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

date	predicted	market price	confidence
2026-08-31	38%	50%	57%
2026-08-24	20%	48%	52%
2026-08-17	46%	48%	58%

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 D Senate + D House, and how does it compare to the standalone Polymarket/Kalshi prices for Democrats winning the Senate and winning the House?
What are current market/forecaster probabilities for Democrats flipping each key Senate seat (NC, ME, OH special, TX, IA, AK, FL special) and holding GA, MI, NH?
What is the current generic congressional ballot average and Trump approval rating, and how do they compare historically to midterms where the out-party gained 4+ Senate seats?
Historically, in midterms where the President's party lost the House, how often did the opposition also win the Senate given a comparably unfavorable Senate map?
Have any major candidate recruitment developments (e.g., Cooper in NC, Brown in OH, Collins retirement in ME, Texas GOP primary outcome) shifted the Senate outlook recently?
How correlated are House and Senate outcomes in current markets, i.e., is P(D Senate and D House) approximately equal to P(D Senate)?

Planner reasoning

This is a joint outcome: Democrats win both Senate and House in 2026. House flip is plausible given midterm dynamics, but the Senate map (Dems need net +4 from 47, defending GA/MI/NH while targeting NC/ME/OH/TX/IA/AK) is the binding constraint. The probability is roughly P(D Senate) since D Senate almost certainly implies D House, so research should focus on Senate race-by-race odds, the generic ballot, and cross-market prices for each chamber individually.

2. Research Parallel tool calls

tool	status	time	items	summary
polymarket_direct	OK	0.4s	1	## This Market's Polymarket Data 2026 Balance of Power: D Senate, D House - Current price (probability): 58.50% - 7-day price change: +9.00% - 30-day price change: +12.00% - Total volume: $3,160,261 (USD notional) - Price range: 41.50% - 58.50% - Data points: 90 days
polymarket_related	OK	0.9s	29	Polymarket search over 5 keyword(s), kept 29 matches. keyword 'Senate control 2026': 6 markets \| keyword 'House control 2026': 6 markets \| keyword 'balance of power 2026': 5 markets \| keyword 'Democrats win Senate': 6 markets \| keyword 'Democrats win House': 6 markets
kalshi_related	OK	8.1s	3	3 related markets / summaries. series KXSENATE: 0 markets (skipped 0 no-signal) \| series KXHOUSE: 0 markets (skipped 0 no-signal) \| series KXCONTROL: 0 markets (skipped 0 no-signal) \| keyword 'Senate control': ok \| keyword 'House control': ok \| keyword 'balance of power': ok
article_search	OK	0.8s	10	Fetched 10 unique articles across 4 queries (lookback=45d). '2026 Senate map Democrats path to majority': 10 articles \| '2026 midterm generic ballot polling': 10 articles \| 'Senate race polls North Carolina Maine Ohio Texas Iowa Alaska 2026': 10 articles \| 'Democrats House majority forecast 2026': 1
claude_news	OK	25.7s	18	## 2026 Midterm Outlook Summary Generic Ballot - Silver Bulletin's likely voter adjusted generic ballot average moves the Democratic lead from D+6.6 to D+7.6 as of late August 2026 (natesilver.net). - Republicans had narrowed the gap, but Democrats again lead by 6 percentage points on Decis
wikipedia	OK	0.6s	3	Fetched 3 Wikipedia entries (0 missing pages).
code_execution	OK	44.5s	0	## Findings: P(D Senate AND D House) — 2026 Midterms Joint Probability Calculation (using illustrative market-implied inputs — replace with live odds when available): - Standalone inputs used: P(D House) ≈ 0.68, P(D Senate) ≈ 0.28 (Senate map is far tougher for Democrats in 2026) - Give

