FiveThirtyEight
WebsiteWikipedia
124
MENTIONS
93
EPISODES
74
PODCASTS
Search complete. 124 mentions across 93 episodes found for "FiveThirtyEight".
Nov 5, 2026
Andrew HarrisonHOST
And our special guest today joins us from Marquette University in Wisconsin.
Andrew HarrisonHOST
She's a professor of politics, a regular contributor to the data journalism website FiveThirtyEight.
Andrew HarrisonHOST
And she's literally what made Milwaukee famous.
Andrew HarrisonHOST
Welcome to the pod, Julia Azari.
Pete DonaldsonHOST
I might be getting confused with the
Luke MooreHOST
guy who used to do FiveThirtyEight.
Luke MooreHOST
You don't know.
Luke MooreHOST
You wouldn't be able to tell anyway, would you? Because you can't do names.
McGraw MilhavenHOST
He's on the cusp of beating Ann Wagner, which would be another shocking upset.
McGraw MilhavenHOST
Just to sort of put this race into perspective, Nate Silver of FiveThirtyEight.com, the well-respected prognosticator, pollster, did put that race, that Missouri second district, on his watch list and said it was one of the most compelling races of the night.
McGraw MilhavenHOST
Missouri's second district.
McGraw MilhavenHOST
He put it on the watch list as one to watch.
Michael KnowlesHOST
On top of that, the New York Times said there was an 85% chance Hillary would win.
Michael KnowlesHOST
On top of that, FiveThirtyEight, Nate Silver's old outlet said there was a 71% chance that Hillary Hillary would win.
Michael KnowlesHOST
And then what happened? Trump won.
Michael KnowlesHOST
So that's a little bit of evidence to not get demoralized by these aggregator numbers where they say, OK, looking at all the races and all the polls, we now conclude there's 100 billion percent chance that the Democrats win everything.
Ken FernandezGUEST
So I think the proof is in the pudding.
Ken FernandezGUEST
And the nice thing is we have organizations like FiveThirtyEight that track the track record of these polls and give them a rating And and and we find that online polls, online panels that might not be random are doing pretty well in predicting elections.
Amanda SturgillHOST
OK, so would you say going to something like 538 that does those ratings of polls would be a good idea if you didn't go to grad school in political science or.
Ken FernandezGUEST
Absolutely, absolutely.
Ken FernandezGUEST
And typically, you know, the nice thing of FiveThirtyEight is that it will link to that organization and you can go dig at their methodological details.
Ken FernandezGUEST
FiveThirtyEight will tell you, is this a live telephone call? Is it online? Is it mixed method? But it won't provide you with all the details of the methodology, but they always place a link to the organization where you can then look at that.
Ken FernandezGUEST
So 538 is a very good, useful tool if you want to start, you know, digging, going down that rabbit hole of polling methodology, because it looks at so many polling results, their track record.
Ken FernandezGUEST
If they typically lean, you know, typically have a Democratic bias or Republican bias and their overall track record and then just an overall letter grade.
speaker_4NARRATOR
Dynamo Faithful is produced by Ian Gregory Graff with marketing and design from Zach Velo.
speaker_4NARRATOR
Dynabot is powered by predictions from FiveThirtyEight Club Soccer Predictions.
speaker_4NARRATOR
Learn more at FiveThirtyEight.com. Music provided by Alex Grohl on Pixabay.
speaker_4NARRATOR
Follow us on Twitter and Instagram at dynamo underscore faithful and send your questions to dynamofaithfulpod at gmail.com.
speaker_4NARRATOR
Thanks for listening and stay faithful.
Nate SilverHOST
I think people overstuff their models sometimes with overly complicated priors and that at the end of the day, if the polling evidence is fairly consistent, right, it should be fairly true to the polls.
Nate SilverHOST
I mean, there are lots of states where we chop off usually a point or so from Democrats and like the advanced flipper versions, but like the polls show Democrats winning in a lot of Also, one thing too, since we used to at FiveThirtyEight evaluate our forecasts, our forecasts were actually like a little bit – have been underconfident slightly historically, right? No one criticizes them for that reason.
Nate SilverHOST
But like I don't know.
Nate SilverHOST
I think it's a good – And it's been ahead of the conventional wisdom.
Aswath DamodaranHOST
And in this particular case, this particular blog post got picked up in four very different places.
Aswath DamodaranHOST
The first was a site called 538.com. What is this site? This is a site maintained by a guy called Nate Silver.
Aswath DamodaranHOST
He used to work at the New York Times, but he's a statistics geek.
Aswath DamodaranHOST
It's a site run for people who like numbers and statistics.
Harry EntenGUEST
Don't make it further along than it needs to be, Harry.
Harry EntenGUEST
A decade ago, I wrote an article for FiveThirtyEight, which is where I was formerly employed.
Harry EntenGUEST
With a very similar thing.
Harry EntenGUEST
And it's one of these issues that I think we sort of have going on right now.
DaveHOST
that polling always over counts or overestimates the left-leaning side of the equation, not just here, but in other places.
DaveHOST
And the other thing about that that's fascinating is it seems like, I don't know how many it is now, but there's been a whole bunch of different polling outlets that have been part of like FiveThirtyEight and Nate Silver and all these other ones that have just been Fraud, just straight up
BrianHOST
fraud.
DaveHOST
And so that's sort of the that's the basis of that.
83 more episodes mention FiveThirtyEight.
Create an account to see the whole feed, search across every transcript, and follow the entities you care about.