Talk:2014 United States Senate election in Kentucky

Internal polls
The consensus on internal polls is not to include them unless there is very little other public polling. With 39 public polls of this race, that does not apply, which is why we don't include them, whether they're from the McConnell or the Grimes campaign. See discussions here for example where this has been debated. Thanks, Tiller54 (talk) 13:44, 13 September 2014 (UTC)
 * Consensus by a handful of people is not consensus. The number of available polls has no bearing on whether or not to include another one. As long as internal polls are clearly marked as such, they should be included. Mellman Group, while partisan, is a very good polling outfit and I think it is important to include their polls, especially when they show contradictory results from another outfit that was in the field for the exact same period (Magellan, in this case). I am all for including McConnell's internals too. But completeness should always be preferred to arbitrary standards. Nonpartisan polling has no monopoly on accuracy. Nevermore27 10:17, 14 September 2014 (UTC)
 * Hell, when a poll by Wenzel is included and they refuse to release their sample size and MoE, let alone their scripts, as referenced in the linked conversation, this whole thing just becomes self-parody Nevermore27 10:20, 14 September 2014 (UTC)
 * It's not a "handful of people", it's agreement across the project. That's why you won't find internal polls on pages where there is an abundance of public polls. Regardless of Mellman's record, the Grimes campaign is only going to release polls by Mellman when they're favourable to Grimes. Hence why, after the last 11 public polls showed McConnell leading, they put one out. These standards are not "arbitrary" and suggesting that including the internal polls that are released because it presents a more "complete" picture of the race ignores the fact that many, many more are never released. Tiller54 (talk) 13:12, 16 September 2014 (UTC)
 * All pollsters conduct polls they don't release for a variety of reasons. It's an arbitrary standard. Nevermore27 05:48, 17 September 2014 (UTC)
 * "The consensus on internal polls is not to include them unless there is very little other public polling."
 * No, that was the "consensus" of the small group who disagreed with me on that thread. They were wrong then, and you are wrong now.
 * Look at this year's race in Illinois. There are many polls, and internal polls are included. If acceptable there, they are acceptable here.
 * Wikipedia articles including internal polls from the Mellman Group can be found here, here, here, here and here. If acceptable in all those other articles, they're acceptable here.
 * Joegoodfriend (talk) 17:06, 2 October 2014 (UTC)

What is the basis for reverting the Mellman edit? "That's why you won't find internal polls on pages where there is an abundance of public polls." This has already been demonstrated not to be true. "Regardless of Mellman's record, the Grimes campaign is only going to release polls by Mellman when they're favourable to Grimes." This is irrelevant. Mellman is cited in many other wikis, it's a reliable source. The fact that you don't like Mellman polls violates No_original_research. And, for the record, it's a highly accurate pollster. They called that North Dakota race correctly after all, didn't they? Joegoodfriend (talk) 15:46, 3 October 2014 (UTC)
 * I don't have anything against Mellman. But you can't just ignore the established consensus because you don't like it. And it was hardly a "small group", it was every other editor who participated in the discussion. If that's not an overwhelming consensus, I don't know what is. There are presently over 40 polls of this race. Adding a single Grimes internal poll adds nothing to this article. Tiller54 (talk) 16:25, 6 October 2014 (UTC)
 * There is no such "consensus." It's a legitimate poll, and Mellman is a reliable source for wikipedia. The arguments you've made, such as not including internal polls "unless there is very little other public polling," simply aren't true. "And it was hardly a "small group", it was every other editor who participated in the discussion." Yeah, do the math on the North Darkota article: 4 against 2. And that doesn't change the fact, again, that internal polls including Mellman are routinely used on wikipedia. I believe that trumps the overwhelming consensus of you and your 3 friends. Joegoodfriend (talk) 17:22, 6 October 2014 (UTC)
 * Yes, there is such a consensus. One that you agreed with: "Four editors, all of whom are clearly nice folks, are against it. Ok, you win." And that was before I'd commented. As for your "you and your 3 friends" crack, was that really necessary? As it happens, I don't know the other contributors that well.
 * And yes, Mellman polls are used, either when they're not internal polls (as half of the "examples" you posted are) or when there's limited polling of a race. Tiller54 (talk) 20:01, 9 October 2014 (UTC)


