Jump to content

Talk:1925 Tri-State tornado

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
This is an old revision of this page, as edited by SportingFlyer (talk | contribs) at 21:08, 16 September 2025 (Requested move 14 September 2025: Reply). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Did you know nomination

The following is an archived discussion of the DYK nomination of the article below. Please do not modify this page. Subsequent comments should be made on the appropriate discussion page (such as this nomination's talk page, the article's talk page or Wikipedia talk:Did you know), unless there is consensus to re-open the discussion at this page. No further edits should be made to this page.

The result was: promoted by SL93 talk 00:02, 12 January 2025 (UTC)[reply]

Track of the 1925 Tri-State tornado
Track of the 1925 Tri-State tornado
  • ... that the 1925 Tri-State tornado (track pictured) had a path length of 219 miles (352 km) across three states?
Converted from a redirect by EF5 (talk). Number of QPQs required: 1. Nominator has 14 past nominations.

EF5 15:27, 17 December 2024 (UTC).[reply]

  • Per WP:DYKSPLIT, splits from non-new articles are not considered "recently created"; instead, they are treated as expansions from the copied material. Since the third revision seems to be the point where you switched from copying to writing new prose, I'll calculate from there. That version had 23601 characters of prose, and the current version has 26147 – nowhere near close to a fivefold expansion. However, I see that you have nominated the article for GA; if it passes, that would make it eligible for DYK. jlwoodwa (talk) 22:04, 17 December 2024 (UTC)[reply]
@Jlwoodwa: GA passed. Do you still mind reviewing? EF5 16:46, 27 December 2024 (UTC)[reply]
General: Article is new enough and long enough

Policy compliance:

Hook eligibility:

  • Cited: No - ALT3 is not present in the article. I don't think the phrase also called the Great Tri-State Tornado (from the "Kids Britannica" source) supports ALT4's claim of often refered to.
  • Interesting: Yes
  • Other problems: No - I'm not sure about ALT0's status as a "definite fact" (WP:DYKHOOK), given the extensive doubts described in 1925 Tri-State tornado § Legacy. Similarly, ALT2 is present and sourced in the article, but then contradicted in the same paragraph.
Image: Image is freely licensed, used in the article, and clear at 100px.
QPQ: Done.

Overall: @EF5: You gave two online sources for ALT1 and ALT5, but the article cites different, offline sources for their claims. This isn't a huge deal, since {{DYKtickAGF}} exists, but it would feel a little silly to use that when online sources are right here. Regardless, once the unsourced sentence is resolved I'm ready to approve either hook. jlwoodwa (talk) 06:08, 28 December 2024 (UTC)[reply]

 Done, removed uncited claim. EF5 06:18, 28 December 2024 (UTC)[reply]
@EF5: Sorry for missing this earlier – I just realized that while ALT1 (deadliest US tornado) is stated in the lead (without citation), I can't find it in the body. 1925 Tri-State tornado § Fatalities compares it with other tornadoes, e.g. more students killed … than in any other tornadic event in U.S. history and killed at least 20 farm owners … more than the combined total of the next four deadliest tornadoes in the history of the United States, but doesn't compare the total deaths. jlwoodwa (talk) 22:54, 28 December 2024 (UTC)[reply]
@Jlwoodwa: Fixed. This week's tornado - EF5 19:57, 3 January 2025 (UTC)[reply]

Hooks ALT1 and ALT2 are approved. ALT5 is also approved with an online source given here and an offline source cited in the article, accepted in good faith. jlwoodwa (talk) 20:21, 7 January 2025 (UTC)[reply]

Edit summary verification check

Small note, I can verify this source does indeed have 798 deaths mentioned for this tornado. The number 798 does not appear, but there is a table mentioning all the deaths and their location. Per WP:CALC, that gives 798. Just making a note here of this, since in this edit summary, Departure– stated it failed verification. I have confirmed 798 deaths would be sourced from this citation, thus, not a failed verification. I am not reverting as this is a controversial edit, but I wanted to clear that fact up. It is not a failed verification, just a controversial edit being reverted. The Weather Event Writer (Talk Page) 01:58, 30 May 2025 (UTC)[reply]

