Закреплённый твит
Gen X Public Defender
467 posts
Gen X Public Defender
@GenXPD
I don’t post here anymore. Follow me on Mastodon: https://t.co/85CMTLi6bo
Denver, CO Присоединился Ekim 2008
3.6K Подписки815 Подписчики
@JyndonBohnson @MattZeitlin Fascinating! Maybe I remember some other prediction market then! Weird: I do remember it, but it’s false.
English
@GenXPD @MattZeitlin PredictIt has existed since 2014. So you do not, in fact, remember this
English
the cool thing about betting markets is that you can consume election "news" well before the polls close
English
@matt_blackwell For those of us who are very slow, your best guess is that Black and Latino support will be more or less what it was in 2020? That's very bullish for Harris, no?
English
Given that I've been collecting this data since the summer, I guess I'll preregister my final thoughts.
Based on validated vote estimates, I think Black support will be around +75 D and Latino support will be around +20 D
Matt Blackwell@matt_blackwell
📊 Here are the (final?) Black and Latino support numbers from the last 3 presidential cycles. Takeaways: - Black support down due to a large group of polls w/ ~30% support for Trump - Latino support still very much in line with 2020 numbers, no large changes post-garbagegate
English
@krazgreinetz Is the case “the other way” that a nationwide overcorrection would produce a greater error in national polling and sun belt states?
English
I’ve thought about this a lot. Hard to explain Trump doing better nationally, better in the Sun Belt, but the same in the Rust Belt—without coming back to polling error as the explanation. But 1) this logic didn’t cash out in 2022 and 2) you could make a case the other way too.
Sean T at RCP@SeanTrende
With national and Sun Belt polls favoring Trump and the Rust Belt remaining stubbornly Dem (though the po). Again, not something I'd bet the farm on at all, but it's a story that holds together surprisingly well and hasn't gotten a ton of attention. 6/
English
@davidsacc12345 Given that thread, why do you feel good about the election in general?
English
@billscher I thought we all concluded several cycles ago that there indeed was bad intent?
English
The current RCP nat'l avg realclearpolling.com/polls/presiden… spans the last 3 weeks of polls
In past years (realclearpolling.com/polls/presiden… realclearpolling.com/polls/presiden… ) final avg covers 1 week
Doesn't mean bad intent. But to only use the last week wld change avg from DT +0.1 to KH +0.6
English
@lxeagle17 I feel like this makes the poll results more credible: their polls were telling a story throughout the cycle, and this is such a departure from that story that it seems unlikely they’d weight their way to these results.
English
The New York Times/Siena college polls today are interesting because they're such a departure from what the Times found previously. Harris has more than 270 electoral votes here, but that is based mostly on a path through the Sun Belt, which has been her worst region all cycle.
Nate Cohn@Nate_Cohn
The final Times/Siena polls of the campaign show a dead-heat, with Harris gaining along late deciders in the Sun Belt while the Rust Belt tightens nytimes.com/2024/11/03/us/…
English
@nick_field90 Are you saying they’re using different weights in the same poll? Or that they’re using idiosyncratic weights to get this split? I *want* you to be right that this a sign the polls are overstating Trump, but I don’t understand the mechanism when we’re talking about the same poll.
English
The NYT/Siena poll of Arizona has Gallego up five and Trump up five. Do we really think we'll see a nine point split-ticket spread in a swing state? nytimes.com/2024/11/03/us/…
English
All SEVEN of the battlegrounds
Politics & Poll Tracker 📡@PollTracker2024
NYT/Siena polls out TOMORROW
English
@matt_blackwell I have been following the debate over herding vs. weighting to reduce variance, but this post is over my head. Can you break it down just a little?
English
Fun fact about the central limit theorem: it holds at a very general level even if the common random sampling assumption is violated. A (weighted) average w/ large N will be roughly normal when:
(a) no one unit dominates in terms of variance &
(b) the data aren't too dependent
English
English
Seems like the self-reported early vote in a wide variety of public polls across multiple states is more Democratic-leaning than the partisan breakdown (or even partisan modeling) of those early voters would suggest? Is that right? @tbonier
Jeffrey Evan Gold@jeffgoldesq
@RonBrownstein @biannagolodryga On the other hand ... 1. No doubt why Trump filed suit in Bucks County today. They know where the suburbs are going. 2. Guarantee you the race is not going to end up 48-48% 3. Pre Trump self own PR October surprise. 4. Interesting actual vote casts in this poll.
English
@nick_field90 Was it actually on camera or just in front of a pool reporter? That could make a difference….
English
I am, once again, begging Joe Biden to stop talking on camera
x.com/gabegutierrez/…
Gabe Gutierrez@gabegutierrez
Pres. Biden tonight: “Donald Trump has no character. He doesn't give a damn about the Latino community…just the other day, a speaker at his rally called Puerto Rico a floating island of garbage?…The only garbage I see floating out there is his supporters.”
English
@nick_field90 Apart from the numbers themselves, do we have reason to think pollsters apply different weights or what-have-you for presidential and down ballot races? E.g., do we know they are weighting on recalled presidential vote for the presidential contest but not the senate contests?
English
ICYMI: My take on the polling gap between Kamala Harris and the Senate Democratic candidates
analysis.decisiondeskhq.com/thoughts-on-th…
English
Lotta people dunking on this but few for misusing the word heuristic
English
English
This POS was a warm up speaker for @KariLake & @bgmasters on the eve of the general election this year.
English
(Obviously, because Dave Matthews Band is the most Basic, normie, flyover act in existence.)
English
Over on Generation X Facebook (XBook?), people are complaining that Taylor Swift isn’t that talented.
1. Just want to say her genius is songwriting in different genres
2. generation X is jealous
3. I don’t care if Twitter is a musky hellsite.
Facebook is infinitely worse.
English
@attorneydad A meh brief in a meh case is always the one that is stressful and time-consuming.
English
Worked so hard and invested way too much time and stress on a brief and in the end it was just meh
English