Talk:The Keys to the White House/Archive 3

Archive 1

Archive 2

Archive 3

Archive 4

Foreign Policy Keys

Hello, in his latest podcast, Litchman didn’t give an official take about those keys and will likely be determined after election day. Litchman with his son gave an official 8 true keys and 3 false keys official prediction, acknowledging his worst case could be 8 true keys and 5 false keys (given Ukraine is losing ground again in the war and ceasefire talks are dead in the waters) . So my suggestion is to live the keys blanks and let Allan Litchman give his official take on his next book. Jimrot45556 (talk) 14:03, 8 September 2024 (UTC)

This is already reflected by the "likely" qualifier for those keys. Lichtman has said that this is the way those keys currently lean multiple times. It would not make any sense for the keys to be determined after election day, since they are supposed to be predictors for the result of the election. TWM03 (talk) 21:19, 8 September 2024 (UTC)

He left a key blank in 2016. He will do it too this time, since Kamala has already 8 so those keys don’t matter that’s his thinking.

He could leave them blank to hedge should the 2024 winner outcome become unpredictable. Jimrot45556 (talk) 10:27, 9 September 2024 (UTC)

Jimrot45556 earlier today engaged in vandalism in this article by changing the status of Key 11 to "Likely False" based on his own judgement rather than what Allan Lichtman has predicted. Keep a close eye on this guy. 192.12.184.6 (talk) 00:55, 13 September 2024 (UTC)

I didn’t engage in vandalism, Litchman literally said those keys are shaky and could flip in his latest podcast. You just didn’t watch it till the end. Plus the prediction is already in so it don’t much difference. Jimrot45556 (talk) 21:49, 24 September 2024 (UTC)

And where did he say that the key was likely false? You changed the key based on your views/judgement in spite of what Lichtman has quite clearly has said about it and ofcourse the key could flip it would otherwise be fully defined. Without an overt indication that Lichtman has shifted his judgment on the key, your edit is going to be vandalism. 2A00:23C5:11E:F901:3883:EF80:6075:256D (talk) 19:31, 28 September 2024 (UTC)

Yeah, I might have made an error in the code, I wanted to put them both blank since Litchman gave a solid 8-3 prediction and a worst case 8-5 prediction. It was the last final call with his son in that podcast. He also said the fortunes of war could turn both false.

I get a feeling he will decide on them after the election, the recent fall of Vuhledar in Ukraine and Israeli troops going into Lebanon makes them both likely false (my opinion).

But until Litchman makes a final call on them, better put them « Undertermined » cause I know he will flip them, he usually does (refer to his official prediction in 2016 and 2008) Jimrot45556 (talk) 04:44, 6 October 2024 (UTC)

Again editing the article on the keys based on your opinion rather than Lichtman's is vandalism. No doubt about it.

And it wasn't an error in the code. You did this intentionally, and frankly should be banned from wikipedia. 192.12.184.7 (talk) 19:04, 9 October 2024 (UTC)

Why are you unhinged like this ? Take your pills geez Jimrot45556 (talk) 21:56, 13 October 2024 (UTC)

I took a look at your edits. It's quite clear that you changed the page from "Likely 4" to "Likely 5" keys down. You also changed key 11 to "Likely False" and changed the explanation for key 11 to

"Cautious call Russian invasion of Ukraine allowed Ukraine to regain territory taken by Russian separatist forces and incursion into Kursk Oblast has stagnated and Russians are making gains again in Dombass and took Nyu York. Ceasefire talks between Israel and Hamas are not going well for now."

This doesn't look to me (and multiple other people here) that it was an honest mistake. You made three changes to make it seem that they keys were 8-5 instead of 9-4 or 8-3-2. It's pretty clear that you changed these keys based on your opinion and there no good reason for that. It's clear cut vandalism.

http://en.wiki.x.io/w/index.php?title=The_Keys_to_the_White_House&diff=1245393883&oldid=1245300511

Also personally attacking another poster is really uncalled for. Take your pills? WTF. If this behavior continues, it will be reported. J12nom (talk) 00:04, 16 October 2024 (UTC)

It is quite clear that you lost hack ! Jimrot45556 (talk) 07:56, 6 November 2024 (UTC)

Restoring neutrality

We've gone through dispute resolution processes. In the RfC and the thread on the BLP Noticeboard, every experienced editor (Classicfilms and notwally) has agreed with me that the more neutral version is better. In addition, LittleJerry twice edited, noting that he was removing obvious POV, though his changes were quickly reverted.

There is no requirement that dispute resolution be continued for 30 days, during which a violation of Wikipedia policies must remain. It's obvious that there'll be no further input on either thread. On that basis, I'm restoring the more neutral version.

Two observations going forward: (1) On the dispute about 2016, I've tried to present both sides, but we can certainly consider how to improve that discussion. The only non-negotiable point is that neither side will be endorsed in Wikipedia's voice. (2) If, in the next month, there's an unexpected influx of editors to either of the discussion threads, this decision can of course be revisited. That remote possibility is not a basis for inaction now. JamesMLane t c 18:07, 4 November 2024 (UTC)

RfCs encourage anyone to engage in the discussion. There is no distinction between experienced an unexperienced editors therein.

The noticeboard was an attempt to circumvent the ongoing RfC, I do not recognize it as legitimate. Despite that, you had a single additional editor weigh in.

The Rfc found the opposition of 4 for these changes. As you noted elsewhere, RfCs are not a majority rule sort of deal. But they most definitely are not a minority wins resolution process either. Overall you did not find consensus for the edit you just made unilaterally. We disagree that your version is the more neutral one. I am reverting your recent edit on these grounds.

I will begin a talk discussion on what actions we can find compromise on tomorrow when I have time (I outlined some initial thoughts when you first proposed the new version and I have some other ideas I've been mulling over in addition to that) as a way forward. If/when we get to what we cannot compromise on in that discussion, then I guess we will have to progress in dispute resolution (for example, to contact an admin). Do not revert these changes back, this is the status quo of the article based on consensus found in the summer. Apprentice57 (talk) 18:41, 4 November 2024 (UTC)

@JamesMLane The article is the farthest thing from neutral. We should absolutely include the information on 2000 and 2016 Kalbome22 (talk) 19:45, 6 November 2024 (UTC)

Emphatic Prediction of Kamala Harris win in 2024

The article should be updated to describe just how emphatic Litchman was about Kamala Harris winning in 2024. Previously Lichtman got away with the claim of a good predictive streak with post-hoc modifications to his intentions in some instances. This election will finally put him on record that his analysis is not much better than looking at poll numbers in the summer as a rough educated guess and then rolling a dice if it's close. Fpbear (talk) 04:56, 6 November 2024 (UTC)

Bit premature there EvergreenFir (talk) 05:55, 6 November 2024 (UTC)

I'm not sure exactly when is the right time to update the article (probably when election raters all agree on a victor?) but when it does we will mention it in a neutral-POV way. If it leads to a lot of outside criticism we can include those with a citation as well in the respective section. Apprentice57 (talk) 06:08, 6 November 2024 (UTC)

We should wait for the dust to settle a little. Tomcleontis (talk) 23:06, 6 November 2024 (UTC)

It is clear this is a major sweep for Trump and the Republicans, with both the popular and electoral majority. In the article this should be contrast with how emphatic Litchman was. This wasn't just Litchman saying it's going to be a toss up either way; he appeared on numerous TV interviews 100% certain of a Kamala Harris win. Fpbear (talk) 04:18, 7 November 2024 (UTC)

Priorities for Updates after 2024 Election

I think everybody can agree that (1) the article requires updating to incorporate the 2024 election results, and (2) finding better consensus around NPOV of the article.

