Thread by @lxeagle17

Created a regression model to analyze 2016->2020 swing based on demographics. Much of the swing in South Texas can actually be explained with education, race, urbanization, and religion when analyzing the Sunbelt.

But even then, there was still a definite underperformance there.

Blue means the area swung more towards Biden than expected given demographics. Red means the area swung more towards Trump than expected based on demographics.

Now, a bit of analysis...and what happened is far more complex than anyone wants to believe. There's no easy answer.

To my eye, the populous centers in South Texas would have seen a pretty big swing right either way. Hidalgo and Webb should have swung 17 points right instead of the 23 and 28 point margin.

So there should have been a big swing right. But a swing of this magnitude? Maybe not.

The rural counties are where the Republicans seriously overperformed. It is here where the Texas Democrats' claims of a lack of canvassing appear to hold up strongest -- places like Zapata/Starr saw incredibly big swings right beyond expected.

But there's not many votes there.

I do question the TX Dems' claim of a lack of canvassing being the key. I don't deny that it made an impact or that it would net them several votes among low-propensity voters.

But the swings in most of the populous South TX areas would still have been huge on demographics alone

That said, even accounting for that, we're seeing a very, very serious underperformance in areas of South Texas -- 11 points in Webb and 6 points in Hidalgo. And that is an underperformance that shouldn't have happened, even with demographics, 2016, and urbanization incorporated.

This needs to be fixed if you want to compete in the state. An underperformance of 6-10 points in places like Hidalgo and Webb matters a lot if races get razor thin, which Democrats certainly hope they will over the next few years as they continue to gain with the White vote.

Is canvassing the answer to restore those votes? That's what Texas Democrats think -- I personally believe that something in messaging certainly seems to have broken, and canvassing could explain it. I'm not a field outreach person, so I'll leave that to the experts on how to win

I added in the 2016 margin as a factor in the regression. Why? Because different areas have different starting points, and it'd be foolish to ignore that, especially as areas begin to vote more in alignment with each other based on demographics.

Turnout increase was not tracked here. My sense is that given the high number of undervotes in the RGV based on the data we do have, the "low-propensity Hispanics" stuff actually did have an impact. I expect it to be rectified in '22, but it may portend long-term slippage.

Now, I'm going to add a couple qualifiers here:

1) Canvassing operations do *not* explain the entire swing in the RGV. Not even close. But Democrats in Texas really, really underperformed even relative to the rest of the sunbelt (!). And there's a lot of things for them to fix.

2) What I did was run Tikhonov regression on the sunbelt -- the question was "given what we saw in the sunbelt this year, how would we have predicted each county to swing based on the demographics of race, education, and religion?"

This is not a *forecast*.

It's funny -- I read the TX autopsy news and thought "ah, canvassing explains it". Thought a bit more and was like "No...why did every other Hispanic area swing right across the nation?"

IMO, it mattered, but wasn't close to the be-all, end-all; TX Dems shouldn't pretend it was.

I put a lot of thought into analyzing, modeling, and mapping this, so if you're going to come into my DMs to argue, I only ask that you be respectful instead of giving dismissive one-liners mocking the work or insulting others. I'm open to constructive criticism and discussion.

Immense thanks to @Mill226 for the incredible dataset, to @MattyGaren for the idea brainstorming, and to @macrotargeting, @kilometerbryman, @PviGuy, @ADincgor, @Thorongil16, @alexanderao, @PoliticalKiwi, and @AndrewPEllison for helping me analyze it.

cc @baseballot, @JMilesColeman, @gelliottmorris, @Nate_Cohn, @NateSilver538, @SenhorRaposa if anyone sees this, I'm curious about your thoughts on this and what your read is on the entire issue.

Should be clear, but another way to read it is: "Blue means area swung more towards Biden than expected, red means less towards Biden than expected" -- they basically mean the same thing, but if something is predicted to be D+17 here and it's D+15, it's a Trump overperformance.

For those who are interested, here’s the South under a regional lens

https://twitter.com/lxeagle17/status/1366424695369732098