By October 26, 2026, a gallon of gas cost about $1.20 more than when Americans last voted, in November 2024. That rise weighed far more on a household that drives long distances on a modest income than on one in a well-off suburb, and it was larger in some places than in others.
One election can hardly tell us how many votes this cost the president’s party, because everyone experienced the rising cost. But I can use these differences to ask a narrower question: did the places where people felt the rising cost most at the pump turn further against the president’s party than comparable places? This page explains how the analysis is set up, what it can and can’t tell us, and why simple comparisons can’t answer either question. I registered the analysis plan on November 2, before the election, and published the code that will produce the results. Results will appear here from November 4.