Firstly, I have made two edits that I hope will be acceptable:

1. I have reverted two changes made by @Jeffs1225 which unambiguously violate NPOV rules. Jeffs altered the article to lead by labelling the system 'discredited' multiple times in the first two paragraphs (particularly citing the 2024 result) without a source.
2. I have added a very brief summary of Lichtman's 2024 prediction to the "Lichtman's prediction record" bullet point list. I do not explicitly state whether the prediction was correct or not, and quote directly from Lichtman using a source that was already in the article. The last sentence claims (for now) that Harris lost the EC & popular vote, but obviously (a) this needs a source, and (b) could still change as votes are counted.

I want to emphasise above that these are edits that I hope will be acceptable to both sides of the POV debate that has been occurring on this Talk page and elsewhere.

I would like to open this topic to form some consensus around priorities for editing the page going forward.

In particular — can we find an agreeable transcript of Lichtman's 2024 prediction? My understanding is that the first 2024 prediction he made was in a video with NYT (which is paywalled and hard for readers to verify), and subsequent affirmations were often also in video format OR were media outlets reporting on what Lichtman had said elsewhere.

It would be great if anyone could find a good source where Lichtman has personally confirmed his 2024 prediction in text, especially if it clarifies the EC vs popular vote prediction matter definitively for 2024. I think it's important to use as much of Lichtman's wording directly as possible so the reader gets clarity.

Hangways1 (talk) 09:34, 6 November 2024 (UTC)

He retroactively fit the keys to make it seem it predicts the winner since 1860 to 1984.

i don’t know what to tell you, when he lies a first time, his credibility is severely destroyed.

since 1984, he is wrong 3 times now (2000, 2016 and 2024) Jimrot45556 (talk) 11:05, 6 November 2024 (UTC)

Jimrot, our job as editors is not for us to insert our opinions about whether someone is credible or not. We can (and should) present the facts of what Lichtman has predicted and the outcomes. We can also provide the facts of what relevant media outlets etc. have stated about Lichtman's credibility (or lack of it). But the article itself should be written with neutral tone and perspective to comply with NPOV requirements.

My interest is primarily in helping reshape this article to lean more on Lichtman's own words and direct statements to help readers assess for themselves if he has been correct or not. If you are certain that Lichtman's own words prove definitively that he has been incorrect, or show obvious inconsistencies, then I don't see why you'd have any issue with this effort. Hangways1 (talk) 13:32, 6 November 2024 (UTC)

I follow another guys model on twitter who predicted correctly with his model a Trump victory in 2024 and 2016. Also his model points to Biden 2020.

I’m thinking he deserve a Wikipedia page, he claims his model points the correct winner since 1916 including Bush in 2000 which Litchman got wrong.

4 points checklist to the White House

1- Incumbent president running

2- Incumbent president has average approval above 50%

3- No recession in last 2 years (here 2024 and 2023)

4- GDP growth in election year higher then first year of incumbent president term.

2 false points❌ = Incumbent party loses

3 correct points ✅= Incumbent party wins

his call for 2024 :

1 - Harris is not sitting president ❌

2- Biden approval average at 41% < 50% ❌

3- No recession in 2023 or 2024 ✅

4- GDP 2024 slower than Biden first year in office 2021 ❌

❌ = 3

incumbent democrats were predicted to lose

2020 election :

1- Trump is sitting President ✅

2- Trump was below 50% approval ❌

3- COVID Recession in 2020 ❌

4- GDP of 2020 slower than Trump first year 2017 ❌

❌= 3

Trump was predicted to lose 2020

2000 election

1- Gore was not sitting president ❌

2- Clinton approval above 50% ✅

3- No recession in 2000 or 1999 ✅

4- GDP growth in 2000 slower than Clinton first year in his second term 1997 ❌

❌ = 2

Gore predicted to lose Jimrot45556 (talk) 13:44, 6 November 2024 (UTC)

Okay. This article is about Lichtman's model. If you'd like to write an article about someone else's model, go for it. Remember that the article will need to comply with Notability requirements. Hangways1 (talk) 14:01, 6 November 2024 (UTC)

Alright.

in regards to Litchman, I think he misjudged the contest key.

Humphrey won 2/3 on first ballots in 1968 but he still marked it as false cause of I quote “vocal divisions with the democratic convention”.

he marked it true for 2024 however… see where I’m getting at ? Jimrot45556 (talk) 14:03, 6 November 2024 (UTC)

About my edits on 7 November 04:11 — These edits are both structural and substantive in nature. My hope is for at least the structural (readability) edits to be preserved regardless of how contested this article is right now.

------

The first structural edit is a slight re-organisation of the introduction. Before my edit, there were two separate paragraphs that began with a comment on the system's methodology, and then moved to discussing Lichtman's track record. I have shifted sentences here so that instead we have a clear "methodology" paragraph, followed by a clear "track record" paragraph.

The second structural edit is a slight re-organisation of the Lichtman's prediction record (1984–present) section. Previously, information about years of disputed accuracy was presented in a bullet point list, which was getting clunky due to the volume of information provided. I have split this out into a lede and two subheadings for 2000 and 2016. I have chosen not to include a subheading for 2024, simply because there is no disagreement about the keys' accuracy in 2024. A source is provided to indicate this.

------

The first "type" of substantive edit, meanwhile, is an effort to passivelyrestore NPOV and consistency in the article.

For example, the introduction made contradictory statements about Lichtman's accuracy — it claimed that Lichtman "failed to predict" the election of George W. Bush in 2000 (which implies that the keys made an EC prediction that year), but then claims that the keys "Lichtman has correctly predicted the popular vote outcome of each presidential election from 1984 to 2012". This is just messy and should be fixed regardless of your stance on whether those predictions were accurate/inaccurate. I have erred on a VERY generic summary in the introduction, leaving the dispute to the later section about Lichtman's track record. I am fully aware that this intro is going to be subject to disruptive edits for a while to come and we'll probably need to discuss wording.

The second "type" of substantive edit is adding greater emphasis on direct quotes from Lichtman about his predictions in 2000 & 2016. This is also an attempt to improve the article's NPOV compliance — my hope is that by rooting the article in what Lichtman actually stated himself, we can avoid some of the dispute about which external source to trust about what Lichtman stated. There are plenty of clear and temporally proximal (to the election) quotes from Lichtman that we can lean on, and I've included some.

The added quotes include both quotes from Lichtman that state the keys predict the popular vote, and quotes from Lichtman that people will no doubt read to imply that Gore & Trump would be elected, not just win the popular vote. Hopefully that 'feels' like balance, but more importantly, because they are all direct quotes from Lichtman himself, surely this is all unambiguously NPOV... right?

Thanks for any feedback you guys can provide.

