Tweet by @NateSilver538

Nate Silver @NateSilver538

Your hypothesis relies on data that's (i) 8 months old (ii) was collected mostly online (fine but not gold-standard) (iii) involves a lot of statistical extrapolation (iv) is calibrated to 2018 rather than 2020 turnout (v) is based on approval rather than head-to-head matchups. — PolitiTweet.org

My hypothesis is that the president's approval rating in Ohio is probably better than this one poll in Quinnipiac,… https://t.co/aHQpJJz8FL

Posted July 25, 2019 Hibernated

Preceded By

Nate Silver @NateSilver538

@joshtpm Q seems a bit outlier-ish, but it's usually a mistake to dismiss a poll as an outlier rather than averaging it in with your other data. My broader point to other Nate is the data is messy/noisy, when Electoral College analysis is something that requires a -=LOT=- of precision. — PolitiTweet.org

Posted July 25, 2019 Hibernated

Followed By

Nate Silver @NateSilver538

All of which is… fine… to get a rough approximation. But the real-world margin of error on those estimates is (at least) several points wide, which is much too large to come to many firm conclusions about the tipping point, especially if you're not working probabilistically. — PolitiTweet.org

Posted July 25, 2019 Hibernated