Senate Forecast
Who controls the Senate after November?
1,000 simulated elections, seeding each race with its polling average, correlated polling error, and a prediction-market prior. NOWCAST — if the election were held today, based on current polling averages blended with prediction-market odds. Not a calibrated election-day forecast.
D keeps/takes control
37%
R control
63%
Mean D seats
49.9
Median D seats
50
Chamber control probability
Democrats 37%Republicans 63%
Seat distribution across 1,000 simulations
Baseline: 47 safe D seats + 48 safe R seats; 5 competitive races simulated. Democrats need 51 seats for control.
Our simulation vs prediction markets — P(Democratic control)
Our model
37%
Polymarket
43%
Market odds also feed the per-race blend (weight 25% market / 75% polls).
| Race | Polling margin | P(D) polls only | P(D) blended | Polymarket | Kalshi |
|---|---|---|---|---|---|
| ArizonaGallego (D) v. Lake (R) | D +4.4 | 79% | 79% | — | — |
| NevadaCortez Masto (D) v. Laxalt (R) | R +1.0 | 43% | 43% | — | — |
| GeorgiaOssoff (D) v. Collins (R) | R +1.0 | 43% | 54% | 86% | — |
| MichiganPeters (D) v. Rogers (R) | D +3.0 | 71% | 71% | 72% | — |
| PennsylvaniaFetterman (D) v. McCormick (R) | R +1.0 | 43% | 43% | — | — |
How the simulation works
- Each race’s polling average margin is converted to a win probability using a normal error model: a national error (σ = 3) shared by every race plus an independent per-race error (σ = 4.5), sized to historical Senate polling misses.
- Prediction-market odds are blended in at 25% weight — markets aggregate information polls miss, and the weight is a tunable model parameter.
- We then simulate 1,000 elections. The shared national error is what makes sweeps (all close races breaking one way) more likely than independent coin flips would suggest.
- Today this answers “if the election were held now”; as backtesting across past cycles completes, it will mature into a full election-day forecast.
Last updated: Jun 20, 2026, 11:56 AM
TRACKER — weighted polling averages only, not a forecast