------ Hangways1 (talk) 04:30, 7 November 2024 (UTC)

Just delete this

The whole topic is just a bunch of lies that adjusted parameters after he had the facts. He's never been able to predict what he claims the keys can predict, before it happens. 185.5.48.148 (talk) 07:56, 6 November 2024 (UTC)

Exactly he retroactively applied the keys to fit the keys Jimrot45556 (talk) 07:58, 6 November 2024 (UTC)

• to fit the predicted winner

Jimrot45556 (talk) 07:58, 6 November 2024 (UTC)

Just because he was (presumably) wrong one time doesn’t mean that this article should be deleted. It’s very very useful for a lot of people. Even if the success rate isn’t 100 percent. Hurricane Clyde 🌀^{my talk page!} 08:42, 6 November 2024 (UTC)

It’s a wrong model.

but I think he misjudged the contest key. Compare the contest key rating of 2024 to 1968 Jimrot45556 (talk) 09:10, 6 November 2024 (UTC)

I agree that the model is deeply flawed highly subjective pseudoscience, for many of the same reasons that you discuss. But Wikipedia has lots of articles about stuff that's wrong, for example flat earth theory. Despite being a bogus theory, his keys have (like it or not) received extensive media coverage, so they're notable. And the article can be a good place to assemble and cite critiques of all of the flaws with his model. -2003:CA:8703:2BFC:9A80:4152:79B:982A (talk) 00:18, 7 November 2024 (UTC)

You have a good point in that this is an extensive article about a method that doesn't give any better results than making an educated guess based on polls. I agree this article and attention to the keys should fade away or be deleted as a pseudoscience. Fpbear (talk) 04:21, 7 November 2024 (UTC)

@Fpbear A theory does not need to be accurate in order to merit an article on Wikipedia. For example, flat earth & moon-landing hoax theories have articles — people want to know about them, and that's the purpose of an encyclopedia! Rather, an article needs to satisfy Notability requirements, and the 13 Keys certainly do so; there has been significant coverage of the Keys by major media outlets over an extended period of time and Lichtman himself is a public figure who regularly appears in interviews.

If people stop being interested in the Keys, the article will "fade away" naturally over time. Notability is not temporary, so the article would remain up, but certainly nobody here should be editing the article to try and force its relevance into the future — don't worry. Hangways1 (talk) 07:29, 7 November 2024 (UTC)

I agree with what @Hangways1 said, the 13 Keys remain notable (there may be an argument down the line for consolidating this and the Allan Lichtman article if interest fades or the level of detail becomes superfluous); it does make it important that an accurate and effective presentation of the system's flaws, critics, and relative strengths is reported here. Caraturane (talk) 23:27, 7 November 2024 (UTC)

Semi-protected edit request on 8 November 2024

Aleksey20022 (talk) 02:23, 8 November 2024 (UTC)

Alan Lichtman has said the primary contest key is False in 9 pm 11/7/2024 litchman live youtube stream for the democratic president in 2024 because the democrats trashed their incumbent president. Also in his final NYT prediction he has not predicted the foreign policy sucess key and in his litchman live key said he is not mike johnson and thinks its leaned true but cant call it.

 Not done: it's not clear what changes you want to be made. Please mention the specific changes in a "change X to Y" format and provide a reliable source if appropriate. Shadow311 (talk) 03:17, 8 November 2024 (UTC)

Really ? I argued about this, you all doubled down on me about the foreign policy success key.

save it bro the gig is up Jimrot45556 (talk) 11:09, 8 November 2024 (UTC)

Why is this on wikipedia?

This is simply pseudoscience and a unimportant model which has failed catastrophically. Jjbomb (talk) 20:17, 6 November 2024 (UTC)

As I noted in my other reply, Wikipedia has articles for lots of other pseudoscience BS, including flat earth. The criteria for inclusion is notability, not whether it's right or wrong. His model/theory has been covered extensively by the media, so it's considered notable, despite its deep flaws. People coming here can read in detail about its flaws and criticisms that have been made. -2003:CA:8703:2BFC:9A80:4152:79B:982A (talk) 00:25, 7 November 2024 (UTC)

I understand I was just puzzled that the article I made on the Kentucky highway shooting would be deemed for deletion but this random theory that was now destroyed stays. Jjbomb (talk) 01:14, 7 November 2024 (UTC)

Did you follow the Notability requirements Jimrot45556 (talk) 22:50, 7 November 2024 (UTC)

It looks like the file you uploaded [1] was because "concern was: Copyvio. No source given, no indication the photo is in PD or CC-BY. No Fair use rationale given." not because of the topic itself. The actual article itself, Interstate 75 Kentucky shooting is still live and not deleted. As far as this article, it meets the criteria for inclusion, and I'm not sure what you mean by "catastrophically", in the sense that this theory could only either succeed or fail, there is hardly and need for hyperbole. TiggerJay (talk) 06:04, 8 November 2024 (UTC)

Yeah yeah I meant that when I made that article a serious discussion to delete it was formed Jjbomb (talk) 18:43, 8 November 2024 (UTC)

The 1892 Election

In the 1892 election there is 7 True keys and 6 False keys. Then it says it would predict Glover Cleveland as the winner, but shouldn't it be Benjamin Harrison as the predicted winner? Therefore an annomality :) 93.161.47.87 (talk) 14:44, 5 November 2024 (UTC)

Good eye, but not quite! From Lichtman's book: "The keys to the White House are stated as conditions that favor reelection of the incumbent party. When five or fewer statements are false, the incumbent party wins. When six or more are false, the incumbent party loses. " Since there were six false keys, the incumbent was predicted to lose.

You probably assumed that if a majority of keys favour the incumbent, that's sufficient to predict they win. But that's not the way the Keys are built. Lichtman's system builds on the principle that elections are primarily a referendum on the incumbent party — they need to do well ENOUGH to justify re-election, or voters will almost inevitably want a change regardless of the challenger. (Hence why there are only two keys about challengers!). His threshold for this "well enough" idea is that the incumbent needs 8 keys, not just 7. Hangways1 (talk) 11:15, 7 November 2024 (UTC)

Ah okay. Thanks for clarifying. :) 93.161.47.87 (talk) 21:20, 8 November 2024 (UTC)

Pseudoscience

Now that the system has demostratably failed, it seems correct to label the system as subjective and pseduoscientfic. Kalbome22 (talk) 17:48, 6 November 2024 (UTC)

I think the way Lichtman has marketed the keys does really hinge its credibility on calling all (or maybe with one miss) elections correctly so I understand the frustration.

But I don't think we should jump to presenting it that way on wikipedia. Models are simplification of real life, and can't reflect it perfectly, and will therefore have misses. I don't think we should portray it as pseudoscience upfront. Arguably would violate wikipedia's requirement on a neutral POV. Apprentice57 (talk) 19:48, 6 November 2024 (UTC)

I agree with @Apprentice57 here that the article itself should remain neutral. There's also more than enough criticism out there of the model (and I'm sure there'll be more to come) that as editors there's absolutely no need for us to take any stance ourselves — we can simply present Lichtman's predictions and any notable support & criticism for them and let the reader decide for themselves. Hangways1 (talk) 02:09, 7 November 2024 (UTC)

I agree, this article should be labeled as a failed pseudoscience. Fpbear (talk) 04:23, 7 November 2024 (UTC)

If we have reliable sources that called it a pseudoscience, it is reasonable to include. I deleted the earlier addition simply because it was uncited in any of the sources I had seen. Jjazz76 (talk) 04:38, 7 November 2024 (UTC)

https://www.thenationalnews.com/news/us/2024/10/21/allan-lichtman-prediction-kamala-harris-donald-trump-reaction/ Kalbome22 (talk) 15:18, 8 November 2024 (UTC)

Ok that's one. I'm not familiar with that particular source and it isn't on Wikipedia's list where they discuss reliable/unreliable sources. Any other articles specifically calling it "pseudo-science"? Jjazz76 (talk) 00:06, 10 November 2024 (UTC)

If the system were pseudoscientific, then it would have been pseudoscientific regardless of whether it correctly predicted the outcome of this election. The fact that it was wrong this time is not a justification to label it as pseudoscience without reliable sources agreeing that this label is appropriate. I think WP:FRINGE/QS is the relevant policy here. TWM03 (talk) 16:12, 7 November 2024 (UTC)

Agree with Apprentice57 and Hangways1. The neutral position is already pretty critical of Lichtman now and speaks as to the successes and failures on its own, there's no need to heap on unnecessarily. Tomcleontis (talk) 16:25, 8 November 2024 (UTC)

I'd agree without a bunch of good reliable sources calling it pseudo-science we don't have enough to make that addition. Jjazz76 (talk) 00:06, 10 November 2024 (UTC)

Earthquakes

The reader deserves to know that Vladimir Keilis-Borok, the co-creator of the Keys, had worked on earthquake prediction and used some of those ideas in developing the Keys model. The merits of the earthquake system itself, however, gets too far afield. The article can do without the (nonpolitical!) pronouncements of the United States Geological Survey.

