Progressive Sen. Elizabeth Warren (D-Mass.) announced Wednesday that there are currently enough votes in the Senate to suspend the filibuster to codify Roe v. Wade and abortion rights if Democrats win control of the House and keep the Senate and White House.
“We will suspend the filibuster. We have the votes for that on Roe v. Wade,” Warren said on ABC’s “The View.”
She said if Democrats control the White House and both chambers of Congress in 2025, “the first vote Democrats will take in the Senate, the first substantive vote, will be to make Roe v. Wade law of the land again in America.”
Was 538 wrong or do people just not understand statistics?
538 is wrong most of the time. Nate Silver has gone back to claiming none of his work is designed to predict outcomes, he’s “just running stats” now 🙄
Whatever you think of him, know his models didn’t get a thing right with regard to elections after the Roe v Wade issue came back to light. The cycle goes like this: his data is wrong, he tells everyone it’s correct, then he writes some bullshit explaining how everyone else is stupid for reading his own published data wrong, but it was actually right in the end.
Just take everything with a handful of salt unless there’s an obvious change affecting the numbers.
I’m not sure that’s correct. 538 was always a polling aggregator, but people treated it like 60% chance means “for sure.” I think what we’re now seeing is we don’t actually have much good polling data due to extremism, and therefore sites like 538 aren’t as valuable.
I distinctly remember Silver refused to make a prediction on who would win in 2016 because he insisted that Trump’s 33% odds according to 538 meant there was a very real chance of a Trump victory. But everyone came out an blamed Silver for calling it wrong.
I don’t actually like the guy, I think his analysis and political savvy is pretty weak and he comes off as incredibly arrogant. But he literally just runs a weighted data aggregator. So if the data is bad, his results will be bad too.
His models use aggregated data to create what he has shifted from calling “predictions”, to now being called “data” (as in, “the data says…”), or more recently just flat out calling them “odds”. Keep in mind he does not open source his analytics at all.
So taking that into mind, he’s just rebranding subtly, sure. His company got bought by Disney, and I’m sure they put the hammer down on the language because they are now an easy target to get sued. Fair enough. My issue is that prior to all of this, he was plainly making predictions, and used those words to say as much. He even talked at length about it, and why he started changing his own words to describe his work.
So he called them predictions the few times his data aligned with real-world outcomes, but on the downslope of his popularity in doing so, is backing away from that attitude.
Lay people will still read exactly what he’s doing as making predictions.