The accompanying images depict the result, by county, of the last two presidential elections in Wisconsin—Obama versus Romney in 2012 (above) and then Clinton versus Trump in 2016 (below). Obama carried Wisconsin rather easily in 2012, but then, in 2016, Trump won the state by about 22,000 votes, a fraction of one percent of the ballots cast. The sources for the images are here and here. The sites themselves are interactive; you can click on a county and see how its residents voted.
I find this kind of data almost endlessly interesting. Also, it seems to me that the actual result refutes common "narratives" that one hears about why Trump won. If you just look at the pictures, you get a pretty good idea why Trump carried Wisconsin: in the small towns and rural parts of the state, he did better than Romney had, and not just by a little. Yet one continually hears that depressed turnout in the Democratic strongholds of Milwaukee and Madison doomed Clinton. This story doesn't stand up to scrutiny, however. Against Romney, Obama carried Milwaukee County—the city and its inner ring of suburbs—by 170,000 votes. Clinton carried the county by 163,000. I attended public schools, but, by my calculations, had Clinton matched Obama's margin in Milwaukee County she still would have lost the state—by 15,000 votes instead of by 22,000. In Dane County, home to Madison, Clinton actually outperformed Obama, winning by 146,000 compared to Obama's margin of 132,000. In the state's two biggest counties, Clinton's margin was slightly wider than Obama's had been.
But look at how much of the rest of the state switched from blue to red! Granted, the visual effect can mislead. Most of these counties have small populations, and if they switched from "narrowly for Obama" to "narrowly for Trump," the difference, in raw vote totals, would be small. But if you click around a little on the two maps I think you will quickly conclude that this is not what happened. Depressed turnout in the population centers did not cost Clinton Wisconsin. She suffered a meltdown across the rest of the state.
Dave Wasserman, of Cook Political Report, has devised a rather dramatic way of demonstrating the tectonic shift in Wisconsin's vote. The state has 72 counties. He eliminates five and looks at the aggregate result in the other 67. The five he sets aside are Dane (Madison and environs), Milwaukee, and three others that are unambiguously "suburban Milwaukee"—Ozuakee, Washington, and Waukesha. Obama outpolled Romney in the 67 counties of "non-metro" Wisconsin by 36,000 votes. Then, four years later, Trump carried the same 67 counties by 228,000. A shift of 264,000 votes! Obama had carried the state by 205,000.
What's happened in Wisconsin has happened across the Great Lakes region, from Pennsylvania to Minnesota. Meanwhile, as Sun Belt metros grow, their states are turning purple or blue—think Denver, Colorado; Las Vegas, Nevada; Phoenix, Arizona; Austin, Texas; San Antonio, Texas; Charlotte, North Carolina; and Atlanta, Georgia. Broadly speaking, the rural parts of the North are voting more like the rural parts of the South (overwhelmingly Republican), and southern metros are voting more like the northern ones (overwhelmingly Democratic). I said it's "interesting" to track how people vote. It's alarming, too. The country is splitting apart, not by geographic region, as in the middle of the nineteenth century, but by geographic region within states.
Comments