The article lists the names of most of the dead bar the huge figures in each major town. A full list of every single name (almost every name) can be found by searching on the findagrave website. In which each name will have their cause of death being attributed to the tornado on their profile and this was partly how individual names were determined. Just want to make it clear the 798 number is not random, it was a very thorough process in which each individual was determined. As the wikipedia page for the deadliest tornadoes in the americas even states, the 695 figure from the red cross is missing many later deaths from injuries. I don't want an edit war either, I just want the information to be all-encompassing and as accurate as possible. For several counties, the respective historical societies provided us with a list of the registered dead (for example in Jackson County where over 300 were listed as tornado victims). Each name and location was provided and mapped out. Dovah12333 (talk) 02:05, 30 May 2025 (UTC)[reply]
I still disagree about it's reliability, several people shown are tornado YouTubers and calling them reliable could very well be a complete stretch. Also, per WP:RSPFINDAGRAVE, Find-a-Grave is WP:GUNREL and thus studies using that can't be verified. — EF5 02:08, 30 May 2025 (UTC)[reply]
Alright then. Let me outline a few thoughts on this.
  • The source is self-published as a WordPress site. As @WeatherWriter points out, this does not mean it's automatically unreliable.
  • The WordPress site as a whole appears to only have this single entry.
  • Of the four authors on "the Tri-Team", the only one with any qualifications to speak of surrounding specifically severe weather is Andrew Hagen (see NHC staff profile). TornadoTRX and June First are entertainment first and foremost and don't appear to inherently be qualified as secondary sources.
    • As for Hagen, he appears to be a subject-matter expert on certain hurricane and tropical cyclone-related topics. This is tornado history. While Hagen also appears to be affiliated with the American Meteorological Society, this "study" doesn't appear to have been done in any sort of official context.
  • The sources cited are in a heap at the end. This is to various libraries and historical societies, as well as to six other seemingly reliable papers. Due to nothing being cited inline, there's no way to know what came from where.
  • WP:REDFLAG seems very relevant to this matter. Extraordinary claims require extraordinary evidence. I think another 103 deaths on top of what's already the deadliest tornado in US history is "extraordinary", and a blog with a single "study" made by people with no qualifications bumping up the figure another 15 percent from literally every other reliable source available isn't going to fly.
  • As @EF5 brings up, Find-a-Grave is unreliable to source this claim. I don't have any proof that this particular study uses it, however, if any particular study did, its reliability would be based on other factors of editorial oversight instead of the cop-out "unreliable".
  • If, by contrast, the source had been published or reviewed or otherwise acknowledged by the relevant authority, for instance the American Meteorological Society, Thomas P. Grazulis, or NWS Paducah (in a way beyond giving vague "the original toll was an undercount" comments), then the veracity of this claim would be less in question.
As is, I'm going to stand by my removal. I am willing to consider any evidence to the contrary for the death toll being higher (didn't Storm Data have the death toll a bit higher too?) but nothing that involves manual sifting through records, tombstones, and self-published blogs and YouTube videos should be anyone's first answer in refuting this. Departure– (talk) 02:37, 30 May 2025 (UTC)[reply]
What about the fact that we acquired lists of the deceased from each historical society from their own records. Whilst the cumulative nature of that number was not 798 it was substantially over 700. And as Doswell, Grazulis and others have already made their statement that the 695 figure is definitely an underestimate and so I am not sure it is an "extraordinary claim especially when it has been agreed upon by experts and known that the death toll is higher. And also considering the original source for 695 is woefully outdated and would fall under that guideline on wikipedia. Being given by the red cross in 1926 and before a lot of later deaths. Dovah12333 (talk) 10:33, 30 May 2025 (UTC)[reply]
Furthermore, no other source has even attempted to sift through the records of deaths until the July 2024 project. Doswell himself in a presentation stated that their team did not focus on the death toll at all. And as for the regional officially listed dead (such as 234 in Murphysboro, 69 in De Soto etc), the cumulative number of all these regions which are already accepted lists a number greater than the aforementioned 695. So based on all of this, the 695 figure should essentially fall under the outdated information category. Dovah12333 (talk) 10:43, 30 May 2025 (UTC)[reply]
I also just want to point out that the article also goes through the death toll at each region as the article progresses. So If you haven't already read the full article in detail it may not be clear what came from what. Dovah12333 (talk) 10:51, 30 May 2025 (UTC)[reply]
I'm just not buying the figure given the source provided. If you have a different source I'd be willing to consider having the claim in the article; as of now this is challenged due to the source seeming unreliable. The outdated figures from reliable sources outweigh the updated figures from unreliable sources in this instance. Departure– (talk) 15:22, 30 May 2025 (UTC)[reply]
If you were involved with this, Dovah, what source was used for finding the toll? If it's solely Find-a-Grave, this discussion's good as over because that website is completely unreliable per WP:RSPS. — EF5 15:25, 30 May 2025 (UTC)[reply]
Yeah, I get the impression Dovah12333 was involved in the creation of this source, and thus this discussion falls under a conflict-of-interest. Pinging @WeatherWriter: cool if we close this? It's apparent to me that this isn't going anywhere as is. Departure– (talk) 15:28, 30 May 2025 (UTC)[reply]
Since this isn’t a formal discussion (like an RFC), it doesn’t have to be closed. If it isn’t going anywhere, then all parties can agree to disagree. The Weather Event Writer (Talk Page) 15:30, 30 May 2025 (UTC)[reply]
Ok, so the process of determining the dead in the project was originally derived from the data collected by Robert H. Johns, Charles Doswell III, Matthew S, Gilmore, Donald Burgess, Steven Piltz,and John Hart. They made several publications in 2013. Their data was given to us by Matthew in the form of a kmz file for google earth which holds an interactive map of the damage and the names of every farm, dwelling and or person who was in the path and recorded. Using this as a basis, we went in and began delving into the files for each respective name, data on the deceased was already partially recorded by Doswell et al, so it was more of filling in the gaps. When it came to finding the dead, we used official records from obituaries, acquired from documents provided by historical societies, individuals, libraries and or newspapers. We then tracked down those who were verified to have been injured by the tornado and inquired as to their cause of death. Often, in cases where it was tornado related, there was written documentation linking it to the sustained injuries. Some historical societies had already begun compiling a list of the deceased in their own regional areas, which we then acquired and used to guide the process further. Through a process that took well over 9 months, we examined thousands of names and researched their cause of deaths and managed to narrow it down. Of note, is that the listed 695 dead by the red cross differs from regional death tolls quite substantially. Further complicating matters, some of those counted, would not be directly linked to the tornado per today's standards, having various degrees of removal (eg. one person died a few days later from crashing their vehicle from the debris of fallen buildings). So for simplicity's sake, we abided by similar standards used in 1925 and included those who had died from secondary causes as well (like fire). Respectfully, before coming to a judgement on the reliability of the article, please take the time to actually read it in full, as it covers the dead as it goes on. If you so wish, I can list every single person included in the 798 figure by name, and provide the lists and documents provided by societies like Jackson County Historical Society. Also of note, many contemporary newspapers did list many more dead (like 828 etc), and the names and locations listed for many victims were often confused and mismatched. This took quite a considerable amount of time on our part to accurately locate the location of each death, which in some instances in large towns like West Frankfort and Murphysboro, proved impossible due to uncertainty as to where they were prior to the tornado. In the case of the latter town, more than 20 were unidentifiable per official records so the names aren't explicitly known, but they are still listed on old records and newspapers. The figure of 798 is not expected to be directly all-encompassing, but is a better representation than the red cross number of 695, which differs from the original figure provided by the weather bureau of 689, so it seems even then there were regional discrepancies. And our findings align with the expectations and statements of figures such as Grazulis and Charles Doswell, who both cited that the death toll was likely far greater than the reported 695, rightly so. I can also provide our online copy of the full interactive damage path. Perhaps it is easier to do so over another form such as discord.
But in general, being reliable was the aim of the article to begin with, as there hasn't been a detailed write up on the full path until the article was published. And when there was, it was focused on specific regions (like Bob Johns and his book on the Franklin, Hamilton and White County areas). I just want to make it clear that the number is not from nowhere, but from months of careful research and documentation from third parties, including individual descendants from some victims, and individuals from local libraries and societies. Angela Mason also aided to an extent with her book and after contacting her directly. Find A grave was not used as a source, rather a tool to compile much of the names into a virtual cemetery. But this was left unfinished due to the fact that it did not have all those names registered anyway. IT WAS NOT USED AS A SOURCE. Dovah12333 (talk) 19:15, 30 May 2025 (UTC)[reply]
The problems with the source have nothing to do with how it may have been assembled but rather due to the fact that on its face it is self-published. By all means, you could pursue some sort of acknowledgement or intent to reassess from the AMS or NWS, but due to the fact that this appears to be from a single-purpose WordPress blog, the claims it makes can't be reliably verified, regardless of how the blog got its data.
If "being reliable was the aim of the article to begin with", then it shouldn't have been on WordPress. Any form of independent verification, such as publication by, again, a subject-area authority, would suffice, but as of now it's just a random website on the internet that a Wikipedia editor that claims to be involved asserts is reliable (or, in layman's terms, "bro, trust me"). Even though its methodology seems semi-professional and would likely be accepted as a source from another body, since WordPress blogs can be made by just about anyone and are not themselves subject to any editorial oversight besides that which the blog's creators themselves subject their work to, they are not a reliable source for claims on Wikipedia.
Even though Wikipedia itself is not a reliable source, the project still has relatively high standards for citing content, and I hate to say but WordPress, in this context, isn't up to that bar. Whether or not the information in the study is factual or not is another topic that isn't within Wikipedia's pervue. Departure– (talk) 19:30, 30 May 2025 (UTC)[reply]
Alright so what about a middle ground. Grazulis is a subject specific expert and so is Charles Doswell, both have stated explicitly that the death toll was in all probability, greater than the established 695. Could we not therefore cite them and include a note that states the toll is potentially greater. That would probably appeal to all parties. Dovah12333 (talk) 19:43, 30 May 2025 (UTC)[reply]
I don't see why we couldn't. Near the death toll a note saying "...this figure is likely an undercount" would be acceptable, sourced to Grazulis or Doswell directly. Departure– (talk) 19:47, 30 May 2025 (UTC)[reply]
Here is Charles Doswell on the death toll in his presentation. At the 45 minute mark roughly. Let me get the Grazulis statement. https://www.youtube.com/watch?v=nEz21uvzXtQ&t=2341s Dovah12333 (talk) 20:40, 30 May 2025 (UTC)[reply]
Also of note is this report from 1925. It lists the deaths as over 740 and is currently on the NWS page. https://www.weather.gov/media/pah/1925_Tornado/March%201925%20Illinois%20Weather%20Bureau%20Climatological%20Data%20Report.pdf Dovah12333 (talk) 21:01, 30 May 2025 (UTC)[reply]
This article here makes a reference to Grazulis and his statement as well. Citing his book Significant Tornadoes, 1993, page 193. https://www.usdeadlyevents.com/1925-march-18-tornadoes-esp-tri-state-tornado-ilinmo-695-alkytn-740-792/ Dovah12333 (talk) 21:20, 30 May 2025 (UTC)[reply]
So are we ok to change the article? In which case I will leave it to you. I appreciate our discussion though. Dovah12333 (talk) 00:18, 1 June 2025 (UTC)[reply]
I changed the article. Please check the source as I want to make sure you're happy with everything. Dovah12333 (talk) 14:57, 9 June 2025 (UTC)[reply]

