Based on 52 data pairs (the 50 states plus DC plus the US as a whole), giving each equal weight, the correlation between the percentage of population 25+ with a bachelor's degree and percentage of vote that went to Trump is -0.83, which is very strong.
69% of the variability in Trump's percentage is explainable by the percentage of the population with a bachelor's degree.
Has any other election features such a strong relationship between educational attainment and support for one candidate?
Source? I'd be interested.
I copied the first table into Excel after sorting by Bachelor's Degree attainment, treating each of the 50 states plus DC plus the US as a whole as 52 equally weighted entities, used vlookup to paste the Trump percentage, and ran a simple linear regression (Trump percentage vs. Bachelor's degree attainment percentage). Took all of 2 minutes.
What I find remarkable is the sheer consistency of the pattern. WV, the least educated state as measured by percentage of Bachelor degree holders, was within a fraction of a percent (behind WY) as Trump's best state. Of course DC is the most educated and Trump's worst area. I suspect a similar pattern of Trump support vs. Post grad degree holders, but not high school graduates (ironically, WY, Trump's best state, also has the highest percentage of HS grads--91.8%).
