This page is within the scope of WikiProject Elections and Referendums, an ongoing effort to improve the quality of, expand upon and create new articles relating to elections, electoral reform and other aspects of democratic decision-making. For more information, visit our project page.Elections and ReferendumsWikipedia:WikiProject Elections and ReferendumsTemplate:WikiProject Elections and ReferendumsElections and Referendums
This page has archives. Sections older than 31 days may be auto-archived by Lowercase sigmabot III.
RfC on sourcing for political candidates in the United States
There's been an ongoing debate on pages for elections in the US about what kind of sourcing you need to move a candidate from the "Filed paperwork" or "Potential" section to "Declared." A lot of editors seem to agree that an FEC filing is not enough, because it is very common for people to file with the FEC without ever actually campaigning. Ideally, we want to cite a news article that says the candidate is running. However, it's easy to find candidates who've filed with the FEC and launched a campaign website but who haven't been mentioned in any news articles. Moreover, some news articles say that a candidate is running purely because they've filed with the FEC, even though the candidate has no online campaign presence and the author of the news article doesn't seem to have reached out to them to confirm they're actually running. In my eyes, there are three proposals for how to deal with this:
1. Maintain the current system. A candidate can only be moved to "Declared" if there is an article from a reliable news source that says they're running.
2. Allow someone to be listed in "Declared" if they have filed with the FEC or the relevant state/local elections agency.
3. Allow someone to be listed in "Declared" with two citations: first, a filing with the FEC or the relevant state/local elections agency, and second, a self-published source from the candidate that says they're running--e.g. a campaign website, campaign social media account, or fundraising page where the candidate explicitly says they're running (so "I'm exploring a candidacy" or something like that wouldn't be good enough).
Status quo (option 1). It's the closest to the GNG right now. Wikipedia Policy has traditionally merited inclusion by due weight and some form of the GNG, and I see no reason why we should deviate. If Mitchell Random Hunt of Charlotte, North Carolina, decides he wants to run for president and only the FEC is filed as a source, but no major news reliable independent websites mention it with significant coverage, then Mitchell Random Hunt shouldn't be included. We follow, not lead. InvadingInvader (userpage, talk) 18:09, 25 July 2025 (UTC) (summoned by YapperBot)[reply]
Question: When you say "it is very common for people to file with the FEC without ever actually campaigning", do you mean that they file with the FEC but don't appear on the ballot paper, or that they appear on the ballot paper but don't campaign (i.e. are a paper candidate)? Number5719:44, 25 July 2025 (UTC)[reply]
The former. Every election you get plenty of people who file a declaration of candidacy with the FEC, but they don't have any online campaign presence, don't really report any fundraising, and don't appear on the ballot. So, in my opinion, an FEC filing is not enough to list someone in the "Declared" section--that's what the "Filed paperwork" section is for. BottleOfChocolateMilk (talk) 20:50, 25 July 2025 (UTC)[reply]
Option 1 until the ballot is finalized. Once the filing period is over, we should use the official record of who will be on the ballot. --Enos733 (talk) 22:20, 25 July 2025 (UTC)[reply]
Comment: Most (all?) states have requirements that candidates must meet for their name to appear on the official ballot -- sometimes it is paying a filing fee, sometimes it is filing a certain number of signatures on petitions, and sometimes it is both. The states then have "filing deadlines" by which those requirements must be met. I suppose some states provide online lists of who all managed to pay the fees/collect and file the signatures. But, also, that some of these people won't actually appear on the ballot because their signatures end up being insufficient. The states might not officially certify a candidate as having made it onto the ballot for some (possibly lengthy) period of time. There's having your name appear on the state's list of people who appear to have met the filing requirements, and then there's actually being certified by the state. I don't know if it is worth it to define those stages.Novellasyes (talk) 13:34, 29 July 2025 (UTC)[reply]
Option 1. We go with what the sources say; digging into FEC filings ourselves to determine who the candidates are is original research. My one caveat is that the source doesn't have to be a news source specifically; in theory other sources could be used to fill out historical elections if eg. a candidate wasn't recognized at the time but later becomes significant and receives coverage as a candidate in other non-news RSes, like a biography or a history book or something. --Aquillion (talk) 03:36, 11 August 2025 (UTC)[reply]
Hiya WikiProject, I'm looking for comments and critique on a new infobox I'm creating at Template:Infobox general election. I do want to make sure this is polished before it moves to beta so I would like comments on it – usability, improvements from other templates, aesthetic improvements, functional improvements, whatever!
I've been watching elections coverage for a long time, and for legislative elections, the most prominent statistic that's being displayed are seat totals and seat changes. See this for example.
For executive (presidential/governor/mayor) elections, it's split between actual raw votes and percentages (no clearcut "winner" vs legislative elections).
It'll actually be great to watch election TV coverage across multiple countries for us to have an idea on what we should present, and which more prominently.
I've moved turnout to the bottom along with the registered parameter to below the party results but above the maps. I do agree that seats won should be the most prominent statistic, though I'm unsure how to achieve such prominence DimensionalFusion (talk · she/her) 20:42, 12 August 2025 (UTC)[reply]
Place seats won at the bottom of the photo with bigger font.
Ditch vote bar graphs. Just show the vote total, then line break, then percentages and swing.
