• Welcome to the new Internet Infidels Discussion Board, formerly Talk Freethought.

Iowa Caucuses (or Cauci?)

Also, what is with the attitude?

I haven't had time to do any analysis, though I'd like to, and well, I am hoping that the IDP and campaigns get their act together to finally publish a rational consensus that is fair and objective. In the mean time, I did note this single comment contained within a larger post and wanted to address it.

So, what exactly are you talking about?
You mean your, "The claim is that the error is unbiased. This graph is inappropriate and misleading to look at that question. Sorry." quip? I seem to be one of the only one trying to put numbers into a truthful unbiased context.

The issue is that this is a full canvas, not a survey or a poll. The numbers are the numbers and there should be no "error" beyond counting errors or reporting errors.

So any attempt to say that the error is higher than the results is bullshit; if the reporting is complete, a .1% difference in a final count can't be hand-waved away by error. The difference is real, and has real results.
 
You mean your, "The claim is that the error is unbiased. This graph is inappropriate and misleading to look at that question. Sorry." quip? I seem to be one of the only one trying to put numbers into a truthful unbiased context.

The issue is that this is a full canvas, not a survey or a poll. The numbers are the numbers and there should be no "error" beyond counting errors or reporting errors.
There will always be errors, elections are usually above the margin of error. This still isn't like Florida, where the error was greater than the margin of victory in a winner take all.

I have seen nothing specific in data as of late pointing to specific existing errors in Iowa.

So any attempt to say that the error is higher than the results is bullshit;
If the margin of error is magnitudes below the margin of victory, complaining about the error is a waste of time.
 
Also, what is with the attitude?

I haven't had time to do any analysis, though I'd like to, and well, I am hoping that the IDP and campaigns get their act together to finally publish a rational consensus that is fair and objective. In the mean time, I did note this single comment contained within a larger post and wanted to address it.

So, what exactly are you talking about?
You mean your, "The claim is that the error is unbiased. This graph is inappropriate and misleading to look at that question. Sorry." quip?

Why are you calling what I wrote a "quip?" I was trying to be respectful and polite about having a disagreement.

Jimmy Higgins said:
I seem to be one of the only one trying to put numbers into a truthful unbiased context.

Firstly, in my experience with data analytics, this isn't how it is done. Maybe an approximate example is bias in residual error. It's not exactly the same, but it should illustrate that one does not need to put error back into the original context. One can look for biases or trends in the "error" itself. In this particular case, we're not talking about residual error but instead campaign "mistakes." Bottom line is there is nothing inherently misleading about labeling a graph as error (or saying it is error) and then examining trends in the error.

Secondly, I did not merely post to you. I also posted to PyramidHead in order to ask him for the numbers that underlie the visualization. This was so I could see if there were non-randomness, i.e. a statistically significant difference. I stated that would make me "halfway" there. I also posted to him skepticism of the data source and that I would have to reproduce it in order to be convinced.

As I wrote a few posts ago, I am supporting principles of democracy and objectivity. I believe it is important to show PyramidHead why he is wrong, IF he's wrong. And to show others why PyramidHead is right, if he's right. In the greater context of the election, it might not mean much once everything is done, but right now, we just have Iowa data to look at or question.
 
