Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Looking for Voter Fraud (In Old Elections) with Data Visualization

probablydance.com

Post reply on HN
Print view View original

31–40 of 467 posts

Re: Looking for Voter Fraud (In Old Elections) with Data Visualization

#31

Post by thristian »

nl wrote: post #6

Am I the only centre-left type person in the world who thinks the Electoral College isn't a bad thing (in theory at least). In the Westminster 2-house system in Australia voters vote for representation from their seat. These seats have approximately the same number of people, but they also vote for an upper-house representative and the smaller states are over-represented in terms of population. This is because otherw…

The Australian lower house is based on the British House of Commons, the Australian upper house is based on the American Senate. Since America already has a Senate to prevent the federal government from optimising for high-population cities, it doesn't need the Electoral College as a second factor.

Re: Looking for Voter Fraud (In Old Elections) with Data Visualization

#32

Post by tastyfreeze »

Glyptodon wrote: post #8

Always keep in mind that however distorted the electoral college is, that distortion is tremendously magnified by the cap on the size of the US house of Representatives and the fact that the Constitution specifies a lower bound for district population but no upper bound.

I have been thinking that maybe states should get an equal number of electoral votes. We would get the protection of the electoral college and every state would have an equal say.

Re: Looking for Voter Fraud (In Old Elections) with Data Visualization

#33

Post by rodgerd »

alisonkisk wrote: post #17
nl wrote: post #6

Am I the only centre-left type person in the world who thinks the Electoral College isn't a bad thing (in theory at least). In the Westminster 2-house system in Australia voters vote for representation from their seat. These seats have approximately the same number of people, but they also vote for an upper-house representative and the smaller states are over-represented in terms of population. This is because otherw…

Why don't people in California deserve a means to influence the presidency as much as Montana or DC?

Because blacks and latinos and liberals are only worth three fifths of a person.

Re: Looking for Voter Fraud (In Old Elections) with Data Visualization

#34

Post by refurb »

nl wrote: post #6

Am I the only centre-left type person in the world who thinks the Electoral College isn't a bad thing (in theory at least). In the Westminster 2-house system in Australia voters vote for representation from their seat. These seats have approximately the same number of people, but they also vote for an upper-house representative and the smaller states are over-represented in terms of population. This is because otherw…

I agree. It’s an intentional feature, not a bug.

If it was purely a popular vote system, then the president would always be decided by the large coastal states. The interior would have zero say in who the president is.

Re: Looking for Voter Fraud (In Old Elections) with Data Visualization

#35

Post by save_ferris »

Rebelgecko wrote: post #26
save_ferris wrote: post #22

Earlier quoted context omitted.

> I don't think it's perfect, but in principle I don't see it as a bad things that people in say Montana have a means to influence the presidency. Under the electoral college system, voters in so-called swing states have much more influence on the outcome of the election than voters in states with established political supremacy. To your point above, in a system where the popular vote determines the outcome, every pe…

I thought the result of the supreme court cases is that the votes of faithless electors are nonbinding?

The ruling grants states the power to enforce the pledge of an elector or penalize them for casting an alternative vote, but it's up to the states to do so.

Re: Looking for Voter Fraud (In Old Elections) with Data Visualization

#36

Post by IfOnlyYouKnew »

grahamburger wrote: post #19
DC-3 wrote: post #13

Earlier quoted context omitted.

Should the vote also be weighted to uphold the interests of other classes of people, like young vs old, or rich vs poor, or bricklayers vs hairdressers? I consider myself more defined by my age and profession than by my hometown but in an electoral college system it is only that latter fact that affects how a vote is weighted.

I think of it as giving some representation to the land itself (and it's natural resources) as represented by the people who work and live there.

The coast of California would then like to send 20 electors (most of them with rather nice beards, some in wetsuits).

Harvard wants to send another 40. Science is, after all, more important these days than natural resources. Just compare Switzerland (absolutely nothing of value in the ground) to Equatorial Guinea (top-10 oil reserves).

