Polymarket resolutions compared with the ten IMD oracle answers

Comparison checked on 2026-10-11 UTC. Original research dated 2026-10-09. This is a historical comparison, not a live attestation for use by a contract.

Result: Six of the ten IMD jobs passed the 70-of-100 consensus threshold. Four accepted IMD answers differ from the Polymarket outcomes reported in the original research: Zelenskyy suit, MicroStrategy Bitcoin sale, US–Iran permanent peace, and Maduro election. Two accepted answers match: Barron/$DJT and Fort Knox. The other four jobs failed to reach consensus and therefore have no accepted IMD resolution, even though the explorer displays a leading answer. These totals are calculated from the primary oracle request records linked in the table.

The Polymarket column comes from the requester's original research job and its published report. It is an attributed historical baseline, not a newly verified Polymarket result. A direct refresh of Polymarket's Gamma API returned HTTP 403; attempts to retrieve the original report's separate Gamma snapshots from its advertised GitHub paths returned HTTP 404. The original report itself was retrieved successfully and its SHA-256 matches the explorer's published artifact hash. Its exact rules, volumes, dispute history and controversy discussion are retained below.

Summary table

All panels have 100 seats; 70 matching answers are required. “Agreed/answered” shows agents agreeing with the displayed answer divided by all recorded answers. “Disagreed” counts recorded opposite answers. “Never / no recorded answer” means a seat has no answer in the saved request record; it does not prove that the agent permanently refused, abstained or never answered later. For example, Fort Knox passed with 70 recorded answers, so its remaining 30 seats are not dissenters. True is mapped to Yes; false to No.

“Pass” means the request record is attested with quorum reached and a recorded signature. “Fail” means disagreed with no attestation. This is the oracle's historical consensus status, not independent validation of the underlying event. The explorer labels the six signed attestations expired at retrieval; their 24-hour validity has elapsed. Signatures were not cryptographically verified in this report.

#	Market	Polymarket (original report)	IMD accepted consensus / leader	Agreed / answered	Disagreed	Never / no recorded answer	Passed? (quorum 70)	Comparison	IMD evidence
1	Will Zelenskyy wear a suit before July?	No	Yes	70/74	4	26	Pass — signed	Differs	Job · API
2	MicroStrategy sells any Bitcoin by May 31, 2026?	No	Yes	70/95	25	5	Pass — signed	Differs	Job · API
3	US x Iran ceasefire extended by April 22, 2026?	No	None; Yes leads	56/94	38	6	Fail — panel disagreed	No accepted result; leader differs	Job · API
4	US x Iran permanent peace deal by June 15, 2026?	Yes	No	70/94	24	6	Pass — signed	Differs	Job · API
5	Ukraine agrees to Trump mineral deal before April?	Yes	None; No leads	54/96	42	4	Fail — panel disagreed	No accepted result; leader differs	Job · API
6	Will the U.S. invade Venezuela by January 31, 2026?	No	None; Yes leads	67/96	29	4	Fail — panel disagreed	No accepted result; leader differs	Job · API
7	Will Nicolas Maduro Win the 2024 Venezuela presidential election?	No	Yes	70/85	15	15	Pass — signed	Differs	Job · API
8	Was Barron involved in $DJT?	No	No	70/74	4	26	Pass — signed	Matches	Job · API
9	Trump declassifies UFO files in 2025?	Yes	None; No leads	66/94	28	6	Fail — panel disagreed	No accepted result; leader differs	Job · API
10	Gold missing from Fort Knox?	No	No	70/70	0	30	Pass — signed	Matches	Job · API

What the comparison establishes

Observed facts (primary oracle records): All ten request records report panelSize=100, quorum=70, boolean answers and a completed status of either attested or disagreed. Every recorded member in these ten snapshots has ok=true; there are no recorded failed or invalid member answers. The member-level counts reproduce the API's agreement clusters and the explorer's “X of Y agreed” displays. The API links in each row identify the underlying request; its request UUID differs from the job UUID in the explorer URL.

Calculated comparisons: Among the six passing jobs, four disagree with the inherited Polymarket baseline and two agree. The leading answers in all four failed jobs differ from that baseline. That does not make eight accepted reversals: only four are accepted alternative answers. Across these ten snapshots there are 872 recorded answers: 663 agree with each job's displayed leader, 209 disagree, and 128 of the 1,000 total panel seats have no recorded answer. These are job-seat totals, not counts of distinct agents across the entire experiment.

Inference: The jobs demonstrate that IMD can return different interpretations of controversial market questions and can withhold acceptance when agreement falls short. They do not establish which interpretation is correct, nor that IMD is more accurate than Polymarket. This is a selected set of controversial markets, not a representative accuracy benchmark.

Differences in inputs and unresolved evidence

The rows are matched by the actual market question in each oracle request, not by the order of the ten supplied links. “IMD resolution” refers to a retrospective answer to that submitted prompt; it did not settle or change the actual Polymarket market.

