Talk:List of close election results

This is the talk page for discussing improvements to the List of close election results article.
This is not a forum for general discussion of the article's subject.

Put new text under old text. Click here to start a new topic.
New to Wikipedia? Welcome! Learn to edit; get help.

Article policies

Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL

Exclude elections that didn't have at least 1,000 total votes cast?

Should we exclude elections that didn't have at least 1,000 total votes cast? For example we now have Akulliq, Nunavut (2008) 0/463 and Ranklin Inlet South, Nunavut (2013) 0/344.

Also, we would otherwise have to include for example all 258 instances where the US Senate was tied (and the US vice-president had to break the tie). Jimsmith1978 (talk) 06:58, 1 July 2017 (UTC)[reply]

Those aren't elections. Timrollpickering 21:50, 15 July 2017 (UTC)[reply]

I've been using the 1000 vote limit, and eliminating local elections and the 0.1% margin threshold, but looking through Massachusetts' very searchable database, I found like 30 elections between 1970 and today. I think we need to up these limits or the list will get unwieldy. I suggest the following: 1. Keep all ties and single vote margin elections (not counting local elections) 2. For all others, set the limit at 1000 votes for the winner and a margin of less than 0.01%. That still leaves a lot of close elections, but makes the list much more manageable. Volcycle (talk) 19:36, 5 January 2018 (UTC)[reply]

UK local elections

It is fairly common in UK local elections for exact ties to occur where two candidates receive exactly the same number of votes. Such an incident occured, for example, in the Osterley ward of the London Borough of Hounslow in 1998. In such cases, the Returning Officer uses 'any random method' to award the casting vote to one of the candidates and thus determine the winner.

A couple of such incidents occur in every set of local elections (i.e. every year) and they'd probably be too numerous to list in this table. Perhaps a List of tied elections article should be set up? --81.19.57.170 10:17, 1 August 2006 (UTC)[reply]

Marcus Morton

This article says Marcus Morton won the Massachusetts gubernatorial election of 1839 by two votes. The article on Marcus Morton says that he won by one vote. According to snopes at http://www.snopes.com/history/govern/onevote.asp the election was by one vote. Nightkey (talk) 17:50, 14 January 2008 (UTC)[reply]

Actually, the snopes article says that Morton recieved 51,034 votes out of 102,066 cast. Doing the math, his opponent must have received 51,032 votes -- 2 votes less than Morton. While one vote may have changed the ultimate outcome, the margin was 2 votes, and while it may be correct to say that "he won by 1 vote," the table's listing of the elector margin is correct. - Sethant (talk) 19:19, 14 January 2008 (UTC)[reply]

102,066 votes were cast. But apparently the rule was that you needed >50% to be declared the winner. So you needed at least 51,034 votes to be declared the winner. "Coincidentally", the final official result was that Morton won 51,034. The second-place candidate Everett won 50,725. And the remaining 307 votes went to other candidates. The margin between Morton and Everett was 309 votes or 0.3%, which is too large a margin to be included on this page. However, one can argue that going by the rules of the election, Morton won by 1 vote or 0.001%, in which case we could include this election on this page. See http://www.ourcampaigns.com/RaceDetail.html?RaceID=233013 and https://wiki.riteme.site/wiki/Massachusetts_gubernatorial_election,_1839 Jimsmith1978 (talk) 05:25, 2 July 2017 (UTC)[reply]

I have now gone ahead and added this entry (1839 Massachusetts governor). Jimsmith1978 (talk) 23:53, 12 July 2017 (UTC)[reply]

Flournoy-Treadway race in Virginia

While there is some documentation stating that this 1847 house race was settled by 1 vote, there's other reports that it was not. Early reports had Flournoy's votes from Patrick county reduced to zero due to a failure to qualify the commissioners at one precinct, but those votes were later restored. Also note that as reported here, the total votes were not correct, because the source reported the county margins not the total votes. [1]

