Scan: 2026 Balance of Power: R Senate, R House

date	predicted	market price	confidence
2026-09-10	13%	14%	69%

1. Planner Fable 5.1 · Fermi decomposition + tool plan

Tools chosen: polymarket_direct polymarket_related kalshi_related article_search claude_news wikipedia code_execution

Sub-questions (Fermi decomposition)

What is the current Polymarket price for 'R Senate, R House' and how does it compare to the implied product of separate Senate-control and House-control market prices on Polymarket and Kalshi?
What is the current generic congressional ballot average and Trump approval rating, and how do they compare to prior midterm cycles where the president's party lost the House?
Which Senate seats are competitive in 2026 (e.g., NC, ME, GA, MI, OH, TX, IA, AK) and how many net pickups do Democrats need to flip the Senate given the current 53-47 split?
How have mid-decade redistricting efforts (Texas, California, Ohio, Missouri, Florida, Virginia, etc.) shifted the expected House seat baseline, and what is the current House margin including vacancies?
What do forecasters (Cook, Sabato, Inside Elections, Silver Bulletin) currently rate as the probability of each party controlling the House and Senate?
What is the historical base rate of the president's party holding both chambers in a midterm, especially when presidential approval is below 45%?

Planner reasoning

This is a joint outcome (R keeps Senate AND R keeps House) in the 2026 midterms; the Polymarket price is the primary anchor, and separate Senate-control and House-control markets on both Polymarket and Kalshi let us check the joint price against the product/correlation of the marginals. Key drivers are the generic ballot, presidential approval, the Senate map (Dems need net +4 with few good targets), redistricting fights, and special elections/retirements. Historical midterm base rates for the president's party losing the House are also highly relevant.

2. Research Parallel tool calls

tool	status	time	items	summary
polymarket_direct	OK	0.3s	1	## This Market's Polymarket Data 2026 Balance of Power: R Senate, R House - Current price (probability): 8.50% - 7-day price change: -4.00% - 30-day price change: -4.00% - Total volume: $3,361,553 (USD notional) - Price range: 8.50% - 19.50% - Data points: 91 days
polymarket_related	OK	0.6s	23	Polymarket search over 4 keyword(s), kept 23 matches. keyword 'Senate control 2026': 6 markets \| keyword 'House control 2026': 6 markets \| keyword 'balance of power 2026': 5 markets \| keyword 'midterms': 6 markets
kalshi_related	OK	5.6s	3	3 related markets / summaries. series KXSENATE: 0 markets (skipped 0 no-signal) \| series KXHOUSE: 0 markets (skipped 0 no-signal) \| series KXBALANCEOFPOWER: 0 markets (skipped 0 no-signal) \| keyword 'Senate control 2026': ok \| keyword 'House control 2026': ok \| keyword 'balance of power': ok
article_search	OK	46.9s	13	Fetched 13 unique articles across 4 queries (lookback=45d). '2026 midterm generic ballot polling': 10 articles \| '2026 Senate map competitive races Democrats path to majority': 10 articles \| 'House redistricting 2026 Texas California Missouri Ohio Florida seats': 10 articles \| 'Cook Political Report
claude_news	OK	26.1s	23	## Key Findings: 2026 Midterm Balance of Power Generic Congressional Ballot - Silver Bulletin: Democrats began September with a generic congressional ballot lead of D +6.4, rising to D +7.5 as of mid-September — a high point for the party this cycle. (natesilver.net) - RealClearPolitics avera
wikipedia	OK	0.5s	3	Fetched 3 Wikipedia entries (0 missing pages).
code_execution	OK	55.9s	0	Historical base rates (midterms since 1946, unconditional on party): - Elections where the president's party held both chambers going into the midterm: 11/20 (1946, 1950, 1954, 1962, 1966, 1978, 1994, 2006, 2010, 2018, 2022). - Of those, they retained both chambers in only **4/11 (36