I appreciate the work you've done here in making a TIE alternative that still looks aesthetically pleasing, but I think we're right back at square one when it comes to the biggest concern with TIE: size. This infobox still absolutely dominates the page, stretching far below its section and (on my screen, anyway) bumping against other tables. The big thing I would suggest is removing the map parameters entirely and leaving it as a table of results. I know that is a controversial suggestion, because We Have Always Included Maps, but scope creep is only making these infoboxes longer and longer and really conflicting with the purpose of an infobox. — Kawnhr (talk) 22:05, 12 August 2025 (UTC)[reply]
This actually does make quite some sense. If the purpose is to summarise, does the map help that if it is included in the infobox and not the results table, because lots of editors do find it to be unnecessary duplication to include it twice. If others agree I would consider doing this DimensionalFusion (talk · she/her) 22:14, 12 August 2025 (UTC)[reply]
I think this is a discussion perhaps beyond the scope of just this template. I do agree that there is some creep in terms of the extent people place maps in the infobox (they are meant to improve the infobox, not simply to show off your work), but election maps can in many cases increase the understanding at a glance, which is in alignment with the point of an infobox. Also, the equivalent templates on the Spanish and French wikis do feature election maps. Gust Justice (talk) 22:43, 12 August 2025 (UTC)[reply]
Several "maps" have been cramming a lot other than maps: graphs, seat diagrams, for examples. These should be reverted back to just maps. We have another parameter for seat diagrams. The template here is already doing its job with graphs. Howard the Duck (talk) 10:58, 13 August 2025 (UTC)[reply]
Agreed. We have lots of infoboxes these days with maps that are unreadable. Take, as an example, the map in the infobox at 2012 Japanese general election. I can see it has lettering on it, but I can't read it and that's on a decent sized desktop screen. This is beyond the original discussion here, but I would like to see an effort to move such maps into Results sections, where there is a chance they can be displayed in a readable format. Bondegezou (talk) 13:41, 13 August 2025 (UTC)[reply]
That's one of the issues I have pointed about maps that aren't just maps. It should just be maps and no graphs to the right of Tohoku. There's even a pie chart(?) at Godzilla's home. Both should be removed. I'd probably concede there should be circles on the party-list vote for seats won, but that's something that has to be thought about more. Howard the Duck (talk) 18:23, 13 August 2025 (UTC)[reply]
Agree with this part. There is a bit of a scope creep for some of these maps. Again to repeat myself, this warrants a broader discussion beyond just this template, but ideally maps in the infobox should not include more information than absolutely neccesary. I would argue a color key is neccesary for most of these maps, but ideally it should be less cluttered. Gust Justice (talk) 22:16, 13 August 2025 (UTC)[reply]
I also completely agree that this needs a separate discussion – map making on Wikipedia has very much become a 'showing off' thing for some editors with ever-more information crammed into them. IMO they should be just that – a map. I think any text (including a key) is best avoided in the image to allow the map to be used on any language version of Wikipedia. 2021 Ugandan general election is a good example of a detailed map that is just a map (with the key in the caption). Number5722:27, 13 August 2025 (UTC)[reply]
FWIW, some of these maps are shared by their creators outside Wikipedia, so it is not created with Wikipedia usage exclusively in mind (so more info crammed, maximizing all pixels, the better). Yes, this can be discussed elsewhere. Howard the Duck (talk) 15:34, 14 August 2025 (UTC)[reply]
Thanks, that is useful to know. I'm not against these maps being used on Wikipedia, but I sporadically move such maps out of the infobox and into the main article, and make them bigger. I am thankful for the work done in creating these maps; it's just pointless putting them in an infobox where nobody can read them (without clicking on them). If others would like to join me in this activity, please do. Bondegezou (talk) 16:45, 14 August 2025 (UTC)[reply]
I suppose we can have MOS-like guideline for maps that are meant to be shown as thumbnails in elections infoboxes. Like just a map. For executive elections, we can use shades for winning margins, but for legislative elections involving constituencies, maybe just one color per party. No graphs, charts and seating diagrams (except for multi-seat constituencies, I suppose?). Howard the Duck (talk) 18:09, 14 August 2025 (UTC)[reply]
This is a horrible infobox which I will completely oppose, it takes up way too much space. The current commonly used infobox template is much better and takes up much less space and I will continue to use it over this. Orca🐋 (talk) 22:13, 12 August 2025 (UTC)[reply]
I don't think calling it horrible is quite warranted. As for your space comment, I appreciate it does take up more vertical space than TIE, but it also takes up less horizontal space and doesn't go 1 -> 2 -> 3, new row, 4 -> 5 -> 6 and instead decends vertically. I do intend for TIE and TILE to continue to be used in their own cases such as elections for a single executive position, or parliamentary elections with lots of parties with seats and this won't displace those use cases, but this is intended for situations that sit inside the spectrum between the two DimensionalFusion (talk · she/her) 22:21, 12 August 2025 (UTC)[reply]
Horrible is warrented and that doesn't fix the problem. There is already an existing template which is less bulky and standardized throughout Wikipedia, this will simply exacerbate issues with template mismatching across pages for an already fine template. Orca🐋 (talk) 22:39, 12 August 2025 (UTC)[reply]
Make the seat result more prominent by using the bar chart for that parameter rather than percentage.
Change the order of seats/votes/% to votes/%/seats
Add a seat change parameter (listed alongside the seat parameter)
Put the leader's name in a separate row to the party and mark them as 'Leader'. Currently it's not clear who the person listed is – are they the party leader or a mayoral candidate?
I notice if all the leader images are missing, they are all replaced with flags. If images aren't available, I think it would be best to hide that column – potentially a better use of space.
Overall, I'd say it's a definite improvement on TIE – being a purely horizontal listing avoids the ridiculous blank spaces we end up with in infoboxes with 5, 7 or 8 parties/candidadate. Also, the claim that 'it takes up way to much space' and is more bulky is laughably wrong. This is what the previous infobox looked like; the new one is noticeably smaller. Number5722:55, 12 August 2025 (UTC)[reply]
All of these suggestions look to be reasoned:
I agree that for parliamentary elections the focus is on the seats and not the vote share which, unless under pretty perfect PR, is not relevant. I agree that the composition bar should display seats and not vote share for parliamentary elections – though I feel that if the template were to be used for elected executive positions for whatever reason it should display vote totals. Works for me
Keeping the actual numbers and the percentages seperate does sound reasonable – Works for me
Seat change parameter existed but was not implemented correctly – Done
Seperating leader from party seems entirely reasonable, people not knowing who the significance of the person being listed is indeed a problem and it shouldn't just be those who already know of the politics and electoral system who are "in the know" – Works for me
On this, I'm mixed: removing the image does sound like a reasonable step when one is not provided as the flag doesn't really add too much. However, I think it might bring it out of alignment of all the other parties/candidates and make it stick out, which is also why such templates as Eppt is used on TIE pages. So for now I'll just say Needs discussion
On the final point, I meant if no images are entered at all, that column should be hidden. I agree that if there is only one missing, it wouldn't make sense to realign that row only. I also agree with Gust that using Eppt would be better than a flag. Number5723:53, 12 August 2025 (UTC)[reply]
Ah, right that does make sense. Hiding all the images when none are provided makes a lot more sense than taking up space with placeholder images. Should be simple enough to write a check to do that – Works for me
I suppose it's just novelty that wants me to keep the flags, but I do see the utility of using the party-specific abbreviation over a generic flag. But if a party did not exist in the political party module, should it use the eppt method of displaying ERR, or not display, or display the flag as a last resort? DimensionalFusion (talk · she/her) 00:02, 13 August 2025 (UTC)[reply]
I don't believe that displaying nothing would really work – as if only one party was missing an image, it would not be in alignment with all the other parties. Maybe a version of eppt could work DimensionalFusion (talk · she/her) 12:35, 13 August 2025 (UTC)[reply]
It seems that the flags/logos/photos are on a separate column altogether. Leaving it blank means the text part of the party name is still aligned.