UK parliament elections

According to http://www.election.demon.co.uk/marginal.html there have been 135 seats in the UK parliament decided by under 100 votes since 1918; assuming an average turnout of 40,000, there are possibly as many as 50 majorities of under 0.1% in that list. Given that it isn't that rare after all, I'm wondering if the criteria for this article ought to be tightened? Or perhaps it needs to take account of the size of the vote; is a margin of 0.1% in a vote of 10 million people more impressive/unusual than 0.1% in a vote of 10,000? This would also take care of the local elections mentioned above, where the electorate is even smaller. Thoughts? — sjorford ++ 17:02, 4 December 2008 (UTC)[reply]

German federal election, 2002

How about the German 2002 elections, with the Social Democrats edging out the Conservatives by 8,864 votes with ~ 48 million cast? 150.127.225.246 (talk) —Preceding undated comment was added at 00:02, 19 December 2008 (UTC).[reply]

It's not really the same thing as that's the total vote in a parliamentary system. Timrollpickering (talk) 16:36, 19 December 2008 (UTC)[reply]

I'm not sure it's as straightforward as that, since the German system is partially proportional making overall vote totals meaningful. Regardless, perhaps I'm looking at the wrong election, but the German federal election, 2002 doesn't appear to be as close as suggested. Perhaps the reference is to the result after coalitions have been formed, however that is a different thing entirely. Debate 木 20:16, 15 February 2009 (UTC)[reply]

The confusion there is because the Bavarian Christian Social Union is formally a separate party from the Christian Democratic Union; however at the federal German level they form a single grouping in the Bundestag known as "the Union" (and indeed in 2002 the Union's candidate for Chancellor was from the CSU). The Union's total is close to the SPD's, but this doesn't tell the full story.

Elections in Germany involve a combination of single-member constituencies and Lander level lists, so the total vote on the federal level isn't so significant as it doesn't indicate an election hinging on that. Also the SPD's coalition partners, the Greens, got significantly more votes than the Union's, the Free Democratic Party, so overall the result was decisive. It's most definitely not the cast that if 4,433 voters had instead voted for the other side the outcome of the election (i.e. the party in government) would have been different. Timrollpickering (talk) 20:38, 15 February 2009 (UTC)[reply]

(Correction to the above. It seems the FDP in 2002 weren't running as the Union's allies but instead as a stand alone force so the SPD-Green coalition had an even larger margin over their nearest rivals. Timrollpickering (talk) 11:39, 31 August 2011 (UTC) )[reply]

That's stupid, at first, the federal election is about the members of the Bundestag, not about forming a government. Officially, that is another election for the government, one voted by the Bundestag members. Also, parties in Germany have never built formal electoral alliances (only exception being the Union sister parties), so looking at the "red-green result" does not make sense, as SPD and Greens are different parties with different lists.

SPD became the strongest party with 38.5%, while the Union (consisting of two sister parties CDU and CSU) was in second place, also with 38.5%. That's what this list is about, not about parliamentary alliances, which might or might not support the parliamental election of a government and can completely change depending on the result. (Oh and btw, the election was actually so close that reporters at first predicted a black-yellow majority and (as we nowadays know) behind the scenes of the first result debate Schröder already said to Stoiber that he could now show what he would be able to do. In Munich, Stoiber even celebrated "We have won the elections", while at the time it began to come clear that he had lost.) --SamWinchester000 (talk) 01:01, 14 December 2017 (UTC)[reply]

List organization and scope

At the moment this list is a bit of a grab-bag and demonstrates a fair degree of geographic bias. I'd like to put some work into improving it, in particular to expanding the results for some other countries such as Australia. Before I do, however, I wonder whether we might be better off splitting the list out into countries/regions. I can see the list getting unwieldy fairly quickly, if indeed it isn't already. I also note the IP's comment 1 August 2006 and agree that close elections are significantly more common at the local level. Consequently I propose to limit the list to national and state/province elections only, which would exclude such items as ABC Unified School District. Debate 木 20:55, 15 February 2009 (UTC)[reply]

