Scan: Will the Democratic party win the House in 2028?

1. Planner Fable 5 · Fermi decomposition + tool plan

Tools chosen: kalshi_direct kalshi_related polymarket_related claude_news code_execution wikipedia

Sub-questions (Fermi decomposition)

What probability does the Kalshi market currently assign to Democrats winning the House in 2028, and how has it trended?
What probability do Polymarket and related Kalshi markets assign to Democrats winning the House in 2026 (midterms) and the presidency in 2028, since these strongly correlate with 2028 House control?
What is the historical base rate of the House flipping or being held in a presidential election year, conditional on which party holds the White House and which party controls the House going in?
How is mid-decade redistricting (e.g., Texas, California, other states) expected to shift the House seat baseline for 2028?
What does the current generic congressional ballot and political environment suggest about the Democrats' structural position heading into 2026-2028?
How strongly do presidential and House outcomes co-move in modern elections (i.e., if a Democrat wins the presidency in 2028, how likely is a Democratic House)?

Planner reasoning

This is a long-horizon structural question about 2028 House control, closing in early 2029. The key drivers are the current Kalshi price, cross-market signals (2026 midterms, 2028 presidency, Polymarket equivalents), historical base rates on House control flips, and structural factors like redistricting and presidential coattails. News and base-rate math will supplement the market anchors.

2. Research Parallel tool calls

tool	status	time	items	summary
kalshi_direct	OK	0.2s	1	## This Market's Kalshi Data Democratic party (CONTROLH-2028-D) - Current price (probability): 72.00% - 7-day price change: +2.00% - 30-day price change: -3.00% - Average daily volume: 259 contracts - Price range: 64.00% - 84.00% - Data points: 27 days
kalshi_related	OK	6.1s	3	3 related markets / summaries. series CONTROLH: 0 markets (skipped 4 no-signal) \| keyword 'House 2026': ok \| keyword 'control of the House': ok \| keyword 'presidency 2028': ok
polymarket_related	OK	6.1s	0	Scanned 100 active Polymarket markets, kept 0 matches. keyword 'House 2028': 0 markets \| keyword 'House 2026': 0 markets \| keyword '2028 presidential election winner': 0 markets \| keyword 'balance of power': 0 markets
claude_news	OK	173.2s	17	## Research Findings: Democratic Prospects for House Control in 2028 Generic Ballot Polling (as of Aug 2026) - Silver Bulletin's likely voter adjustment moves the Democratic lead from D+6.6 to D+7.6 as of late August 2026. (natesilver.net) - Decision Desk HQ's polling average shows Democrats
code_execution	OK	118.2s	0	## Historical Base Rates: US House Control in Presidential Election Years (1968–2024, n=15) - Same-party alignment: The winning president's party also controlled the House after the election in 8/15 (53.3%) of cycles since 1968; restricting to the post-1994 realigned/polarized era (1996–202
wikipedia	OK	6.1s	3	Fetched 3 Wikipedia entries (2 missing pages).