If, however, it were to be included, then like anything else it's subject to WP:NPOV. The linked USGS page was not unreservedly negative about Keilis-Borok's work. I've changed the reference to a verbatim quotation from the USGS.

My preference, though, is to delete the whole thing. JamesMLane t c 02:54, 8 November 2024 (UTC)

But the entire method of predicting earthquakes simply didn't work. Jjazz76 (talk) 23:58, 9 November 2024 (UTC)

1. Per the USGS site itself, there's no "simply" about it. You want to cherry-pick the USGS's negative comment. If you insist on addressing this subject here, what's wrong with a verbatim quotation of the full statement, including "a legitimate approach to earthquake prediction research"?

2. I still don't see why we need to address it here, though. There's a wikilink to Keilis-Borok's bio article. That article includes the "unproven" statement. Any reader who cares to know more about earthquake prediction can find it with one click. The validity or lack of validity of the Keys is not affected by earthquake records. By way of comparison, check out Benzene#Ring structure. The pioneer of a major scientific advance "said that he had discovered the ring shape of the benzene molecule after having a reverie or day-dream of a snake biting its own tail (a symbol in ancient cultures known as the ouroboros)." The theory of the benzene ring doesn't collapse just because it was arrived in part through nonscientific methods.

3. Merely reporting that Keilis-Borok used methods that he had developed for other purposes doesn't imply that his earthquake model was accurate. JamesMLane t c 03:03, 11 November 2024 (UTC)

Which track record summary do we prefer?

@Kalbome22 and I have recently disagreed (via conflicting edits) on the best way to introduce the article. Additionally, @JamesMLane has his own preferred version which has been discussed on various talk pages. I'm collating three different versions here to solicit preferences.

---

Hangways' Preference: Lichtman has successfully predicted a large majority of the eleven presidential elections held since 1984 by using the keys. However, he incorrectly predicted that Kamala Harris would win the 2024 election, and the nature and accuracy of his predictions for Al Gore in 2000 and Donald Trump in 2016 have been disputed.

Kalbome's Preference: Lichtman has made predictions for all of the last 11 eleven presidential elections held since 1984 by using the keys. However, he incorrectly predicted that Kamala Harris would win the 2024 election, and the nature and accuracy of his predictions for Al Gore in 2000 and Donald Trump in 2016 have been disputed.

JamesMLane's Preference: Lichtman is credited with a high degree of accuracy in predicting the outcomes of the elections from 1984 through 2020 using the system. Assessments state variously that he got them all right, or that he was correct in all but 2000, or that he was correct in all but 2016. It is undisputed that his prediction of a win for Kamala Harris in 2024 was incorrect.

---

Note: "majority" is equally acceptable to me in place of "large majority", as would be the use of some other descriptor than "large". Indeed, my version initially read "majority" until I edited in "large" to try to reach compromise with James.

My primary objection to Kalbome's version is that it omits any direct mention that Lichtman has made any successful predictions:

• To my knowledge, every single reliable source in the article (including those directly after the sentence) agrees that a majority of Lichtman's predictions from 1984-2024 have been correct. There are really only 3 disputed years; other sources dispute the impressiveness of his predictions, but I don't think any source disputes that 8+ of 11 predictions were correct. At the very least, "majority" seems uncontroversial.
• The paragraph explicitly mentions years that Lichtman definitively got incorrect or which are controversial. I believe it is best to then also explicitly mention successes (somehow) to strike a neutral tone overall.

My primary objection to JamesMLane's preference is that it introduces the article with a higher emphasis on coverage of his track record until 2020 than of his track record overall (which now includes the 2024 election).

• The choice to summarise perceptions of a slightly truncated version of his record seems pretty arbitrary. Lichtman did not change the keys between 2020 & 2024, nor was 2024 the first prediction he is alleged to have gotten incorrect, nor did Lichtman have an undisputed streak of accuracy for a long period of time before that point. I can't see a good reason to particularly emphasise perceptions of Lichtman's track record as of four years ago in the prose.
• In comparison, summarising his track record from 1984-2024 inclusive is intuitive (we simply summarise his entire track record), agreeable (every reliable source now agrees he has gotten the majority but not all of his predictions correct), and easier to read (since we don't need to step readers through three conditional alternatives).

Of course, "high degree of accuracy" should probably also be discussed in comparison to my own "successfully predicted the majority" and Kalbome's "made predictions for". Which of these phrasings is most NPOV? Hangways1 (talk) 17:25, 8 November 2024 (UTC)

Selecting out 1984-2020 isn't arbitrary. As a practical matter, we have many reliable sources that assess the system's performance over those years. My subjective impression is that the single most common view is "9 out of 10, with Gore wrong" and there are several in that camp. Presumably those commentators will at some point be saying "9 out of 11, with Gore and Harris wrong," but the election is too new and they haven't written those stories yet.

There's also a substantive reason. 2024 is the only election in which all sources, including Lichtman himself, agree that the prediction was wrong. NPOV doesn't require that we treat each year identically. 2024 is more important than any of the other years, so it's not POV for that failure to be highlighted in a sentence of its own.

The problem with "majority" is that it obscures important information. It's like saying that a majority of climate scientists believe in anthropogenic global warming. That statement is true. It would be very misleading to say it, however, because of the connotation of a close division. In the case of the Keys, "majority" could mean six out of 11, and not even Lichtman's most vehement critics allege that.

Another problem with both the Hangways1 version and the Kalbome2 version is that neither acquaints the reader with the undeniable fact that some sources assess Lichtman as having been correct in all the 1984-2020 elections. Editors who think that "balance of reliable sources" weighs against that view are not entitled, on that basis, to proclaim that view to be wrong, or to excise it from the discussion.

I'm not wedded to "high degree of accuracy" but the fact is that many independent media have hailed him as a predictor. Above I suggested mentioning the "Nostradamus" comparison, which has been very common. Even the McFall article in Newsweek, overall very critical of Lichtman, refers to that term in its title. What about including that as a neutral indication of his reputation, without asserting that the reputation is justified? As McFall notes, "Allan Lichtman's 'Keys to the White House' model has long been hailed as a reliable predictor of U.S. election outcomes." That's an accurate statement of the mass media's most common attitude toward the Keys. We could convey that fact with the "Nostradamus" language. Or we could convey it just by citing McFall:

Lichtman's "model has long been hailed as a reliable predictor of U.S. election outcomes." For the elections from 1984 through 2020, assessments state variously that he got them all right, or that he was correct in all but 2000, or that he was correct in all but 2016. It is undisputed that his prediction of a win for Kamala Harris in 2024 was incorrect.