Endorse the restriction. If it gets unwieldy, suggest we cut out the "least marginal" elections. MikeHobday (talk) 21:16, 15 February 2009 (UTC)[reply]

Endorse restriction. Timrollpickering (talk) 21:48, 15 February 2009 (UTC)[reply]

Great to see quick responses! Can I assume from this that the proposal to split the list into countries/regions also has support? Debate 木 23:00, 15 February 2009 (UTC)[reply]

Not at this stage from me. I agree the list will get unwieldy, but prefer the approach of restricting entries by significance of election (as above) and by majority, rather than by country. No problem with your creating List of narrow elections in Australia for individual countries. MikeHobday (talk) 07:26, 16 February 2009 (UTC)[reply]

I've been making a lot of updates to this. In addition to suggesting we limit the scope (larger, closer elections, ignoring local elections - see above), I've made a changes to the table itself. I've replaced the column for total votes cast with one for votes for the winner. The reason is that some votes involve multiple candidates, sometimes running for multiple offices. There are also blank ballots or write-ins etc... But what matters is how many votes were decisive, so I think that means votes cast for the two candidates who tied or who were close. In addition I've defined the margin % as the margin divided by the votes cast for the two close candidates (meaning that votes for 3rd candidates or blank ballets won't be counted). A 3 way race that is 1001-1000-800 is not closer than one then a two-way one is 1001-1000. Anyway, let me know if there is any opposition to this. Volcycle (talk) 19:54, 5 January 2018 (UTC)[reply]

The restrictions are reasonable, but the intro says the significance of this list is to address the issue of when and where individual votes matter. That issue applies more often in local elections than state and national, because there are more local elections and they have fewer voters. To begin to address this without making the table too long, I added a short section at the end where there is room to give simple counts and percents of close elections at all levels. I hope others will add more geographic variety. Numbersinstitute (talk) 02:41, 20 April 2018 (UTC)[reply]

Dead heats

Currently some elections where the two were exactly equal are listed as 1 vote majorities on the basis of however the tie was broken. I think it would be better to list these as 0 vote exact results, as the tie breaking vote is separate from the normal ballot - winning an election by having one more person supporting you is different from winning on a coin toss. Timrollpickering (talk) 10:30, 16 February 2009 (UTC)[reply]

That seems reasonable. Debate 木 11:08, 16 February 2009 (UTC)[reply]

Australia: to do list

Need to find good references for the following:

Federal elections

Ballaarat 1919 1 vote (13,569/13,568) votes 0.00369%
Werriwa 1914 7 votes 0.027%
Macquarie 1917 9 votes 0.033%
Stirling 1974 12 votes 0.022%
Hawker 1990 14 votes 0.021%
Brisbane 1917 15 votes 0.038%
Wannon 1954 17 votes 0.046%
Corio 1975 20 votes 0.032%

State elections (very incomplete)

Victorian_Legislative_Council 1985 Nunawading Province (draw, name drawn out of a hat)* court later orders new election
South Australian House of Assembly 1968 Millicent (draw, casting vote of returning officer)* court later orders new election

Debate 木 11:38, 16 February 2009 (UTC)[reply]

Corresponding list of lopsided elections?

I concede that this might not be the best place to discuss this, but since I did not see one that already existed, I figured I would see if there was any interest in making such a list. My (current) concept of such a list would ignore Saddam Hussein/Kim Jong Il 100% turnout/100% of the votes received "elections" in favor of those that were at least nominally-contested in a democratic setting.

I welcome your thoughts. 24.149.110.44 (talk) 15:44, 8 July 2009 (UTC)[reply]

"Narrow elections"?

