Nate Silver

@NateSilver538 ↗

Deleted No

Hibernated No

Last Checked May 23, 2021

Created

Thu May 06 14:58:44 +0000 2021

Likes

184

Retweets

Source

Twitter for Android

View Raw Data

JSON Data

View on Twitter

Likely Available

Nate Silver @NateSilver538

I'll confess I believed the opposite until a few years ago, when I actually did the work to study this in updating our Congressional model in 2018. We collected a ton of special election results, were excited to have an metric of the political environment that incorporated... — PolitiTweet.org

Posted May 6, 2021

Preceded By

Nate Silver @NateSilver538

This isn't true, though, empirically. Special election results don't add predictive power in forecasting Congressional outcomes once you have polling data (e.g. generic ballot polls). Certainly an *individual* race doesn't tell you much. But even collectively they don't add much. — PolitiTweet.org

Karl Rove, making the case that election results are more revelatory than polls in anticipating the political futur… https://t.co/KFHsbG30DX

Posted May 6, 2021

Followed By

Nate Silver @NateSilver538

...both special elections and the generic ballot. Tried a bunch of different model specifications, etc. The special election results added nothing, contradicting my priors. They're sort of fool's gold. — PolitiTweet.org

Posted May 6, 2021