The one export market where the US dominates is culture. Mind if Hollywood sends in a few representatives?

Re: Looking for Voter Fraud (In Old Elections) with Data Visualization

#37

Post by mistermann »

IfOnlyYouKnew wrote: post #20
ve55 wrote: post #12

What I dislike about a lot of these potential fraud analysis techniques is that if a group that was performing fraud was aware of them, then they would simply perform the fraud 'properly' such that the given technique(s) don't show evidence (since they know what others are looking for and hoping to find, they simply ensure it isn't there). The same goes for similar techniques like Benford's law: once you know about t…

The system is secure and robust, at least with regards to the mechanics of individual votings. Among the main factors is its distributed nature. You just can't run a scam with senior citizen volunteers in thousands of individual precincts. Then, there are observers. Famous case from this week's election was Republicans complaining that their two dozen(!) observers weren't allowed closer than 6 feet to the ballots. It…

> The system is secure and robust, at least with regards to the mechanics of individual votings.

Not a single vulnerability or shortcoming?

How would one know such a thing?

> You just can't run a scam with senior citizen volunteers in thousands of individual precincts.

Is this the only possible exploit? Is corruption of thousands of precincts even needed in a tight race?

> It would be impossible for fraud to happen on any regular basis and not coming to light.

How would one know such a thing?

Is it possible on an irregular basis?

Is it impossible that it does get noticed, but no serious followup?

Have there been no examples of whistleblowers who were not listened to in the history of the USA, election related or not?

> So the incidence of 0.00002% (https://www.vox.com/policy-and-politics/2017/4/26/15424270/v...) is somewhat informative.

How might we know that that impressive looking number is accurate? Does vox have an omniscient on staff?

I wonder if when the discussion is about a non-politicized topic (like standard, run of the mill security), might there be a difference in mindset when people are formulating their opinions. If the system in question was for anything other than voting, I suspect IT folks would be a little less accepting of evidence-lite assurances that a distributed, largely manually implemented (by non-experts) system of this scale and frequency of usage is 100% up to snuff. It seems rather unlikely that an inconsistent, somewhat cobbled together and non-expertly manned system of this scope is absolutely perfect. And yet, this seems to be not only the overwhelming narrative, but the fact passed down from the media - the same one that spent the last 4 years assuring us that Russia had hacked the prior election.

Re: Looking for Voter Fraud (In Old Elections) with Data Visualization

#39

Post by joshuaheard »

refurb wrote: post #28

The author’s analysis seems a bit one sided. He speaks of poor turn out in LA, because of the electoral college system which means additional democratic votes don’t matter. Then speaks about the same lack of Democratic turnout in Texas, because it always go Republican (again, due to the electoral college). But wouldn’t those same effect apply to Republicans as well? They don’t bother to vote in CA because there is a…

I saw the same thing. He says Democrats won't vote in California because they know they are going to win, and Democrats don't vote in Texas because they know they are going to lose. I think it would apply equally to both parties.

I was surprised to see this article on HN, since most other statistical election fraud analyses I have seen are being dismissed as conspiracy theories. Facebook even censored one of them. I'm glad to see they are getting some serious consideration.

Re: Looking for Voter Fraud (In Old Elections) with Data Visualization

#40

Post by ekimekim »

kart23 wrote: post #23

The author says about turnout: "One clarification: The turnout number at the bottom is slightly wrong: You’d think that you can easily get these statistics for past elections, but I can’t find the turnout number anywhere. Instead I was able to get the population of the counties, so this is the number of votes divided by the population." Isn't this quite inaccurate? I think counties release registered voter numbers, s…

I'd argue registered voter turnout also isn't a true turnout statistic, as many people will not be registered due to apathy/feeling it doesn't matter/etc.

The statistic you really want is votes / eligible voters, not registered voters. But that's a hard number to figure out in the US.

Post reply on HN