Chamber control · 2026 cycle

Who controls Washington?

20,000 Monte Carlo simulations with a national-environment swing — each iteration draws a national shift and applies it to every race in logit space before drawing outcomes. Within a sim races are independent; across sims they correlate via the shared swing.

Senate

35 races
P(D CONTROL)
1.1%
P(R CONTROL)
98.9%
1.0999999999999999 D0 TOSSUP98.9 R
EXP D-SEATS 4580% CI 4146
384945

House

443 races
P(D CONTROL)
0.0%
P(R CONTROL)
100.0%
0 D0 TOSSUP100 R
EXP D-SEATS 15880% CI 135179
132182158

Governors

38 races
P(D CONTROL)
94.0%
P(R CONTROL)
6.0%
94 D0 TOSSUP6 R
EXP D-SEATS 2880% CI 2529
223228

Original vs Bayesian-shrunk thresholds

From the calibration page
Chamber P(D) original P(D) recalibrated Δ pt Exp D-seats orig. Exp D-seats recal.
Senate 1.1% 0.3% −0.8 44.6 44.0
House 0.0% 0.0% +0.0 157.5 147.8
Governors 94.0% 91.1% −2.9 28.1 27.6