2026 US balance of power

What exactly resolves?

A chamber-control outcome is not the national popular vote. Polymarket counts seats with stated exceptions. Kalshi’s generic series source list includes unrelated macroeconomic sources; this study rejects that list as evidence for election settlement and relies on the actual contract components.

Election day is November 3, 2026, but runoffs and organization can take longer. Polymarket has a January 4 caucus rule; the selected Kalshi contract references February 2027. An API election-day end timestamp does not settle every exception.

Are the two platforms asking the same question?

Polymarket spells out election-seat attribution, independents, runoffs, and tie-breaking control. Kalshi combines House and Senate CONTROL components for February 2027, with component-specific and accelerated resolution rules. The full CONTROL documents are needed before asserting equivalence.

Is there room at the quoted price?

Depth below is measured from the captured book within 2¢ of each side’s best ask. The 100- and 1,000-share examples consume successive ask levels. These are snapshots of visible entry liquidity, not reservations, total market capacity, or exit guarantees. Kalshi asks are derived as 1 minus the opposing bid with its available quantity.

Reported 24h activity: Polymarket parent event 349,085 USD; selected contract 30,232 USD. Kalshi parent event 672,890 contracts; selected contract 102,760 contracts. These units are not interchangeable.

Both sides: upside is not the same as likelihood

One hypothetical share at each captured ask, before fees. The price is not a measured win probability. A winning share pays one unit; a losing share pays zero. Football ties and exceptional settlement can pay other amounts.

Return if win is conditional net gain ÷ entry cost. It is not expected return, annualized return, or a recommendation. Fees lower the gain and increase the amount at risk.

Evidence and forecasting thesis

The selected price already expresses a joint outcome. It should not be multiplied by another chamber probability. No district or state simulation was fitted in this study; the interactive cases are sensitivity demonstrations.

• Seat-level competitiveness and common national polling errors, not a single national poll.
• Correlation between House and Senate outcomes; the same turnout surprise can move both.
• Runoffs, independent caucus decisions, vacancies, and the exact control tie-breakers.

What would invalidate the assessment: Any change in which elections count, attribution rules, or caucus information calls for a fresh rules review. A generic political sentiment score cannot replace that review.

Risk profile: Correlated polling error, concentrated political exposure, recounts, runoffs, and control-definition mismatch. Several related election positions can behave like one large exposure.

Probability model and payoff sensitivity

The starting benchmark is the captured Polymarket bid–ask midpoint, 63.50%. No validated independent model justifies changing it. Lower and higher scenarios move this assumption by five percentage points; they are not confidence bounds or recommendations. Figures use 100 shares at the best ask plus an illustrative 1¢ per share total fee allowance. Larger entries require the depth simulation.

Formula: expected net = N × [p(win) + 0.5 × p(tie) − entry − fee]. The binary table assumes p(tie)=0. Break-even under that assumption is 65.00%. A better-looking result after increasing p proves only that the calculation responds to p.

Every tool’s contribution

Observed inputs, calculated outputs, actual provider responses, and tools that cannot support this study are identified separately. All sixteen learning tools are accounted for.

Actual provider outputs: inspect before trusting

Jev returned different for the rule comparison, with classification confidence 51.00% using jev-1.13.0. This is not the event’s probability. The request rubric and full distribution are in the evidence download.

Drift forecast window: 2026-09-21T23:42:33.141436Z to 2026-11-02T14:42:33.141436Z. Quality flag: low_data. Rejected for probability modeling. Near-zero variation in a projected band is not proof of low risk.

Not applicable: a stock-market ticker cannot establish chamber control. The financial-evidence integration is deliberately omitted rather than given a misleading proxy.

No financial price series was used for this case.

{
  "research": {
    "status": 200,
    "body": {
      "topic": "2026 Balance of Power: D Senate, D House",
      "keywords": [
        {
          "term": "Democratic Senate control",
          "score": 85,
          "trend": "rising"
        },
        {
          "term": "2026 midterm elections",
          "score": 78,
          "trend": "stable"
        },
        {
          "term": "House of Representatives contr",
          "score": 70,
          "trend": "rising"
        },
        {
          "term": "Republican Senate majority",
          "score": 65,
          "trend": "declining"
        },
        {
          "term": "voter dissatisfaction",
          "score": 58,
          "trend": "rising"
        },
        {
          "term": "key Senate races",
          "score": 45,
          "trend": "stable"
        },
        {
          "term": "Trump approval ratings",
          "score": 35,
          "trend": "declining"
        },
        {
          "term": "cost of living",
          "score": 25,
          "trend": "stable"
        }
      ],
      "coverageIndex": 75,
      "coverageTrend": "accelerating",
      "coverageChangePct": 40,
      "sentimentScore": 25,
      "sentimentTrend": "improving",
      "signalScore": 68,
      "signalDirection": "bullish",
      "interpretation": "Recent news indicates a growing possibility of Democrats gaining control of both the Senate and the House in the 2026 midterm elections. Several polls show Democrats making inroads in traditionally Republican states and a strong preference for Democratic control of Congress among voters. Voter dissatisfaction with the current administration and economic pressures are contributing to this shift.",
      "leadingIndicators": [
        "Democratic candidates polling well in toss-up Senate races",
        "High voter dissatisfaction with the current Republican administration",
        "Increased focus on economic issues like cost of living"
      ],
      "watchNext": [
        "Upcoming polls in battleground states (e.g., Florida, Ohio, Maine)",
        "Further shifts in generic congressional ballot polling",
        "Impact of ongoing economic conditions on voter sentiment"
      ],
      "confidence": "Medium",
      "dataWindow": "7 days"
    }
  },
  "analyzer": {
    "status": 429,
    "body": {
      "error": "You've used your free AI analysis for today. Upgrade to Premium for unlimited access.",
      "usage": {
        "feature": "ai_analyzer",
        "used": 1,
        "limit": 1,
        "remaining": 0,
        "resetAt": "2026-09-22T00:00:00.000Z",
        "isPremium": false,
        "enforced": true
      },
      "isPremium": false,
      "requestId": "2362ed52-a364-4c29-be11-f3487fd6b020"
    }
  }
}

Costs, participation mechanics, and limits

• Entering: choosing a side, size, and price is different from receiving a fill. A limit can remain unfilled; crossing available quotes can consume multiple levels. These pages place no orders.
• Exiting: today’s entry depth says nothing certain about future bids. Early exit introduces a future-price assumption and may add fees. A marked gain is not a realized payout.
• Fees: all five captured Polymarket markets report fees enabled; the legacy fee endpoint returns base_fee 1000. This study does not interpret that as a flat 10% charge. Use the current market-specific schedule and rounding. Kalshi’s series fee type is quadratic; overrides and execution type can matter.
• Time and settlement: distinguish the event time, order close, resolution, dispute period, and withdrawal availability. Holding time and currency/funding costs are excluded from the scenario.
• Eligibility and access: platform and regional rules apply. Educational access to these pages does not establish eligibility to transact on either venue.
• Concentration: related topics may move together. The five studies are a learning set, not a diversified portfolio or an allocation proposal.

Polymarket fee documentation · Kalshi fee schedule