That quickly puts the reader in the picture. Without getting into "majority" or "large majority", the reader can see the numbers, can see where the disputes are, and can see the system's single most notable failure. JamesMLane

I like Handway's way of wording it. Simply stating he made predictions and then listing the ones he may have got wrong doesnt sound neutral either. Wikiman5676 (talk) 02:33, 9 November 2024 (UTC)

Also, why don't we just say "most" instead of large majority? Its a neutral and accurate term. nobody disputes he got most of the calls right. Wikiman5676 (talk) 05:10, 9 November 2024 (UTC)

Yes I am also a supporter of 'most'. Also feel like Handway's does the best job of striking the balance of NPOV. @JamesMLane's comes off as too sympathetic towards Lichtman and @Kalbome's a bit too critical.I want to thank all three though for their hard work on this topic! Jjazz76 (talk) 00:02, 10 November 2024 (UTC)

Having read this some more and watched the debate, Handways's proposal does a good job as well. I probably lean towards Kalbome's overall but both are fine. Caraturane (talk) 00:05, 10 November 2024 (UTC)

Ok as of this point in time, I'm reading:

• Hangways, James, Kalbome presumably all support our own version
• Jazz & Wikiman & TWM03 support an amendment of Hangways' version to use "most"
• Caraturane supports Kalbome's version

All of these views have some level of overlap (e.g. it seems we are all at least partial to one of the other phrasings); I'm just summarising the primary preference.

For now, I'm going to make a small edit to restore "most" instead of "large majority".

--------

I continue to not think that summarising 1984-2020 is useful for the article, since this is an incomplete version of Lichtman's track record:

• 2024 being unanimously incorrect is important to note distinctly, but the very next sentence already accomplishes that goal. This therefore isn't an argument for also collating perspectives on 1984-2020 before 2024 occurred.
• Similarly, the presence of sources summarising 1984-2020 isn't by itself an argument for highlighting that portion of a track record any more than the abundance of sources summarising 1984-2016 would be. The assumption here appears to be that if we don't have a wide range of sources specifically summarising the whole of 1984-2024, then we should default to selecting & summarising views of the closest time period we can get wrap-ups for (1984-2020) — but I view the leap from "summarise Lichtman's entire track record" to "as editors, select a different timeframe and summarise that" is actually quite a large difference in principle. Omitting specific years/data points from a track record and conveying conclusions made for the rest of it is the quintessential means of "using statistics to lie", right? I stress that nobody here appears to be doing this, but I don't think the article should head in this direction. We can simply summarise the entire track record and point out individual years of note without having to do this.
• On an almost identical note, the failure to "acquaint the reader with the undeniable fact that some sources assess Lichtman as having been correct in all the 1984-2020 elections" just strikes me as arbitrary. Do we also need to acquaint the reader with the undeniable fact that some sources think he was right from 1984-2016? Or 1984-2000? The same implied assumption is made that if extensive views wrapping up 1984-2024 aren't available, we should default to those that wrap up 1984-2020.

--------

Where I *do* agree with James is that this sentence does not convey the general sentiment that Lichtman has broadly been regarded favourably.

I wonder if something like this is acceptable to all?

Lichtman has successfully predicted most of the eleven presidential elections held since 1984 by using the keys. The system has been frequently regarded as a reliable predictor of U.S. election outcomes. However, he incorrectly predicted that Kamala Harris would win the 2024 election, and the nature and accuracy of his predictions for Al Gore in 2000 and Donald Trump in 2016 have been disputed.

This is a modification of one of James' sentences, with the primary replacement of "hailed" with "regarded" to be a bit more NPOV. (I know that "hailed" is a direct quote, but removing it allows us to add more than 1 source to justify the sentence.)

I like this version for three reasons:

1. The combination of those 1st and 2nd sentences conveys (to my mind) what James feels is missing from my original version (the vibe that Lichtman isn't just thought to have gotten 6/11 right, even by his critics), but without the need to step into summarising outdated views on the 1984-2020 period.
2. It very cleanly separates the factual stance that everybody agrees on (Lichtman has gotten most correct) from the overall impression of the model's value (which is much more contentious).
3. Despite the introduction of another positive sentence for Lichtman, this still feels balanced overall to me — perhaps because the second, more critical sentence had two parts to it. This paragraph as a whole is basically 50-50 positive/negative in tone for Lichtman by wordcount, so I think it's still fine.

I haven't made this edit yet as it hasn't been discussed. Hangways1 (talk) 04:21, 10 November 2024 (UTC)

I think this version is good, as it combines the unobjectionable parts of the two versions. However, I think we should be careful to avoid WP:RECENTISM when suggesting that the 1984-2020 record is "outdated" and not worth commenting on. Say if hypothetically the model gets the next three elections wrong, so that it becomes clearly worse than using the polls as a predictor based on all of its predictions. I still think it would be worth putting something in the article about how it was seen as reliable from 1984-2020, to explain why so many people cared about it, rather than just saying "it only got 9/14 right". I think it may be worth considering something similar here. TWM03 (talk) 09:40, 10 November 2024 (UTC)

I support this. Caraturane (talk) 16:18, 10 November 2024 (UTC)

I think Kalbome's version is a bit too harsh and kind of frontloads the 2024 election as more important than any other election/missed call from Lichtman. I don't know that I'd call it non neutral-POV but it's on the edge of that.

James' version I supported the earlier version of as a compromise before the 2024 result this week, but with that miss to the keys I think it just isn't apt anymore. Sources obviously disagreed on which elections to score in the key's favor but on the lowest end were at 8/10, and updating that to 8/11 (the miss here is undisputed, even by Lichtman so I think that's a safe update) I just don't think that reaches "high degree of accuracy". They adjusted this by separating out pre 2020 to post 2020 but like you I think that's an arbitrary choice.

On the balance, I like the version you've presented the most of the three hangways. I also prefer it with just the "majority" language, or with "most" as wikiman5676 just proposed.

Implicit and important here is that a lot of those 11 elections weren't really hard to call. Some we can debate that on (for instance 2008, which was not a close election but Lichtman made his call pretty early before that was obviously the case), but for others I hope that's not under dispute ('84 for instance). Some sources have definitely brought up this dynamic before if I need to argue against this being original research. In any event, I think leaving it at just "majority" is a way to neutrally reflect this dynamic in the intro. Apprentice57 (talk) 09:40, 9 November 2024 (UTC)

Kalbome's suggestion strikes me as the most neutral and accurate description of what has actually happened and what is reality (especially as his track record is no longer highly accurate from a comparative standpoint). Caraturane (talk) 19:06, 9 November 2024 (UTC)

I'm of the position that you have to highlight both his successes and failures, or not mention it at all to keep it neutral. Saying he made predictions and only pointing out the ones he got wrong is not NPOV. We should say something along the lines of "successfully predicted the outcome of most presidential elections. However, he missed...". Rather than just saying he made predictions, heres the ones he got wrong. Thats no better than saying, he successfully predicted most elections and completely ignoring the ones he got wrong in the lead. Wikiman5676 (talk) 22:53, 9 November 2024 (UTC)

I would be in favour of Hangways' preference, but I would want "a large majority of" replaced by "most of" so that we are not making subjective judgements about what fraction would be a large majority. Kalbome's phrasing says nothing about what he got right and only focuses on what he got wrong, which is not an fair reflection of the sources (the media only cares about the system because it has been right so much, and most articles emphasise this).

