Detailed approval estimates from G. Elliott Morris suggest Trump’s autocratic behavior suggest a sharp drop his support in the swing regions, Deep South, and exurban Yankeedom
By Colin Woodard
Data journalist G. Elliott Morris – who headed FiveThirtyEight until ABC shut it down last year – recently posted highly detailed estimates of Trump’s approval ratings, and they’re catnip for anyone trying to figure out if the republic is going to survive its 250th birthday.
The estimates are based on Morris’ latest Strength in Numbers/Verasight poll, conducted January 14-20, as our NATO allies were mobilizing troops to deter Trump from invading Greenland and lawless federal agents were executing people on the streets of Minneapolis. The poll – which shows Trump underwater by 18 points nationally – was then cross-referenced with detailed community profiles Morris built from the results of surveys of 12,000 individual and precinct-level demographic data, using statistical techniques to create inferential estimates of his approval ratings in every Congressional District, and in the country’s 2400 Public Use Microdata Areas, an obscure Census Bureau unit containing roughly 100,000 adults. Definitely check out his post and the results of the underlying poll.
Here are my two additional cents.
The regional patterns, are clear and striking, and they give some of the best indications yet of how Trump’s fully unveiled authoritarian ethnonationalism is being received by the public in the distinct regional cultures identified in American Nations. If Morris’ estimates are accurate, Trump and his enablers are in very deep trouble.
The American Nations model is powered by county-level data, so we couldn’t calculate approval estimates for each region with Morris’s less-detailed estimates, but the map tells a clear story on its own. At the top of this post is a map of his estimates with the American Nations boundaries overlayed in bold.
For comparison, these are the county-level results of the November 2024 presidential election:
Notice the dramatic shift in Trump’s support across El Norte, Tidewater, the Deep South, Yankeedom, and swaths of the Midlands and Far West. Morris’s estimates have Trump faltering in areas he won just over a year ago, including almost the entirety of Iowa, the Yankee Upstate of New York, the southern tier of Yankee Wisconsin, the Far Western interiors of California and Washington State, the Carolina lowlands, and Texas’ Lower Rio Grande Valley. From southern New Jersey to Central Alaska, the president is hemorrhaging support.
At the same time, Trump remains popular across Greater Appalachia, the most populous of the regional cultures, whether one is in southern Illinois and Ohio, northwestern Arkansas, northern Alabama or Central Texas. For historical reasons discussed in detail in Nations Apart, ethnonationalism finds fertile soil in this region, which has always been the core of Trump’s support. (He won there by more than 20 points in all three of his elections.) He’s holding on in parts of the Far West and Midlands – check out the High Plains, Rocky Mountains proper, and Central Ohio and Illinois – but if these were election results, Democrats would likely be flipping both these regions into their column.
Of course, one should take this with several grains of salt. These are inferential estimates based on a statistical model and driven by a national poll of less than 1600 Americans – a solid sample for gauging national-level public opinion, but by no means definitive for sub-state analysis. The model is based on assumptions that may not turn out to be true. (Who would have predicted in mid-2020 that Tejanos would flock en masse to a president who thinks Americans of Mexican descent shouldn’t sit on the federal bench?) The margin of error for most of the individual microdata areas is between 7 and 12 percent, theoretically enough to switch scores of them into positive territory. And a lot of people might disapprove of Trump, but might support him anyway because they dislike his opponents even more.
On the other hand, Morris’ estimates are consistent with what we’ve seen in recent months whenever voters have gone to the polls, with Democrats greatly over-performing 2024 across the country, including special election wins in ruby red legislative districts in suburban Fort Worth, the Virginia Tidewater, Pennsylvania Midlands, and Deep Southern Georgia. If his approval estimates are even half right, Trump may finally be uniting much of the country: against him, and in defense of democracy and the rule of law.
— Colin Woodard, author of Nations Apart: How Clashing Regional Cultures Shattered America and American Nations: A History of the Eleven Rival Regional Cultures of North America, is director of Nationhood Lab.