Comment by soundwave106 | Hacker News Reader

Comment by soundwave106 | Hacker News Reader

Parent Full thread

I'd be cautious of this report. This is an American politics advocacy organization. Not to say that this report is incorrect per se, it may be 100% right. But it also may be anything else, ranging from wildly exaggerated to outright wrong. It's very difficult to find sober, factual information in the national American political scene.

I looked up one of the referenced incidents -- the botched poll in Arizona -- and based on other articles (see: https://www.washingtonpost.com/news/the-fix/wp/2016/03/31/ar...) it's hard to tell from a bird's eye perspective whether even that was "rigged".. or whether it was mere "incompetence".

I know FiveThirtyEight, referenced several times in the paper, (http://fivethirtyeight.com/features/the-system-isnt-rigged-a...) lean against the "rigged" argument, which is where I stand from a big picture perspective. Political shenanigans always happen, but regardless of this, demographics explain the result of the Democratic primary more than anything else. (That being said, in my opinion, the same demographics should frankly tell the Democratic Party to handle the Sanders supporters much better than they have been so far.)

I think this comment lowers the quality of the discussion. The comment asks us not take an exhaustively prepared article seriously without providing any evidence.

It's as though the comment is trying to prevent overracting, however weigh it against the threat of underreacting. The harm in taking these allegations too seriously (false positive) is doing research and finding out the statistics are invalid (extra research time). The harm in taking these allegations not-seriously-enough (false negative) is mass-scale election fraud.

Given how hard it is for such allegations to even be considered at all in our social climate (even when we all know they're much more technically feasible than people would like to admit) I think a defusing this article detracts from pursuit of truth in exchange for peace of mind.

> The harm in taking these allegations too seriously (false positive) is doing research and finding out the statistics are invalid

That's one possible harm, but I think it's also possible that people will take the allegations too seriously and withdraw from the political process entirely. One of the paradoxes of modern political life is that the elections that get the most attention aren't necessarily the ones with the greatest impact on our day-to-day lives. If you don't show up to the polls in November because {Clinton,Trump,Satan} will win anyways because of the rigged system, then you won't vote for mayor, school board, and local bond measures that can much more directly affect the policing policies in your town, the education your kids get, and the transportation infrastructure to get to your places of work or leisure.

The difference a few individual voters can make in local elections is certainly huge. Over time I've seen multiple initiatives and candidates I liked or disliked decided on only a handful of votes.

However, I don't think this is a sufficient argument to motivate people who would refuse to vote because of the state of the presidential election. I think they would feel that local educational or policing policies are insignificant in comparison to the big-picture changes they now have no say in because their favored candidate was eliminated by corrupt party and media practices. They might still feel that they will be more affected by the year-to-year impact of foreign policy and trade policy than the day-to-day impact of a .1% chance increase of getting a bogus traffic ticket.

Apologize for not referencing more evidence for the counterview.

So let me bring up some exhaustively prepared articles from the past. A lot of these are from the late 2000s when the "voter ID" issue was in play, but I do not think the electoral landscape has changed that much since then.

A report on voter fraud in 2007 from the Brennan Center concluded a very low rate of voter fraud in 2004. http://www.brennancenter.org/sites/default/files/legacy/The%...

News21 (part of the Carnegie Knight media initiative) created a database of voter fraud. They found little. http://votingrights.news21.com/article/election-fraud/

We have this Washington Post reporter who tracked voter fraud, and found little. https://www.washingtonpost.com/news/wonk/wp/2014/08/06/a-com...

And... another paper on how "voter fraud is in the eye of the beholder." (Harvard Law Review) http://www.harvardlawreview.org/wp-content/uploads/pdfs/anso...

So... that's a lot of data (I could link many more reports from this era) that indicates that, frankly, voter fraud from any angle really hasn't been a big issue before (and that any perception may be due to political bias, perhaps). I do believe there's plenty of inefficiencies regarding the American electoral process, and that might be right ("inaccurate, costly, and inefficient" as the Pew Center on the States alleged in 2012, that I buy -- http://www.pewtrusts.org/~/media/legacy/uploadedfiles/pcs_as...). But there's a difference between incompetence and awful systems, and outright fraud.

I must acknowledge that this isn't the only paper alleging fraud in the 2016 primaries (https://drive.google.com/file/d/0B6mLpCEIGEYGYl9RZWFRcmpsZk0...). I will note however that, for this one, Snopes was skeptical in that the numbers had not been peer reviewed by independent parties (http://www.snopes.com/stanford-study-proves-election-fraud-t...) and note counter-viewpoints ala Joshua Holland of The Nation (who does not think much of "exit poll conspiracies": https://www.thenation.com/article/reminder-exit-poll-conspir...)

This is why I'm very cautious. Bias is huge in politics; it may be exhaustively prepared, but if they are using bunk statistics, it's not worth much.

What are you talking about? Those articles are generic studies about past elections, there's no theoretical way those studies could have found whether or not there was an issue in the 2016 election.

It's okay to take back something you said in error in HN, you don't always have to double-down.

I don't think it's in error or "doubling-down" to wait for the "peer review", which is all I'm asking. :)

If it's not clear, I'm not saying that the paper is wrong per se.

There are ongoing lawsuits relating to this where many people of varied interests are going to be pouring over this, and other data, way more than I can at the moment.

If the results of the lawsuits validate this paper, then this paper is important.

But it's also possible that the lawsuits will not be successful.

Historically most claims of voter fraud have been wildly exaggerated, which is why I bring up the past. Including the fact that bias unfortunately colored many fraud claims in the past.

Maybe it's different this time. Maybe not. We'll see.

The debate around exit polls will only be settled with the publication of raw data by Edison Media Research. With so much polemic on the topic you would think that it would be in their interest to make these data public if they had nothing to hide. This isn't even a new issue. Exit polls data transparency was already the object of a debate during the 2004 Presidential Elections: http://electiondefensealliance.org/frequently_asked_question...

On one hand, we have an electronic vote count which can't be verified and on the other, we have raw exit polls data that are kept secret. What kind of Democracy is that??? Isn't transparency one of the most fundamental principal in a Democracy?

The American election process isn't transparent at all... How can we claim our elections to be democratic?

The authors of this report can be taken seriously as they put their professional reputation on the line. I am not sure that Fritz Scheuren, professor of statistics at George Washington University and the 100th President of the American Statistical Association (ASA), would risk his reputation by making statements we can't defend. Here is what he says: “as a statistician, I find the results of the 2016 primary voting unusual. In fact, I found the patterns unexpected [and even] suspicious. There is a greater degree of smoothness in the outcomes than the roughness that is typical in raw/real data.”

Reply on news.ycombinator.com