The Polls Got It Right In 2022. Here Are The Pollsters To Trust. | FiveThirtyEight
HTML-код
- Опубликовано: 9 мар 2023
- The pollsters got it right in 2022, at least some of them did. Senior Elections Analyst Nathaniel Rakich breaks down what goes into FiveThirtyEight’s pollster ratings.
Website: fivethirtyeight.com/
Merch: fivethirtyeight.com/store
Twitter: / fivethirtyeight
Facebook: / fivethirtyeight
Podcast: itunes.apple.com/us/podcast/f...
“Of course, I was a straight-A student”
*Holds up essay predicting a McCain victory in 2008
*snort*
How old was Nathaniel Rakich in 2008? Even twenty?
@@racookster Not if he was in high school getting letter grades.
🤣
If that's a one dollar pizza slice, I'll take it!
Don't diss out my $1 pizza slices like that😭
This was very informative, and fun to watch. Thank you.
Hello Professor
Thank you so much for your interesting News,
i do appreciate your job. I wish you peace and happness under the sky of prosperity. All the best. Take care and have a good time.
Your Student from Algeria.
Don't trust polls? well, we have the polls that show you what polls are the polls to trust. I think that I shall simply die.
That pizza looks really good
I thought Trafalgar was more biased than that, but I'm not crunching the numbers on every poll they conduct. Nathaniel is. I'll take his word for it. Trafalgar seems to be more than two points off in races I care about, but that's a form of selection bias on my part.
I think they got a break from the Trump effect in 2016 and 2020. If I'm not mistaken they were a little off in 2022.
@@matthewdrews Trafalgar was off in 18, 20 and 22. They were most off in 22. In WA and CO they were off by double digits. I remember they were off by like 18 points in our CA recall election too. It feels like Trafalgar makes most of their polling statistical ties while giving Republicans the edge. The saying “a broken clock is right twice a day” is the best way to describe Trafalgar’s 2016 accuracies. If being within the margin of error but getting the winning candidate wrong still counts as being accurate then I guess that’s how Trafalgar gets a passing grade. It’s clear as day they don’t do real polling though.
This is great!
I luv how poor pollsters get a slice of pizza rather than 💩
So they did poll ratings on polls rating. Interesting.
Why would you use a very edible food, that’s likely more appealing to a decent portion of viewers, to illustrate that point? 🤷🏻♂️ Just as in polling, presentation is important.
One dollar pizza has saved many lives, lets use some other metaphorical junk food.
@FiveThirtyEight $100 says that wasn't $1 pizza. That was the slice you were fiending to eat after the shoot. If I'm wrong, I'll pay you at the bell house 😉
I guess most of the final generic ballot polling was close. Some of the state polls were pretty off though. It probably is a lot like 2016 where most of the polls were pretty close but we remember the media declaring a very different outcome for months so it just feels like the polling was way off. I remember even 538 was forecasting a red wave in most outcomes too though. So you all played a part in the misperception of the polls being so off. I guess from now on I’ll just ignore what the media’s punditry, take the polling with a grain of salt and pay closest attention to recent special elections. The special elections from the summer were completely ignore once the media decided there was going to be a red wave. That was a big mistake.
#1: Rasmussen
"The more reliable pollster..." seems like an opinion presented as a fact.
Race to the Wh >
Ok
Y’all lying bruh
That was a terrible analogy, because most people eat things based on how they taste & not on how healthy they are. The vast majority of people would choose the $1 pizza slice. Also, the polls were not accurate in 2022, and trying to pretend they were is truly cringe.
Most people also prefer to believe polling that reinforces their beliefs (the pizza) than that which is actually accurate (the good polls). The polls were as accurate as they get in 2022. It was media punditry that was wrong.
BRO YOU PEOPLE GOT THE SENATE AND 50 SEATS WRONG LMAOO
You clearly didn't watch the video. The way they judge pollsters (and themselves) is based on difference from net margin, not picking the right winner. Their example: in a race that ends D+1%, a poll that predicted R+1% is more accurate than a poll that predicted D+10%. Your metric "got the result right" would credit is less accurate poll and punish the more accurate poll.
Was just thinking to myself that I appreciate this channel but 538's best content is long form and discussion based. But then watching a video that is less than five minutes is still too much of an ask for a particular pack of fools.
All they had was polling which were biased towards republicans. Basically every poll had nevada going red near the end
I mean, they got all of the elections local to me that they predicted correctly. They also gave the Dem senate win a 41% chance, which is way more of a coin flip then you think it is. Ask google to pick a random number between 1 and 10 a few times - how many times is it 4 or less? Clearly not zero - so a Dem senate win was totally possible and 538 knew that.