 * Agree Polls are either acceptable or not. You can't pick and choose which article it's okay to have an internal poll on. When you (Tiller) say "Adding a single Grimes internal poll adds nothing to this article.", it's ridiculous. Because the poll released shows something very different from other polls. It's important to have the context. And since I did not participate in that North Dakota discussion, the argument is now 4 editors to 3. Nowhere near consensus. If you want, we can have a true Project-wide discussion on the matter, but your cited conversation ain't gonna cut it anymore. Nevermore27 08:38, 7 October 2014 (UTC)
 * The point is that internal polls aren't generally accepted unless there's little public polling available. This isn't just a Wikipedia policy, it's something news organisations note when reporting on internal polls that are released to them. For example, here, where National Journal comments on a Thom Tillis internal poll of the NC Senate race: "The poll—which, like all internal surveys leaked to the media, should be viewed warily..." That's because campaigns only leak internal polls when they're favourable to their side. This is something which is laid out here, where User:Rxguy quotes Nate Silver who says: "the subset of polls which [campaigns] release to the general public is another matter, and are almost always designed to drive media narrative... What we've found is that is that polls commissioned by campaigns and released to the public show, on average, a result that is about 6 points points more favorable to their candidate’s standing than nonpartisan polls released at the same time." It's also the reason why polling aggregators like Real Clear Politics and FiveThirtyEight don't include internal polls (whether from Mellman or otherwise) in their list of polls that they analyse. Tiller54 (talk) 20:01, 9 October 2014 (UTC)
 * "Four editors, all of whom are clearly nice folks, are against it. Ok, you win." Yeah, I gave up on that one because I was tired of arguing for the moment.
 * Ok let's review.
 * 1. There are hundreds, if not thousands, of articles on wikipedia that include polls.
 * 2. Are internal polls routinely used? Yes.
 * 3. Are Mellman polls routinely used, meaning Mellman is a reliable source for wikipedia? Yes.
 * 4. Is there a consensus on this talk page that internal polls and/or Mellman should not be used? No?
 * 5."The point is that internal polls aren't generally accepted unless there's little public polling available." Is this true for wikipedia? No, as has been demonstrated, in article such as the 2014 Illinois gubernatorial.
 * "National Journal comments on a Thom Tillis internal poll of the NC Senate race: "The poll—which, like all internal surveys leaked to the media, should be viewed warily..." Excuse me, I believe that's "should be viewed" not "should not be viewed." Joegoodfriend (talk) 20:20, 9 October 2014 (UTC)


 * Internal polls are only used, as I've said several times now, when there aren't about dozens of public polls available, as in the case of this election and others like this one and this one. Yes, Mellman polls are sometimes used - when they're not internal polls, or, if they are internals, when there's little other polling of an election. I'm sorry, but you haven't presented any specific reasons for why internal polls should be included other than "but they're on some other pages", which basically amounts to WP:OTHERSTUFFEXISTS, which is not a viable reason for including something. Tiller54 (talk) 20:42, 9 October 2014 (UTC)
 * No, it's "should be viewed warily", which is what it says. Which is why internal polls are not regarded as reliable sources by any media outlet. Which is why Nate Silver pointed out their unreliability and usefullness only to drive a media narrative. Which is why aggregators like Real Clear Politics and FiveThirtyEight don't even include them in their analysis. If they're not regarded as reliable by literally everyone else other than Wikipedia, why should Wikipedia regard them as reliable? Tiller54 (talk) 20:42, 9 October 2014 (UTC)
 * To quote you: "The poll—which, like all internal surveys leaked to the media, should be viewed warily..." They can't even view it warily if someone is trying to deprive them the opportunity to see it. The internal polls here are clearly marked, you should allow people to "view them warily". Nevermore27 14:13, 11 October 2014 (UTC)
 * See below for reply on indiscriminate information. Tiller54 (talk) 18:01, 18 October 2014 (UTC)


