this post was submitted on 17 Jul 2024
99 points (100.0% liked)

Politics

10180 readers
185 users here now

In-depth political discussion from around the world; if it's a political happening, you can post it here.


Guidelines for submissions:

These guidelines will be enforced on a know-it-when-I-see-it basis.


Subcommunities on Beehaw:


This community's icon was made by Aaron Schneider, under the CC-BY-NC-SA 4.0 license.

founded 2 years ago
MODERATORS
 

No matter what the polls ever say, the most important thing to do is vote and encourage others to vote.

all 13 comments
sorted by: hot top controversial new old
[–] theangriestbird@beehaw.org 43 points 4 months ago (3 children)

Personally, I don't like to trust commentary on FiveThirtyEight's tracker from anyone other than FiveThirtyEight. They do a fantastic job of tempering expectations and not putting too much stock in momentary changes in the tracker. The current numbers are basically the same as they were on May 11:

It hasn't changed much, even after all the wild shiz that has happened since May 11. Furthermore, the polls usually work on a delay, so they haven't yet factored in the assassination attempt, or the RNC. And often, the tracker stays at a dead-even heat largely because their statistical model factors in uncertainty from the 4 months between now and Election Day.

All of that is to say: the race is still essentially dead-even. Like Chris said in the OP, the most important thing to do is vote and encourage others to vote, no matter what the polls say.

[–] remington@beehaw.org 18 points 4 months ago (2 children)

The RNC has 6 Million LESS viewers than 2016. That's quite a swing!

[–] snooggums@midwest.social 22 points 4 months ago

Republicans still show up to vote and reliably vote party line. The only way to beat them is motivating people to vote against them in battleground states.

[–] adespoton@lemmy.ca 2 points 4 months ago* (last edited 4 months ago)

Less viewers because fewer people are engaged or because people already know who they’re voting for?

Republicans vote Republican. 2016 was an anomaly because Trump was an unknown factor and a voice for change in the RNC. I doubt many people’s opinions on Trump have shifted since 2020 and the Republican candidate was a foregone conclusion.

[–] Ranvier@sopuli.xyz 13 points 4 months ago* (last edited 4 months ago) (1 children)

Exactly, I think because races have been so close lately, and the probabilities are ending up close to 50% often, people sometimes unintentionally conflate them with poll numbers. 53% to 46% would be a massive poll lead. For probabilities though in this situation it's the same as saying they have even odds of winning. Look at those massive 95% confidence intervals, the race is in a statistical dead heat. It's kind of remarkable how steady it has been despite all the wild events that have happened.

[–] Zaktor@sopuli.xyz 5 points 4 months ago (1 children)

538 has an explanation on their site about why it's not moving. Essentially they're so far out they heavily discount current polls and events. We're effectively seeing their baseline result with some previous races and economic measures tweaking things a little.

[–] Ranvier@sopuli.xyz 3 points 4 months ago

Yes, that's true. The poll averages themselves haven't moved much either though. And the reliance on the fundamentals forecast has me nervous, but they definitely do it for a reason. When they developed the models and looked at poll history the pattern they found was the fundamentals had a big influence on what the polls would look like closer to the election and the eventual result. Polls closer to the election are more predictive than the fundamentals. Polls farther away from the election less so. There's at least some reason to think things have changed enough maybe the fundamentals aren't as fundamental for this race, but I guess we won't know until afterward.

[–] chaos@beehaw.org 6 points 4 months ago

It's a new model this year, as Nate Silver took his with him when he left 538. The new one seems to put a lot of emphasis on "the fundamentals" this far out, that is, it "thinks" that the general environment and economy and such is pretty good for the incumbent and that the polls might move in that direction by the time election day comes along. And since it's fitted to historical data, it's also implicitly assuming that this election will be similar to past elections (like, say, including a competent campaign by a candidate who can get out there and effectively communicate accomplishments and a plan for their term).

I personally think those assumptions are pretty clearly wrong this year and so I'm more inclined to base my perception of the race on pure polling averages, which are looking quite bad for Biden.

[–] Zaktor@sopuli.xyz 15 points 4 months ago (1 children)

538's win projections aren't really based on current polling or really current events at all, that's why nothing's moved despite crazy things happening in the race. They've got some other fundamentals in the model that makes them think it's all going to revert to norm and Democrats will win, but it's just their own guesswork, not anything particularly well supported.

[–] huginn@feddit.it 3 points 4 months ago (1 children)

Assassination attempts don't correlate with wins per se. Neither does debate performance.

[–] adespoton@lemmy.ca 1 points 4 months ago

Popularity doesn’t even correlate with wins. There are people who will do whatever they can to prevent Trump from being elected, there are people who will do whatever they can to prevent the radical liberal cabal from being elected, and there are those who won’t vote (for one of a multitude of reasons). Everyone else is statistically insignificant.

[–] autotldr@lemmings.world 1 points 4 months ago

🤖 I'm a bot that provides automatic summaries for articles:

Click here to see the summaryDonald Trump's chances of winning the November 2024 presidential election are declining, according to a new aggregation of polls.

It comes after a Presidential Voting Intention poll of 3,601 swing state voters by Redfield & Wilton Strategies, found that Trump's margins over President Joe Biden have narrowed since June in two key swing states: Florida and North Carolina.

In a post on their website, FiveThirtyEight said their latest predictions can be explained by the flurry of swing-state polls that show Biden is encroaching upon Trump's lead.

In the week before the shooting, national polls had Trump as the favorite to win the White House, leading narrowly in the six key battleground states of Arizona, Georgia, Michigan, Nevada, Pennsylvania, and Wisconsin.

An ABC News/Ipsos/Washington Post poll released Thursday found Biden and Trump are tied among registered voters.

Biden's fitness for office has been called into question since his shaky debate performance against Trump in June in which he muddled his words and appeared to lose his train of thought.


Saved 75% of original text.