The {{{noimage}}} is now Implemented. I'm working on adding an automatic check so that if no images are provided the parameter will be automatically set to true instead of having to set the parameter manually to supress the flag images.
The seats have now been moved below percentage and vote, Implemented
Re the last election, I'd make it a single row (rather than having the percentage and seats on different rows, which creates a lot of wasted space) and replace 'In [year]' with 'Last election' so it's clearer. I'd also have the percentage listed first so that people don't think it's a percentage of seats won. Number5710:57, 13 August 2025 (UTC)[reply]
The collapsing of the last election field isn't (wasn't) intentional, I've Fixed now so it'll all be displayed on one row. I agree that the "Last election" is clearer than "In [year]" so that is Implemented.
I'm saying the percentage should go before the seats, not the vote totals (I agree percentages should follow vote totals), so it would be "Last election 50.60%, 66 seats" not "Last election 66 seats, 50.60%" to avoid people thinking that 66 seats = 50.6% of the seats. Number5712:39, 13 August 2025 (UTC)[reply]
I would like this template to be an option, regardless of the question of whether it should be the preferred option. That said it needs better documentation. I couldn't get it all working when testing it out. I would like for the alliance parameter to be useful as well. Preferably, it should also be possible to by default have a text containing the party abbreviation, in place of a generic flag. Like the Eppt template. Gust Justice (talk) 23:22, 12 August 2025 (UTC)[reply]
Yes, the documentation is very much a work in progress and not entirely useful, as I've yet to do some parameter consolidation (such as with the alliance field which only generates the title of an unbulleted list, the contents of which are given by partyX coalitionX) – and I do hope that soon the template can become accessible to people that don't know the internal workings of the source code. To your second point, I'm a bit torn because a flag with the party's colour and an abbreviation of the flag both seem roughly equal in usefulness DimensionalFusion (talk · she/her) 23:56, 12 August 2025 (UTC)[reply]
Ideally it might be made flexible enough that either can be chosen depending on what is appropriate. I am just not the biggest fan of using flags for it. Gust Justice (talk) 10:00, 13 August 2025 (UTC)[reply]
As per others, I think they look better than what they are replacing, so good stuff! But they retain the same problems as what they are replacing (too big, unreadable maps, etc.). Bondegezou (talk) 13:35, 13 August 2025 (UTC)[reply]
@Bondegezou I'd disagree there. Only 5 parties won seats which is, in my opinion, far too few to justify TILE which is mainly used in countries with far, far more parties winning seats. A TILE infobox here would have the opposite problem in my opinion, being far too small to adequately summarise the election DimensionalFusion (talk · she/her) 19:27, 14 August 2025 (UTC)[reply]
MOS:INFOBOX is clear that infoboxes should be small. I quote from MOS:INFOBOXPURPOSE: "The less information that an infobox contains, the more effectively it serves its purpose, allowing readers to identify key facts at a glance. Some infoboxes need to use more than a handful of fields, but information should be presented in a short format wherever possible, and should exclude unnecessary content." The infobox you have at 2025 West Northamptonshire Council election is huge, it goes on for several screens. It includes fairly minor information (leaders' seats).
MOS:IBI talks about adding "an image". That infobox has 9 images. MOS:IBI continues, "When adding an image to an infobox, thumbnails should NOT be used." That infobox has lots of thumbnails. It has a council crest in the flag_image field. MOS:INFOBOXFLAG says, "Generally, flag icons should not be used in infoboxes, even when there is a "country", "nationality" or equivalent field: they could be unnecessarily distracting and might give undue prominence to one field among many." A council crest also fails MOS:NONSOVEREIGN.
So, that's six clear violations of MOS:INFOBOX. A TILE infobox would summarise the key facts: what parties won what seats, what vote share, changes in seats, changes in leader of the council. I think it could still include the map and the arc of council seats images. Bondegezou (talk) 09:40, 15 August 2025 (UTC)[reply]
I still feel TILE does not summarise the article, which is MOS:INFOBOXPURPOSE. For an election article, the infobox should summarise the article, not solely the results of the election. The leaders’ seat in the article, for example: the infobox displays some important context in that the leaders of lots of parties lost their seats DimensionalFusion (talk · she/her) 10:04, 15 August 2025 (UTC)[reply]
I agree that TILE is insufficient for elections like this, however I do not believe that leader's seats should be in the infobox, especially as leaders do not usually lose their seats; if we implement such an idea, then it would be hard to find a threshold in which it would work. However, this specific point is potentially out of the scope of this discussion, as it affects many TIE infoboxes too. Quinby (talk) 10:12, 15 August 2025 (UTC)[reply]
The leaders’ seats is about the 39th most interesting fact about an election. It is trivia that isn’t even mentioned in the article text of most council election articles.
Well as I say, this infobox is indev so parameters like leaders’ seat could be eliminated in these early stages. If a couple of people agree then it can be removed very easily!