3. Evidence Brief Sonnet 5 · 6586 chars

# Event Kalshi/Polymarket market: will Republicans control BOTH the U.S. Senate AND the U.S. House after the 2026 midterms ("R Senate, R House"). # Outcomes to forecast - Yes (R Senate + R House) - No (any other combination) # Kalshi market anchor Kalshi-direct data not separately returned in this pull; Polymarket-direct (same event, ticker matches) shows current YES price **8.5%**, down from a 30-day/7-day high of 19.5%, trending down (-4pp both 7d and 30d). Volume $3.36M over 91 days — a liquid, actively traded contract with a clear downward trend against Republicans holding both chambers. # Resolution source Not run (no resolution_source tool result provided). Official resolution will rely on state election authorities, Congress, and consensus reporting per the market's own rules; status quo control is R Senate 53-47, R House 218-213 (4 vacancies) as of Jan 2026 (Wikipedia/Cook via claude_news). # Sub-question answers 1. **Polymarket R/R price vs. implied product** — Polymarket "R Senate,R House" = 8.5%. Standalone Polymarket markets show P(R Senate)=40.5%, P(R House)=10.5%; independence-implied product ≈4.25%, roughly half the actual 8.5% joint price — consistent with positive correlation between Senate/House outcomes (both driven by shared national environment). No separate Kalshi standalone Senate/House control markets were returned for direct comparison. 2. **Generic ballot & approval vs. history** — Generic ballot: Democrats +6–8.6 (Silver Bulletin D+7.5 mid-Sept; RCP D+8.6; CNBC/FiftyPlusOne D+6). Trump approval ~37-38% approve, net -20 to -22 (Silver Bulletin, Wikipedia aggregator, Sept 2026). This approval level and ballot deficit resemble 2006 and 2018 — both cycles where the president's party (also holding both chambers) lost at least one chamber. 3. **Competitive Senate seats / pickups needed** — Democrats need net +4 to flip the Senate (53R-47D). Competitive races: Ohio, Iowa, Alaska (toss-up); Texas, Maine, Michigan (lean D); North Carolina, Georgia (likely D) per Silver Bulletin ratings (Aug 2026, via X/Politics tracker); CBS Battleground Tracker also flags 9 key Senate races. GOP defends 22 of 35 seats up. 4. **Redistricting & House baseline** — Finalized maps: Texas (+5 GOP), Missouri (+1 GOP), North Carolina (+1 GOP), Ohio (+2 GOP), Utah (court-drawn), vs. California (+5 Dem, ballot-approved). Net effect roughly cancels out (~+7 GOP vs +5 Dem), per claude_news/Ballotpedia. Current House: R 218-213 with 4 vacancies (Wikipedia/claude_news, Jan 2026 baseline before new maps take effect). 5. **Forecaster probabilities** — Silver Bulletin (Sep 18, 2026, Deluxe/FLIPR model): Senate — Dem 51 seats/Rep 49, 57-59% chance Dems win Senate (i.e., ~41-43% R Senate); House — Dem 233/Rep 202, 85-90% chance Dems win House (~10-15% R House). Prediction markets (Polymarket) show Dems 54%, Reps 48% for Senate (rounding artifact) on $4.45M volume (Newsweek/Silver Bulletin). 6. **Historical base rate** — Since 1946, when a president's party held both chambers entering a midterm (11 cycles), it retained both only 4/11 (36%). Conditioning on approval <45% narrows to 6 cycles, retaining both only 2/6 (33%) — and all three most recent low-approval cases (2006, 2018, 2022) lost at least one chamber. Current approval (~37-38%) sits in this adverse historical band. # Key facts (high-confidence, factual) 1. [Wikipedia] Senate: 53R-47D (incl. 2 ind.), 35 seats up in 2026 (incl. OH, FL specials); House: 218R-213D, 4 vacancies (pre-election baseline). 