The elections were not narrow. The margins of victory may have been. There has to be a better title. The best I can come up with is "List of elections decided by narrow margins". Any other ideas? -- 202.142.129.66 (talk) 04:28, 4 June 2010 (UTC)[reply]

"List of narrow (or close) election results". Sussexonian (talk) 20:56, 4 June 2010 (UTC)[reply]

OK, I've moved it to "List of close election results". Cheers. -- Jack of Oz ... speak! ... 12:46, 21 June 2010 (UTC)[reply]

Intro

The current introduction claims that the close Canadian results are due to boundaries being set by an independent commission without party interference - but so are the boundaries in many other countries. And there's no source given for it. Could it be that the high number of Canadian entries is actually down to one of the following:

As a federal country with thirteen provinces & territories, plus a common habit of short hung parliaments and early elections, Canada has a disproportionately high number of results overall?
The Canadian electorate has a lower proportion of strong party identification compared to similar countries and so elections are far more volatile, throwing up many more close results?
The boundary commission goes out of its way to make seats marginal?
Canadian past election results are easier to access than those for many other countries?
Some or all of the above?

Timrollpickering (talk) 11:49, 31 August 2011 (UTC)[reply]

Yes, this content is rather silly. I'm deleting it. --76.69.45.64 (talk) 05:39, 3 February 2016 (UTC)[reply]

Another one to add to the list

Count and preference distribution has just finished in the House of Representatives seat of Fairfax, in the Australian Federal Election, 2013. Margin is 7 votes, out of a total of 84,667 formal votes (or 89,173 including informal votes), which is 0.00827% (or 0.00785% including informal votes). Because of the narrow margin an automatic recount is under way.

http://vtr.aec.gov.au/HouseDivisionFirstPrefs-17496-160.htm http://www.abc.net.au/news/2013-10-01/electoral-commission-orders-recount-for-seat-of-fairfax/4992130

Not sure if this article wants the total of formal votes, or the absolute total that includes informal votes. For clarification, formal votes are valid votes that were cast, informal votes are votes that were cast but can't be counted for a variety of reasons, such as them being illegible, blank, incorrectly filled out, etc.

14.2.28.244 (talk) 12:49, 1 October 2013 (UTC)[reply]

Mississippi legislative district 79, 2015

Another tie: Mississippi legislative district 79 -- total of 9178 votes.

^[1]

Dtilque (talk) 07:59, 29 December 2015 (UTC)[reply]

References

^ http://www.nytimes.com/2015/11/21/us/mississippi-house-race-comes-down-to-one-deciding-straw.html?_r=0

Austria 1999

I would add the Austrian legislative election, 1999 where the difference between FPÖ and ÖVP was only 415 votes (total of 4,695,225 votes = 1:11,313 or 0.0088%). They were the 2nd and 3rd placed parties, but their rank order was highly relevant as the following cabinet coalition consisted of those two parties.----Bancki (talk) 20:30, 15 November 2016 (UTC)[reply]

BC, Canada Provincial Election 2017

While the results are not yet finalized (we have yet to count absentee votes), BC has one riding with the BCNDP declared victorious with a 9 vote margin over the BC Liberals. There are 87 ridings in total, and the current count is 43 BC Liberals, 41 BCNDP and 3 BC Greens. If that 9 vote margin riding flips, it hands the BC Liberals a majority government with 44 seats. At present, with 43, they will have to form a minority government.

While a 9 vote gap is not particularly special in Canadian politics (apparently), the fact that this change could actually spell the difference between how the province is governed, rather than just more or less one MLA for a given party suggests that this is one of the closest elections in history (assuming the absentee vote doesn't significantly disturb the balance).

Does anyone else know if any of the other entries on this list were actually deciding factors in the overall election results? 24.69.201.248 (talk) 23:04, 13 May 2017 (UTC)[reply]

External links modified

Hello fellow Wikipedians,

I have just modified 2 external links on List of close election results. 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:

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

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 21:11, 20 May 2017 (UTC)[reply]

Should the 2017 BC election as a whole count?

