Joe really needs to hide in his fucking basement for the next 19 days.
2 Likes
They learned late last night, so it was at the most a week after the debate… Well within the incubation period.
This would be relevant if they polled the entire electorate. But they’re not. They are:
- Sampling some portion of the electorate, but not simply taking the % of Biden/Trump/undecided voters in that sample.
- Applying weights to different groups in that sample to construct a pseudo-sample population that corresponds to the expected voting public.
So suppose their sample of 500 includes 50 (10%) white non-college educated males, and 40 of those 50 (80%) choose Trump. Of the remaining 450, 250 (55.5%) choose Biden. So the in-sample split is 260/240 (52%/48%) Biden. But they don’t report the 52%/48% Biden split as their result. Instead, the pollster then needs to estimate the percentage of the final electorate made up of white non-college educated males and apply the 80% to that estimate. So if they think white non-college educated males are actually going to make up 20% of the final votes, their split would be: (20%)(80%)+(80%)(44.44%) = 51.55% Trump/48.45% Biden.
What that means is that two pollsters using literally the same polling data could generate Biden/Trump/undecided splits of:
50/42/8
45/45/10
42/45/13
etc.
depending on what they think the final voting composition will be.
So I know it’s tempting to say that anything over 50% Biden is bulletproof, because no amount of undecided Trump lean matters. But that 50% number is just an estimate.
This is a really good article talking about this issue:
Punchline:
Polling results rely as much on the judgments of pollsters as on the science of survey methodology. Two good pollsters, both looking at the same underlying data, could come up with two very different results.
How so? Because pollsters make a series of decisions when designing their survey, from determining likely voters to adjusting their respondents to match the demographics of the electorate. These decisions are hard. They usually take place behind the scenes, and they can make a huge difference.
To illustrate this, we decided to conduct a little experiment. On Monday, in partnership with Siena College, the Upshot published a poll of 867 likely Florida voters. Our poll showed Hillary Clinton leading Donald J. Trump by one percentage point.
We decided to share our raw data with four well-respected pollsters and asked them to estimate the result of the poll themselves.
2 Likes
I have to say, while he did not persuade me, he was more persuasive than I expected. That said, it seemed to me like his basic premise is still something like the “Shy Trump Boter.” He goes into other things, but the two things that he said that I admit kind of scare me are
- that if you ask a few good questions (like “Are you in favor of fracking” in Pennsylvania and other similar questions that, if answered Yes, would show you’re aligning with Trump), you’ll find that a lot of so-called undecided voters are clearly aligned with Trump
- If you ask people who they think their neighbors are voting for, you sometimes see a significant (and he says predictive) swing. He says this is the reason why he accurately called the Gillum/ DeSantis election in 2018. Because the “neighbor” question showed a 5-point swing (due to racism, I guess) while there was no similar swing in the FL Senate race that he was also polling.
ETA: I’m gonna leave the Boter typo, duh
1 Like
Shy Trump Boater
2 Likes
SC is an inelastic state. Trumps numbers here aren’t relevant. I’m interested in Jaime.
Wow. Wonder what he said 
1 Like
This is the longest fucking month or my life.
8 Likes
By the way, I should mention that the title of this thread is absolutely A++ would read again.
3 Likes
I think the same every time I see it. Gets a chuckle from me without fail.
you laugh but we’re one election result away from that guy replacing nate silver.
It’s happening 
Political Polls (@PpollingNumbers) Tweeted:
National Poll:
Biden 54% (+13)
Pence 41%
University of Massachusetts Lowell (LV, 10/5-10/12)
I didn’t know Joe’s middle initial was R.
JRBs!
1 Like
That’s some quality Vincin’ (this is now a new verb FYI).
2 Likes