Last edited:
2020 Iowa Democratic Caucus Election Results | Des Moines Register
SDE = State Delegate Equivalent
[TABLE="class: grid"]
[TR]
[TD]Candidate
[/TD]
[TD]SDE's
[/TD]
[TD]Pct
[/TD]
[TD]Delegates
[/TD]
[TD]Proportional
[/TD]
[/TR]
[TR]
[TD]Pete Buttigieg
[/TD]
[TD]564
[/TD]
[TD]26.2%
[/TD]
[TD]13
[/TD]
[TD]11
[/TD]
[/TR]
[TR]
[TD]Bernie Sanders
[/TD]
[TD]562
[/TD]
[TD]26.1%
[/TD]
[TD]12
[/TD]
[TD]11
[/TD]
[/TR]
[TR]
[TD]Elizabeth Warren
[/TD]
[TD]388
[/TD]
[TD]18.0%
[/TD]
[TD]8
[/TD]
[TD]7
[/TD]
[/TR]
[TR]
[TD]Joe Biden
[/TD]
[TD]340
[/TD]
[TD]15.8%
[/TD]
[TD]6
[/TD]
[TD]6
[/TD]
[/TR]
[TR]
[TD]Amy Klobuchar
[/TD]
[TD]264
[/TD]
[TD]12.3%
[/TD]
[TD]1
[/TD]
[TD]5
[/TD]
[/TR]
[TR]
[TD]Andrew Yang
[/TD]
[TD]22
[/TD]
[TD]1.0%
[/TD]
[TD]0
[/TD]
[TD]0
[/TD]
[/TR]
[TR]
[TD]Tom Steyer
[/TD]
[TD]7
[/TD]
[TD]0.3%
[/TD]
[TD]0
[/TD]
[TD]0
[/TD]
[/TR]
[TR]
[TD](uncommitted)
[/TD]
[TD]4
[/TD]
[TD]0.2%
[/TD]
[TD]0
[/TD]
[TD]0
[/TD]
[/TR]
[TR]
[TD](other)
[/TD]
[TD]1
[/TD]
[TD]0.0%
[/TD]
[TD]0
[/TD]
[TD]0
[/TD]
[/TR]
[/TABLE]
No SDE's: Michael Bennet, Michael Bloomberg, John Delaney, Tulsi Gabbard, Deval Patrick

For proportional allocation, I used three algorithms: D'Hondt highest averages, Round-down (Hare) then D'Hondt, Round-down then Largest Remainder. All three agreed.

There was some departure from proportionality, like some thresholding somewhere in the count, that reduced how many delegates AK got.
 
2020 Iowa Democratic Caucus Election Results | Des Moines Register
SDE = State Delegate Equivalent
[TABLE="class: grid"]
[TR]
[TD]Candidate
[/TD]
[TD]SDE's
[/TD]
[TD]Pct
[/TD]
[TD]Delegates
[/TD]
[TD]Proportional
[/TD]
[/TR]
[TR]
[TD]Pete Buttigieg
[/TD]
[TD]564
[/TD]
[TD]26.2%
[/TD]
[TD]13
[/TD]
[TD]11
[/TD]
[/TR]
[TR]
[TD]Bernie Sanders
[/TD]
[TD]562
[/TD]
[TD]26.1%
[/TD]
[TD]12
[/TD]
[TD]11
[/TD]
[/TR]
[TR]
[TD]Elizabeth Warren
[/TD]
[TD]388
[/TD]
[TD]18.0%
[/TD]
[TD]8
[/TD]
[TD]7
[/TD]
[/TR]
[TR]
[TD]Joe Biden
[/TD]
[TD]340
[/TD]
[TD]15.8%
[/TD]
[TD]6
[/TD]
[TD]6
[/TD]
[/TR]
[TR]
[TD]Amy Klobuchar
[/TD]
[TD]264
[/TD]
[TD]12.3%
[/TD]
[TD]1
[/TD]
[TD]5
[/TD]
[/TR]
[TR]
[TD]Andrew Yang
[/TD]
[TD]22
[/TD]
[TD]1.0%
[/TD]
[TD]0
[/TD]
[TD]0
[/TD]
[/TR]
[TR]
[TD]Tom Steyer
[/TD]
[TD]7
[/TD]
[TD]0.3%
[/TD]
[TD]0
[/TD]
[TD]0
[/TD]
[/TR]
[TR]
[TD](uncommitted)
[/TD]
[TD]4
[/TD]
[TD]0.2%
[/TD]
[TD]0
[/TD]
[TD]0
[/TD]
[/TR]
[TR]
[TD](other)
[/TD]
[TD]1
[/TD]
[TD]0.0%
[/TD]
[TD]0
[/TD]
[TD]0
[/TD]
[/TR]
[/TABLE]
No SDE's: Michael Bennet, Michael Bloomberg, John Delaney, Tulsi Gabbard, Deval Patrick

For proportional allocation, I used three algorithms: D'Hondt highest averages, Round-down (Hare) then D'Hondt, Round-down then Largest Remainder. All three agreed.