3. Evidence Brief Sonnet 5 · 6609 chars

# Event Will the Democratic party win control of the U.S. House of Representatives after the 2028 elections? # Outcomes to forecast - Yes (Democrats win House control in 2028) - No (Republicans retain/win House control in 2028) # Kalshi market anchor **CONTROLH-2028-D current YES price: 72.00%** — 7-day change: +2.00%; 30-day change: -3.00%; 27 days of data; range 64-84%; average daily volume 259 contracts (moderate liquidity). Market has been range-bound in the high-60s/low-80s, currently trending slightly up over the past week but down over the month. [kalshi_direct] # Sub-question answers 1. **Current Kalshi probability & trend** — 72% YES, up 2pts in 7 days, down 3pts in 30 days; has ranged 64-84% over 27 days, suggesting moderate but not extreme confidence in a Dem House in 2028. [kalshi_direct] 2. **Polymarket/related Kalshi on 2026 House & 2028 presidency** — No direct Polymarket matches found. Related Kalshi market "2028: Democratic Sweep" (Presidency+House+Senate) prices at 52%, implying a very high conditional probability of Dem House given the sweep-only bundling (down 3pts in 7 days, up 3pts in 30 days). No standalone Kalshi 2026 House market surfaced in this pull, but Speaker market (Hakeem Jeffries becoming next Speaker) prices at 87%, implying strong odds Democrats control the House sooner than 2028 (likely 2026). [kalshi_related] 3. **Historical base rate** — Since 1968, the House has NEVER flipped control in a presidential-year election (0/15); persistence rate from prior midterm is ~94-100%. Post-1994, if Dems win the presidency, P(Dem House)=50%; if Reps win presidency, P(Dem House)=0%. The dominant driver is the 2026 midterm outcome, not the 2028 presidential race. [code_execution] 4. **Redistricting shifts** — Texas's 2025 GOP gerrymander (upheld by SCOTUS for 2026, uncertain beyond) aimed for +5 GOP seats but Brookings estimates only ~+2 realistically, with backfire risk. California's new Dem-favorable map (approved Nov 2025 special election, litigation pending, hearing Dec 15 2025) could yield +5 Dem seats. Net effect nationally: six states have redrawn maps (vs. historically only 2 since 1970), and the traditional Democratic map bias has flipped — Republicans can now lose the popular vote by 1% and still hold the House. [claude_news] 5. **Generic ballot/environment** — As of Aug 2026, Democrats lead generic ballot by ~5-8 points (Silver Bulletin D+7.6, DDHQ D+6, PollingSource D+4.9); Trump approval ~40%/57% disapproval. This is favorable for Democrats but only sets up the 2026 midterm, not 2028 directly. [claude_news] 6. **Presidential/House co-movement** — Strong post-1994: if a Democrat wins the presidency in 2028, P(Dem House)=50%; if a Republican wins, P(Dem House)=0% (small n=8 sample). Coattail effects exist but are secondary to the 2026 midterm outcome per the base-rate model. [code_execution] # Key facts (high-confidence, factual) 1. [kalshi_direct] CONTROLH-2028-D trades at 72% YES. 2. [code_execution] House control has never flipped in a presidential-year election since 1968 (0/15). 3. [claude_news] Republicans hold a 220-215 House majority entering 2026; Dems need net +3 seats to flip. 4. [claude_news] Texas and California have both enacted aggressive new congressional maps in 2025; litigation ongoing (CA hearing Dec 15, 2025; TX map upheld for 2026 by SCOTUS, future uncertain). 5. [claude_news] Generic ballot in Aug 2026 shows Democrats leading by ~5-8 points across major trackers. 6. [kalshi_related] Kalshi's "Democratic Sweep 2028" (Pres+House+Senate) market prices at 52%. 7. [kalshi_related] Kalshi's "Next Speaker — Hakeem Jeffries" market prices at 87%, implying strong odds Democrats already control the House pre-2028 (i.e., after 2026). # Cross-market signals - Kalshi related: Democratic Sweep 2028 (52%) and Speaker=Jeffries (87%) both point toward markets pricing in Democratic House control well before/independent of the 2028 election, consistent with (but somewhat below) the standalone 72% on CONTROLH-2028-D. - Polymarket: No relevant active markets found for House 2028/2026 or balance-of-power. [polymarket_related] - Sportsbook implied: None available. # Analyst opinions and speculation - Cook Political Report (Erin Covey/Matthew Klein): GOP majority "dangerously thin" entering 2026; Republicans could gain up to 3 seats via Ohio redistricting control, partially offsetting Dem generic-ballot lead. [claude_news] - Brookings: Texas gerrymander likely underperforms design (only ~+2 GOP seats, not +5), with backfire risk in a strong Dem year. [claude_news] - Constitutional law commentary: SCOTUS's 2026 Texas ruling doesn't guarantee the map survives for 2028. [claude_news] - claude_news synthesis: No dedicated 2028-specific House forecasts exist yet from major handicappers; all current analysis focuses on 2026 as the pivotal, structurally decisive election for the 2028 baseline. # Directional lean per outcome - **Yes (Dem House 2028)**: Supported by strong Dem generic ballot lead (5-8pts), historical ~94-100% persistence of House control from midterm into presidential year, thin GOP majority, and elevated Kalshi-implied odds (72% on target market, 87% Jeffries-Speaker market, 52% full sweep market). Opposed by GOP-favorable redistricting shift (net map bias now favors Republicans nationally) and uncertainty in Texas map's post-2026 legal status. - **No (Rep House 2028)**: Supported by redistricting math structurally favoring the GOP (can lose popular vote by 1% and still win House), possible GOP gains via Ohio redistricting, and the post-1994 finding that P(Dem House)=0% conditional on a GOP presidential win. Opposed by strong current Dem polling environment and historically near-certain persistence once a party wins the House at a midterm. # Gaps / unknowns - No confirmed 2026 midterm House result yet (the single largest driver per base-rate model) — brief relies on mid-2026 polling only. - No 2028 presidential nominee/matchup data provided. - Legal status of Texas map beyond 2026 and California map litigation outcome (hearing Dec 15, 2025) unresolved. - No Polymarket coverage for comparison/arbitrage. - No explicit 2026 Kalshi House-control market price was retrieved directly (only inferred via Speaker and Sweep markets). # Calibration anchors - Kalshi CONTROLH-2028-D current YES price: **72%** (primary anchor). - Historical precedent: House has flipped in a presidential year 0/15 times since 1968; ~94-100% persistence rate from prior midterm control. - Related Kalshi cross-checks: Democratic Sweep 2028 at 52%; Speaker=Jeffries at 87%.