Something like JamesMLane's preference would also be OK with me, but I think the first two sentences should be changed to the past tense if we go with this to make it clear that these assessments were not based on the most recent information, and the last sentence should be stated more directly (i.e. "he got it wrong", not "it is undisputed that he got it wrong"). TWM03 (talk) 01:16, 10 November 2024 (UTC)

I think we should mention in the lead that Lichtman used the system to predict elections months, or sometimes years in advance. That's actually one of the things that makes his system noteworthy compared to other things like polling, which obviously can't predict much of anything until a few weeks before the election. Wikiman5676 (talk) 06:04, 11 November 2024 (UTC)

@Wikiman5676 Can you find a reliable source that expresses that? (ie that he generally makes these predictions early and that's notable)

One alternative if we can't is to note the year for certain individual predictions. eg I'm pretty sure his Obama prediction came at least 1 year out or maybe 2. We could add stuff like "In September 2007, Lichtman predicted that..." to help the reader decide which individual predictions were more/less impressive Hangways1 (talk) 06:20, 11 November 2024 (UTC)

Well. its literally stated in the body. The body discusses some of his predictions which came years in advance. I was thinking we just mention it in the lead. Wikiman5676 (talk) 06:38, 11 November 2024 (UTC)

Okay. so when i read the body now it seems like that isnt true anymore. But i remember that was the case in a prior version. "Lichtman's model received significant media coverage after he released his forecast for the 2012 election, predicting that Barack Obama would win re-election."

This used to also say why he received significant media coverage for his 2012 forecast. Namely that he released the prediction in 2010. And thats why it was covered. We should probably mention that in the body because just saying he got significant media coverage for his forecast of 2012 when he's a professional election forecaster is kinda silly if you dont explain why. Wikiman5676 (talk) 06:45, 11 November 2024 (UTC)

These sources all mention the Obama prediction.

https://www.usnews.com/news/blogs/washington-whispers/2011/08/30/never-wrong-pundit-picks-obama-to-win-in-2012

https://www.american.edu/media/news/20100712_lichtman_predicts_obama_wins_reelection_2012.cfm

https://www.historynewsnetwork.org/article/allan-lichtmans-13-keys-predict-obama-will-get-re- Wikiman5676 (talk) 06:52, 11 November 2024 (UTC)

As for sources that explicitly state he generally makes predictions months in advance. This one says it.

https://www.usatoday.com/story/news/politics/elections/2024/11/03/allan-lichtman-nate-silver-election/75993624007/

This is primary source but it also mentions years. Realistically we have to mention months or years. because even if the secondary source says months, we literally have an example in the body that says years.

https://www.socialstudies.org/system/files/publications/articles/se_720110.pdf Wikiman5676 (talk) 07:05, 11 November 2024 (UTC)

I recently found this source. Video from the NY Times. This mentions that Lichtman used the keys to predict Reagan would win reelection two years prior to the election.

https://www.youtube.com/watch?v=mp_Uuz9k7Os Wikiman5676 (talk) 07:47, 12 November 2024 (UTC)

 t c 22:57, 8 November 2024 (UTC)

References

JamesMLane t c 22:57, 8 November 2024 (UTC)

Changing "In Dispute" to be more specific

How does everyone feel about changing the description of the 2000 and 2016 election from "In dispute" to something similar to "He did not predict the outcome of the Electoral College in 2000 and 2024, or the popular vote in 2016 and 2024." This is probably more objective, especially because there seem to be basically no media sources beside Lichtman himself who contest the 2000 election. Kalbome22 (talk) 19:11, 11 November 2024 (UTC)

You mean "correctly predict" right? Jjazz76 (talk) 22:14, 11 November 2024 (UTC)

I think thats just too wordy. Saying in dispute is a good summary because there are complexities the to it the reader can read more about in the body Wikiman5676 (talk) 04:30, 12 November 2024 (UTC)

Well, there has been pushback from JamesMLane on thePostrider, but if accepted that would be once source that accepts his call for 2000 (as thePostrider's overall point is that it was always a popular vote model until after the 2016 election).

I also think that Lichtman can define what the model is predicting (so long as he does so ahead of the outcome of the election). And the important metric for a prediction is: did it match the variable he said it was predicting? That variable is the popular vote until 2016 (though I think there were claims from media sources for 2000 that he was insufficiently clear in his paper publication that it was popular vote; I'd have to double check those sources) and EV for 2020 and 2024. I don't think denoting the alternative (EV up to and including 2016, PV for 2020 and 2024) is very useful even if it is more objective. Apprentice57 (talk) 10:54, 12 November 2024 (UTC)

@Jjazz76 makes a good point that we should use the phrasing "correctly predict" where we can, I know a few times that's been fixed on these pages as just saying "he predicted" doesn't carry any meaning.

Good note on the Postrider source, may re-read that now that we have the 2024 election results. I forgot that they did the deep dive into the 2000 election too and have defended his prediction record in 2000. Actually may funnily enough be something worth including in the criticism section to add some more nuance to the criticism but that seems to be a hot topic again so push any buttons. Tomcleontis (talk) 19:12, 15 November 2024 (UTC)

Organisation of Criticism Section

This Talk page earlier debated two versions of the article, with "Version 1" being the status quo and "Version 2" being @JamesMLane's suggested major edit.

That debate unfortunately got framed almost entirely around NPOV disputes — but one of the major virtues of James' rework (which I *hope* we can all agree on!) was re-organising the Criticism section entirely by topic. The article previously demarcated an entire section for Nate Silver's criticisms specifically, which was pretty messy since some of his criticisms (e.g. on subjectivity) overlap with those of others'.

I have accordingly made a large edit to reorganise the Criticism section by topic.

---

The article has changed a lot since that debate, particularly around presentation of the popular vote vs electoral college content — so my headings & sorting aren't quite the same.

I also really like that James broke some of the statistical content down into further subheadings, eg "correlation with vote margins", "statistical fitting". I'm very open to integrating those.

However, I think we should also chat about whether presenting so much of Silver's statistical critique is actually required and NPOV. There's a lot of it and I wonder if we can summarise it more concisely to avoid giving it undue weight. I didn't want to cut any major content while re-organising, but I would personally favour a moderate cut here.

Point being... don't take the headings I put in as an indication that I much prefer them over James' Version 2 breakdown or anything!

---

I have added some small bits to Reception, notably:

• Lichtman's favourable reception by some outlets in general (not just for 2012 & 2016), including the Nostradamus terminology. @JamesMLane, hoping this goes some way to address your concerns...?
• More details of The Atlantic criticism of Lichtman's claim that he changed his methodology in 2016.
• A few sources here and there. (Some of the Silver critiques could have more sources added, even if it's the same one repeatedly being used.)

The biggest thing I tried to do was limit the Prediction of Popular Vote section to ONLY contain media commentary on the keys' predictions here.

Recall that the "Lichtman's prediction record (1984–present)" section already outlines exactly what Lichtman stated his predictions to be and his subsequent defence of his methodology. The kind of mental split I'm using is that the "Prediction Record" section currently contains only what Lichtman has predicted & stated about his popular vote predictions, while the "Reception" section currently contains only what others have said about it.

Is this a reasonable split? Is it redundant? Thoughts?

Hangways1 (talk) 06:34, 10 November 2024 (UTC)

I actually really like what you've done here, I think this does a good job reorganizing and siloing off the general critiques and points against his system. Caraturane (talk) 16:20, 10 November 2024 (UTC)

Yeah second my compliments to your re-organization. Like it a lot! Jjazz76 (talk) 00:30, 11 November 2024 (UTC)

Still no "Support" section, which my version included.

Still treating the Postrider bloggers as if they were worth quoting. They're not.

