meta_pixel
Tapesearch Logo
Log in
People I (Mostly) Admire

117. Nate Silver Says We're Bad at Making Predictions

People I (Mostly) Admire

Freakonomics Radio + Stitcher

Society & Culture

4.61.9K Ratings

🗓️ 28 October 2023

⏱️ 43 minutes

🧾️ Download transcript

Summary

Data scientist Nate Silver gained attention for his election predictions. But even the best prognosticators get it wrong sometimes. He talks to Steve about making good decisions with data, why he’d rather write a newsletter than an academic paper, and how online poker led him to the world of politics.

Transcript

Click on a timestamp to play from that location

0:00.0

My guest today, Nate Silver, is the founder of the Data Driven Website 538.

0:09.9

He's also the author of the best-selling book, The Signal and the Noise.

0:13.8

Silver's best-known for his shockingly accurate election predictions, but that's just

0:18.6

the tip of the iceberg.

0:20.5

I think I have strength in dealing with imperfect information and dealing with uncertainty

0:24.7

and kind of refining best guesses.

0:30.3

Welcome to People I Am Mostly Admire, with Steve Levin.

0:36.5

What I love most about Nate Silver is that he has such amazing instincts both for analyzing

0:42.5

and writing about data.

0:44.4

Almost every time I read something he's written, I have the same reaction.

0:48.2

Damn, I wish I had written that.

0:52.8

So, let's start with the topic you're most famous for, and that's predicting election

1:02.0

outcomes.

1:03.0

In 2008, your first foray into political prediction, you correctly predicted 49 out

1:08.6

of 50 states in the electoral college, and then you against all odds, did even better

1:13.9

in 2012, getting every single state right.

1:17.9

Those two election cycles led people to believe you were some kind of a messiah or an

1:23.2

oracle, and I have to imagine that's both the blessing and the curse, right?

1:27.6

Very much so.

1:28.6

I remember telling my little agent that I am being set up to fail here, inevitably there

1:33.7

will be a time when the low probability outcome comes up, although in 2016, it wasn't

1:38.7

that low actually.

...

Please login to see the full transcript.

Disclaimer: The podcast and artwork embedded on this page are from Freakonomics Radio + Stitcher, and are the property of its owner and not affiliated with or endorsed by Tapesearch.

Generated transcripts are the property of Freakonomics Radio + Stitcher and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.

Copyright © Tapesearch 2025.