I think Biden polling 8%+ behind in swing states that he won four years ago is meaningful. It's not so decisive that it means Trump is absolutely going to win, but it's definitely not a good sign for the Dems.

[–] AlkaliMarxist@hexbear.net 35 points 9 months ago (1 children)

It does give an indication of who'd win if the election were held right now, but it's only so reliable. Right up till the last minute in 2016 the polls had Hillary far ahead of Trump, but we know how that turned out. Still, that year was somewhat of an outlier and they're usually not that far off. Really it depends on the methodology of the individual poll, inherent biases, sample sizes, demographics, etc.

These things can also be deliberately manipulated by pollsters to give the results they want.

[–] WalrusDragonOnABike@reddthat.com 7 points 9 months ago

Up until the last minute? Even exit polling in numerous states were far different from the actual tallies, enough that the results were expected the be the opposite based on that data the day of election.

[–] tactical_trans_karen@hexbear.net 30 points 9 months ago

Depends, when the tea leaves are correct, they matter. But when they are incorrect, they do not matter. think-about-it

[–] emizeko@hexbear.net 20 points 9 months ago

I think comparing the same polls between months can have some useful info, assuming nothing else has been changed in bad faith by the pollster (hard assumption once you've seen through the fake puppet dance of electoralism and how all of the institutions are really only for bourgeois self-defense)

[–] an_engel_on_earth@hexbear.net 20 points 9 months ago* (last edited 9 months ago)

An aggregate of several polls taken around the same (more or less) timeframe can certainly give a pretty realistic picture of how it's looking out there for a certain candidate. You shouldn't base any conclusions around one single poll

[–] Egon@hexbear.net 15 points 9 months ago

Toonworld rules: It matters if it's funny.