To your second point, I definitely agree that these are not being followed, but those seem to be more article/technical than relevant to the infobox itself – not using a thumbnail is a technical problem that can be easily fixed so this is to do, and your point on notsoverign is definitely valid but that’s more relevant to the article than the infobox template DimensionalFusion (talk · she/her) 12:04, 15 August 2025 (UTC)[reply]
I am conflicted over the existence of this template and its utility in most scenarios (especially how the bar charts could be unwieldy), however in this situation I am of the view that if it is to be used for many elections, it would be preferable to use party logos. Parties are not one person (well, in most cases, looking at you PVV). In the example given of Northamptonshire, Reform's local presence was very little known and their campaign joined with their national success, so stressing the leader seems improper. In scenarios where it is about Prime Ministerial candidates, that works, but elsewhere, less so. Quinby (talk) 10:22, 15 August 2025 (UTC)[reply]
Agreed. It is misleading to emphasise little known council leaders. (Indeed, some council party leaders aren’t even chosen until after the election, so it’s misleading to show them.) It is very clearly not the job of an infobox to show headshots of a large number of people. Bondegezou (talk) 10:45, 15 August 2025 (UTC)[reply]
I don't think it's misleading, because the decision of how "personality/leader-based" an election is is incredibly unquantifiable, and deciding to include/not to include the leader based on the quality of being "little-known" would be OR DimensionalFusion (talk · she/her) 11:50, 18 August 2025 (UTC)[reply]
If that's the case, let's just TILE instead. This was just basically TILE with photos of leaders, and people refuse to use TILE because it's ugly. Howard the Duck (talk) 14:30, 13 August 2025 (UTC)[reply]
@Bondegezou:, how small can a thumbnail be for you to not consider it to be encompassed by MOS:ICON? I'm wondering this because ultimately, every photo of a party leader or party logo will be considered as "decoration". Howard the Duck (talk) 22:30, 15 August 2025 (UTC)[reply]
There’s no problem with having pictures of party leaders or logos in the main article text, where they can be shown at a sensible size. But having these tiny headshots in an infobox as above comes under MOS:ICON. MOS:LOGO has further advice on logos. Bondegezou (talk) 10:40, 16 August 2025 (UTC)[reply]
My comment is that I don't like the horizontal bar charts that only have one item in them. A chart makes sense when it has the data for all parties together, but it's weird to have multiple related charts that add up to ~100% but are split into several individual data points. Reywas92Talk04:31, 14 August 2025 (UTC)[reply]
I think some of the difficulty here is that whilst some have been saying for quite a while that en-wiki should adopt an infobox more similar to FR/ES, when it does make its way here, we find that it wouldn't meet policy. See the Spanish version of 2025 Bolivian general election; the infobox there is incredibly long and if implemented in en-wiki would get shot down for its length DimensionalFusion (talk · she/her) 12:06, 18 August 2025 (UTC)[reply]
I think we could make things shorter than the Spanish wiki by cutting down on having parl charts in the infobox, not having two maps (or have a switcher), not including candidates/parties below the electoral threshold, and not including other non-essential information, like the length of the campaign period and list of debates. Gust Justice (talk) 23:35, 18 August 2025 (UTC)[reply]
Currently, there's been issues with maps that are placed on infoboxes where aside from geographic info, it also include things such as graphs, charts and parliamentary diagrams.
Ideally, there map at the infobox should be simple and should only be a map excepting on specific circumstances. A seat diagram can be added on another parameter.
Here are my suggestions:
Executive positions (presidents, governors, mayors) or single-winner elections or referendums:
Single color for winner in a specific place
Can use gradients if there are 3 or less candidates, and if the colors are distinguishable.
If maps are to be created for a series of elections, it's recommended to be consistent for all, either use gradients or single color for all.
As for legends, I suppose a simple legend of what color stands for who/what on the map itself can be used. If there are to be gradients I suppose it can be also added on the map per se or on the description page. Howard the Duck (talk) 22:49, 15 August 2025 (UTC)[reply]
There are too many map creators and too many types of election for a standard to be completely established in my view. There can be guides to work towards, but it cannot be shifted to standardisation with code like infoboxes.
Regarding a guide around gradients, less than 4 candidates rules out practically all elections. I understand not necessarily having a gradient if there are many parties (e.g., 5+), but 4 would mean even many US races could not have a gradient.
Apologies, for clarification the 5+ meant 5+ parties with relative success, i.e., winning seats or getting over 5%. It could also be framed in terms of how many parties win districts, as often this is many fewer than the number of competitive parties. Quinby (talk) 23:11, 15 August 2025 (UTC)[reply]
I was referring to a party winning a specific location (e.g. county). This means most US maps will only have 2 parties winning a county, so they can have a gradient. I don't think there third parties in the US are that strong in which more than one of them wins a county each. I can understand changing the maximum number of winning parties to more than 3. I also understand this can be different on primary elections. It also depends if the two major parties had the similar colors (see 2016 Philippine presidential election for an example). Howard the Duck (talk) 23:36, 15 August 2025 (UTC)[reply]
It is probably not achievable to have some universal rules on this. Even within countries that on the surface have a similar electoral system (e.g. same district magnitude) might substantially be quite different from each other. I do agree I do agree that if there are multiple acceptable maps, preference should generally be given to one which is consistent for the series of elections. I think party colors in general should be the same as the one used on Wikipedia, but this can't be an absolute rule, just because there will be instances where multiple parties have extremely similar "official" colors, and so using those will make the map hard to read. Gust Justice (talk) 22:32, 16 August 2025 (UTC)[reply]
I don't think we need to get this into the weeds, since graphics need to be edited manually and we're already knee-deep in thousands of maps — it would be a herculean effort to bring everything in-line. I think what we really need to do is establish a consensus of what sort of thing belongs in the infobox map, and what doesn't. My view is that they should be kept as simple as possible: a map colour-coded for the results, and that's it. A summary of the vote/seats is acceptable if it can be tucked into otherwise-empty space, but things like a list of constituencies or a circle graph that compares vote count to seat count to turnout, or a badge declaring the outcome (???) really shouldn't be there. Also, the seating plan of the legislature after the election can be used as an alternative to a map, but not in addition to — we don't need two visuals for the election outcome, especially since the latter is going to be used on the relevant Xth Foo Parliament page.