 * "Internal polls are only used, as I've said several times now, when there aren't about dozens of public polls available" This is simply not true.
 * "you haven't presented any specific reasons for why internal polls should be included other than "but they're on some other pages", which basically amounts to WP:OTHERSTUFFEXISTS, which is not a viable reason for including something." I find this baffling. Yes, other articles on elections include polls to indicate the state of the race, but that's no reason to include polls here? Why have wikipedia at all if the existence of relevant information is no reason to include relevant information? Joegoodfriend (talk) 21:28, 9 October 2014 (UTC)
 * "This is simply not true" No, it is true. Look at these elections pages                 , all of which feature lots of public polls of the general election and no internal polls. Why? Because internals are only useful when there isn't such an abundance of public polls. In this election for example, where the only polls that we have are 2 internal polls.
 * "Yes, other articles on elections include polls to indicate the state of the race, but that's no reason to include polls here? Why have wikipedia at all if the existence of relevant information is no reason to include relevant information?" Because, as the core policy WP:N states, we avoid including indiscriminate information, especially in circumstances where that information (eg: internal polls) is self-promotion and indiscriminate publicity. With dozens of public polls available, all but one of which from the last 3 months shows McConnell leading, the internals from the Grimes campaign have only been released to attempt to counter the media narrative that the race is moving away from her: . Thus, they do not provide anything except for the Grimes campaign to be able to say "I'm not dead yet!" Not forgetting the fact that polling aggregators don't include internal polls, news sources are dismissive of them and polling experts like Nate Silver point out that they're biased and only used to drive a media narrative. Now do you see why we should only use them when there's little other polling available? Tiller54 (talk) 18:01, 18 October 2014 (UTC)


 * Well, you defined 4 against 2 as consensus, and we have the same ratio here, 2 against 1. So consensus it is I guess Joegoodfriend (talk) 21:32, 9 October 2014 (UTC)
 * No, it is not "consensus". Getting one other person to agree with you does not overturn the overwhelming consensus that was established two years ago on the ND elections page. As WP:CONLEVEL clearly states: "Consensus among a limited group of editors, at one place and time, cannot override community consensus on a wider scale." Thus, the community-wide consensus as established on literally every other elections page from the 2014 and 2012 election cycles, and the much larger discussion from 2012 which included contributions from 7 editors (5 against 2, not 4 against 2) is not overturned by 2 of those same editors being joined by one other editor who didn't take part in the original decision. Tiller54 (talk) 18:01, 18 October 2014 (UTC)
 * "Consensus among a limited group of editors, at one place and time, cannot override community consensus on a wider scale."
 * Thanks for making my argument for me. Here's a search of the top 500 wikipedia articles reference "internal poll." Among the articles linking internal polls, we find internal polls for Tim Huelskamp, Andy Tobin, Mike Collins, Edwards in Florida, Eric Schertzing, Lawrence J. Hogan, Schauer (MI), Angel Taveras, Ralph Abraham, Abbott (TX), Osborn (NE), Lindsey Graham, Bill Brady (IL), Lamar Alexander (TN), Spilka (MA), .... I can literally cite hundreds and hundreds of examples.
 * "consensus as established on literally every other elections page from the 2014 and 2012" I'm sorry you could not be more wrong. Again, look at the search. Just for 2012 and 2014 we see the gubernatorials for MD, RI, TX, IL, and Senate races in NE, SC, TN, AZ, OK, GA, IA, HI, OR. I see practically every single major race in just the first few links.
 * The idea that a 4 against 2 argument in 2012 on one article about one race somehow trumps this is ridiculous. And I don't see anyone agreeing with you here. The consenses here is that you are wrong. Joegoodfriend (talk) 19:33, 18 October 2014 (UTC)
 * Once again, it wasn't "4 editors" or "4 against 2", it was a discussion with 7 contributors where 5 agreed that internal polls shouldn't be included and only 2 disagreed.
 * The consenses here is that you are wrong So please explain how a 2-1 consensus can override a 5-2 one. It's not "ridiculous", it's POLICY, established so that a little-seen discussion like this one can't trump a wider discussion that happened previously.
 * As for the search you did, the fact that an article for a politician includes a line like "Two weeks before the March 12 primary, an internal poll showed Romney losing to Nixon by a six-to-one margin in New Hampshire" doesn't have anything do with the consensus established about the use of internal polls on elections pages. Internal polls exist on elections articles where there is either no or little public polling. For example, here where the only 4 polls available are all internals; here where across all 9 congressional districts there are only 10 polls, of which 7 are internals; here where 4 of the 7 polls are internals; here where 25 congressional districts have 23 polls, of which 11 are internals; here where 5 of the 11 gubernatorial primary polls are internals. On Senate elections pages like this one, this one and this one, there are internal polls - of the primary elections, not of the general elections. The usage of internal polls is pretty clear - in elections (mostly House general elections and gubernatorial/Senate primary elections) where there is no or little public polling available. Tiller54 (talk) 19:39, 23 October 2014 (UTC)


