Chick-fil-A, out of the main culture-war conversation for a few years now, came roaring back this week with a bit of advocacy from the senior senator from South Carolina.

       Incensed that students at the University of Notre Dame were objecting to the introduction of one of the chain’s franchises on campus, Sen. Lindsey O. Graham (R) tweeted his strong support for the chicken sandwich purveyors.

       Support our journalism. Subscribe today. arrow-right

       “Going to war” for the principles of a fast-food chain is quite a bit of hyperbole, even understanding that the chain is more political than most (leading to the sort of blowback seen at Notre Dame). But Graham might be surprised to learn that counties that have welcomed Chick-fil-A franchises are home to more than three-quarters of Americans — and that they voted for President Biden by a 10-point margin.

       The chain’s website lists all of its franchises, nearly 2,500 in total. More than half are in states former president Donald Trump won last year, thanks in large part to the nearly 500 stores in Texas and Florida alone. But the state with the most stores per capita was won by Biden: Georgia, where there are 22 stores for every million residents.

       Looking at the distribution of the stores at a county level (using an admittedly imperfect translation of Zip codes to county codes), we see how heavily weighted the distribution of stores is to Democratic-voting areas. In state after state, there are far more Chick-fil-A franchises in Biden-voting counties than Trump-voting ones.

       Shown visually, we see an interesting complicating factor. There’s a small positive correlation between the number of stores in a county and the margin by which it supported Biden in last year’s election.

       So we have about 54 percent of the country living in one of those counties that both has a franchise and backed Biden.

       Advertisement

       Story continues below advertisement

       But you are not dumb, and you are already thinking: Well, sure, they put a lot of stores in big cities, places that were more likely to vote for Biden. And you are correct.

       If we adjust for population at the county level, the correlation flips. There’s a slight positive relationship between the number of stores per person in a county and the margin by which they preferred Trump. That graph looks like this.

       When we shift the map to represent that per-resident number of stores, we end up with this much-redder map. Some of those big circles are a function of small populations, but this is how such ratios work.

       The politics here are weird. Counties without a store voted much more heavily for Trump in 2020, with eight Trump counties for every Biden one. Counties that have a franchise preferred Biden by a smaller margin, even though there are two Trump counties for every Biden one. That’s because there were two Biden votes for every Trump vote in those places.

       In counties that had a below-median number of stores (for those counties with a store), Trump was a narrow winner. But counties with a below-median number of stores per resident, Biden won.

       Then there’s South Carolina. Less than a fifth of counties in the state even have a franchise, and the counties that do cast about 30,000 more votes for Biden than Trump.

       Yet Graham probably doesn’t intend to become a champion of the blended values held by the areas of his state in which Chick-fil-A is actually located.