As a source of general information, not just "Manifold Markets has a market with over 2000 traders [citation: link to manifold]".
Must be clearly accepted as a general Wikipedia policy or norm, not just one rogue editor adding one in and nobody caring enough to contest it.
Update 2025-03-11 (PST) (AI summary of creator comment): Additional Clarification on Accepted Sources:
Accepted as a source means the prediction market must be recognized under general Wikipedia policy as a reliable source, not merely cited once.
The citation must meet a higher threshold than one-off mentions (for example, being integrated into multiple articles or discussed by multiple experienced editors).
A single citation (or self-citation) does not suffice to demonstrate that a prediction market is generally accepted as a source.
1,000Wikipedia article LK-99 mentions prediction markets in the following manner:
>A video from Huazhong University of Science and Technology uploaded on 1 August 2023 by a postdoctoral researcher on the team of Chang Haixin,[41] apparently showed a micrometre-sized sample of LK-99 partially levitating. This went viral on Chinese social media, becoming the most viewed video on Bilibili by the next day,[56][41] and a prediction market briefly put the chance of successful replication at 60%.[57] A researcher from the Chinese Academy of Sciences refused to comment on the video for the press, dismissing the claim as "ridiculous".[56]
https://en.wikipedia.org/wiki/LK-99
>Prediction markets show very accurate forecasts of an election outcome. One example is the Iowa Electronic Markets. In a study, 964 election polls were compared with the five US presidential elections from 1988 to 2004. Berg et al. (2008) showed that the Iowa Electronic Markets topped the polls 74% of the time.[11] However, damped polls have been shown to top prediction markets. Comparing damped polls to forecasts of the Iowa Electronic Markets, Erikson and Wlezien (2008) showed that the damped polls outperform all markets or models.
https://en.wikipedia.org/wiki/Political_forecasting
>In 1995, Sale agreed to a public bet with Kevin Kelly that by the year 2020, there would be a convergence of three disasters: global currency collapse, significant warfare between rich and poor, and environmental disasters of some significant size. The bet was turned into a claim on the FX prediction market, where the probability has hovered around 25%.[2][18] Sale and Kelly agreed that William Patrick would be the judge of the outcome. Patrick stated that Kelly had won the bet. Sale then refused to acknowledge the loss, and did not pay the $1000 that had been previously agreed.[19]
https://en.wikipedia.org/wiki/Kirkpatrick_Sale
Some other potential inclusions:
>https://en.wikipedia.org/wiki/Kuo_Jung-tsung#cite_ref-10 (a prediction market is used as a source for the likelihood of this politician winning)
>As of January 18, Brown led Coakley in the Intrade prediction market by high double-digit margins.[93][94] Statistician Nate Silver of FiveThirtyEight.com projected on January 18 that there was a 75% chance that Brown would defeat Coakley.[95]
>The disaster occurred only seven weeks prior to the 2011 general election (which took place on 26 November 2011) and partially affected the campaign.[54] On 14 October it was reported that the disaster had caused a 4% drop in the governing National Party's polling on the iPredict prediction market.[55]
https://en.wikipedia.org/wiki/Rena_oil_spill
In most of these cases, the usage of a prediction market as a source has persisted over numerous Wikipedia edits, for example the prediction in the LK-99 article has been there since Jan 2024. I believe this closely aligns with the example that
gave to clarify the resolution criteria: "Prediction markets indicate that the origin of Covid-19 was likely from an accidental laboratory release."
Also, even if citing prediction markets is not commonplace, Wikipedia editors clearly accept using them as a valid form of evidence to use for their articles, given the persistence of these examples.
@spiderduckpig Hmm, the election ones are interesting, since they cite the probability and do seem to be presenting it as informative in and of itself.
Is there any prediction market that's listed on the list of accepted reliable sources?
@IsaacKing There isn’t an official list of accepted reliable sources. The closest is https://en.wikipedia.org/wiki/Wikipedia:Reliable_sources/Perennial_sources which lists sources that have been discussed repeatedly and the conclusions from those discussions. However there has been little to no discussion on citing various prediction markets, hence no entry. If you were particularly curious you could try to make a post on the noticeboard to gauge consensus.
Here's a Polymarket market being cited on Wikipedia as a primary source about itself.
@marvingardens It was added in September, meets the Wikipedia criteria for primary sources, and evidently didn't trip any bots that look for unapproved sources.