2. [Polymarket] Sub-market prices sum consistently: D Senate/D House 59.5%, R Senate/D House 30.5%, D Senate/R House 1%, R/R 8.5%, Other 0.25%. 3. [Silver Bulletin/claude_news] Generic ballot ~D+7 to +8.6; Trump net approval ~-20 (Sept 2026). 4. [claude_news] Redistricting largely nets out nationally (GOP TX/MO/OH/NC gains roughly offset by CA Dem gains); FL and IN maps unresolved/rejected. 5. [Wikipedia] Two incumbent GOP senators (Cassidy-LA, Cornyn-TX) lost renomination; Maine and SC nominees replaced — signals of primary volatility complicating some GOP-held seats. # Cross-market signals - Kalshi related: No direct KXSENATE/KXHOUSE standalone series returned; unrelated markets only (AI takeover, 2028 sweep odds). - Polymarket: D/D House-only 90.5% Yes; D Senate-only 60.5% Yes; R/R combo consistently priced lowest (8.5%) among four combos, confirming market sees House as near-lock for Dems and Senate as competitive but leaning D. - Sportsbook implied: none provided. # Analyst opinions and speculation - Silver Bulletin: House "everything points the same direction" toward Dem takeover (85-90% model); Senate model has swung toward Dems (+8pp in two weeks) on TX/OH/MI movement. - CNN/NPR/WaPo (Sept 2026): consensus framing is a "Democratic wave," but some Democrats fret the generic-ballot lead isn't wider given depressed Trump approval. - CNBC: markets increasingly pricing divided/Democratic Congress risk (debt ceiling gridlock). # Directional lean per outcome - **Yes (R Senate, R House)**: Opposing evidence dominates — poor approval (-20 net), D-favorable generic ballot (D+7-9), historical base rate (only 25-36% retention in comparable low-approval cycles), and converging forecaster models (R Senate ~41-43%, R House ~10-15%) all point against. Minor supporting factors: redistricting rough parity, some GOP resilience noted by CNN despite bad topline numbers, Senate map still has multiple GOP-favorable toss-ups (OH, IA, AK) that could all break red. - **No**: Strongly favored — virtually all cross-market and forecaster signals price House as near-certain Democratic and Senate as toss-up-to-lean-Democratic, making the compound "both R" outcome a low-probability tail scenario. # Gaps / unknowns - No direct Kalshi standalone Senate/House control prices retrieved for arbitrage check against the 8.5% joint price. - Independent-candidate caucus decisions (per market rules) in KS, SD, ID, NE could add late resolution uncertainty. - Florida/Indiana redistricting outcomes still unresolved, could marginally shift House baseline before Nov 2026. # Calibration anchors - Polymarket YES price (anchor): 8.5%, trending down from 19.5% high. - Implied product from current standalone Polymarket marginals (R Senate 40.5% × R House 10.5%) ≈ 4.25%, below the actual 8.5% joint — market pricing meaningful positive correlation. - Historical precedent: president's party with unified control + low approval (<45%) retained both chambers only 2/6 times (33%) since 1946; all three most recent instances (2006, 2018, 2022) lost at least one chamber.

