Tools chosen: polymarket_direct kalshi_related polymarket_related claude_news claude_news gdelt_news wikipedia code_execution

Sub-questions (Fermi decomposition)

Planner reasoning

This is a Polymarket question on 2026 House control, so the Polymarket price is the primary anchor, with Kalshi's parallel market as an independent crowd check. The key empirical drivers are generic ballot polling, presidential approval, redistricting/mid-decade map changes (TX, CA Prop 50, MO, NC, OH, etc.), special election swings, and retirements — plus historical midterm base rates for the president's party losing House seats.

# Current state Republicans currently hold the House 222–213 (9-seat majority); Democrats need a net gain of ~3-5 seats (accounting for redistricting shifts) to win the 218-seat majority in the 120th Congress. Polymarket's direct market on this exact question prices Democratic control at 87.5% YES, with generic-ballot polling, Trump approval, and nonpartisan ratings all currently favoring Democrats, while enacted mid-decade redistricting has added a net ~8-seat structural advantage for Republicans that partially offsets the polling environment. # Timeline of key events - 2025-09-28 (confirmed): Missouri enacts new GOP-favorable map (~+1 R seat), targeting KC-area Democratic seat. - 2025-11-04 (confirmed): "Blue wave" off-year elections — Dems flip VA governorship, hold NJ by landslide, win CA redistricting referendum (Prop 50), gain seats in VA House of Delegates and NJ Assembly; Mamdani wins NYC mayor. [Wikipedia] - 2025-12 (confirmed): Federal court allows North Carolina's new GOP-favorable map to be used for 2026 (~+1 R seat). [NPR] - 2026-01 (confirmed): Kansas GOP redistricting effort fails for lack of votes; no new map. - Early 2026 (confirmed): Virginia Supreme Court strikes down Democratic effort to redraw map via constitutional amendment (would have been ~+4 D seats) — effort collapses. - 2026-04-27 (confirmed): US Supreme Court reinstates Texas's GOP-engineered map (~+5 R seats), clearing it for 2026 use. [Al Jazeera] - 2026-05-04 (confirmed): Florida Gov. DeSantis signs new GOP map aiming to flip 4 Dem-held seats. - Louisiana v. Callais ruling (2026, confirmed): SCOTUS 6-3 strikes down LA's second majority-Black district, likely costing Democrats ~1 seat. - Ohio (confirmed, date unspecified): GOP-Dem deal on map nets GOP only 1-2 seats (less than hardliners wanted). - Utah (confirmed): Court-ordered map redraw yields ~1 seat favoring Democrats. - Indiana (confirmed): GOP-controlled legislature rejects a proposed favorable redraw — no change. - 2026-05 to 2026-08 (reported): Generic ballot climbs from R+3.3 (Jan 2025) to a cycle-high D+6.9 (May 28, 2026); Sabato's Crystal Ball initial ratings (mid-2026) call Democrats favored but not certain to flip the House, with 19 toss-up districts. - 2026-06 to 2026-08 (reported): Nate Silver publicly estimates Democrats' House odds at 85-90%, above then-current prediction-market pricing; NYT expert panel (July 2026) unanimously predicts Democrats reach 218+ seats. - 2026-08 (reported): Generic ballot averages D+5 to D+8 across trackers (Silver Bulletin, DDHQ, RCP, Verasight); Trump net approval around -19.5 to -20.6 (historic lows), ticking up slightly to 40% approval in Gallup's late-August read. # Event Will the Democratic Party control the U.S. House after the 2026 midterm elections (resolves Yes/No based on which party holds a majority of voting seats in the 120th Congress). # Outcomes to forecast - Yes (Democratic control) - No (Republican control / other) # Kalshi market anchor No direct Kalshi YES price was returned in this research pull (kalshi_direct tool data absent). The closest available cross-market anchor is **Polymarket's own market on this identical question: 87.5% YES (Dem control)**, roughly flat over 7 days (+0%) and up +2pts over 30 days, trading in an 80.5%–88.5% range over 91 days on $5.5M volume — a fairly stable, high-conviction consensus. Adjacent Kalshi markets (Speaker Jeffries @ 83%, 2028 House-D @ 65%, 2028 Dem Sweep @ 55%) are consistent with a broadly pro-Democrat structural lean across related products, though not direct substitutes. # Sub-question answers 1. **Polymarket/Kalshi price & 90-day trend** — Polymarket direct: 87.5% YES, range 80.5-88.5% over 91 days, +2pt/30d trend, high volume ($5.5M). No live Kalshi price captured for this exact ticker. [polymarket_direct] 2. **Generic ballot** — D+5 to D+8 across trackers as of Aug 2026 (Silver Bulletin D+6.9-8.1 with likely-voter adjustment, DDHQ D+6, RCP