The "Prediction of popular vote" section is entirely one-sided. Points that are properly encyclopedic that are worth including, even if some editors disagree with them:

• Lichtman's interviews shortly before the 2016 election, predicting a Trump win without reference to the electoral vote.
• Reliable sources (more so than the bloggers!) that credit Lichtman with a correct 2016 prediction.
• Lichtman's explanation of the switch.

The same is true of the "2016 U.S. Election" section, very nonneutral. JamesMLane t c 02:00, 11 November 2024 (UTC)

I am fine with including some points in Lichtman's defense but this is in the Criticism section at large, so I don't know that it needs to be restated if we do it in the 2016 U.S. Election section.

You continue to be the only one (other than Lichtman) calling them "bloggers" and protesting their inclusion. The Postrider journalists (we have already been through this, the two are legitimate journalists with a background in these areas are are cited for it) are frequently cited across sources as the most prominent investigators and critics regarding Lichtman's record (probably the most prolific other than Nate Silver himself at this point). It should be beyond dispute that they're notable for their criticism of Lichtman at this point, especially given how frequently they're quoted or cited on this dispute (or, honestly, how often Lichtman talks about them). Much of the actual substance of the Atlantic article is based on their work, in fact a third of that article talks about Lichtman's relationship with them, and Lichtman has repeatedly referred to them for their criticism. Time and time again, Lichtman, and you, have singled them out, and it has become very transparent.

Overall, I think @Hangways1 work is sufficient, with some minor touch ups here and there. But this is a Criticism section, it can be a little repetitive, but doesn't need to fully rehash every single argument. I think the most obvious part that could be reduced is some of the Nate Silver-statistical accuracy in predicting margins, as that's not something Lichtman has necessarily said the system is useful for and just seems to be some superfluous Nate Silver math (I know he's known for that, but trying to lightly tease :D) Caraturane (talk) 03:12, 11 November 2024 (UTC)

@JamesMLane

• Support section: I actually had a look at the Wikipedia policy on Criticism sections and it usually advises against dividing articles into Support v Criticism for NPOV reasons. Obviously the Media Coverage section is extremely favourable to Lichtman but that's why I didn't rename it to something like Support. Maybe it's possible that we just scrap the Criticism heading entirely too, but what would the alternative layout be, that wouldn't be highly redundant with the Track Record section?
• Correct 2016 prediction: There are a few sources in there that do credit him with a successful 2016 prediction. Do you want this more directly stated, or..? Edit: Scratch that, just looked at the page any there is a direct sentence already stating that he was labelled one of the few forecasters to correctly predict 2016. This looks fine to me.
• Lichtman's Words: As above, I tried to keep Reception mostly to reception of Lichtman's work, not his own statements. His claim that he switched methodology, for example, is already in the article, just under his track record section — so I didn't add it in again under Criticism. Do you think it's worth the risk of redundancy to repeat this sort of stuff? If so, we'd obviously also need to double up on other content (eg Lichtman stating in 2016 that he was predicting the popular vote) to keep it balanced. (In your favour, there ARE some of Lichtman's words/defenses even in the Criticism section; though this is primarily because they were already in there and I tried not to cut sourced content during my reshuffle, rather than because I'm advocating for inclusion.)

Hangways1 (talk) 03:27, 11 November 2024 (UTC)

1. I agree that "Support versus Criticism" is often disfavored. In the present case, however, there are editors bound and determined to include every conceivable criticism. If we must have a separate Criticism section, then we should have Support.

An alternative: Reception section, first subsection is media coverage (just about how prominent the Keys have been), second subsection is "Issues" or the like. Within "Issues", one subsubsection is "Popular vote versus electoral vote" and it presents all the arguments about 2000 and 2016. I personally favor a summary presentation but it appears that that won't fly. People would continue to re-insert any criticism they could find. Therefore, as a practical matter, the only way to preserve balance against constant assaults would be to go kitchen-sink and have an unduly long section, so that no one could plausibly object to any omission.

Your first comment also highlights the overlap between the prediction record and a separate, dedicated section on the issue of popular vote versus electoral vote. Of course, in the current heavily POV version, the point of the overlap is to reiterate the anti-Lichtman position.

One alternative that I've thought about is to move all the substantive discussion to the dedicated section, and confine the prediction recap to a bare-bones mention with a wikilink.

2. The discussion of 2016 is unbalanced. The POV that says he got 2016 right is supported by Lichtman himself, by his explanation, and by multiple independent reliable sources. That POV is mentioned, as you note, but is not fairly presented. The current version of the article quotes two no-name nonnotable bloggers, but can't find room to report that an established reliable source, The Washingtonian, has agreed that Lichtman got 2016 right.

If you think it looks fine to say only that he was one of the few credited with getting 2016 right, then it should be equally fine to say only that some commentators say he got it wrong. Elaborate on both POVs or neither.

3. As I noted above, I don't like the redundancy. We have an advantage that someone writing a book or article decades ago did not have: We can wikilink. In the prediction recap, we could just give a bare-bones summary mentioning the various POVs, each with a couple of sources footnoted, and conclude with "See #Popular vote versus electoral vote" (or however the section is titled.) JamesMLane t c 15:45, 11 November 2024 (UTC)

"An alternative: Reception section, first subsection is media coverage (just about how prominent the Keys have been), second subsection is "Issues" or the like. Within "Issues", one subsubsection is "Popular vote versus electoral vote" and it presents all the arguments about 2000 and 2016."

I don't mind the idea of replacing "Criticism" with "Issues".

However, I don't support merging in a lot of the content currently in the prediction record section and merely wiki-linking to it, for a few reasons. Top three are:

1. A separate "track record" section (as it stands) separates out exactly what Lichtman has stated himself, both before and after his predictions. IMO that's a big virtue on a topic where the actual substance of his predictions is disputed, and helps the reader clearly identify what his predictions actually were.
2. We're presumably keeping the track record table, and it makes sense to keep the substance of what Lichtman stated his predictions to be near the table that summarises his predictions.
3. It might not be viable to use a "summary" section and merely link later, because exactly what do you say in that summary about 2016, for example? You obviously can't settle on a single judgement ('right', 'wrong', 'unclear') while you're also presenting the dispute later in the article. And the introduction of the entire article has already summarised his record at a high level. So if you're going to start getting into the nuance of what his prediction record is, a "summary vs issues" section split makes less sense to me than a "what Lichtman said vs reception of it" split.

"The discussion of 2016 is unbalanced. The POV that says he got 2016 right is supported by Lichtman himself, by his explanation, and by multiple independent reliable sources. That POV is mentioned, as you note, but is not fairly presented."

I don't remotely agree with this.

The "2016 U.S. Election" section contains:

• 1x obviously neutral paragraph
• 1x paragraph with Lichtman's statements that (seem to) support him predicting the Electoral College
• 1x paragraph with Lichtman's statements that (seem to) support him predicting the popular vote

I say 'seem to' because not only is this as neutral as it gets, but it's entirely Lichtman's own words! There's no actual POV at all here. Presenting the facts of what Lichtman stated in 2016 is not unfair to him, and the section doesn't include anything like "therefore he was right/wrong". This is absolutely neutral.

You state that the section excludes what other sources have thought of his record — well, yes, because the article has a Reception section and that content is accordingly in Reception instead. That section leads with a "Media Coverage" section which is entirely positive towards Lichtman and there are four sources there at the end of a sentence that specifically says he was labelled as one of the few forecasters to get 2016 right. You want to add the Washingtonian as a fifth source for it? Go ahead, I guess. But, frankly, the article in its current shape *is* neutral and contains what you claim it doesn't.

