The archive
What changed
Every published run is archived. When the forecast actually moves (the chamber odds shift, a rating flips, a race drifts, new polls land), it's noted here in plain language. Runs where nothing meaningful happened keep quiet, and that's most of them.
Aug 27, 2026, 12:43 a.m. ET
Texas moves from Toss-up to Lean R
Democrats' chance of the seat fell from 41 to 40 in 100.
University of New Hampshire (Aug 20–24): Ed Markey 49, John Deaton 30.
Aug 26, 2026, 8:01 p.m. ET
New Hampshire moves from Likely D to Lean D
Democrats' chance of the seat fell from 86 to 77 in 100.
South Carolina shifts toward Democrats
Democrats' chance of the seat rose from 6 to 12 in 100.
University of New Hampshire (Aug 20–24): John E. Sununu 45, Chris Pappas 43.
Impact Research (Aug 18–24): Annie Andrews 41, Darline Graham 41.
Aug 26, 2026, 2:03 p.m. ET
Democrats' chance of Senate control fell to 38 in 100
Down from 39 in 100 in the previous run.
Aug 26, 2026, 10:33 a.m. ET
South Carolina shifts toward Republicans
Democrats' chance of the seat fell from 11 to 6 in 100.
Nebraska shifts toward Democrats
Democrats' chance of the seat rose from 20 to 23 in 100.
Aug 25, 2026, 2:29 p.m. ET
First published forecast
This is the model's first published run; future runs will note what moved.