Thread by @lxeagle17

A suspicion I have, based on polling and name ID, is that a lot of the reason the "ancestral Democrats" effect persisted downballot is because most people didn't even know who those candidates are but considered themselves Democrats still when name ID was stripped away.

These folks would say "yeah, the national Democrats definitely don't represent me" etc, and that's because they know who those candidates actually are (based on Fox News etc).

But when you strip away name ID, it's down to party ID, which likely tends to take longer to switch.

@JMilesColeman @gelliottmorris @baseballot @SenhorRaposa does anyone have any data to back this up or contradict it? Would love to see some actual research on it but I haven't seen anything, so this is informed only by polling and name ID.

Been thinking about this some more -- there's a few angles to this. On one hand, name ID for hyper-localized races may explain some of it.

On the other hand, when you consider things like union endorsements, that may actually explain voting Dem locally almost instinctually.

I actually think the take holds up better than I thought it initially did. Voter awareness is often low, so name ID helps make up for it. In its absence, you fall back to endorsements, party ID, and local party stances, which are way more in line with the median voter in the area

Another thing to remember is that a Democrat in West Virginia was not a national Democrat, and they took great pains to draw that distinction and took some pride in it. I mean, that's how you saw people elect George W. Bush by double digits while sneering at the R running for Gov

And so when you vote, you're not voting for the national Dems, but for the candidate who your union endorsed, for the candidate who is part of your party, the "true Dems" instead of the "national ones" etc.

I do wish I had some data on name ID and how it changes with the race.