The British Columbia general election, 2017 has a mere 1,566 votes/0.8% separating the Liberals from the NDP, and has been refered to as the closest election in BC history.. On the other hand, there were more than two parties running, and the popular vote doesn't determine seats in BC. What does everyone think? Kawnhr (talk) 15:39, 25 May 2017 (UTC)[reply]

0.8% wouldn't meet the "1 in 1000" criterion anyway. Jimsmith1978 (talk) 07:52, 17 June 2017 (UTC)[reply]

External links modified

Hello fellow Wikipedians,

I have just modified 9 external links on List of close election results. 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:

Corrected formatting/usage for http://www.clarkcountynv.gov/Depts/election/Pages/ElectionHistory.aspx?ecode=11P&ecat=1
Corrected formatting/usage for http://redrock.co.clark.nv.us/electionresults/Results_pv.aspx?c=11G&e=2011%20MUNICIPAL%20GENERAL%20%2806%2F07%2F2011%29
Corrected formatting/usage for http://melrosemirror.media.mit.edu/servlet/pluto?state=3030347061676530303757656250616765303032696430303435363830
Corrected formatting/usage for http://www.santacruzsentinel.com/archive/2006/December/02/local/stories/01local.htm
Corrected formatting/usage for http://www.news-register.net/community/story/1125202003_com02.asp
Corrected formatting/usage for http://www.jsonline.com/story/index.aspx?id=220680
Corrected formatting/usage for http://www.herald.ns.ca/cgi-bin/home/displaypackstory?1999%2F08%2F18+191.raw+PE99Aug18+2
Added {{dead link}} tag to http://sudbiblioteka.ru/vs/text_big2/verhsud_big_30224.htm3
Added archive https://web.archive.org/web/20121029053540/http://www.electionresults.act.gov.au/ElectorateParty.html to http://www.electionresults.act.gov.au/ElectorateParty.html
Added archive https://archive.is/20120731112218/http://www.edmondsun.com/opinion/local_story_315180205.html to http://www.edmondsun.com/opinion/local_story_315180205.html

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

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 02:45, 19 September 2017 (UTC)[reply]

External links modified

Hello fellow Wikipedians,

I have just modified 2 external links on List of close election results. 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:

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

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 10:07, 29 December 2017 (UTC)[reply]

Pennsylvania's 18th congressional district special election, 2018

Does this election (Pennsylvania's 18th congressional district special election, 2018) merit inclusion in the article? Thanks. 32.209.55.38 (talk) 04:46, 17 March 2018 (UTC)[reply]

Not close enough. It would have to be about 4 times closer for it to included in this list. HotdogPi 15:03, 3 September 2018 (UTC)[reply]

Stari Grad mayoral election, 2017

Does this election merit inclusion in the article? The first vote had no candidate gettin a majority, so a run-off was held in which both candidates got exactly 721 votes. A third vote was held in which one candidate won by 1%.

Source: https://www.izbori.hr/arhiva-izbora/data/lokalni/2017/izabrani/i_06_17_0000.pdf (pages 42-44)

The election is also mention in the Cities table of this article

--95.168.124.179 (talk) 22:29, 6 May 2019 (UTC)[reply]

close election

The term currently redirects here, but it may be notable on its own. There are plenty of mentions in passing of the term, but I am struggling to find any in-depth discussion and analysis. Some sources: Unpublished manuscript from 2011, the term used in a title of an academic paper from 1971; another paper from 1960, here is something from 2017 on [2]. Overall, I think the lead of this list should be split into a proper article. Thoughts? Ping User:Volcycle (on a side note, recent Polish presidential elections should be added here: [3], and most people considered 48.5/51.5 from last one close too 2015 Polish presidential election, so was 1995 Polish presidential election, others had at least 2-3% difference, some much more). PS. Anyway, digging into the scholarly works a bit more, the more common term I get more hits for is election closeness: [4] --_{Piotr Konieczny aka Prokonsul Piotrus| reply here} 05:55, 13 July 2020 (UTC)[reply]