I also think we need to keep infoboxes clear of maps that are not results but just the location of the place the vote took place. That might sound obvious, but this is is, bizarrely, standard practice for UK by-elections, eg. I actually went through and deleted all of them a while ago, but they've returned for the new by-election articles that have popped up since my sweep. I really don't understand what's going on here. — Kawnhr (talk) 17:14, 17 August 2025 (UTC)[reply]
For colors, there can be exceptions. For example, US maps do not use the exact hexcodes for Democrats and Republicans. This should be fine if there is a two-party system for at least a century, where no other party claims a similar color.
For parties with similar colors, this where gradients aren't useful, at least on an infobox. Maybe it can be created in a results section, but not for the infobox.
Yeah, those examples on Australia and Canada are indeed egregious. It should only be a map, with insets. As for by-elections, in a case in the Philippines special elections, a background section where the district (constituency) is, and the infobox shows the result per party. See 2023 Cavite's 7th congressional district special election, for an example. I'd probably concede that prior to election day, the location map can be added on the infobox, but after that, it should be a results one. Howard the Duck (talk) 20:32, 17 August 2025 (UTC)[reply]
I agree with the general take on this. We should prefer the map in the infobox (but not neccesarily those lower down in the article) to not have too many features that aren't immediately relevant to understanding the election. While some features can be justified in some instances, there does seem to be a scope creep among some maps, where a lot of superflous features are included, just because they can be. Generally I would say it should be limited to showing a color key, a map of constituencies/administrative divisions potentially shaded by vote share of largest party, and circles indicating seats won. But again, to repeat myself, no country is exactly the same, and there may be exceptions. Gust Justice (talk) 08:12, 18 August 2025 (UTC)[reply]
Yes, I should clarify that I'm not opposed to more detailed maps elsewhere in the article. In theory, we could replace the current infobox maps (or edit them to remove the extraneous features), but move the originals down to the results section. — Kawnhr (talk) 21:51, 18 August 2025 (UTC)[reply]
We should follow MOS:INFOBOX and have compact infoboxes that provide a brief summary, and push back against the trend to try to include everything in the infobox. Results maps are great, but much of the time, they should be in the article text instead.
Along similar lines to Gust Justice & Kawnhr, I don’t think we need detailed guidance. Instead, I think we need to push basic principles. As per Kawnhr, that means a small number of images only (generally 0 or 1 image); and no empty maps before results are in; but, most importantly, images have to be sufficiently readable when viewed in an infobox.
I suggest the latter is the biggest problem. This is far from the worst example, but consider 2023 Argentine general election. On a reasonable sized desktop screen, I can’t read most of the words. I can’t tell what the side blobs are. I don’t know why there are two Argentina-shaped bits. But I can see that Massa’s strength is in north central Argentina. On a smartphone screen, it’s even worse. That is simply unacceptable. I think that graphic should be moved from the infobox to the article, where it can be displayed at a larger size. Or try 2008 Pakistani general election. I can’t read any of the words. I can’t make out the legend. I have very little idea of what is going on here. Or 2024 Slovak presidential election. I can’t read most of the words. The colours don’t match those used higher up. The bar chart on the right of the graphic is unnecessary. 2024 Lithuanian presidential election, 2024 French legislative election, the same.
Instead of writing long guidance, I think we need editors enforcing good practice. Can we commit to going around and removing inappropriate maps, moving them to the Result section where suitable? Bondegezou (talk) 08:00, 18 August 2025 (UTC)[reply]
I agree that there are problems with the maps you mention, but I don't know if it is always achievable that the map can be easily viewed without opening it in full screen. Especially if it's e.g. 2024 United Kingdom general election where there's just no way to view all 650 constituencies if the image is only 300px wide. Gust Justice (talk) 08:18, 18 August 2025 (UTC)[reply]
Not being able to "read words" seems like overstreched criticism to me: just click on the image to get a close up of it. Otherwise, this argument would mean that we should discard maps altogether, even in the article text (and not just for election articles!), because maps will typically not be shown at screen size due to these being unmanageable.
My two main criticisms of maps come with 1) consistency among maps of different elections, at least within the same country; and 2) consistency of colours with those used in the rest of the article.
Beyond that, if we agree that infoboxes are indeed intended as summaries, maps do actually fulfill that purpose quite well, so a commitment to remove them from there seems quite an exaggeration. But it will also be nigh to impossible to agree to any set of rules for maps (aside of encouraging good practice), because:
Each type of election is different and has its own peculiarities (i.e. "vote blobs" may have less sense for parliamentary elections, but they indeed do for presidential elections where the electoral system is directly based on the popular vote each candidate gets).
Each country is different and has its own electoral/political dynamics.
Electoral dynamics are not static, but quite fluid. What may seem to work today may not work tomorrow.
You cannot agree on a set of rules to be applied on all elections for all countries all the time. This will be prone to (senseless) edit warring, and will require a continuous surveillance effort which, frankly, I don't think anyone of us will be willing or able to maintain.
