely Republican — modest late GOP stabilization in a few districts (reported). # Event Will Republicans hold between 220 and 224 House seats (inclusive) after the 2026 midterms, as seated in the 120th Congress? # Outcomes to forecast - Yes (final GOP seat count 220–224) - No (final GOP seat count outside 220–224) # Kalshi market anchor No direct Kalshi price returned for this ticker in raw research; Polymarket is the primary quantitative anchor. Polymarket prices this exact bucket at **8.2%** YES, up +2.15% over 7 days but down -2.45% over 30 days (range over 116 days: 0.3%–19.5%; volume $59,267). No Kalshi-specific "control of House" market found via keyword search, though adjacent Kalshi markets (2028 House control, next Speaker) exist but are not directly comparable. # Sub-question answers 1. **Polymarket adjacent buckets** — Not returned in raw research (no bucket breakdown data for 215-219, 225-229, etc.); only the 220-224 bucket price (8.2%) is available. 2. **Kalshi/Polymarket House control pricing** — No direct Kalshi "GOP retains House" market was surfaced; Kalshi's 2028 House control market shows Dem favored (64%), suggesting general Dem-leaning sentiment but not directly informative for 2026 seat count. 3. **Generic ballot** — RCP average: Dem+6.8 (48.5%D–41.7%R); Silver Bulletin likely-voter adjusted: Dem+8.1 as of 2026-08-11. Historically, a mid-to-high single-digit generic ballot lead for the out-party has correlated with double-digit seat losses for the president's party (claude_news synthesis). 4. **Redistricting net effect** — Sabato's Crystal Ball estimates net R+5 to +9 (expected R+6) from Texas/Missouri/NC/Ohio gains offset by California/Virginia Dem-favorable maps; Kondik argues net map bias resembles 2018, insufficient to offset a wave (Wikipedia, claude_news). 5. **Historical base rate for 5-seat windows** — Code execution analysis of 1946-2022 midterms: a 5-seat window centered on the eventual outcome has never been hit historically (0/20); even a 10-seat window succeeds only 15% of the time. Seat swings average -25.6 (SD 21.6) for the president's party. 6. **Expert forecasts** — Cook Political Report and Sabato's Crystal Ball both initially favored a narrow Democratic takeover (GOP falls below 218), with no reversion toward Republicans through June 2026; Everything Policy (March 2026) projects Democrats gaining ~10 seats. # Key facts (high-confidence, factual) 1. [270toWin/Center Square] Current House: 220R–213D, 2 vacancies. 2. [RCP, Silver Bulletin] Generic ballot: Dem+6.8 to Dem+8.1 (Aug 2026). 3. [Sabato's Crystal Ball] Redistricting net effect estimated at R+5 to +9 (expected +6). 4. [Cook/Sabato, via claude_news] Both major handicappers' initial 2026 ratings favored a narrow Dem takeover. 5. [Polymarket] 220-224 bucket priced at 8.2%, trending down over 30 days. # Cross-market signals - Kalshi related: No direct House-control-2026 market found; 2028 House control market favors Dems (64%). - Polymarket: 220-224 bucket = 8.2%, implying most probability mass is elsewhere (likely below 218 given Dem-favorable fundamentals). - Sportsbook implied: None found. # Analyst opinions and speculation - Kondik (Rasmussen, 2026-07-27): 2026 map bias ≈ 2018, redistricting gains alone won't save GOP majority. - Cook/Sabato consensus (through June 2026): steady drift toward Democrats, no reversals. - Cook (Aug 13, 2026): minor late GOP stabilization in 3 districts (CO-03, OH-10, OH-15). # Directional lean per outcome - **Yes (220-224)**: Supported by redistricting gains (+5-9) partially offsetting Dem wave; opposed by strong generic ballot deficit (D+6.8-8.1), historical base rate showing narrow windows are rarely hit (0% historically for 5-seat bands), and consensus forecaster view that GOP falls below 218. - **No**: Favored by generic ballot, historical swing magnitude (mean -25.6 seats for president's party), and forecaster consensus pointing to GOP majority loss (or, less likely, a bigger-than-224 GOP number if redistricting dominates) — mass likely split between <218 and possibly >225 scenarios, squeezing the 220-224 band. # Gaps / unknowns - No bucket-by-bucket Polymarket distribution (215-219, 225-229, etc.) to confirm modal outcome location. - No direct Kalshi price for this specific ticker was returned. - Outcome of Maryland's redistricting amendment (voter approval) unresolved, affecting final net seat math. - Numerous toss-up redistricted seats remain uncalled, adding variance. # Calibration anchors - Polymarket YES price: 8.2% (anchor). - Historical base rate: 5-seat window hit rate ≈0% (1946-2022); 10-seat window ≈15%. - Code-execution Bayesian estimate: under favorable (tight SD, mean~218-220) conditions, ~18-19%; under realistic wider uncertainty, ~9-12%.

