What I implied is that new the law probably doesn't increase the length of lines, and could just as easily shorten lines.
I am not implying that lines don't matter to turnout (the article, in fact, says long lines affect turnout), or to human beings
When it comes to the effect on in-person voting, there are provisions cutting in both directions.
An example in the negative direction: ending the Fulton County bus voting stations, which could directly alleviate a problem
An example in the positive direction: expanding early voting days and the provision requiring that large precincts with long lines (over an hour) either add staff, add machines or split apart.
I won't go into depth to try and game it all out. There are too many provisions of debatable and cross-cutting effects to confidently ascertain whether more or fewer people will stand in long lines
What I am really convinced of, though, is that long lines really matter, and the reformers may be making a mistake by focusing on expanded voting options (say, assuring no-excuse mail voting) as opposed to protecting the fundamentals (making sure there aren't long lines)
HR1 includes both kinds of provisions (there's a requirement for lines under 30 minutes iirc). But yet again, it's overlooked--and the proposed narrower versions of the bill don't include this kind of basic protection
Expanding/adding convenience voting is usually accompanied by a reduction in the availability of in-person/election day voting.
In the extreme, you get a state like Washington that's nearly all mail but just a single in-person site per county
As a consequence, there are lot of the states with all three voting methods that manage to have awful lines, while there are (or were pre-20) states without advance voting but very few lines
For one, there's just a basic resource overstretch problem.
For another, optimizing resources can be more challenging when the way voters can turn out varies so much
You may recall, for ex, back in 2016 that the Arizona dem primary was horrible, as the state cut back on in-person options in part bc mail had been increasing steadily... then they crushed (no electoral gain, just bad administraiton) cbsnews.com/news/phoenix-l…
And to be clear, there's no reason it can't work to have multiple voting methods! If you can make it all work, that's great. Some states do, even though others suck at it year after year. But multi method voting does create administrative challenges of its own
• • •
Missing some Tweet in this thread? You can try to
force a refresh
The new Georgia law increases the risk of partisan interference in election administration, even if it doesn't make it easy.
Election subversion is a risk outside of Georgia, too, and it's a risk that's being overshadowed by focus on mail voting nytimes.com/2021/04/06/ups…
As an aside, the response here to the my piece on the voting half of this law was series of absolutely unhinged--which is not to say there weren't several valid critiques, some of which I'm sympathetic to.
One of the most wrong headed critiques was the notion, often implicit, that there's something problematic about examining individual aspects of a law at a time (especially when there's more coming later on those very aspects)
The Georgia election law's restrictions on voting are unlikely to discernibly affect turnout or the result nytimes.com/2021/04/03/ups…
I've limited my analysis to the provisions affecting whether and how people can vote, not those that empower the state legislature to play a larger role in election administration. More on those later, but those provisions don't inherently affect voting access in a particular way
I also like to say that I do think it's important for journalists to report about the consequences of these laws, not just their intent or morality, for a few reasons.
I do think there’s a credible, somewhat counterintuitive case that Democrats would have been better off if the GA GOP got rid of no excuse absentee voting altogether, v make it more difficult and risk higher rejections
This supposes that no excuse absentee voting does little to nothing to increase turnout, which I think is probably right. It’s less clear whether the GA law would increase rejections by more, but it’s uncertain enough to be imaginable
And the GA law does quite a bit to make it harder, between ID requirements and the huge dropbox restrictions. Not hard to imagine the law trapping Dem voters in a more difficult voting method
One thing I've been thinking about lately: how different is the optimal voting system in a low-trust and high-trust society? What about in a society where partisans will play no-holds barred to win, versus one where democratic norms are strong?
To take an easy example from 2020: maybe it's not optimal for the vote count to last three weeks in a low trust society. There's nothing wrong with it, strictly speaking. It could have advantages. But maybe it's not worth the risk if there are bad actors and low trust
I can imagine taking this to more extreme places where there would probably way more debate: say, arguing against multiple forms of voting with varying eligibility, as it creates distinct categories of voters/ballots that can be targeted by law, election admin, courts, etc
One of the best hope for polling is the theory that the error/bias was mainly just about the coronavirus, for instance. Our Oct. 2019 polls were way better than Oct. 2020! The poll averages in Feb/Mar 2020 were way better too. But I think the evidence is pretty inconclusive
That said, history offers plenty of reason to hope that polling could leave the hospital. So if that's the main @NateSilver538 position here, then he's right that we may not disagree as deeply as I think
I'd boil down the disagreement to one fundamental thing: I think low systemic bias is far, far more important than important for thinking about the polls than average error, while I think @NateSilver538 looks a lot more at average error
You can see both the magnitude of systemic bias in this chart, along with the case that there's a trend toward greater systemic bias. And fwiw, I think the D+5 bias is probably mitigated by some 'nonpartisan' firms that, tbh, aren't so nonpartisan or above the board