Discussion about this post

User's avatar
Daniel Echlin's avatar

Wait, aren't you arguing that primaries are a more natural setting to see how good polls are at polling? In a general election, polls can report the last election, plus what vibe the pollster thinks the country is in relative to last election. In a primary election, you can't do that, so you get to see the value-add from the poll, which is apparently that it gets wrong about the fraction of people you would imagine were undecided in the first place.

Markets were also really, really bad at this race. They were 80/20 Cuomo on most of the race, 50/50 on the eve of the race, and it was Mamdani +12. So if polls just report prior election, and markets don't really have the guts to decouple from polls, we should maybe go back to that whole "do markets do more good than harm" question.

Expand full comment
Andrew Xu's avatar

Ceteris Paribus, RCV probably helps explain how someone like Lander was able to get to 11% 🤷‍♂️

Expand full comment

No posts