Hacker News new | ask | show | jobs
by vsssk 2015 days ago
> How do we have a higher mail in ballot rejection rate in non-pandemic times than we do in pandemic times?

I mean, here's an explanation for a specific case

> According to the nonprofit, nonpartisan organisation Ballotpedia, Georgia rejected 6.42% of mail-in ballots in total in the 2016 general election and 3.10% in total in the 2018 midterm (here). These totals include rejections because of signatures, but also include, for example, ballots received late or past deadlines, problems with return materials or a voter having already voted in person.

> It may be that Trump was referring to the 0.15% of ballots specifically rejected for "missing or non-matching signatures" when saying that ballots rejected in 2020 were "almost zero", but this percentage is consistent with past years. The higher percentage he mentions for past years is likely based off the total rejected ballots (here) which can not be compared with 2020, as this information is not available.

https://www.reuters.com/article/uk-factcheck-georgia-rejecte...

https://sos.ga.gov/index.php/elections/number_of_absentee_ba...

Edit: And just for fun, I pulled data for 2020 and 2018 statewide, November elections from https://elections.sos.ga.gov/Elections/voterabsenteefile.do

I looked at a largeish file (didnt want to deal with the 1gig state wide file), ended up being CHATHAM county. (file name 025).

In 2018, there were 351 rejections for 40683 mail in ballots. In 2020, there were 321 rejections for 110831 mail in ballots.

The interesting thing to me was that in 2018, 286 ballots were rejected for being late, but in 2020, only 191 were rejected for being late.

Which tells me at least some of the higher acceptance rates can be attributed simply to people being more diligent about mailing their ballots in.

1 comments

> The higher percentage he mentions for past years is likely based off the total rejected ballots (here) which can not be compared with 2020, as this information is not available.

This is also the same style of reasoning that helps people explain away the abnormally high voter turnout in WI. Please understand I mean no disrespect in this - I just think changing definitions to cure the abnormalities isn’t necessarily correct (and is frighteningly Orwellian in the bad way).

If I had ever turned out 90% of voters measured as (votes/total registered voters) I would have more money than Elon. The way these shocking statistics get diluted is by placating to people with no experience in the process. The goal is to make them feel better by offering an alternative (however misguided) representation, that will stop the line of questioning.

So when confronted with an extremely implausible voter turnout, people say the real turnout is votes/total eligible voters... which is very different than every other time I’ve experienced voter turn out

> So when confronted with an extremely implausible voter turnout, people say the real turnout is votes/total eligible voters... which is very different than every other time I’ve experienced voter turn out

Not sure how whatever conjured view of voter turnout YOU have is relevant. You can go to https://elections.wi.gov/elections-voting/statistics/turnout grab `Voter Turnout Partisan-NonPartisan Through August 2020.xlsx` and see that turnout percentage is computed as `number of votes cast` / `voting age population`.

Which puts Wisconsin turnout at 72.3% [1]

We were talking about ballot rejection, and you start going off about "shocking" voter turnout statistics because you had no retort to the rejection stuff. Either you have an axe to grind or you are a troll. Or maybe you're a masked free speech crusader. The hero we deserve. But I kind of doubt that.

1: https://www.usatoday.com/story/news/factcheck/2020/11/05/fac...

> Not sure how whatever conjured view of voter turnout YOU have is relevant. You can go to https://elections.wi.gov/elections-voting/statistics/turnout grab `Voter Turnout Partisan-NonPartisan Through August 2020.xlsx` and see that turnout percentage is computed as `number of votes cast` / `voting age population`.

This is funny -- here's a single example from the WI.gov page to show how little care is given to cultivating and maintaining this data...

HINDI: 56184

County: ST. CROIX COUNTY

Municipality: VILLAGE OF SPRING VALLEY

Total Electors 2012: 4

Voting Age Estimate 2012: 6

Voter Turnout 2012: 66.67%

Total Electors 2016: 319

Voting Age Estimate 2016: 10

Voting Turnout 2016: 3190.00%

So I can find slivers that support wild claims too. I think the more productive discussion is just doing a side by side of signatures and rolls - not that hard, no real argument about it - and everyone can stop trying to dig for little things and becoming unbearably pedantic with each other.