4. Forecasters Fable 5 + GPT-5.6 Sol · in parallel

claude-fable-5

55% conf

Yes	74%
No	26%

The Kalshi anchor is 72% YES, and the causal chain runs primarily through the 2026 midterm: the Jeffries-Speaker market at 87% implies markets expect Democrats to take the House in 2026 amid a D+5-8 generic ballot lead and a razor-thin 220-215 GOP majority. Historically, House control has never flipped in a presidential-year election since 1968 (0/15), so a 2026 Dem win would translate to ~90-95% odds of retaining in 2028; chaining ~85% (2026 flip) × ~92% (persistence) plus a small chance of a 2028-only flip yields roughly 75-80%. Offsetting this is the structural redistricting shift favoring the GOP (Republicans can now lose the popular vote by ~1% and still hold the House) and unresolved TX/CA map litigation, which justifies staying near rather than well above the market. I land slightly above the anchor at 0.74, consistent with the Jeffries and persistence signals.

gpt-5.6-sol

68% conf

Yes	73%
No	27%

I anchor near Kalshi's 72% YES price, with its recent range and moderate liquidity arguing against a large adjustment. Democrats' 5–8 point generic-ballot lead, the GOP's narrow 220–215 majority, and the 87% Jeffries next-Speaker price point toward Democrats entering 2028 with House control. Since House control has persisted from the preceding midterm through every presidential election since 1968, a likely 2026 Democratic takeover would strongly favor YES. I raise the estimate only slightly to 73% because GOP-favorable map bias, unresolved redistricting litigation, and uncertainty about the 2026 result remain meaningful risks.

5. Devil's Advocate Sonnet 5 · red-team critique

Both forecasts converge almost identically (74% and 73%) within 2 points of the Kalshi anchor (72%), suggesting minimal independent signal extraction beyond the market price itself — this reduces the value of having two "independent" forecasts.

Neither forecaster reconciles the 15-point gap between the Jeffries Speaker market (87%) and the CONTROLH-2028-D market (72%): if Dems are 87% likely to hold the House post-2026, but only 72% likely to hold it in 2028, that implies a meaningful "flip-back" probability (Dems win House control in 2026 but lose it again by 2028) that neither rationale explicitly models or quantifies.

The brief's key base-rate finding — "if a Republican wins the presidency [in 2028], P(Dem House)=0%" (n=8 sample) — is not addressed by either forecaster; both skip modeling the 2028 presidential race probability entirely, treating the 2026 midterm as the sole driver despite the brief flagging presidential-House co-movement as a real (if noisy) factor.

Both treat the "0/15 since 1968" base rate as strongly reliable, but the brief notes six states have redrawn maps in this cycle "vs. historically only 2 since 1970" and that the traditional Dem map bias has flipped — this is a structural regime change that could weaken the historical persistence base rate, and neither forecast discounts the reference class for this discontinuity.

Neither forecast stress-tests a tail scenario where TX's map is extended/expanded post-2026 (SCOTUS only upheld it "for 2026... future uncertain") combined with CA litigation invalidating the Dem gerrymander (hearing Dec 15, 2025) — a combined adverse redistricting outcome that could materially lower P(Yes) below both current estimates.

The "Democratic Sweep 2028" market at 52% is used as a soft corroboration by the brief but neither forecaster does the arithmetic to check consistency (52% sweep requires wins in Presidency+House+Senate simultaneously, which mechanically implies P(House)≥52% but doesn't by itself validate 72% vs., say, 65% or 80%) — this cross-market check is underexploited as a calibration tool.

6. Aggregator Fable 5 · final calibrated forecast

72% (62% confidence)

Yes: 72% No: 28%

The Kalshi anchor at 72% is well-supported and neither forecast adds material independent signal beyond it. The critique's strongest points cut against the forecasters' slight upward tilt: the 87% Jeffries vs. 72% CONTROLH gap already reflects meaningful flip-back and presidential co-movement risk, the '0/15 since 1968' persistence base rate is weakened by an unprecedented redistricting regime change (six states redrawn, net map bias now GOP-favorable), and adverse combined TX/CA litigation outcomes present unmodeled downside. With the 2026 result still unknown and the market showing a stable range with moderate liquidity, I revert to the market anchor rather than trading above it. The Sweep market at 52% only mechanically bounds P(House)≥52%, consistent with 72% but not justifying more.

Pipeline Timing

Total pipeline time: 262.0s

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