🗳 2026 Election Simulation

Monte Carlo · 100,000 runs · Polymarket + Kalshi probabilities

🎲

Running simulations…

Fetching Polymarket + Kalshi prices, then running 100,000 scenarios

First load can take a few minutes — cached for 6h after that

Senate (100 seats total · Dems need 51, R VP breaks ties)

Projected Dem seats

—

p10–p90: —

Projected GOP seats

—

P(Dem ≥51)

—

Monte Carlo

P(GOP ≥50)

—

Monte Carlo

Distribution of Democratic Senate seats (out of 100 total)

GOP controls (≤50 D — R VP wins ties) Dem majority (≥51 D)

Senate contested races —

State	Current	D win %	Mkt margin	Poll avg	'24 Pres	PVI	Fundraising (CoH)	Pivotal	Probability bar	Source

House (435 seats total · majority = 218)

Projected Dem seats

—

p10–p90: —

Projected GOP seats

—

P(Dem ≥218)

—

Monte Carlo

P(GOP ≥218)

—

Monte Carlo

Distribution of total projected Democratic House seats (out of 435)

GOP majority (<218 D) Dem majority (≥218 D)

Safe seats: non-market races keep current party (D=215×— proportional). Competitive pool: — races with active markets are simulated.

Competitive House races with market data —

District	D win %	Mkt margin	Poll avg	'24 Pres	PVI	Fundraising (CoH)	Pivotal	Probability bar	Source

Dummymander Watch (partisan-engineered seats · backfire = the drawing party fails to hold or capture the seat)

Dummymander distribution (per-sim count of engineered seats backfiring · market-priced seats only)

Engineered seats —

Seat	Drawn by	Map	Mkt margin	Poll avg	'24 Pres	P(backfire)

Historical Trends (one snapshot per day · builds forward from first run)

Senate — P(D control) & P(Pivotal Candidate)

House — P(D control) & P(Pivotal Candidate)

Projected Democratic Senate seats (of 100)

Projected Democratic House seats (of 435)

Senate seat distribution over time (one ridge per day, newest at front · amber line = 51-seat majority)

House seat distribution over time (one ridge per day, newest at front · amber line = 218-seat majority)

Market-implied Senate seats over time (Kalshi bracket ladder · one ridge per day · amber = 51)

Market-implied House seats over time (Kalshi bracket ladder · one ridge per day · amber = 218)

D-win probability of the 5 closest Senate races over time

Dummymander Watch — expected backfires over time (mean engineered seats lost by the party that drew them · market-priced only)

🗳 2026 Election Simulation

Senate (100 seats total · Dems need 51, R VP breaks ties)

House (435 seats total · majority = 218)

Seat Markets (Kalshi + Polymarket bracket ladders — the markets' own seat distributions vs. our model)

Dummymander Watch (partisan-engineered seats · backfire = the drawing party fails to hold or capture the seat)

Historical Trends (one snapshot per day · builds forward from first run)