3. Evidence Brief Sonnet 5 · 7172 chars

# Current state Market resolves on whether Democrats simultaneously win House AND Senate majorities in Nov 2026 midterms. As of campaign fall 2026, Democrats are heavily favored for the House (generic ballot D+5–8, Trump approval ~38%) but the Senate remains a genuine toss-up due to a structurally unfavorable map (53R-47D headed in), with several red-state races (NC, ME, OH, TX, IA, AK) now competitive. # Timeline of key events - 2025 (various): Senate retirements open seats — Durbin (IL-D), Peters (MI-D), Tina Smith (MN-D), McConnell (KY-R); Vance/Rubio resignations trigger OH and FL special elections (confirmed, Wikipedia). - 2026-Q1–Q2: Cassidy (LA) and Cornyn (TX) lose GOP renomination primaries — first such double-incumbent loss since 2010 (confirmed, Wikipedia). - 2026-04/06: Cook Political Report and Sabato's Crystal Ball move NC Senate race from Toss-up to Lean/Likely Democratic as Cooper leads polling by 6-8 pts (confirmed, ncnewsline/thehill). - 2026-07-10: Graham Platner withdraws as ME Democratic Senate nominee amid misconduct allegations; Troy Jackson selected via convention 2026-07-25 (confirmed, claude_news/Wikipedia). - 2026-08-03/07: Polling shows TX (Talarico-Paxton), IA (Turek), OH (Brown-Husted), AK all competitive; DSCC touts "multiple paths to majority" (reported, CNN). - 2026-09-08 to 09-16: Generic ballot averages settle around D+5–8; Cook moves NH from Lean D to Toss-up (Sununu strong GOP recruit) (confirmed/reported, multiple). - 2026-09-13: CBS Battleground Tracker: GOP holds 53-47 Senate edge entering election; Dems need net +4 (confirmed, CBS News). - 2026-09 (ongoing): Polymarket joint "D Senate, D House" price rises from ~46.5% to 58.5% over 30 days (confirmed, polymarket_direct). # Event Will Democrats win control of BOTH the U.S. Senate and U.S. House in the 2026 midterms (per Kalshi resolution criteria on chamber majorities/caucus affiliation)? # Outcomes to forecast Yes / No # Kalshi market anchor No kalshi_direct price was returned in research (tool did not surface a live YES quote for this ticker). Best available anchor is the identical Polymarket market ("2026 Balance of Power: D Senate, D House"), trading at **58.5% YES**, up +9pts (7d) and +12pts (30d), on $3.16M volume — a clear upward trend toward Democratic sweep pricing. Treat this as a strong proxy for consensus; actual Kalshi price should be checked directly before finalizing. # Resolution source Not run (no resolution_source tool output provided). # Sub-question answers 1. **Joint vs standalone Polymarket prices** — Joint D Senate+D House = 58.5%. Standalone: D controls Senate ≈58%, D controls House ≈87.5% (polymarket_related). Independent product would be ~51%; actual joint (58.5%) exceeds this, implying strong positive correlation — essentially, most D-Senate-win scenarios already include a D House win. 2. **Key Senate seats** — NC: Cooper favored (Lean/Likely D, Cook/Sabato, +6-8pt polling). ME: toss-up, Jackson-Collins within margin of error (48-45 CNN). OH: toss-up, Brown vs. appointed incumbent Husted. TX: toss-up-leaning-competitive, Talarico trails Paxton narrowly (48-51 Fox poll). IA: Turek leads by 4 (toss-up). AK: toss-up (Cook). FL special: not covered in research (gap). GA (Ossoff) and MI, NH: Ossoff seen as competitive incumbent in Trump+2 state; MI complicated by progressive nominee El-Sayed over establishment pick; NH downgraded from Lean D to Toss-up due to Sununu candidacy. 