There was some departure from proportionality, like some thresholding somewhere in the count, that reduced how many delegates AK got.

That's a problem, too.

According to NBC link I gave earlier that I've been discussing it's worse because it gives Pete 14 and Bernie 12 but it's also got a different total, 41, not 40. In tabular form, here is what NBC has as quoting the IDP:

[TABLE="class: grid"]
[TR]
[TD]Candidate
[/TD]
[TD]SDE's
[/TD]
[TD]Pct
[/TD]
[TD]Delegates
[/TD]
[/TR]
[TR]
[TD]Pete Buttigieg
[/TD]
[TD]564
[/TD]
[TD]26.2%
[/TD]
[TD]14
[/TD]
[/TR]
[TR]
[TD]Bernie Sanders
[/TD]
[TD]562
[/TD]
[TD]26.1%
[/TD]
[TD]12
[/TD]
[/TR]
[TR]
[TD]Elizabeth Warren
[/TD]
[TD]388
[/TD]
[TD]18.0%
[/TD]
[TD]8
[/TD]
[/TR]
[TR]
[TD]Joe Biden
[/TD]
[TD]340
[/TD]
[TD]15.8%
[/TD]
[TD]6
[/TD]
[/TR]
[TR]
[TD]Amy Klobuchar
[/TD]
[TD]264
[/TD]
[TD]12.3%
[/TD]
[TD]1
[/TD]
[/TR]
[TR]
[TD]Andrew Yang
[/TD]
[TD]22
[/TD]
[TD]1.0%
[/TD]
[TD]0
[/TD]
[/TR]
[TR]
[TD]Tom Steyer
[/TD]
[TD]7
[/TD]
[TD]0.3%
[/TD]
[TD]0
[/TD]
[/TR]
[TR]
[TD](uncommitted)
[/TD]
[TD]4
[/TD]
[TD]0.2%
[/TD]
[TD]0
[/TD]
[/TR]
[TR]
[TD](other)
[/TD]
[TD]1
[/TD]
[TD]0.0%
[/TD]
[TD]0
[/TD]
[/TR]
[/TABLE]

One can see how this is a problem, but the other thing here is that those numbers in the first column (SDEs) are supposed to be finalized today, not by yesterday when the IDP jumped to this weird math.

The column is actually unaltered from the previous reports, even after Warren, Sanders, and Buttigieg all submitted complaints about 5% of the data. The IDP had said:
“The IDP has initiated the process of review of the requested precincts, and any corrections will be released before Monday at 12pm CST in a single update – reported prior to the allocation of national delegates,” an unsigned statement from the state party read. “The top priority of the IDP continues to be ensuring the accuracy of reported data as the process moves towards completion.” Troy Price, the Iowa Democratic Party chair, is also slated to hold a press conference on Monday.
https://www.politico.com/news/2020/02/08/iowa-democratic-party-reviewing-results-112532

But as PyramidHead gave us an update on, they seemed to have backtracked using a lawyer statement later:
https://www.chicagotribune.com/elec...0200209-vun7qmw54rdx5hitwm4zrg3m2m-story.html

They also argue that a recount is impossible.
 
Trying to figure out why there is a discrepancy here, not sure why...but I have tracked this part down...

Here is the official website for IDP Iowa caucus:
https://results.thecaucuses.org/

It has Buttigieg at 14 national delegates and Sanders at 12 with total for all national delegates at 41.
 
I have created some consolidated counts for ideologically similar candidates.
[TABLE="class: grid"]
[TR]
[TD]Candidates
[/TD]
[TD]Delegates
[/TD]
[/TR]
[TR]
[TD]Bernie Sanders + Elizabeth Warren
[/TD]
[TD]20
[/TD]
[/TR]
[TR]
[TD]Pete Buttigieg + Joe Biden + Amy Klobuchar
[/TD]
[TD]20
[/TD]
[/TR]
[/TABLE]
So the left and center lanes are tied.
 
lpetrich, there are 41 delegates though. So, it is 21 v 20. Also, there are 8 super delegates not counted. Likely, they will be in slight favor of the non-left, making maybe 26 v 23.
 
Jebus! So I'm trying to run numbers, but first thing I've noticed is, finding numbers is hard... and potentially inaccurate. I looked here (Des Moines Register) but the numbers don't make any sense. Firstly they list SDEs per county, but it appears to actually be votes, when looking at the LA Times site. I sort the numbers and the vote totals are well off. Then I looked here, this appears to be the more recent after the weekend update. The numbers change notably! Van Buren County (small county) saw total votes drop by half for all candidates.
lpetrich, there are 41 delegates though. So, it is 21 v 20. Also, there are 8 super delegates not counted. Likely, they will be in slight favor of the non-left, making maybe 26 v 23.
The Super Delegates don't vote in the first national convention vote.
 
IDP Caucus 2020 - Iowa Democratic Party
I redid the proportional allocation for 41 total delegates.
[TABLE="class: grid"]
[TR]
[TH]Candidate[/TH]
[TH]SDE's[/TH]
[TH]Del's[/TH]
[TH]HA[/TH]
[TH]LR[/TH]
[/TR]
[TR]
[TD]Bennet[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Biden[/TD]
[TD]339.678[/TD]
[TD]6[/TD]
[TD]6[/TD]
[TD]7[/TD]
[/TR]
[TR]
[TD]Bloomberg[/TD]
[TD]0.21[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Buttigieg[/TD]
[TD]564.302[/TD]
[TD]14[/TD]
[TD]11[/TD]
[TD]11[/TD]
[/TR]
[TR]
[TD]Delaney[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Gabbard[/TD]
[TD]0.114[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Klobuchar[/TD]
[TD]263.883[/TD]
[TD]1[/TD]
[TD]5[/TD]
[TD]5[/TD]
[/TR]
[TR]
[TD]Patrick[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Sanders[/TD]
[TD]561.528[/TD]
[TD]12[/TD]
[TD]11[/TD]
[TD]11[/TD]
[/TR]
[TR]
[TD]Steyer[/TD]
[TD]6.619[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Warren[/TD]
[TD]388.48[/TD]
[TD]8[/TD]
[TD]8[/TD]
[TD]7[/TD]
[/TR]
[TR]
[TD]Yang[/TD]
[TD]21.856[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Other[/TD]
[TD]0.693[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Uncommitted[/TD]
[TD]3.732[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[/TABLE]

  • SDE: State Delegate Equivalent
  • HA: Highest-averages proportional allocation - D'Hondt with and without initial round-down
  • LR: Largest-remainder proportional allocation
 
The head of the Iowa Democratic party filed his resignation Wednesday, as the organization is still picking up the pieces from last week's caucus debacle.

Troy Price had been head of the state party since 2017, but after his role in overseeing a process widely panned as disorganized and opaque, it became an open question whether he would stay on in his job.

"While it is my desire to stay in this role and see this process through to completion, I do believe it is time for the Iowa Democratic Party to begin looking forward, and my presence in my current role makes that more difficult," Price said, in a letter to the Iowa Democratic Party State Central Committee.

Prior to being elected head of the state party in July 2017, Price worked on the presidential campaigns of Barack Obama and Hillary Clinton in Iowa.

Last week's contest was Price's first presidential caucus as chairman, and he had long made it clear he hoped to use technology to make the caucus process more accessible and quick.
https://www.npr.org/2020/02/12/8054...troy-price-resigns-after-caucus-night-debacle

It just keeps getting worse.
 
Alexandria Ocasio-Cortez on Twitter: "Candidate preferences aside, the integrity of our primaries are incredibly serious. We cannot afford to be pointing fingers- we need to pinpoint solutions.
If the DNC was responsible for security and there were security failures, we need to address that. https://t.co/ix4cK71v7W" / Twitter

noting
The Iowa Caucuses App Had Another Problem: It Could Have Been Hacked — ProPublica
While there is no evidence hackers intercepted or tampered with the results, a security firm consulted by ProPublica found that the app lacks key safeguards.

...
The IowaReporterApp was so insecure that vote totals, passwords and other sensitive information could have been intercepted or even changed, according to officials at Massachusetts-based Veracode, a security firm that reviewed the software at ProPublica’s request. Because of a lack of safeguards, transmissions to and from the phone were left largely unprotected.

Chris Wysopal, Veracode’s chief technology officer, said the problems were elementary. He called it a “poor decision” to release the software without first fixing them. “It is important for all mobile apps that deal with sensitive data to have adequate security testing, and have any vulnerabilities fixed before being released for use,” he said.
So this app had some elementary mistakes in it.
 
Alexandria Ocasio-Cortez on Twitter: "Candidate preferences aside, the integrity of our primaries are incredibly serious. We cannot afford to be pointing fingers- we need to pinpoint solutions.
If the DNC was responsible for security and there were security failures, we need to address that. https://t.co/ix4cK71v7W" / Twitter

noting
The Iowa Caucuses App Had Another Problem: It Could Have Been Hacked — ProPublica
While there is no evidence hackers intercepted or tampered with the results, a security firm consulted by ProPublica found that the app lacks key safeguards.

...
The IowaReporterApp was so insecure that vote totals, passwords and other sensitive information could have been intercepted or even changed, according to officials at Massachusetts-based Veracode, a security firm that reviewed the software at ProPublica’s request. Because of a lack of safeguards, transmissions to and from the phone were left largely unprotected.

Chris Wysopal, Veracode’s chief technology officer, said the problems were elementary. He called it a “poor decision” to release the software without first fixing them. “It is important for all mobile apps that deal with sensitive data to have adequate security testing, and have any vulnerabilities fixed before being released for use,” he said.
So this app had some elementary mistakes in it.
Okay, DNC’s involvement only centered around security... but the app was terribly insecure. Who in the hell is in charge of IT for the DNC, Gomer Pyle.
 
Hardly known for leftist conspiracy theorizing, USA Today offered some chilling reflections the morning after:

‘What’s this about Shadow and where did the app come from? The app was created by a company called Shadow Inc., and issued by Jimmy Hickey of Shadow Inc., metadata of the program that the Des Moines Register analyzed Tuesday shows. A LinkedIn profile for James Hickey lists him as COO of Shadow and an engineering manager for Hillary Clinton’s 2016 presidential campaign. Two other former Clinton campaign workers, former Gerard Niemira and Krista Davis, co-founded Shadow…The New York Times has reported that ACRONYM – a Democratic nonprofit founded in 2017 “to educate, inspire, register, and mobilize voters,” according to its website – supported Shadow. Its founder and CEO is Tara McGowan, a former journalist and digital producer with President Obama’s 2012 presidential campaign, The Los Angeles Times reported….Iowa Democratic Party Chairman Troy Price, who also worked as Clinton’s 2016 Iowa political director, did not immediately respond to requests for comment Tuesday about the relationship between the party and Shadow, which it paid $63,184 for website development and travel expenses…’

It gets worse. According to the Los Angeles Times, in an article titled “Tech Firm Started by Clinton Campaign Veterans Linked to Iowa Caucus Debacle”: “Among Shadow’s clients is Pete Buttegieg’s presidential campaign, which paid $42,500 to the firm in July 2019 for ‘software rights and subscriptions,’ according to disclosures to the FEC.”
https://www.counterpunch.org/2020/02/07/the-game-is-rigged/
 
All,


I have been watching the official results at the link I gave and I noticed recently that they have changed.