4. Forecasters 5 draws · Fable 5.1 + GPT-5.6 Sol

claude-fable-5-1

66% conf

Yes	9%
No	91%

The Polymarket anchor for R Senate/R House is 8.5%, trending down from a 19.5% high, and standalone marginals (R Senate ~40.5%, R House ~10.5%) with strong positive correlation imply a joint probability in the 8-11% range. Silver Bulletin's model gives Democrats an 85-90% chance of the House and 57-59% for the Senate, with a D+7 to D+9 generic ballot and Trump net approval near -20 resembling 2006/2018 environments where the president's party lost at least one chamber. The compound outcome essentially requires a large systematic polling miss toward Republicans; in that world the Senate would very likely follow the House, so the joint sits close to P(R House) ≈ 10-12%. Redistricting roughly nets out and offers little offset. I stay near the market anchor with a slight upward nudge for correlation and polling-error tail risk.

claude-fable-5-1

63% conf

Yes	9%
No	91%

The Polymarket anchor for R Senate + R House is 8.5%, trending down from a 19.5% high, and standalone marginals (R Senate ~40.5%, R House ~10.5%) imply a correlated joint near 8-10%. Silver Bulletin's model (Sept 2026) gives Democrats 85-90% to win the House and 57-59% for the Senate, consistent with a D+7-9 generic ballot and Trump net approval around -20, an environment resembling 2006/2018 when the president's party lost at least one chamber. Redistricting roughly nets out nationally, so the House baseline offers little structural cushion for the GOP. Residual upside for Yes comes from polling error correlated across chambers (a 2022-style GOP overperformance) and GOP-favorable Senate toss-ups (OH, IA, AK), which justifies staying slightly above the independence-implied product but near the market price.

claude-fable-5-1

62% conf

Yes	9%
No	91%

The Polymarket joint 'R Senate, R House' contract trades at 8.5% (down from a 19.5% high), and this is consistent with standalone marginals (R House ~10.5% market / 10-15% Silver Bulletin; R Senate ~40-43%) once positive correlation is applied: conditional on Republicans holding the House, the national environment would almost certainly be strong enough for them to hold the Senate too, so the joint price should sit only modestly below P(R House). Trump's net approval near -20 and a D+7 to +9 generic ballot resemble 2006/2018, cycles in which the president's unified-control party lost at least one chamber, and the historical retention rate under low approval is only ~33%. Redistricting roughly nets out, so no structural offset is being missed by markets. I place YES at ~9%, essentially at the market anchor with a slight nudge upward for correlation and residual polling-error risk in the House.

gpt-5.6-sol

78% conf

Yes	8%
No	92%

The 8.5% Polymarket YES price is the best available consensus anchor, with high volume and a sustained decline from 19.5%. Democrats’ 6–8.6-point generic-ballot lead, Trump’s roughly -20 net approval, and historical midterm patterns strongly imply Republicans will lose at least one chamber. Silver Bulletin’s estimates of only a 10–15% Republican House chance and roughly 41–43% Republican Senate chance also support a low joint probability, although positive correlation between chamber outcomes keeps it above the 4.25% independence-implied product. Near-offsetting redistricting changes and a still-competitive Senate map provide insufficient reason to move materially from the market anchor.

gpt-5.6-sol

79% conf

Yes	8%
No	92%

I anchor on the liquid Polymarket YES price of 8.5%, which has declined sharply from 19.5%. A Democratic generic-ballot lead of roughly 6-9 points, Trump approval near 37-38%, and adverse midterm history all point toward Republicans losing at least one chamber. Silver Bulletin gives Republicans only about a 10-15% chance of holding the House and 41-43% of holding the Senate, while the joint market appropriately prices positive correlation between the two results. Redistricting appears roughly offsetting nationally, so there is insufficient evidence to move materially away from the market anchor.

6. Final plain mean of 5 draws

9% (70% confidence)

Yes: 9% No: 91%

Mean of 5 draws (0.09, 0.09, 0.09, 0.09, 0.09). The Polymarket anchor for R Senate/R House is 8.5%, trending down from a 19.5% high, and standalone marginals (R Senate ~40.5%, R House ~10.5%) with strong positive correlation imply a joint probability in the 8-11% range. Silver Bulletin's model gives Democrats an 85-90% chance of the House and 57-59% for the Senate, with a D+7 to D+9 generic ballot and Trump net approval near -20 resembling 2006/2018 environments where the president's party lost at least one chamber. The compound outcome essentially requires a large systematic polling miss toward Republicans; in that world the Senate would very likely follow the House, so the joint sits close to P(R House) ≈ 10-12%. Redistricting roughly nets out and offers little offset. I stay near the market anchor with a slight upward nudge for correlation and polling-error tail risk.

Pipeline Timing

Total pipeline time: 129.4s

Per-tool research timings shown in the Research section above.