 * You really lose track of your own argument when you say the North Dakota discussion was "overwhelming consensus". It was a grand total of 6 people in a discussion. Six people. If your best argument is linking to a conversation among six people where - not to put too fine a point on it - the people arguing against internal polls ended up being wrong, then you're losing.  Nevermore27  (talk) 17:48, 20 October 2014 (UTC)
 * I'm not in sync with joegoodfriend's argument that this discussion here establishes a new consensus, I'm just saying the original discussion is not consensus at all.  Nevermore27  (talk) 17:52, 20 October 2014 (UTC)
 * Again, it was seven, not six. And no, they weren't "wrong". The fact that Heitkamp won the election should have no bearing on whether or not her internals should be included. Tiller54 (talk) 19:39, 23 October 2014 (UTC)
 * "Internal polls exist on elections articles where there is either no or little public polling." Explain to me how the 2014 Illinois gubernatorial is a race with, "either no or little public polling." Seems to me it's got a quite a lot of public polling. While you're at it, maybe you'd like to explain why, weeks after your first contribution to that article, you suddenly decided to start deleting some of the polls there, without explaining yourself on that article's TALK page. Did you really think the deletions would go unnoticed because you inaccurately labeled your Edit Summaries as "add poll"? Joegoodfriend (talk) 18:54, 24 October 2014 (UTC)
 * Seven people or six doesn't make your argument any better. And currently, you are the only person on this vendetta, and employing shady tactics to try and get your way. Why exactly are you trying to stand in the way of more information? On an encyclopedia?  Nevermore27  (talk) 07:23, 25 October 2014 (UTC)
 * So seven editors does not a consensus make, but three does? How exactly does that work? And in case you were unaware, Wikipedia is not an indiscriminate collection of biased information, polls or otherwise. Furthermore, sources that are biased should be handled with care and as such internal polls should not presented with equal validity to public polls. More to the point: why are you so determined that these slanted polls should be included in an article of an election that has over 50 public polls available? What possible service to the article does their inclusion serve? There exists a consensus against them, news articles warn against taking them seriously, polling aggregators don't include them in their calculations and polling experts pretty much dismiss them out of hand. Just because information exists doesn't mean Wikipedia should include it. Tiller54 (talk) 02:14, 2 November 2014 (UTC)

Ignoring the issue of consensus, it makes sense to include internal polls. It's data. Someone paid for it. Also, a technical correction in the listing of these polls -- some have been asking specifically about "some other candidate", and while Patterson is that other candidate, asking it in that way (not giving Patterson by name) skews results. As such, Some Other Candidate should be listed under "Other", not under Patterson, as Patterson was not included, by name, in that poll. 192.200.93.225 (talk) 19:55, 25 October 2014 (UTC)
 * Which poll are you referring to here? I agree the data should be as accurate as possible, so I'd be happy to change it for you, or you're welcome to do it too.  Nevermore27  (talk) 04:06, 28 October 2014 (UTC)

Patterson in infographic; polling
Patterson had been polling at or above 5% until September. Being on the ground here in KY, please let me explain what's going on.

First, several media articles (intentionally?) confused Patterson with the neo-nazi on September 18th. That is the event that dropped Patterson's support from 8%-ish to 3%-ish. ( http://www.courier-journal.com/story/politics-blog/2014/09/17/anti-semite-runs-for-senate/15791301/ )

Polls that use "Some Other Candidate" without specifically naming all ballot-listed options are not legitimate as they do not give respondents their actual choices on election day and thus have an inherent bias against those not listed. I have been involved in market research off-and-on for decades and have recently started my own market research company.

A lot of the polling that is being done now is actually pushing against Patterson; pollsters are saying that people can only vote for Grimes or McConnell. If the respondent tries to answer Patterson, the respondent is told that Patterson isn't an option.

I'd like to see where the "5% rule" exists, officially, as I haven't been able to find it. Additionally, does the rule say "actively polling"? If a candidate is on the cusp, should the article be edited based on the poll of the day?

