Inspired by the questions raised by this graph posted by ZeroHedge:
First, I'll acknowledge a few things about the chart. The y-axis range is bound limited which has the effect of exaggerating the change over the years. Also, others have pointed out correctly that the 2024 year doesn't have complete data yet.
Nevertheless, none of those critiques take away from the core rhetorical question: what explains the gap of votes for Biden over 2016 in the 2020 election?
The question that intrigued me:
The NYT visuals that show the "shift left/right" vectors on a map got me thinking about how to use these techniques to identify where those votes came from:
