Era V · 2008–2024
The Polarized Era
Calcified coalitions, the diploma divide, and county landslides under close national margins
The modern era's paradox is visible in one comparison: national margins have been close in every election since 2008 — nothing beyond Barack Obama's 7.2 points — while the typical county has become a landslide. In 2008 Obama carried the county-summed two-party vote 54–46; by 2024 nearly four in five counties were decided by 20 points or more. The map stopped moving even as the elections stayed close, because the sorting went local: place itself became partisan.
The engine of the era is education. Run the almanac's county regression for 2000 and the college-share coefficient is close to zero; by 2016 it rivals race as the strongest predictor on the board, and in 2020 and 2024 it leads outright — counties with more college graduates lean further Democratic, holding income, race, age, and rurality constant, in nearly every state. The diploma divide replaced the class politics of the New Deal within two decades, and the scatter view shows it reorganizing the whole cloud of three thousand counties.
National popular-vote margin, 2008–2024. No election since 1984 has been decided by double digits — ten straight, the longest such run in the record.
Each cycle contributed a chapter. 2008 put together the high-water Democratic coalition — young, diverse, metropolitan — and 2012 held it. 2016 broke the blue wall on the strength of the non-college white counties of the industrial Midwest: Donald Trump won the electoral college 304–227 as cast (306–232 pledged, with seven faithless electors, the most defections since 1872) while losing the popular vote by two points — the era's second split verdict. Evan McMullin's conservative protest run that year left a genuine third-party mark in the returns: over a third of the vote in a handful of Utah and Idaho counties, the largest 'other' shares in the modern file.
2020 set a turnout record — more than 158 million votes — and moved the map barely an inch; the almanac's swing view is a study in small arrows. 2024 moved it more: Trump won the popular vote outright, 49.8 to 48.3, with broad small swings toward Republicans across urban and Latino counties, and the record's newest geographic quirks arrived with it — Alaska reporting by state-house district and Connecticut by planning region.
Two-party Democratic share in Virginia and Colorado against West Virginia and Ohio, 2000–2024. High-education states climb as lower-education states fall away — the county regression's coefficient, drawn on the map.
Whether 2024 closes this era or extends it is not a question data can answer yet. What the record shows is the shape of the present: two coalitions of nearly equal national size, sorted by education and density down to the county line, contesting a shrinking handful of places — while the other 3,000 counties vote like they always have, only more so.