D+6.4, Verasight D+8); above the D+6.5 margin that preceded Dems' 2018 40-seat gain. [claude_news] 3. **Trump approval** — Net approval ~-19.5 to -20.6 (historic lows) as of late Aug 2026; Gallup shows 40% approval (up from 37%). Historically, midterms under unpopular presidents produce large House losses for the president's party. [claude_news] 4. **Redistricting net effect** — Roughly **R+8 net seats** from enacted maps (TX+5, MO+1, NC+1, OH+1-2, FL+4 for R; CA+5, UT+1 for D; LA-1 for D via Callais); VA Dem effort and Indiana GOP effort both failed, Kansas stalled. [claude_news/multiple] 5. **Base rate since 1946** — President's party lost House majority in 7 of 12 midterms where it held the majority (58.3%), averaging a ‑29.7 seat swing when losses occurred — far exceeding the 3-5 seats Dems need in 2026. [code_execution] 6. **Seats needed & ratings** — Dems need a net ~3-5 seat gain from 222-213 GOP majority; Sabato's Crystal Ball initial ratings favor Dems with 19 toss-up districts (Dems need 9 of 19, GOP needs 11); Cook Political Report has been shifting multiple seats (CO-03, OH-10, OH-15) toward Democrats through August 2026. [claude_news] 7. **2025 special/off-year elections** — "Blue wave" Nov 2025 results: Dems flipped VA governorship, held NJ by landslide, flipped 13 VA House of Delegates seats and 5 NJ Assembly seats, won CA redistricting referendum — signaling a favorable Democratic swing environment driven by economic discontent. [Wikipedia] # Key facts (high-confidence, factual) 1. [Wikipedia] Current House: 222 R–213 D; 218 needed for majority. 2. [claude_news] Generic ballot averages D+5 to D+8 as of Aug 2026. 3. [claude_news] Trump net approval ~-20 as of Aug 2026. 4. [claude_news] Redistricting nets ~R+8 seats from enacted maps. 5. [code_execution] Historical base rate: incumbent House majority lost 58.3% of time in comparable midterms; avg swing ‑29.7 seats. 6. [polymarket_direct] Polymarket prices Dem House control at 87.5%. # Cross-market signals - Kalshi related: Speaker Jeffries (implying Dem House control) 83% YES, +4pt/30d; 2028 Dem House control 65% (falling -9pt/30d); 2028 Dem Sweep 55% (+7pt/30d) — directionally consistent but not direct proxies for 2026. - Polymarket: 87.5% YES, stable/rising, high volume — strong consensus anchor. - Sportsbook implied: none available. - Note: code_execution's cited "Polymarket 66¢/Kalshi 63¢" figures are explicitly labeled illustrative placeholders (no live data), contradict the actual polymarket_direct reading of 87.5%, and should be disregarded as stale/synthetic. # Analyst opinions and speculation - Nate Silver: Dems 85-90% to retake House, called markets "a little low." [claude_news] - NYT expert panel (July 2026): unanimous Dems reach 218+. [claude_news] - Sabato/Kondik: Dems favored but "won't be easy" given competitive map; 19 toss-ups. [claude_news] - Gateway Pundit-cited analysis: redistricting deficit "favor[s] Democrats to overcome" but could let Republicans "save their majority under the right circumstances." [claude_news] # Directional lean per outcome - **Yes (Dem control)**: Strong support — generic ballot D+5-8, historic base rate (58%) plus swing-model (~90%) both favor flip, poor Trump approval, favorable 2025 off-year results, expert consensus (Silver, NYT panel, Sabato). Polymarket at 87.5% reflects this. - **No (GOP retains)**: Supported by ~R+8 net redistricting edge shrinking the effective margin Dems must overcome, small structural House majority requiring only modest GOP resilience, and possible late-cycle approval recovery (Trump approval ticked up 3pts in Aug). Toss-up-heavy map (19 districts) leaves room for GOP to hold with a good break. # Gaps / unknowns - No live Kalshi YES price captured for this specific ticker (only Polymarket direct available) — true Kalshi consensus unconfirmed. - Outstanding redistricting litigation (Ohio, additional VRA challenges) could still shift 1-3 seats before Nov 2026. - Independent-caucusing member uncertainty could affect final seat count post-election per market rules. # Calibration anchors - Polymarket YES price (primary anchor): 87.5%, stable/rising, $5.5M volume. - Base rate: president's party loses House majority ~58% of time in similar midterms (1946-2022), average swing ‑30 seats when it happens — historically large relative to the 3-5 seats needed here. - Model blend (generic ballot + base rate) in code_execution: ~74-90% range depending on method, all pointing same direction as Polymarket though with more uncertainty than the market itself implies.