Requested move 14 September 2025

1925 tri-state tornadoTri-State TornadoTri-State Tornado – As per the last discussion, the tornado is widely referred to as simply the "Tri-State Tornado." WP:COMMONNAME CutlassCiera 17:26, 14 September 2025 (UTC)[reply]

I really don't like that we waited until others involved in previous discussions were sanctioned to do this (it's really unfair and a little trashy) but strong support per my other rationales above. EF5 17:32, 14 September 2025 (UTC)[reply]
Here's the first ten pages of sources on the event via a Google search by looking up "tri-state tornado", in complete lowercase. Entries are ordered by where they showed up: Tri-State Tornado, The Tri-State Tornado, Tri-State Tornado of 1925, Tri-State Tornado, Tri-State Tornado, "Tri-State Tornado", Tri-State Tornado, Tri-state Tornado, The Great Tri-state Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado of 1925, Tri-State Tornado, Tri-State Tornado, 1925 Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, The Great 1925 Tri-State Tornado, Tri-State Tornado, Tri-State Tornado of 1925, Tri-State Tornado, Tri-State Tornado, Great Tri-State Tornado, 1925 Tri-State Tornado, The Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Deadly Tri-State Tornado, Tri-State Tornado, 1925 Tri-State Tornado, Tri-State Tornado of 1925, Tri-State storm, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-state Tornado, Tri-State Tornado, 1925 Tri-State Tornado, the tri-state Tornado, Tri-State Tornado, Great Tri-State Tornado, 1925 Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, 1925 Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, Tri-State Tornado, 1925 Tri-State Tornado, Tri-State Tornado outbreak, The Tri-State Tornado, Tri-State Tornado (one here is excluded as it is about the 2021 event but is still uppercased), Tri-State Tornado, he Tri-State Tornado, and a LinkedIn and YouTube video which are uppercased, Tri-State Tornado, (the next source is excluded as it is unrelated but still uppercases "Tri-State"), Tri-State Tornado, (the next source is excluded as it's unrelated), (the next source is excluded as it's unrelated and doesn't contain "Tri-State/other variants"), (the next source is excluded as it's unrelated and doesn't contain "Tri-State/other variants"), (the next source is excluded as it's unrelated and doesn't contain "Tri-State/other variants"), (the next source is excluded as it's unrelated and doesn't contain "Tri-State/other variants"), and another unrelated source. It is laughably obvious and exceedingly clear which casing news outlets use. Almost every single source is uppercased to some degree, and people saying that it's not usually uppercased are out of their mind. EF5 13:18, 15 September 2025 (UTC)[reply]
The first source I clicked at random doesn't use the phrase tri-state at all. Another one you claim is capitalised, but only the headline is: [1] This one is also lowercase in spite of the list: [2] Also, I'm not sure how many of these are actually reliable sources. There's at least some confusion about headlines, where every word is capitalised, versus what's actually in the text. SportingFlyer T·C 20:54, 16 September 2025 (UTC)[reply]
TWC and Safe Rooms both capitalize "Tri-State", although I admit I didn't check whether they were headlines. I just used the first ten pages of non-YouTube sources I could find. EF5 21:01, 16 September 2025 (UTC)[reply]
I'm fine with capitalising Tri-State. I don't think it's technically grammatically correct, but almost everyone seems to treat it like a proper noun. SportingFlyer T·C 21:08, 16 September 2025 (UTC)[reply]

Support — Almost always called Tri-State Tornado, in all caps, including the official name of this tornado from the National Weather Service seen here. The Weather Event Writer (Talk Page) 19:10, 14 September 2025 (UTC)[reply]