3. **Generic ballot/approval vs history** — Current D+5–8 generic ballot with ~38% Trump approval most closely resembles 2018 (D+7-8, ~40% approval), a cycle that produced a House flip but only a modest, unfavorable Senate map outcome (GOP net +2), not a Senate flip. 4. **Historical House-flip → Senate-flip correlation** — In full House-flip midterms (n=4: 1946,1994,2010,2018), Senate also flipped 50% of the time (1946,1994), but this drops sharply on unfavorable maps (2018 is the closest analog and Senate did NOT flip). In all midterms with House losses (n=18), out-party gained 4+ Senate seats only 44% of the time (code_execution synthesis). 5. **Recruitment shifts** — Cooper's dominant NC primary win, Cornyn's primary loss to Paxton (weaker general candidate), and competitive polling in OH/IA/AK have expanded the Senate map favorably for Democrats versus early-cycle expectations (CNN, claude_news). 6. **House-Senate correlation** — Joint price (58.5%) is nearly equal to standalone Senate price (58%), implying P(D House | D Senate) ≈ 100%; Senate win is treated as a near-strict subset of House win scenarios (code_execution, polymarket_related). # Key facts 1. [Wikipedia] Senate class 2 up in 2026; GOP holds 53-47 entering cycle; OH and FL have special elections. 2. [CBS News] Dems need net +4 Senate seats for majority; 9 key Senate races, 42 key House races. 3. [Silver Bulletin/DDHQ] Generic ballot D+6-8 late Aug/Sept 2026. 4. [Polymarket] D House standalone 87.5%; D Senate standalone ~58%; joint sweep 58.5% (up from ~46.5% a month ago). 5. [CNN] Democrats' Senate map expanded into TX, IA, OH, AK but face internal recruitment friction in MI (El-Sayed) and defense risk in NH (Sununu). # Cross-market signals - Polymarket direct (same market): 58.5% YES, strong uptrend. - Polymarket sub-markets: D Senate 58%, D House 87.5%, D Senate/R House 31.5%, R sweep 10.5%. - Kalshi related: POWER-28-DH-DS-DP (2028 D sweep) trading 51%, not directly informative for 2026. - No sportsbook data available. # Analyst opinions and speculation - Nate Silver's model: ~56.5-59% D Senate control, clear D House majority (claude_news, dates vary Sept 2026). - CNN's Aaron Blake: Senate math remains complex despite expanded map; independent-candidate strategy (KS, SD, ID, NE) unproven. - CBS/Cook: House flip to Dems seen as likely baseline; Senate genuinely competitive but GOP-favored on paper. # Directional lean per outcome - **Yes (D Senate + D House)**: Supported by rising Polymarket trend (+12pts/30d), expanded red-state Senate competitiveness (NC favorite, TX/OH/IA/AK toss-ups), strong generic ballot/approval environment. Opposed by unfavorable Senate seat math (need net+4 on tough map), NH downgrade to toss-up, MI nominee friction, and historical precedent that Senate flips are less likely than House flips even in wave years with bad maps (2018 analog). - **No**: Supported by structural map disadvantage, historical base rate (Senate flipped in only 50% of full House-wave midterms, worse on unfavorable maps), narrow polling margins in ME/TX/AK. Opposed by consistent multi-poll evidence of an unusually broad, energized Democratic map for a red-leaning cycle. # Gaps / unknowns - No direct Kalshi YES price captured — must verify live. - FL special Senate election dynamics not covered. - Independent-candidate caucus uncertainty (KS, SD, ID, NE) could delay/complicate final seat counts under Kalshi's caucus rule. - No sportsbook data. # Calibration anchors - Polymarket proxy YES: 58.5% (anchor, pending direct Kalshi confirmation). - Historical: Senate flip alongside House flip occurred in 2/4 postwar full-flip midterms (50%), but 0/1 in most map-comparable case (2018).

