The races that decide the chamber
Each card shows our forecast for the key 2026 races — who wins, in what share of 50,000 simulations, and the median margin — built from the weighted polling average, fundamentals and the national environment. Click a card to chart the race; toggle the experimental NYT vibes adjustment or compare against Polymarket and Kalshi. For the chamber-control total, see the Senate Forecast.
Our forecast has Dan Sullivan winning over Mary Peltola in 67% of simulations, with a median margin of R+2.2 (80% range R+9.0 to D+4.5).
Our forecast has Jon Ossoff winning over Mike Collins in 88% of simulations, with a median margin of D+6.2 (80% range R+0.6 to D+12.8).
Our forecast has Ashley Hinson winning over Josh Turek in 62% of simulations, with a median margin of R+1.5 (80% range R+8.1 to D+5.3).
Our forecast has Troy Jackson winning over Susan Collins in 77% of simulations, with a median margin of D+3.6 (80% range R+3.1 to D+10.4).
Our forecast has Haley Stevens winning over Mike Rogers in 86% of simulations, with a median margin of D+5.5 (80% range R+1.3 to D+12.3).
Our forecast has Chris Pappas winning over Scott Brown in 93% of simulations, with a median margin of D+8.2 (80% range D+1.5 to D+15.0).
Our forecast has Roy Cooper winning over Michael Whatley in 91% of simulations, with a median margin of D+7.1 (80% range D+0.3 to D+13.8).
Our forecast has Sherrod Brown winning over Jon Husted in 64% of simulations, with a median margin of D+1.7 (80% range R+5.1 to D+8.4).
Our forecast has Ken Paxton winning over James Talarico in 55% of simulations, with a median margin of R+0.6 (80% range R+7.3 to D+6.2).
How the NYT “vibes” component works
The vibes layer is an experimental adjustment on top of the base polling average. It tries to capture momentum and scandal effects that polls pick up only with a lag:
- We pull each candidate’s recent New York Times coverage (Archive API) and extract every sentence that mentions them.
- A sentiment model scores each mention, producing a five-point tone bucket from overwhelmingly negative (−2) to overwhelmingly positive (+2), plus a 0–1 scandal-severity score from curated triggers (indictment, ethics investigation, resignation, …).
- Each candidate’s effect on the margin is
0.4 × tone − 2.5 × scandalpoints; the race adjustment is the Democrat’s effect minus the Republican’s, capped at ±3 points so vibes can nudge — never overturn — what the polls say.
The coefficients are conservative priors, not fitted values; they will be backtested as labelled cycles accumulate. When coverage data hasn’t been fetched yet, the toggle is disabled and the base average is shown.
A work in progress from the team at Policy y Peaches. Learn more here.
Last updated: Sep 20, 2026, 12:01 AM