Definition of close

Isn't the current definition ("a margin of victory of less than 0.1%") of close WP:OR? Many other elections with a bit wider margins were often called close. Of course, increasing the margin makes the list bigger and bigger, but I think it should be considered for at least 1% if not 2% of a difference. --_{Piotr Konieczny aka Prokonsul Piotrus| reply here} 06:05, 13 July 2020 (UTC)[reply]

Margins

Hi @Volcycle:, I wonder why you removed multiple Ukrainian close elections I have added. Ukrainian elections are almost never between two but always between multiple candidates, so margins are calculated compared to all participants and not just the first two. For instance, in 79th constituency (Vasylivka) in 2014 there were 78,839 voters, out of whom 17,197 (or 21.8128%) voted for Bandurov and 17,180 (or 21.7912%) voted for Hryhorchuk. I thus get a margin of 0.0216% which was reflected in the article. You replaced it with a margin of 0.0495%, what does it correspond to? Thanks — NickK (talk) 10:28, 29 July 2020 (UTC)[reply]

Hi @NickK:. The reason that I remove some of the elections you added is that they were not close enough. When using an election with multiple candidates we only consider the closest two. So for an American election, for example, we would only count the winner and first place finisher, we don't add in all the other votes cast. Likewise a race with 8 candidates and 3 winners, we only consider the votes for the 3rd and 4th place finishers. In that case if they have 10000 and 9989 votes each, it's not close enough - even if there are a million other votes cast. It makes logical sense. That race is no closer than a 2 candidate race with the same vote breakdown so it shouldn't be treated differently. To calculated the margin, I use the equation (vote margin/(vote for winner+vote for loser)) so for that race it would be 17/(17197+17180). Hope that helps. Volcycle (talk) 18:35, 30 July 2020 (UTC)[reply]

@Volcycle: Thanks but this is a very counter-intuitive definition of margin. Election margins are not calculated this way, a typical way of calculating an election margin is by dividing the majority (in number of votes) by the total number of votes. For instance, North East Fife (UK Parliament constituency) says that the margin in 2017 was at 0.004% (a usual way of computing this margin), while this article provides 0.00727%. Do we have a source on such computation? It looks rather WP:ORIGINAL in my view — NickK (talk) 19:22, 30 July 2020 (UTC)[reply]

@NickK: Routine calculations are not ORIGINAL. I don't think it's counter-intuitive, but even if it is, that doesn't make it wrong. In the election you cite, Gethins won by two votes over Riches. The argument you're making is that had 1000 more people voted, but all voted for Garton, somehow that would've made it closer. But that's illogical. The margin is the difference between Gethins and Riches, and the number of votes Garton gets is irrelevant. Or do you disagree. Do you think the number of votes won by minor candidates changes the margin between the top candidates? Anyway, this was discussed above under "List organization and scope" two years ago and while only one other person chimed in there has been a lot of time for others to argue against it.Volcycle (talk) 21:55, 31 July 2020 (UTC)[reply]

@Volcycle: The definition of the margin as % between the top two candidates only looks WP:ORIGINAL.

Here the margin of victory is defined as the smallest number of votes that need to be changed to change the election outcome. A highly-cited article defines it as the smallest number k such that k voters can change the winner by voting differently. One more article defines it as the number of ballots that must be changed in order to change the outcome. Ballotpedia defines it as the difference between the share of votes cast for the winning candidate and the second-place candidate. Similar definition is given by The Hindu: the percentage point lead that a candidate or party has in an election over the candidate or party ranked second. All this definitions have one thing in common: they consider all voters and not just those who voted for the top two candidates. I would call the 'scholarly' definition gives the smallest number of votes that need to be changed, while the 'popular' definition gives the difference in %.