I also don't think that limiting creativity is actually positive: sure, sometimes it leads to messy situations, but I have frequently found useful information from particular designs in maps that I hadn't thought of myself. Creativity is a plus and should not be diminished, but encouraged. I think, however (and I agree with Quinnnnnby here) that we can (and probably should) establish some kind of guide, similar to naming conventions and the such, to harmonize the design of election maps as much as possible without excessively limiting creativity. On here, I also agree with Kawnhr and Gust Justice's comments in favour of simplicity, consistency and practicality when it comes to maps' design and what to (and what not) to have in infoboxes. Impru20talk08:27, 18 August 2025 (UTC)[reply]
Re: articles like 2024 Lithuanian presidential election, 2024 French legislative election: I would also say that I don't see the point of including results for both rounds if the infobox is only displaying the results of the second round. The switcher thankfully keeps the infobox from ballooning to two images, but this still strikes me as shoving more information into the infobox than it actually requires. — Kawnhr (talk) 22:04, 18 August 2025 (UTC)[reply]
Would you agree that switchers are a bad practice as well? There were also switchers in some articles I had edited on which did not work and the people who insisted on doing that had no alternatives (LOL). Howard the Duck (talk) 18:22, 20 August 2025 (UTC)[reply]
I think there's a time and place for them. They work well in location infoboxes, for example: Toronto has a switcher showing its location in Ontario, and then in Canada as a whole; the first is more specific but the latter probably more helpful for international readers, so both serve a purpose. But too often I feel like they're being used to add something where it doesn't really belong. — Kawnhr (talk) 20:27, 20 August 2025 (UTC)[reply]
The purpose of an infobox is to give a quick overview. If I have to click on an image to be able to make out what is going on, then that image is not suitable for an infobox. It has clearly failed the purpose of an infobox. An image in the main article can be bigger and, there, if someone needs to click on it to see it even larger, that’s fine.
I am absolutely going to push back on the idea that not being able to “read words” is some minor consideration. It is a fundamental necessity that readers can see what is in the image. Bondegezou (talk) 18:27, 20 August 2025 (UTC)[reply]
The words issue is quite fundamental IMO, not just because of readability (and violating WP:ACCESSIBILITY, which states that no text in an article should be less than 85% of the default font size) but also because maps should be designed to be used on all language versions of Wikipedia, and having text on the maps in a particular language is limiting.
Going back to the original point, I do think we need some rules around maps, but I would not include shading, but rather things like:
Text should not be used as maps should be able to be used on all language wikis
Keys should not be included; they should instead be included in the image's description so they can be added to the map_caption
Parliamentary diagrams (which are almost always included in the article itself) should not be included
No other form of graphs, pie charts etc should be included; it should simply be a map
Territory claimed by the country but not recognised as being part of it by almost all other countries and in which the election should not be held should not be included (e.g. this map would not be allowed as it shows the Venezuelan claim on Guyana)
I don't have a problem with switchers; they are great for situations like showing the first and second round results in presidential elections, showing results at different levels of jurisdictions, showing different sets of results in mixed systems, or showing presidential and parliamentary results. Number5720:48, 20 August 2025 (UTC)[reply]
My personal view would be that the amount of text should be minimized, but not having any text at all might not be realistic for all maps. For instance I do think there is a benefit of showing constituency names for proporitonal elections. This usually isn't a language issue since most place names don't need to be translated.
I agree that having an overly large key like in 2022 Latvian parliamentary election should be avoided. However a key may be neccesary if e.g. it's an election where a lot of different parties win a plurality of the vote somewhere. Moving the key to the infobox itself could easily be unwieldy. For elections with a two-round system, the key should be in the infobox, as it can fit there.
I agree on this. They basically never add anything useful to the map.
I would say preference should be against including these as additional elements, but I wouldn't have a hard rule against it.
In principle I agree, but this could give cause to problems for how to implement it. What if the country itself records votes being cast in the disputed area? Or it's a separate constituency? There could be weird edge cases that would need to be handled.
Re constituency names, I see your point, but maps could be used in a Wiki that doesn't use the Latin script. I think it's best to be truly universal. Re the last point, if the country controls the area and votes are cast in it (e.g. Crimea/Russia) then it can be shown but highlighted differently; I'm more thinking about situations like Venezuela/Essequibo, Guatemala/Belize (I edited a load of the Guatamalan maps to remove Belize), Argentina/Falklands where the countries in question have zero control over the claimed area and no votes are cast. Number5714:42, 21 August 2025 (UTC)[reply]
Text: This can be useful on insets, more so if there are multiple.
Keys: I can understand on having non-Latin scripts, but this means we'd have to put the legend on the infobox per se. Having the reader go to another page to see the legend is a disservice. If the hex values are identical to the infobox, then we'd have the argument that all parties that are on the map should be on the infobox.
Parliamentary diagrams are another parameter on the infobox that nobody uses.
I've seen it with multiple candidates on other articles (can't remember where) and it didn't take up that much space. Number5715:44, 21 August 2025 (UTC)[reply]
I present the ugliness that is the old version of the 2022 Philippine presidential election. Broken switcher, has gradient legend, the "results by margin" map has the legend that cannot be read on thumbnail.
