Ignoring baseline partisanship and going solely by race and education-based splits, Democrats performed about 8% below expected in Florida in 2020.
The struggles with the Cuban community are well-documented, but they also perform *way* worse than expected among college whites.
I see the white retiree issue, but here's the thing: it's not just Democrats struggling with them; that's not the underperformance area that stands out to me. There's a persistent underperformance among college-educated whites that cuts against the state's demographic profile.
I don't think there's honestly much to instill confidence in me that we reverse this, but if they wanted to do it, there's a path, even if it's very hard. 60% of Floridans are pro-choice, 73% are pro-climate action. 67% are majorly pro-criminal justice reform.
There's a clear path to flipping these voters, even accounting for the retirees issue, by focusing on those policy areas and building a competent statewide machine that ties FL Republicans to the national GOP platform and positions Dems well, but it's Florida Democrats, so...
Yeah, and also an under-discussed factor is how the panhandle is zooming left because it behaves like the rest of the Deep South instead of the rest of Florida.
Given educational polarization, what if you tried to predict 2020 2-party margin based on nothing other than education split by race (white college, white non-college etc)?
You'd still get a pretty strong correlation. The resulting over/underperformance map is interesting to see
Some thoughts...
[1/] Secular/non-evangelical voters are ones that Democrats perform really well with, which obviously isn't captured with just educational data
[2/] An area's baseline partisanship plays a huge role in determining immediate margins, so I wouldn't take this as gospel or whatever (none of the "OMG Wisconsin's sliding 20 points right in 2022" dooming) -- just a fun experiment to see what'd happen if the trend continues
I've reconstructed the 2014 and 2018 electorates by demographic, thanks to @DKElections and @Catalist_US data and a lot of math
-Midterms are whiter, more educated (~2pt Dem boost with whites on education-based turnout differential)
-Minority turnout is a crucial wildcard
[1/]
Midterms are generally whiter + more educated; whites are ~2% more favorable to Dems on educational splits alone. This was more pronounced in 2018 than it was in 2014. If R voters are increasingly tied to Trump ballot presence, it could complicate things for the GOP. [2/]
I calculate the electorate demographic composition for 2014/2018 myself and project 2020 support by demographic onto each electorate to get an idea of what its partisanship would be now. 2012/16/20 demographic composition & 2020 2-way support by demographic are from Catalist [3/]
THREAD: Simulating an RCV election in Alaska, we see that running a Democrat probably helps Kelly Tshibaka more than anything. But the value Lisa Murkowski provides to a Democratic majority is minimal, and the expected value of running a Democrat is still higher, IMO
Let’s construct a grossly simplified scenario where we have Tshibaka (R) at 40%, Galvin (D) at 30%, Murkowski (R) at 30%, and Murkowski loses the second spot by a hair to Alyse Galvin. Now you go to the H2H...
Does Galvin get 66% of Murkowski’s voters to back her as the second choice? Possible...but a tall order...so you’ve just given Tshibaka a huge boost here.
Conversely, would 66% of Galvin’s voters rank Murkowski as a second choice? That’s much easier to imagine.
If the goal of the Democratic Party is to retain the majority in 2022, then funding or helping Lisa Murkowski makes absolutely no sense, because the value-over-replacement she provides to a Senate Democratic majority is minimal.
If Democrats think they've certainly lost the Senate in 2022, then helping to keep Murkowski might make a lot of sense.
If they think they've got a good chance to retain the Senate (as they do, given the map they're playing in), then keeping Murkowski provides no utility at all.
Any bill Murkowski goes for, all Democrats would have already supported, including Manchin. There is no use to a Democratic *majority* here, especially when you have a lane to elect Galvin, who'd be at the party median and cut the reliance on Manchin/Sinema by a fair bit.
Murkowski's 2022 odds are honestly not nearly as high as everyone thinks they are and I think it's not unreasonable to say that come November, she may not be the favorite to make it out of the field.
You can call Alaska likely/safe R, but it's not likely/safe Murkowski.
.@EScrimshaw breaks it down here, but because of the way RCV works, Tshibaka poses a very, very serious threat, especially given the amount of campaigning Trump will do for her against Murkowski. scrimshawunscripted.substack.com/p/2022-murkows…
That *does* open up an outside lane for a Democrat (I think @ElpisActual has discussed it as well) in which you could have Tshibaka (R) at 40%, Galvin (D) at 30%, Murkowski at 25%, and a random Independent at 5%.
And Galvin could edge out Tshibaka in a H2H there with RCV.
Also if you like things like this go follow @notkavi and our bot @bot_2024 — kavi does a lot of great modeling and work and programmed the bulk of that bot.
As I said in the replies to the original thread, I think the lack of demographics available to our bot (because of a lack of data) makes this estimation a bit susceptible to favoring Dems too much in some elections, but the overall picture is largely correct.