For example, the previous numbers had more of a distance between Sanders and Buttigieg:
Buttigieg's SDE (as you can see in lpetrich's table) was 564.302 while Sanders had SDE of 561.528:

IDP Caucus 2020 - Iowa Democratic Party
I redid the proportional allocation for 41 total delegates.
[TABLE="class: grid"]
[TR]
[TH]Candidate[/TH]
[TH]SDE's[/TH]
[TH]Del's[/TH]
[TH]HA[/TH]
[TH]LR[/TH]
[/TR]
[TR]
[TD]Bennet[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Biden[/TD]
[TD]339.678[/TD]
[TD]6[/TD]
[TD]6[/TD]
[TD]7[/TD]
[/TR]
[TR]
[TD]Bloomberg[/TD]
[TD]0.21[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Buttigieg[/TD]
[TD]564.302[/TD]
[TD]14[/TD]
[TD]11[/TD]
[TD]11[/TD]
[/TR]
[TR]
[TD]Delaney[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Gabbard[/TD]
[TD]0.114[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Klobuchar[/TD]
[TD]263.883[/TD]
[TD]1[/TD]
[TD]5[/TD]
[TD]5[/TD]
[/TR]
[TR]
[TD]Patrick[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Sanders[/TD]
[TD]561.528[/TD]
[TD]12[/TD]
[TD]11[/TD]
[TD]11[/TD]
[/TR]
[TR]
[TD]Steyer[/TD]
[TD]6.619[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Warren[/TD]
[TD]388.48[/TD]
[TD]8[/TD]
[TD]8[/TD]
[TD]7[/TD]
[/TR]
[TR]
[TD]Yang[/TD]
[TD]21.856[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Other[/TD]
[TD]0.693[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[TR]
[TD]Uncommitted[/TD]
[TD]3.732[/TD]
[TD]0[/TD]
[TD]0[/TD]
[TD]0[/TD]
[/TR]
[/TABLE]

  • SDE: State Delegate Equivalent
  • HA: Highest-averages proportional allocation - D'Hondt with and without initial round-down
  • LR: Largest-remainder proportional allocation


HOWEVER, here are new values from the official website:
[TABLE="class: grid"]
[TR]
[TH]Candidate[/TH]
[TH]SDE's[/TH]
[/TR]
[TR]
[TD]Buttigieg[/TD]
[TD]563.207[/TD]
[/TR]
[TR]
[TD]Sanders[/TD]
[TD]563.127[/TD]
[/TR]
[/TABLE]

There is currently a .08 difference in SDEs, but the national delegates remain disproportionate for some reason.

I will add that I examined a couple of the cases that PyramidHead brought to the thread and the complaints were mathematically valid.

For example, I spot checked randomly 2 of the complaints submitted by PyramidHead, the first 2:

I spot checked the second one in the new results and see it has been adjusted to be mathematically correct.

I don't know that these are the final numbers now and for the last week or so I have been delving into the numbers to reproduce the issues. So I actually have both the old file with old results and new file with new results and can compare.

While I will report back any issues or non-issues I see, it is important to note that (1) there were actually many validly complained about problems because the Sanders campaign submitted 25 issues and the Buttigieg submitted 57, but overall Sanders decreased the distance and (2) the national delegate assignment remains disproportionate.
 
Here is the worst offending row in the table as far as I can tell.

County: Scott
Precinct: (D34) City of Davenport
Benefited: Other
Harmed: Sanders

The category of Other had ZERO Final votes. But for some reason ended up with SDEs. (.4133)

I looked at final votes and constructed an ideal proportional allotment to candidates based on the sum of all delegate assignments in that precinct. Then, I looked at which candidate was furthest from the ideal (Sanders) which one benefited the most (Other). I then determined the root mean square between the ideal proportional assignment and the actual assignment. Afterward, I took the root mean square of the putative correction which would give the delegate to Sanders to see which fit of discrete delegate assignments would have lesser error from the ideal and the answer was assigning it to Sanders by far. So when I wrote that this was the most offending data point I had meant that the RMS was furthest away from RMS of putative correction which I hope you will all agree is a fairly reasonable and objective way to measure error.

This is only a single data point as there are others where the error is less obvious.
 
Back
Top Bottom