Demslide or republican resilience? A political history forecast of the 2026 us congressional elections

The history book on the shelf is always repeating itself," an oft-cited line from ABBA's song Waterloo, captures the intuition that politics often shows persistence and is governed by identifiable patterns or "laws."Building on this, we take a structural approach to election prediction, moving away from the customary focus on short-term rational choice explanations, and instead incorporate a perspective rooted partly in Historical Institutionalism, emphasizing long-term party performance and governance trends, to see if Political History (PH) offers reliable guidance on the Democratic Party's prospects in U.S. congressional races.We demonstrate that ex-post models that rely on PH provide credible, accurate forecasts of both party control of Congress and Democratic seat counts in each chamber from 1946 to 2024.We extend the PH model to forecast the 2026 elections.It forecasts that the Democratic Party is likely to win Control of Congress, with a 78% likelihood of winning the House and a 66% chance of winning the Senate.While the PH model forecasts the Democratic Party is on course for victory, it looks set to fall short of a DEMSlide and instead predicts a consequential but bounded gain for the party.