As a meta-note, I've also personally edited in several of your requests (the Brandeis source, the Nostradamus quote, more quantification of the 9/11 number, etc.) to try to find compromise, and there are several of us actively removing vandalism along with you (e.g. claims of pseudoscience). I also took you up on the offer to help improve your Sandbox version some time ago.

That's not to say there aren't more improvements to make. But, frankly, I don't see a way to bridge our perceptions on the article's current framing and overall editorial approach to it. You seem to be mired in this conspiratorial sense that everybody here (on a regular basis) except you is out to get Lichtman, and I'm just not seeing it. Hangways1 (talk) 17:00, 11 November 2024 (UTC)

The article as it now stands fully presents one POV about 2016. Furthermore, to make sure the reader doesn't miss the criticism of Lichtman, it's presented (with some redundancy) in two different places.

The other POV is a throwaway with string cites.

Putting aside the redundancy problem that hasn't attracted as much attention, if we want the fuller discussion to be NPOV, there are two ways to go:

1. Change "Criticism" to "Issues"; subsection on "Popular vote versus electoral vote"; in that subsection, fully present both sides. This is what I prefer.

2. Leave "Criticism" as totally negative and biased. Counter it with a totally positive and biased "Support" section. Sources in these two sections will be arguing with each other but the reader will have to look in both places to see the full discussion. This would be a disservice to the reader but at least it wouldn't be as bad as the current version.

From what you wrote, I'm not clear if you insist on approach #2 or if you're open to a reorganization along the lines of approach #1. JamesMLane t c 19:30, 16 November 2024 (UTC)

I think your rewrite and reorganization is really good. Agree with Caraturane about the criticism section re: Nate Silver and inclusion of Emerson and Lovito (Newsweek and Atlantic articles, where they are the prime quotes and citations, targets of Lichtman's ire, put that to some rest). I agree with you Hangways about avoiding repetition and redundancy, and yes I think Lichtman's defense in the criticism section is pretty much already there, we did a lot of work on that months ago.

Happy to see we're moving on to some more substantive stuff, thanks for your work @Hangways1. Tomcleontis (talk) 15:15, 11 November 2024 (UTC)

Focusing on the question of redundancy

In the current version, the section on 2016 recounts the popular vote versus electoral vote issue, doing so in a very biased way. To make it neutral, there would have to be a better presentation of the opposing POVs.

The redundancy problem is that the same issue is addressed again under Criticism. (The neutrality problem is that this section is also biased and, if standing alone, would also need to be augmented, but in this comment I'm focusing on redundancy.)

Both treatments omit the numerous reliable and independent sources that agree with Lichtman that his 2016 prediction was correct. One way to balance them would be to include these sources in both places, thus making the redundancy even worse.

This problem was present in the neutral version that I posted in my Sandbox. Both Hangways1 and Caraturane made some edits to that version, but what's there now still has heavy duplication.

I've come to think that the best approach is to have an extremely terse summary in the "prediction" section, with a wikilink to a dedicated "Popular vote versus electoral vote" section under Criticism.

For now, I'm going to work on a neutral version of the subsection under Criticism. If the consensus is to keep an extensive discussion under the track record as well, then the bias there can be addressed. JamesMLane t c 17:52, 15 November 2024 (UTC)

I think we need to get out of the habit of creating a new topic on this page for everything, since hasn't the Criticism section come up in like four other sections where there were already conversations about this. Personally, the Criticism section doesn't read "biased" to me, it reads like criticism, which is what it is. Tomcleontis (talk) 19:03, 15 November 2024 (UTC)

It's clearly POV to present only criticism, as the current version does. One alternative is to have a subsection in which everything negative is reported, and a separate subsection in which everything positive is reported. I think it makes more sense, though, to divide by issues. Thus, a subsection about "popular vote versus electoral vote" would fairly summarize all the different points of view on that score. Another subsection would report the criticism that the system uses qualitative judgments, and of course would also report Lichtman's response.

As for the "habit" of creating new sections, you are of course free to post your opinions in whatever format seems good to you. I will do the same. As my headline indicates, I want to focus specifically on the question of addressing 2000 and 2016 at length in two different places. JamesMLane t c 21:36, 15 November 2024 (UTC)

Yes, you have a habit of creating too many new sections rather than responding to the same. If we all did the same thing, it would be an amplification of the same problem, not a solution. Case in point: almost all of this could've been said in response to Hangways last and fairly comprehensive reply to you in the last convo. Apprentice57 (talk) 22:16, 15 November 2024 (UTC)

We just don't agree with you on the POV issue that keeps being rehashed. And you're mistaken about there not being enough sources about him calling 2016 right. There's *four* of them, at a certain point it becomes subtly non-neutral POV to over-list sources.

Frankly the article already has an issue at presenting these sources as equally deep as the deep dives into his record. That's a compromise that resolved in your favor. This has already been a long an arduous road to get where we are, and I didn't fight the current version because it seemed like we reached an acceptable compromise.

I cosign hangway's pushback to this that they wrote above, please respond to them there instead, and build consensus before making an edit. Apprentice57 (talk) 22:23, 15 November 2024 (UTC)

No, we have not reached consensus.

As for starting this subsection, the previous sections that you mention were generally about neutrality-related issues. I thought it was obvious that I was being more specific -- addressing the redundancy issue and wanting to focus just on that. Apparently it was not obvious. JamesMLane t c 02:56, 16 November 2024 (UTC)

Never stated we had.

And you commented on much more than just the redundancy in your OP. Apprentice57 (talk) 14:38, 16 November 2024 (UTC)

I wanted a discussion of the specific question of redundancy. No one has commented on that. The comments have instead reverted to the favorite sport of criticizing my conduct. Duly noted. JamesMLane t c 17:25, 16 November 2024 (UTC)

Ok let's SPECIFICALLY discuss redundancy. go ahead. but please be concise as much as possible. Jjazz76 (talk) 19:51, 16 November 2024 (UTC)

Thanks for your interest in considering my actual idea about improving the article!

Here's an example of what might be written under predictions. All the 2000 and 2016 material now in that section and not included in this summary would be moved to a dedicated section, "Popular vote versus electoral vote", which would have all the arguments on all sides.

=== Lichtman's prediction record (1984–present) ===

Using the 13 keys, Lichtman has correctly predicted the outcome in the majority of the eleven presidential elections from 1984 to 2024. He incorrectly predicted that Kamala Harris would win the 2024 election, which was instead won by Donald Trump. Disagreement arises concerning 2000 and 2016, the years when the popular vote and the electoral vote diverged. One point of view is that Lichtman predicted the popular vote; hence his prediction for 2000 ("Al Gore wins") was correct, but his prediction for 2016 ("Trump wins") was incorrect.* A second point of view is that Lichtman predicted who would become president; hence 2000 was incorrect but 2016 was correct.* A third point of view is that he was predicting the popular vote in 2000 but switched to predicting the electoral vote in 2016; hence both predictions were correct.* See #Popular vote versus electoral vote. It is undisputed that his eight other predictions in the period 1984-2024 were correct.

==== 2024 election ====

Lichtman incorrectly predicted that Kamala Harris would win the 2024 election. He attributes this to three unprecedented events: the Democrats "trashing" their sitting president after the first presidential debate, their eventual nominee not participating in any primaries or caucuses, and belief in disinformation during the election cycle being at a very high scale.

Each place where I have an asterisk would have just one or two footnotes, citing a source that supports that particular POV -- not content footnotes, just a link to show the reader that there are sources for each. JamesMLane t c 23:22, 16 November 2024 (UTC)

References

JamesMLane t c 23:22, 16 November 2024 (UTC)