Finally, I can understand where in a state where there might be 10 candidates on the ballot, there needs to be a legitimate criteria set; however, in a race with 3 people, versus the two listed, does it make much sense to prevent the 3rd from being seen?

I only placed Patterson into the infographic twice, so please don't hold the actions of others against me. I can't find my old Wikipedia account -- maybe it was purged or maybe I've just forgotten my password or something. I've been here for a long, long time. Usually, my edits are just accepted and life goes on which is why I don't remember my login/password any more. Bnewmark42 (talk) 03:37, 21 October 2014 (UTC)

More polls
With the page locked down, someone with access needs to add the latest NBC/Marist poll: http://newscms.nbcnews.com/sites/newscms/files/kentucky_october_31_2014_annotated_questionnaire_nbc_news-marist_poll.pdf

Not logging in, because I don't want to. — Preceding unsigned comment added by 192.200.93.225 (talk) 04:36, 3 November 2014 (UTC)

Results Box
I am unable to edit the page myself, so I will post a results box here.

Results
I didn't make use the percent change column, but kept it in because it is used on all other elections I've checked for this seat. I am unsure of what its purpose is; it doesn't appear to reflect the change in the number of votes for either that candidate or the candidate's party from the last election, nor the change in percent from the last election, nor the margin of victory/defeat separating any of the candidates from any identifiable mark. The column seems like it should have something in it, as other Kentucky U.S. Senate election pages do, but I cannot figure out exactly where that number should come from. SusanBroil (talk) 16:15, 5 November 2014 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on United States Senate election in Kentucky, 2014. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive https://web.archive.org/20131112192835/http://www.therepublic.com/view/story/80c4e3a0b9f844b7b9144da2a7a60001/KY--Kentucky-Senate-Bevin to http://www.therepublic.com/view/story/80c4e3a0b9f844b7b9144da2a7a60001/KY--Kentucky-Senate-Bevin

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers.—cyberbot II  Talk to my owner :Online 23:24, 25 February 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 5 one external links on United States Senate election in Kentucky, 2014. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive http://web.archive.org/web/20140225013524/http://www.wenzelstrategies.com/blog/wp-content/uploads/2014/02/Final-KY-Poll-US-Senate-2-13-2014.pdf to http://www.wenzelstrategies.com/blog/wp-content/uploads/2014/02/Final-KY-Poll-US-Senate-2-13-2014.pdf
 * Added archive http://web.archive.org/web/20140925160138/http://www.ipsos-na.com:80/download/pr.aspx?id=13904 to http://www.ipsos-na.com/download/pr.aspx?id=13904
 * Added archive http://web.archive.org/web/20140903021541/http://www.wenzelstrategies.com/blog/polls/ws-poll-mcconnell-continues-lead-over-grimes-in-kentucky/ to http://www.wenzelstrategies.com/blog/polls/ws-poll-mcconnell-continues-lead-over-grimes-in-kentucky
 * Added tag to http://atfactionfund.org/wp-content/uploads/2014/06/Kentucky-Poll-Toplines-6-23-14.pdf
 * Added archive http://web.archive.org/web/20140508025204/http://buildkxlnow.org/wp/wp-content/uploads/2014/05/KY-topline-tables.pdf to http://buildkxlnow.org/wp/wp-content/uploads/2014/05/KY-topline-tables.pdf
 * Added archive http://web.archive.org/web/20130218044451/http://www.runswitchpr.com:80/2013/02/13/runswitch-harper-polling-survey-shows-mcconnell-up-9-over-judd/ to http://www.runswitchpr.com/2013/02/13/runswitch-harper-polling-survey-shows-mcconnell-up-9-over-judd/

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 11:18, 21 July 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 external links on United States Senate election in Kentucky, 2014. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20140904163827/http://rothenbergpoliticalreport.com/ratings/senate to https://rothenbergpoliticalreport.com/ratings/senate
 * Added archive https://web.archive.org/web/20130704054931/http://www.lex18.com/news/grimes-to-run-against-mcconnell-in-2014-senate-race to http://www.lex18.com/news/grimes-to-run-against-mcconnell-in-2014-senate-race/
 * Corrected formatting/usage for http://www.courier-journal.com/viewart/20131105/NEWS010605/311050034/Mitch-McConnell-Alison-Lundergan-Grimes-clash-over-right-work-measure

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 04:16, 13 January 2018 (UTC)