Four times in the modern era, a midterm stopped being an election and became a weather event: 1994 (Republicans +54 House seats, both chambers flip), 2006 (Democrats +31, both chambers), 2010 (Republicans +63, the largest swing since 1948), and 2018 (Democrats +40). Each felt shocking in the moment. None was actually a surprise — every wave broadcast the same four signals well in advance, which makes them a useful checklist for reading 2026.
Signal one: presidential approval underwater
The non-negotiable ingredient. Clinton in 1994, Bush in 2006, Obama in 2010, and Trump in 2018 all sat in the low-to-mid 40s or worse by election day. No president with healthy approval has ever absorbed a wave; no president deep underwater has ever fully escaped one. It is the closest thing midterm forecasting has to a physical law (why it works).
Signal two: the enthusiasm chasm
In every wave year, polls showed the out-party’s voters caring dramatically more — and the specials confirmed it with actual ballots. The 2017-18 cycle was the textbook: Democrats overperforming baselines in special after special, a full year before the House flipped (how to read those signals). 2009-10 ran the identical pattern in red.
Signal three: the retirement stampede
Incumbents can smell the water rising. Wave years are preceded by abnormal retirement waves in the exposed party — members who would rather leave than lose — and every retirement converts a defended fortress into an open field. Watching which party’s members head for the exits has historically been as predictive as any poll.
Signal four: one party recruits, the other apologizes
Waves are won a year early in candidate recruitment. In 2006 and 2018, Democrats landed veterans, prosecutors, and local celebrities in districts they had no business contesting; in 1994 and 2010, Republicans did the same. The tell is the mismatch: one party fielding its dream class while the other fails to fill ballot lines or nominates liabilities.
Run this cycle through the four signals and the picture is genuinely split. The approval and enthusiasm signals point toward Democrats; the retirement and scandal attrition on the GOP side (see NC-11) fits the pattern too. But two things no previous wave faced push back: a House map engineered mid-decade specifically to be wave-proof (the scoreboard), and an out-party whose primaries produced insurgent nominees the GOP intends to disqualify wholesale. Every prior wave crashed onto a fixed shoreline. This one meets a seawall built for it — which is exactly what makes November the most interesting stress test in decades.