Everything Crosstab publishes is computed from public data with the rules below. The same code runs in the app, the emails and the social posts.
Crosstab compiles publicly released polls from pollsters' own releases and public poll listings. Each poll is shown with the pollster that ran it and its release date. Partisan-sponsored polls are included and labeled when identified.
National (approval, generic ballot): the most recent poll from each pollster over the past 14 days, averaged. Races: each race's latest six polls from the past 45 days. Trend charts apply the same rule to every day.
For each pollster, the average gap between its polls and other pollsters' polls of the same contest within 30 days (14 for approval), shrunk toward zero so firms with few polls can't earn large leans. A lean describes a firm's house style relative to other polls, not its accuracy. Adjusted averages subtract each poll's lean.
Each race's lean-adjusted average is blended with a starting rating from published expert ratings (worth 1.5 polls), so a single poll can't swing a race. Uncertainty includes about 5 points of normal polling error on the margin plus more the further we are from Election Day. Senate control comes from 20,000 simulations with a shared national error, so races can miss together. Republicans keep control at 50 seats because the Vice President breaks ties.
From the forecast: Toss Up when either side's chance is 40–60%, Lean 60–80%, Likely 80–95%, Solid above 95%. Races without polls keep their starting rating.
Betting odds come from Polymarket's public prices (midpoint of the order book when it's tight), shown for information only. A gap is a signal to look closer, not proof either side is right.
Congressional records and photos from public-domain government data; campaign finance from the FEC; bills from Congress.gov; voting-record scores from Voteview (UCLA); background from Wikipedia and campaign sites. Positions are quoted from candidates' own sites, and every quote is checked word for word against the page. Crosstab doesn't rate candidates.
Straight news and pollster releases about tracked races, never opinion pieces. Articles about the same event are grouped; "Breaking" means three or more established outlets reported it within two hours. Headlines link to the original reporting; the line under each is Crosstab's own data.
AI writes some summaries and social posts. It only receives the numbers above, and any draft containing a number that isn't in the data is rejected.