I would actually rather the legend be included in that map; the legend being done in text means that the infobox is now six lines longer. That's not massive bloat (particularly for this Ugandan box, which covers multiple elections; those six lines are not what's making it long), but most maps will have the space to tuck a legend in there. United States elections, like 2024 United States presidential election, always fit them in nicely. — Kawnhr (talk) 16:55, 21 August 2025 (UTC)[reply]
"The purpose of an infobox is to give a quick overview. If I have to click on an image to be able to make out what is going on, then that image is not suitable for an infobox" Then you are basically ruling all UK election maps as unsuitable for infoboxes, because the country's own geography and electoral system makes it nigh to impossible to obtain a "quick overview" from most districts without either clicking and expanding the images or setting a very large size for the maps (2024 United Kingdom general election, 2019 United Kingdom general election, 1918 United Kingdom general election, 1885 United Kingdom general election...). Common sense is warranted. Also, I don't think that MOS:SMALLTEXT applies to images as such but to plain text (which you cannot expand), unless the text is too small within the image itself. Impru20talk11:19, 21 August 2025 (UTC)[reply]
Yes, I'm sympathetic to the view that text should be readable without having to click on the map, but the realities of geography means that simply isn't always possible. Basically unless we're looking at a country that's very square or perhaps a horizontal rectangle; anything very vertical, or diagonal (Japan) is going to necessarily struggle with legibility at a small size. I agree that map makers should pay mind to how it will appear at a small size… but sometimes there's only so much that can be done. — Kawnhr (talk) 17:05, 21 August 2025 (UTC)[reply]
Not at all. All I'm saying is that it would be nice for a map to be perfectly readable when reduced to 300px, and so that's something we should strive for, but because geography rarely conforms to UI design, that means it's not always possible and we will have to accept some maps will require zooming in to fully read. One reason would be, indeed, that some constituencies are very small. — Kawnhr (talk) 16:14, 1 September 2025 (UTC)[reply]
No, no, I did not set out any condition. I just countered the claim that "The purpose of an infobox is to give a quick overview. If I have to click on an image to be able to make out what is going on, then that image is not suitable for an infobox". If we require every aspect of every infobox image to be visible without clicking it, then we will be structurally excluding some countries from having any infobox images at all. Impru20talk12:00, 3 September 2025 (UTC)[reply]
Looking at the discussion here, it was Bondegezou who said that. I don't think what was meant was that all districts (or subdivisions) be visible at any size, although if the purpose of the map is for it to be in the infobox, the text has to be readable at the intended size. Howard the Duck (talk) 12:19, 3 September 2025 (UTC)[reply]
For FPTP elections, it doesn't matter at all as it is winner takes all. Maybe for a separate map at the results section yes, but for the map at the infobox? No. Howard the Duck (talk) 11:34, 1 September 2025 (UTC)[reply]
Yes, and the graphic about the Samoan election that we are originally discussing about here has a lot more information than being simply just a map but people here don't want that, at least on the infobox. Howard the Duck (talk) 17:24, 1 September 2025 (UTC)[reply]
Here's another problematic map: 2025 Singaporean general election's map has the opposite colour scheme as all previous Singaporean election results maps, and even the infobox it's supposedly an illustration for. Disclaimer: I removed it from the article and it was promptly restored, but I am not trying to canvas to find someone else to remove it for me. I bring this up only because I think this showcases a problem we have with some map designs. The re-insertion was justified because (paraphasing) "that's what the legend is for", but at the size the map displays in the infobox, that legend is unreadable, because it's small-white-text-on-colour-block. While it's good practice, I think, for maps to have their legend in the image, those legends should be something like the US example of a block of colour beside the name, because that's much more easily read at smaller sizes. — Kawnhr (talk) 17:11, 8 September 2025 (UTC)[reply]
MOS:INFOBOXPURPOSE states that infoboxes should "readers to identify key facts at a glance". What a glance entails is perhaps flexible, but having to open up things in the infobox to understand them feels clearly in not-a-glance territory, so the image should be understandable at default display size. Per MOS:TEXTASIMAGES, we should be relying on captions to host legends etc. rather than the images themselves, so any image in the infobox too complicated to be understood by the infobox's caption field is probably inappropriate (and harder to use on other language wikis). Within those constraints, some more subjective thoughts after reading above: Real-world constituency maps have their purpose but they can greatly skew perceptions of election results due to different population densities, imply homogeneity within constituencies that doesn't exist (not including of course the constituencies shaded by vote share), and don't work for non-constituency elections anyway, so they seem a dubious default inclusion in infoboxes. Political parties having consistent colours is probably a decent idea, but the idea of consistent colours is very modern, and a reader should be able to understand the article without looking at others, so the specific colour choice is a small issue compared to other considerations. CMD (talk) 01:37, 9 September 2025 (UTC)[reply]
US election articles, use of Race to the WH predictions.
On United States election articles, predictions from notable organizations are usually included in a section near the bottom of a page. Example seen here:
2026_New_Mexico_gubernatorial_election#Predictions
The "Race to the White House" source has been added to most of these 2026 pages, despite the organization not having a Wikipedia page or even a redirect. I have done searches but cannot find any sources that would establish notability for this. Including with a link it may violate WP:ELLIST, which I brought up in a discussion on the talk page of an editor who added it, but they disagreed. Since it was added in such a widespread manner, I wanted to start a discussion on this. I personally oppose it's inclusion unless notability can be demonstrated
From this discussion at the Senate election page, there is opposition to it as well.
I support Race to the White House staying on the numerous pages that it is featured on all throughout Wikipedia for multiple reasons which I will list after addressing two of the points made above. Firstly, this was not recently added, Race to the WH has been a part of Wikipedia since at least April, so this has been around for quite a while. Secondly I was the editor who you wrote about it violating WP:ELLIST, I wrote this back justifying why I included it "I looked at the rules and I believe this qualifies as a high-value external link, providing additional information that is useful to the article. If you click on the link it shows an image/model which could not be uploaded to Wikimedia, this is one of the exceptions to this rule. Lastly, its up to the person who wants to add the external link for it's justification, and I believe that this adds to the article and is an acceptation to what would be a policy violation." You never wrote back, leading me to believe that you understood my reasoning and agreed with it.
However these are the reasons I believe it should stay a part of Wikipedia:
1. It is notable:
While it obviously is not as historically notable, famous, or proven as other predictors like Crystal Ball or Cook, it is not correct to say it is currently not notable. Nate Silver's website, The Silver Bulletin, released an article about 2026 praising WH as a "major forecaster" and for its prediction model. This is very big considering the websites owner, Nate Silver, is probably the most notable person in this field. Also, Politico did a joint interview with Lakshya Jain (co-founder of Split Ticket) and Logan Phillips who, is the founder of WH and calling them two of the nation's "top forecasters" and stating they have "muscled their way" into the conversation of polling analytics. Split Ticket predictions are used all throughout Wikipedia and they are being treated as equals to Race to the WH by the Politico article. Last part of this point, 270toWin (which I'm sure everyone reading this knows) features the WH predictions for Senate alongside the likes of Cook, Inside elections, and Crystal Ball for their major 2026 senate forecasts.
2. It is accurate.
I believe a feature of notability is accuracy and while it is obvious that we don't add predictors for their accuracy (since you could have a random Joe get an accurate prediction and then you get Joe's 2026 predictions on here), but Race to the WH has a record of accuracy. Their previous predictions are up on their website with their track records comparing to other major predictors. Again, I know this is not the reason we add predictors on here but I think it is an important point we overlook in these discussions.
3. There is support to add this.
Except for me, the most prominent voices in this discussion have for the most part been negative, but that does not mean people aren't for this. In this archived chat there are 3 different people open to, or outright supporting the idea of adding WH. In the spain 4 months there have only been 3 people who have been against it on the Senate pages, and there has been no one on the Governor pages who have spoken out at all. I think if adding this was not a popular idea or if it wasn't a good idea, there would be a lot more pushback.
