You just can’t draw the article’s statements as even a reasonable conclusions without ignoring a whole host of things. They’re largely comparing a general election to a midterm election, Rs show up for midterms more reliably than Ds, a drop is basically expected. Even compared to the previous midterm, that was with a very polarizing R in office, so we’d also expect a higher D turnout than one with a D in office. Plus add in consideration for the effects of voter suppression that’s been building basically specifically targeted at the group in question.

This article is just bait created by someone who is either ignorantly or purposely misrepresenting the actual study’s implications.