Strange piece and report, perhaps reflecting some of the inconvenience of Trump's gains among nonwhite voters in a high turnout election.
But I'd like to note some of the numbers in the report, which speak to real '20 data challenges
But I'd like to note some of the numbers in the report, which speak to real '20 data challenges
https://twitter.com/RonBrownstein/status/1446139963397926915?s=20
To take an obvious case, this report appears to think that Democrats made gains among Hispanic voters in most states--including Florida! 
This is clearly based on pre-2020 voter file partisanship modeling, like party registration or primary vote history, without any effort to incorporate polling or results from 2020
That interpretation yields weird things in other sections, as well. This has Coloradan whites leaning R, though Biden won them. They have Dems losing among ground white Pennsylvanians, bc of voter reg shifts, even tho Biden gained
Anyway, this voter file data useful is great for weighting and individual level targeting. But the story from aggregate voter file data, based on old measures of partisanship like registration, may yield a weird picture post-2016
In the end, this is an example of how some serious data limitations / questions about the Democratic coalition post-2020 make it more difficult for the party to game out its future
Clearly, pre-2016 data supported the assumption that a high turnout, mobilization election would be great for Democrats. It didn't turn out that way, as we all know. And it's not clear that holds going forward.
Given overall 2020 polling errors, we don't have much reason to be confident that we have a grip on low turnout voters right now. We don't know exactly why Trump surged with Latinos.
It makes it harder for Democrats to craft a strategy today
It makes it harder for Democrats to craft a strategy today
• • •
Missing some Tweet in this thread? You can try to
force a refresh