1. Both forecasters converge to nearly identical numbers (9.0%) essentially by anchoring on Polymarket, but neither addresses the "Gaps/unknowns" caveat that Polymarket's 8.2% is a single-bucket price with no visible adjacent-bucket data — meaning they cannot confirm the 220-224 band is actually the local mode-adjacent trough rather than mispriced, so their confidence in matching the exact anchor may be misplaced. 2. Neither forecast grapples with the redistricting-driven bimodality risk: since Cook noted late GOP stabilization in 3 districts (Aug 13) and net map effects are R+5 to +9, there's a plausible scenario where Republicans land *above* 224 (comfortable majority) rather than just below 218 — both rationales treat the "No" outcome as if it's overwhelmingly a Dem-takeover scenario, understating the probability mass that could sit above 225. 3. The calibration anchor section explicitly gives a range of Bayesian estimates (9-12% realistic, 18-19% favorable-conditions), yet both forecasters settled at 9%, hugging only the low end of the "realistic" band without justifying why the more optimistic redistricting-favorable scenario (18-19%) should be discounted so heavily given Sabato's expected R+6 figure sits almost squarely in the 220-224 range. 4. Neither critique addresses the Maryland redistricting amendment (uncertain voter approval, per "Gaps/unknowns") which could flip a GOP seat and mechanically push the final count toward the lower boundary or below 220 — this is a discrete, dateable event that should shift some probability mass out of the 220-224 band specifically, yet it's not mentioned in either rationale. 5. Both forecasts state confidence=0.62 despite giving identical point estimates and reasoning chains — this lack of independent variation (same number, same anchor, similar phrasing) suggests herding on the Polymarket price rather than independent synthesis of the historical base-rate model (0% for 5-seat windows historically, which if taken literally would argue for an even lower probability than 9%). 6. Neither model discusses the vacancies (2 currently) or how special elections before Nov 2026 might alter the starting 220R count baseline, which is a mechanical detail affecting whether "220" is even the right center-point for the redistricting-adjusted estimate.

10% (62% confidence)

Yes: 10% No: 90%

The 220–224 band is a narrow 5-seat window inside a distribution with large residual variance (uncalled toss-ups, Maryland amendment, unresolved maps), so even if the center of the distribution were exactly 222, this bucket would capture only ~15-20% of mass; but the center is likely lower. A D+6.8 to D+8.1 generic ballot, Trump approval near lows, and unreversed Cook/Sabato/Inside Elections drift toward Democrats place the modal outcome below 218, with redistricting (net R+5 to +9) providing the main path into this band rather than the baseline. The critique's point that some mass sits above 225 is fair but only reinforces No, since it further squeezes the 220–224 window; the Maryland seat-flip risk also pushes mass out of the low end of the band. Polymarket's 8.2%, trending down over 30 days, is the only direct quantitative anchor, and the code-execution Bayesian range (9–12% realistic) brackets it. I settle marginally above the market at 9.5% to account for the genuine redistricting-offset scenario without over-crediting a bucket that historical base rates say is rarely hit.

Total pipeline time: 265.8s