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

claude-fable-5-1

52% conf

Yes	56%
No	44%

The binding constraint is the Senate: Democrats need net +4 on a 53-47 map, and the joint sweep price (Polymarket 58.5%) is essentially identical to the standalone D-Senate price (~58%), since a D House (87.5%) is a near-certain component of any D-Senate scenario. The Senate case is genuinely competitive — NC (Cooper) leans D, and ME/OH/TX/IA/AK are toss-ups, with Nate Silver's model at ~57-59% D Senate — but the 2018 analog (D+7 generic ballot, ~40% approval, Senate did not flip) and the need to run the table in several red states argue for shading slightly below the market. Additional friction in NH (Sununu, now toss-up) and MI (El-Sayed) adds downside risk to the Democratic path. I anchor near the Polymarket 58.5% consensus but trim modestly for structural Senate difficulty, landing at 56% Yes.

claude-fable-5-1

52% conf

Yes	56%
No	44%

No live Kalshi quote was surfaced, so the identical Polymarket market at 58.5% YES (up 12 pts in 30 days) serves as the consensus anchor, corroborated by Nate Silver's model at ~56.5-59% D Senate control. The House is near-locked for Democrats (~87.5% standalone, D+5-8 generic ballot, Trump ~38% approval), so the sweep effectively reduces to the Senate, where Democrats need net +4 on a GOP-favorable map but have expanded competitiveness into NC (Lean D), OH, TX, IA, AK, and ME. Historical precedent (2018 analog: strong wave, no Senate flip; Senate flipped in only 50% of full House-wave midterms) and defensive risks in NH (Sununu) and MI argue for shading slightly below the market's uptrend-driven price. I land at 0.56, essentially the market with a small discount for structural map difficulty and momentum overshoot.

claude-fable-5-1

47% conf

Yes	55%
No	45%

No direct Kalshi quote was available; the identical Polymarket market trades at 58.5% YES with a strong 30-day uptrend, and Nate Silver's model puts D Senate control at ~56.5-59%, with the House flip near-certain (87.5%) so the joint essentially equals P(D Senate). The Senate path is real but tight: Democrats need net +4, plausibly NC (Cooper favored) plus ME and two of OH/IA/TX/AK while defending GA, MI, and a newly toss-up NH against Sununu. Historical precedent (2018 analog with similar generic ballot and approval yielded no Senate flip on a bad map; Senate flipped in only 50% of House-wave midterms) argues for shading slightly below the market, and some of the recent price momentum may be overshoot. I land modestly under the Polymarket anchor at 0.55.

gpt-5.6-sol

69% conf

Yes	57%
No	43%

The 58.5% Polymarket YES price is the best available consensus anchor, while the 87.5% House and 58% Senate standalone prices show that the Senate is the binding constraint. Democrats’ D+5–8 generic-ballot advantage, low Trump approval, favorable North Carolina polling, and expanded opportunities in Ohio, Iowa, Texas, and Alaska point toward Yes. However, needing a net four Senate seats, alongside risks in New Hampshire, Michigan, and Maine, makes the map less favorable than the national environment; the 2018 precedent also shows that a strong House wave need not flip the Senate. I therefore shade slightly below the 58.5% proxy anchor while retaining a narrow Democratic-sweep favorite.

gpt-5.6-sol

72% conf

Yes	58%
No	42%

The 58.5% identical-market Polymarket price is the best available consensus anchor, while the lack of a direct Kalshi quote argues against a meaningful adjustment. The House appears highly likely to go Democratic, and the near-equality of the joint and standalone Senate prices indicates that the Senate is effectively the deciding leg. Democrats benefit from favorable polling in North Carolina and expanded opportunities in Ohio, Iowa, Texas, and Alaska, but they still need a difficult net gain of four while defending competitive seats including New Hampshire, Michigan, and Georgia. The D+5–8 environment supports a wave, although the unfavorable-map 2018 precedent and unresolved Florida and caucus-affiliation dynamics keep the sweep close to a toss-up rather than a strong favorite.

6. Final plain mean of 5 draws

56% (58% confidence)

Yes: 56% No: 44%

Mean of 5 draws (0.56, 0.56, 0.55, 0.57, 0.58). The binding constraint is the Senate: Democrats need net +4 on a 53-47 map, and the joint sweep price (Polymarket 58.5%) is essentially identical to the standalone D-Senate price (~58%), since a D House (87.5%) is a near-certain component of any D-Senate scenario. The Senate case is genuinely competitive — NC (Cooper) leans D, and ME/OH/TX/IA/AK are toss-ups, with Nate Silver's model at ~57-59% D Senate — but the 2018 analog (D+7 generic ballot, ~40% approval, Senate did not flip) and the need to run the table in several red states argue for shading slightly below the market. Additional friction in NH (Sununu, now toss-up) and MI (El-Sayed) adds downside risk to the Democratic path. I anchor near the Polymarket 58.5% consensus but trim modestly for structural Senate difficulty, landing at 56% Yes.

Pipeline Timing

Total pipeline time: 135.2s

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