Greg Dworkin
I am so glad Nate Cohn wrote this

Like his inclusion of Patrick Murray, it's so refreshing to speak with honest pollsters who just want to get it right.

That includes how polls can miss, how pollsters don't want them to, and what steps you might take to prevent it.

Nate Cohn
3h
How One Polling Decision Is Leading to Two Distinct Stories of the Election
https://nytimes.com/2024/10/06/upshot/polling-methods-election.html (archived: https://archive.ph/hH7fa )

Greg Dworkin
It makes you trust the pollsters *without* blindly trusting the polls.

I and a few other observers think it's closer to a 2022 environment than 2020 and this puts teeth in the argument, but we shall see.

cc
@reesetheone1
12:48 PM · Oct 6, 2024

NYT (archive) - How One Polling Decision Is Leading to Two Distinct Stories of the Election

A methodological choice has created divergent paths of polling results. Is this election more like 2020 or 2022?

By Nate Cohn
Oct. 6, 2024, 5:03 a.m. ET

Over the last month, one methodological decision seems to have produced two parallel universes of political polling.

In one universe, Kamala Harris leads only narrowly in the national popular vote against Donald J. Trump, even as she holds a discernible edge in the Northern battlegrounds. The numbers look surprisingly similar to the 2022 midterm election.

In the other, Ms. Harris has a clear lead in the national vote, but the battlegrounds are very tight. It’s essentially a repeat of the 2020 election.

This divide is almost entirely explained by whether a pollster uses “weighting on recalled vote,” which means trying to account for how voters say they voted in the last election.

/snip