On the talk page, we can see ~10 active editors in good standing, including one former Wikipedia administrator, discussing various aspects of the article after the September edit. One section of discussion is dedicated to the exact portion of the article with the citation in question. At no point does anyone raise any questions or objections about the citation. At no point in the article's history has the citation been removed, while the aforementioned editors made various changes all around it.
@marvingardens based on the section linked and the list of references, I don’t see anything citing the market itself. I only see citations to news articles discussing the odds on polymarket.
@GleamingRhino The link is directly to the citation of a Polymarket market. You can see the citation number in the URL.
@marvingardens I think just ping @IsaacKing -- until the market is closed, moderators are unlikely to intervene here. Isaac shows up from time to time so keep pinging to get a review.
@marvingardens This doesn't seem to me like the question author's intention (especially given "likely requiring a certain threshold" in the description). I think the author meant citing markets themselves as a source of ground truth (e.g. "Ultimately, LK-99 was found to not be a room-temperature superconductor[1]" and then later "[1] [Link to Polymarket question about LK-99]".
(I also think that it's clear enough that this was the intention that the Polymarket citation should not cause the market to resolve YES.)
@EricNeyman Certainly we can agree that your example is one kind of citation that Isaac was thinking of. I do not see any reason to say it's the only kind of citation acceptable for this market resolution.
@marvingardens Hmm. I think "accept as a source" means something more than just "is cited at least once". If there were a Wikipedia page about me, they might cite my personal website for the statement "Isaac maintains a personal website", but that doesn't mean my articles are accepted as sources for whatever it is I'm writing about.
@IsaacKing I believe that the policy in question you are referring to is this one: https://en.m.wikipedia.org/wiki/Wikipedia:Verifiability#Self-published_or_questionable_sources_as_sources_on_themselves
Since this policy has been in place before market creation I’m assuming most claims to resolve YES based on it are not in the intention of the market or else its premise would be moot.
If there were a Wikipedia page about me, they might cite my personal website for the statement "Isaac maintains a personal website", but that doesn't mean my articles are accepted as sources for whatever it is I'm writing about.
I believe that would contravene their rules on self-published primary sources for living people. That would not be an acceptable citation as I understand it.
I'm not aware of any kind of clause in Wikipedia's citation policies that uses how many times something is cited as a criterion for acceptable sourcing.
Any "accepted source" will be acceptable to substantiate certain kinds of things, and not acceptable to substantiate other kinds of things. Your blog included, and prediction markets included. If your blog was accepted as a source for some statement about the Judge program, it still wouldn't be accepted as a source of information about yourself. But it would still be an accepted source -- just not on every topic for every kind of statement. As with every source.
@marvingardens I’m not sure where you got that impression from. Self published material by the person in question would be the only form of self published source explicitly allowed for BLP. See relevant policy https://en.wikipedia.org/w/index.php?title=Wikipedia:BLPSPS and https://en.wikipedia.org/w/index.php?title=Wikipedia:BLPSELFPUB
@GleamingRhino Okay, then in that case it would mean that Isaac's blog would be an acceptable source. Not for all topics in all ways but for some topics in some ways, like all sources.
@GleamingRhino I agree that that is one of the relevant policies that would govern this, and that it was already in place.
This policy has a list of five requirements for acceptability under those circumstances. The Polymarket citation is apparently seen by reputable editors to pass those criteria, along with any other criteria from other relevant policies.
That is not "moot". That is evidence for resolving this market.
Your position appears to be that only acceptability under novel Wikipedia acceptability policies would count, not merely the application of existing policies. I don't see why that would be assumed on this market, and I think such a standard would be unintuitive and fatally damaging to many markets on this website.
@marvingardens that’s not what I’m saying at all. What I’m saying is that market creators do not make markets off of things that are already known. It would not make sense for me to make a “will I eat lunch today” market if I had already finished my lunch. Markets that appear to fall under this category usually indicate a misinterpretation of creator intent or an error on the part of the creator. I also believe that you vastly overestimate the amount of scrutiny that Wikipedia editors give to individual citations.
Regardless, this is not my market and not my choice to resolve (I don’t even hold a position). But ultimately I think you have simply jumped the gun here and this case is not nearly as open and shut as you make it out to be.
@GleamingRhino But it wasn't already known. The standards in the policies in question had not yet been tested against a real-world instance -- that we know of, or that Isaac knows of.
You're saying that a reading of the policies would have been able to predict this eventuality... Great. Might as well go to a Trump lawsuit market and point out that the laws, precedents and standards that will decide the case already exist.