So in short I believe we should add this because it is notable, accurate, and is wanted. Nothing can be notable or be proven until it is given the chance and I believe WH has made the team, but is just on the sidelines waiting to be put in the game. And I feel like it has made all the qualifications to be added. I think this cycle is pushing WH over the sides into a predictor that deserves to be on WH and many other people agree to. MAWJ The Awesome (talk) 23:57, 22 August 2025 (UTC)[reply]
The sources you provide would not make this website notable, only show that it has been mentioned in the past. Race to the WH would fail WP:GNG. As accuracy goes, it is probably irrelevant in conclusions, but listing New Mexico as "Tilt D" is far from what any other predictor would refer to it as such. And I acknowledged previous discussions, but none of them seemed very widespread, which is why I started this. Yoblyblob (Talk) :) 00:01, 23 August 2025 (UTC)[reply]
I would agree these articles on their own would not make the website notable but together they do. Politico is probably one of, if not the most notorious website for pure political news, Nate Silver is the creator of 538 the most well-known and used poll aggregator before its closure (and this is his new website), and 270towin is very notable and prominently features WH. These are all major sites and should be taken seriously and not be written off. They are definitely notable sources and "mentioning something in the past" is not a common consensus of discarding sources.
Also while accuracy is most likely irrelevant in these conclusions, it is simply false to say New Mexico's "Tilt D" is far from what the other predictors say. Cook and Sabato both have it as competitive while Inside Elections literally has it as a toss-up. Also, since WH uses a model, the tilt D translates to a 73.6% lead for the Democrat.
For your last point, I am glad there is a discussion here because I agree that there should be widespread conversation on these matters. But I would also argue the implementation by many people of WH into Wikipedia is also widespread and the culmination of pushback was only some chatter on the 2026 senate page. I'm basically saying while the implantation has been widespread, the questions of its inclusions have concentrated, and if it were a huge issue, those questions too would be widespread. MAWJ The Awesome (talk) 00:39, 23 August 2025 (UTC)[reply]
I wouldn't take the lack of heavy pushback on the 2026 Senate or Governor elections as an apparent acceptance of notability. These elections are still over a year away, so the pages are low-traffic, both from a viewership and an editing standpoint, so our sample size of opinions is still rather low. For me, at least, I have tried removing WH predictions multiple times, but they were usually reverted rather quickly, which is why I started taking to the talk pages to try to reach a consensus instead of starting an edit war. As such, I reject the notion that the pushback was "only some chatter." That being said, I still oppose the inclusion of WH predictions. OutlawRun (talk) 03:07, 23 August 2025 (UTC)[reply]
Even though I have added WH predictions before, I don't remember reverting them after they've got deleted. I've seen it happen but I don't think I usually reverted them. While I agree that we should have discussions here instead of going into editing wars, the fact that they got reverted means the people have been adding it back, like the change, and want it to remain on Wikipedia.
I also don't believe the lack of heavy pushback now means necessarily means there will be heavy pushback later. For starters the midterms have been all over the news and some of the senate races, like Ohio, North Carolina, Georgia, Maine, and Michigan have all been getting Heavy news coverage. Most people click on the first thing that comes up when they search these races online, and that is usually Wikipedia. These articles have not had any pushback
Also, if there is a lack of pushback, that's great! We shouldn't use speculation that these changes will be unpopular in the future to get rid of them right now. But I will say I didn't include everything I found on my quick search on the comments above. I found many other articles praising WH, like a Newsweek article about the 2024 senate races comparing it directly to 538, and praising it. Just because it's new doesn't mean its something we shouldn't add.
Its like I said above, it a random Joe made a prediction and happened to get it right, but there was no coverage of it anywhere and was very obscure, I would totally take your side and say we shouldn't add it. Its just that the dominos have fallen into place already. If we are looking to the heavens for a sign for this to stay in, we're not gonna get one. But we have had many other signs indicating it should be added, and a lot of them from notable sources too. MAWJ The Awesome (talk) 03:33, 23 August 2025 (UTC)[reply]
In 2022, the Census Bureau, at the request of the state of Connecticut, changed its county-equivalents from the eight historical counties (whose governments were abolished in 1960) to the nine Councils of Government (COGs). This has resulted in occasional edit warring between editors who disagree over which map should be included in election infoboxes, and—if both—which should be displayed first. I think it would be prudent for there to be consensus on which of these maps should be used in displaying election results; this would only apply to elections held since 2022.
As I see it there are four ways this can be resolved:
Option A — Only the COG results should be displayed in the infobox. The old county results may be placed elsewhere in the article.
Option B — Only the old county results should be displayed in the infobox. The COG results may be placed elsewhere in the article.
Option C — Both should be displayed in the infobox, but the COG results should be displayed first.
Option D — Both should be displayed in the infobox, but the old county results should be displayed first.
B. Although the county governments no longer exist, they are still understood by the public, and have stable boundaries. For the most part, only people inside the government and people who are politically active are aware of Councils of government, and since the public has no attachment to these creations, there is little reason to expect the boundaries will be stable. Jc3s5h (talk) 13:52, 31 August 2025 (UTC)[reply]
Option A is the most reasonable of the presented options. The counties no longer play any relevance when it comes to election administration or even statistics. Presenting them is wrong when they are now about as relevant as Roman Catholic dioceses. There's an argument to be made that neither map should be in the infobox, and instead a map of municipalities should be the only one used from 2022 onwards. Reason being that it seems it is at this level — and only at this level — that the Connecticut Secretary of State and thus the media actually presents the results at. Gust Justice (talk) 13:53, 31 August 2025 (UTC)[reply]
Option A is my personal preference. For all intents and purposes, the old counties no longer even exist. Their inclusion in the infobox is therefore trifling. — Mad Mismagius (talk) 14:01, 31 August 2025 (UTC)[reply]
.jpg)
.svg)
.png){kind=link}
