Text polling, with a simple one-option-or-the-other answer, would actually be a lot more accurate than the way they do the polls in reality, I think.
I’m not saying that the polls are necessarily biased either for or against Biden, just that the methodology is so laughably poor that the polls don’t particularly mean anything. I dug into this at some length a few days ago and found that for a handful of recent randomly selected elections, the polls were off by an average of 16 percentage points.
Ehhhhhhh it’s pretty much always been the case that a single poll by itself is very sketchy data. You gotta look at multiple independent polls. If you want to actually predict the election they had all better be from the last few days before voting ends, because most voters don’t make up their mind until the last second.
I’m not talking about single polls, I’m talking about the aggregate of all the polling for the election (although there was one election with only one poll existing for it, the ones that were aggregates of multiple polls were just as bad). It’s not like one poll said +26 and one said +4. They’re usually within a few percentage points of each other.
And these were polls from much much closer to the election than all these polls the media is treating as a big deal for the November election.
Imagine if those were the polls that show Biden behind.
Text polling, with a simple one-option-or-the-other answer, would actually be a lot more accurate than the way they do the polls in reality, I think.
I’m not saying that the polls are necessarily biased either for or against Biden, just that the methodology is so laughably poor that the polls don’t particularly mean anything. I dug into this at some length a few days ago and found that for a handful of recent randomly selected elections, the polls were off by an average of 16 percentage points.
Ehhhhhhh it’s pretty much always been the case that a single poll by itself is very sketchy data. You gotta look at multiple independent polls. If you want to actually predict the election they had all better be from the last few days before voting ends, because most voters don’t make up their mind until the last second.
I’m not talking about single polls, I’m talking about the aggregate of all the polling for the election (although there was one election with only one poll existing for it, the ones that were aggregates of multiple polls were just as bad). It’s not like one poll said +26 and one said +4. They’re usually within a few percentage points of each other.
And these were polls from much much closer to the election than all these polls the media is treating as a big deal for the November election.