1. Both forecasts converge almost identically on 0.88 with near-identical rationale, essentially just restating the Polymarket price rather than independently stress-testing it—this is herding on a single cross-market anchor (not even a confirmed Kalshi price for this exact ticker, per "No live Kalshi YES price captured for this specific ticker") rather than true triangulation. 2. Neither forecast grapples with the tension between the historical base-rate swing (-29.7 seats average when the majority flips) and the redistricting-compressed margin — if a flip happens it's likely to be large, but the R+8 redistricting shift effectively makes this cycle's "true" needed swing more like 11-13 seats rather than 3-5, which is a meaningful raw-vote-to-seat conversion risk understated in both rationales. 3. Both cite Nate Silver's 85-90% and the NYT panel as corroborating evidence, but neither notes these are not independent of each other or of the polling data — they're derivative of the same generic-ballot inputs, so citing three sources that agree isn't really three independent confirmations (correlated evidence, overweighted as if diversified). 4. Neither forecast addresses the "possible late-cycle approval recovery" flagged in the brief (Trump approval ticked up 3pts in Gallup to 40%) or built in any scenario for closing momentum before November 2026 — three months remain between the Aug 2026 data cutoff and the election, an eternity in midterm dynamics, yet both treat current polling as near-final. 5. Outstanding litigation risk ("Ohio, additional VRA challenges could still shift 1-3 seats") is mentioned in the brief's gaps section but not incorporated into either forecast's uncertainty band — given the closeness of the needed seat margin, a few more seats shifting either way from pending court rulings could be decisive and warrants explicit downside/upside scenario weight. 6. Both forecasters land at exactly 88% with 0.78 "confidence," but neither explains why they deviate slightly upward from Polymarket's 87.5% rather than settling at or below it, nor do they seriously entertain the low end of the model-blend range cited in the brief (74-90%), which would suggest more uncertainty/wider distribution than the tight consensus implies.

86% (78% confidence)

Yes: 86% No: 14%

The direct Polymarket price of 87.5% on this identical question, with $5.5M volume and a stable-to-rising 91-day range, is the strongest available anchor, and fundamentals independently corroborate it: a D+5 to D+8 generic ballot, Trump net approval near -20, a 2025 off-year blue wave, and a base rate in which the president's party loses the House ~58% of the time with a ~30-seat average swing when it does — far more than the seats Democrats need. The critique is right that both forecasters herded on one anchor and treated correlated expert estimates (Silver, NYT panel, ratings shops) as independent confirmation, and that the ~R+8 redistricting shift effectively raises the true required swing to roughly double digits, making seat conversion riskier than the raw 3-5 seat framing implies. With ~3 months to the election, a slight Trump approval uptick, and 1-3 seats of pending litigation risk, the low end of the brief's own model blend (74-90%) deserves some weight. I therefore settle marginally below the market rather than above it, at 0.865 Yes, which preserves the strong directional signal while widening the uncertainty band the consensus understated.

Total pipeline time: 300.1s