Most of those articles are defining MOV as the difference in votes (a whole number), not percentages. They aren't concerned with closeness. The 2nd article listed states that "For the plurality rule, the margin of victory is very easy to compute, which is the difference between the highest and the second-highest plurality scores divided by 2." thus explicitly ignoring all other candidates other than the top two. Only the last two use percentages and thus all candidates. None seem particularly concerned with determining the closeness of elections for comparison. The Guiness Book of Records has declared a "closest election" but not defined how they determined it.

North East Fife (UK Parliament constituency) is a really bad example here (just two votes and I don't know how the Returning Officer would have voted), so I would rather pick the next closest one, Perth and North Perthshire (UK Parliament constituency). The 'scholarly' definition says that we need to find the smallest number of votes, which would be 11 Wishart voters casting a vote for Duncan instead (Duncan would have won with 21,794 votes vs 21,793 for Wishart), giving 11 / 51,525 = 0.0213%. The 'popular' definition looks for the difference between their scores, which would be (21,804 / 51,525 - 21,783 / 51,525) = 0.0408%. However, none of these definitions gives the 0.0482% currently in the article. Because of course having more voters for third- or four-placed candidates would make an election even closer, as this would make the difference between the first- and second-placed candidates even lower in terms of % of votes — NickK (talk) 15:38, 1 August 2020 (UTC)[reply]

I disagree that having more third party candidate votes makes an election closer. It doesn't change the smallest number of votes that need to be changed to change the election outcome, for example. Nor does it change the number of ballots that must be changed in order to change the outcome. It only means there were more votes cast. It does not make logical sense that adding votes to Roemmele's total, in your example, in any way makes the race "closer". Volcycle (talk) 22:26, 3 August 2020 (UTC)[reply]

On your first point, I can find way more sources that fall into either category: taking the margin between the first two candidates for the plurality rule or that same margin divided by two. What about your definition? Do you have any source defining the margin as the difference between the top two candidates divided by votes for these two candidates only?

If we are talking about the smallest number of votes that need to be changed, we need to divide this number by something. This something changes. Convincing 10 people out of 20,000 or 10 people out of 100,000 is significantly different. In addition, third-party candidates may be allies or opponents of the top two candidates, thus affecting votes distribution — NickK (talk) 11:19, 7 August 2020 (UTC)[reply]

Should we add a column for "Margin of Recounts"?

I think, due to a lot of these elections having recounts, there should be a percentage difference shown between the inital count and the final count

.

Nexxl (talk) 18:51, 15 April 2022 (UTC)[reply]

How to categorize 2019 Boston City Council election

The race for the fourth at-large Boston City Council seat was decided by a single vote in 2019. I'm just not quite sure which category this falls under. -A-M-B-1996- (talk) 17:12, 23 May 2022 (UTC)[reply]

2024 California's 16th congressional district election primary (2nd/3rd place tie in two-round election)?

I'm not thinking of posting this yet as there is a recount pending, but would 2024 California's 16th congressional district election qualify if the recount came with the same results or sufficiently close result? I ask this because this tie wasn't to determine the winner but which of the next two who would qualify for the next round; should it remain tied, all three would advance to the the general election. In general, would primary or first-round elections and/or elections with a tie lower than first but with a significant effect qualify for this list? T.c.w7468 (talk) 02:34, 11 April 2024 (UTC)[reply]

Close referendums

Should a category for close referendum results with the same criteria as for close elections be added? Matveytabakh (talk) 17:31, 20 November 2024 (UTC)[reply]

I don't see why not. Alaska Measure 2 2024 would be a great candidate for inclusion. LV ^{✉ ✎} 21:56, 20 November 2024 (UTC)[reply]

[1] ttp://www.nytimes.com/2015/11/21/us/mississippi-house-race-comes-down-to-one-deciding-straw.html?_r=0

[1]