  • Oppose and procedural objection. This article has had three previous RMs within the last year, and two move reviews, and the new proposal is identical to the last one, which was rejected. Prevailing by nagging is not how Wikipedia is supposed to work. The community already had a recent thorough discussion of this. The proposed title is contrary to the MOS:HYPHENCAPS guideline. "State" is not a proper noun. —⁠ ⁠BarrelProof (talk) 20:01, 14 September 2025 (UTC)[reply]
What is there to procedurally object BarrelProof? WP:IAR exists for a reason: "If a rule prevents you from improving or maintaining Wikipedia, ignore it. An overwhelming number of sources capitalize the full name, including the word "State". National Weather Service (Official name) (Twice), National Severe Storms Laboratory, Evansville Museum of Arts, History & Science, Illinois State Government, Associated Press, Missouri Life, USA Today, and so many secondary reliable sources with an overwhelming consensus. Even international sources such as the BBC and spanish sources like Diario AS ([3]) capitalize all parts, as that is the official name. Interestingly, The Weather Channel actually calls it "Tri-State tornado". Either way, there is a very clear case to be made for WP:IAR to circumvent MOS:HYPHENCAPS, given the overwhelming consensus, especially on "Tri-State" being capitalized. The Weather Event Writer (Talk Page) 20:24, 14 September 2025 (UTC)[reply]
The objection is to the repeated opening of RMs on the same issue after thorough discussion, repeating the same arguments, which aggravates friction and wastes time and energy. —⁠ ⁠BarrelProof (talk) 00:00, 15 September 2025 (UTC)[reply]
The "thorough" discussion was hampered by bludgeoning which eventually made its way to ArbCom and led to several people being sanctioned. EF5 13:23, 15 September 2025 (UTC)[reply]
I would like to request you provide multiple examples of sources using the lowercase version of “tri-state tornado”, given the majority of sources, based on my search, as well as the ton of sources listed above, have it capitalized. Until I see a lot of evidence, based on my own searches, I have to consider your statement false, as I actually can’t find sources for it being lowercase myself. The Weather Event Writer (Talk Page) 12:39, 15 September 2025 (UTC)[reply]
N-grams show lowercase 'Tri-state' when referring to a tornado' does not exist. Randy Kryn (talk) 14:51, 15 September 2025 (UTC)[reply]
N-grams is problematic, then. [4] [5] [6] [7] [8] [9] - all the academic research at least uses lowercase. The book "The Great Tri-State Tornado" directly on the tornado also does not capitalise it, as does [10]. I'm more concerned about dropping the year. The first reference almost always includes the year. I absolutely don't think the phrase tri-state tornado is common enough to ignore the inclusion of the year, and at least the academic journals show it's not a proper noun. SportingFlyer T·C 20:49, 16 September 2025 (UTC)[reply]
Ref 2 you provided uses "The Tri-State Tornado", ref 3 uses "1925 Tri-State Tornado", ref 4 uses "The Tri-State Tornado", ref 5 you are correct about, ref 6 capitalizes "Tri-State" and ref 7 also capitalizes "Tri-State". EF5 20:57, 16 September 2025 (UTC)[reply]
After looking into this, I'd be fine with 1925 Tri-State tornado. The year is absolutely part of it as the year is always introduced before or concurrently with the tornado. There seems to be confusion as to whether it's a proper noun as publications that have clear editing standards (academia) don't use it as such, and because there are other tri-state tornadoes, even though they show up far less often. Also be careful. You are looking at titles, which are always capitalised. 2 uses "Tri-State tornado" within the text, as does 3. You are correct the headlines are capitalised, but so is every other word that is longer than three letters. SportingFlyer T·C 20:59, 16 September 2025 (UTC)[reply]
Note: WikiProject Severe weather, WikiProject Indiana, WikiProject Disaster management, WikiProject Missouri, WikiProject Illinois, WikiProject United States, WikiProject Death, and WikiProject Weather have been notified of this discussion. Aviationwikiflight (talk) 12:47, 16 September 2025 (UTC)[reply]
  • information Note: The {{Old moves}} template on this talk page previously listed only one of the three prior RMs, all of which have taken place in the last 9 months. I have updated the talk page banner and am highlighting these here for visibility during this discussion. There were two move reviews for the December 2024 RM, as already mentioned. --MYCETEAE 🍄‍🟫—talk 20:36, 16 September 2025 (UTC)[reply]