• #1 — Zelenskyy: The accepted answer is Yes; the selected definition treats a suit as a coordinated formal jacket and trousers with no tie required. The oracle prompt omits the original rule's final sentence requiring “a consensus of credible reporting.” This is a material input difference, so the comparison is not an identical-rules adjudication. The aggregate sources also contain inconsistent descriptions and dates of appearances; individual citations were not independently audited. See the row's API question, agreement.definitions and member notes.
• #2 — MicroStrategy: The accepted Yes treats any completed Bitcoin disposal as a sale and relies on a June 1 filing describing sales before the May 31 deadline. The original report says Polymarket's controversial clarification required confirmation within the deadline. That clarification is absent from this oracle prompt, and its exact wording and timestamp remain unverified. This compares interpretations of the supplied base rules, not verified identical clarification sets.
• #3 — US–Iran extension: Yes leads 56–38 but fails. The selected leader cites Trump's extension announcement, whereas the supplied rules require mutual agreement or the specified media-consensus fallback and exclude unilateral pauses. Whether that fallback was satisfied remains unresolved; the failed panel supplies no accepted answer.
• #4 — US–Iran permanent peace: The accepted answer is No, 70–24. The selected interpretation distinguishes a temporary ceasefire or interim agreement from a lasting end to hostilities. The original report's conflict over agreement dates and qualifying language remains unresolved by this comparison; a vote tally alone does not authenticate the agreement or its timing.
• #5 — Ukraine minerals: No leads 54–42 but fails. The selected definition excludes merely agreeing to negotiate and requires explicit involvement of rare earth elements. The question's rule that an announcement can qualify still matters. The failed panel does not resolve that interpretive dispute.
• #6 — Venezuela invasion: Yes leads 67–29 but fails, three votes short of quorum. The selected interpretation treats an announced intention to run Venezuela as intent to establish control. Polymarket's reported later clarification distinguishing a raid from an invasion is absent from this oracle prompt. The failed panel leaves that dispute open.
• #7 — Maduro: The accepted answer is Yes, 70–15, for Maduro only. The original report also includes the González market; no separate González oracle answer was supplied, and none is inferred here. The selected API definition says inauguration occurred by December 31, 2024, while sources in that same record describe a January 10, 2025 inauguration. That internal chronology conflict is flagged rather than accepted as a fact. The role of the official proclamation versus credible reporting remains the underlying interpretive issue.
• #8 — Barron/$DJT: The accepted answer is No, 70–4, matching the original report's final Polymarket No. The original report separately describes Polymarket criticizing UMA and refunding Yes holders; this table does not conflate a refund or platform intervention with an independent oracle verdict. The selected oracle definition adds an evidentiary interpretation of direct involvement; the underlying source claims remain unaudited here.
• #9 — UFO files: No leads 66–28 but fails. The oracle prompt restricts sources to the opening–closing interval ending December 10, 2025, while the market's written deadline is December 31, 2025. Some source strings in the API nevertheless reference 2026 reporting. Thus compliance with the submitted source restriction is unverified, and there is no accepted resolution.
• #10 — Fort Knox: The accepted No is unanimous among the 70 recorded respondents, not all 100 panel seats. The selected oracle definition ends the qualifying confirmation window at the market's March 11, 2025 close, although the written rule runs to June 30, 2025 and allows an audit-based early No. The appended prompt says to use sources between opened and ended dates; its application to the early close is ambiguous. Matching outcomes do not establish that the early closure was justified or identify the actual source UMA used.

Uncertainty about independence: Each recorded response has a distinct agent token ID within its job, but several IDs share an owner wallet. For example, the 70 Fort Knox respondents span 50 wallets. “70 agents agreed” therefore means 70 recorded agent IDs, not necessarily 70 independent people, owners, models or research processes. The records do not establish independent sources or reasoning.

Unanswered questions: Which missing seats would have answered if collection continued? How would all ten panels resolve with identical historical Polymarket clarifications and evidence windows? Are the member source claims correct and within each prompt's limits? Which interpretation is substantively correct? This report does not answer those questions. Original unanswered questions about lawsuits, UMA tallies, clarification timestamps and historical rule edits remain unanswered as well.

Evidence and local verification

Primary oracle JSON, explorer HTML and readable explorer text are saved under evidence/, indexed by job UUID. evidence/oracle-manifest.json links each job UUID to its request UUID and records the API retrieval time. evidence/comparison.json contains the derived counts. evidence/original-report.md preserves the original published report exactly; its SHA-256 is 10726ae3ab857bd65cef0a4719f2e523e68e7156a3656e3f17817c010264ee73.

python3 tools/build_report.py rebuilds this report offline. python3 tools/check_report.py checks all ten vote counts against the member records, quorum status against attestation status, count totals, the original report hash, row coverage and report reproducibility. These local checks assess consistency and provenance, not event truth, signature validity or research quality. No independent review was performed.

Original research context (2026-10-09)

The following is retained from the requester-linked report. Its “FACT (primary)” labels describe that original author's API-based evidence, not fresh verification by this update; its “I” statements refer to that author. Press claims and inferences remain attributed historical research. The original references to evidence/gamma-api/ describe that author's files, which could not be retrieved here. The unchanged original bytes are saved separately in evidence/original-report.md.

Top 10 most controversial Polymarket resolutions (in the vein of "Suitgate")

Research date: 2026-10-09. Covers markets resolved from June 2024 to June 2026.

How to read this report

Each entry is labeled by the kind of claim it makes:

• FACT (primary): taken directly from Polymarket's own public Gamma API (gamma-api.polymarket.com/markets?slug=…&closed=true), fetched 2026-10-09. This covers the exact question, the rules text (the description field), dates, volume, the final outcome (outcomePrices, where ["1","0"] means Yes won and ["0","1"] means No won) and the UMA proposal/dispute history (umaResolutionStatuses). Raw JSON snapshots are in evidence/gamma-api/.
• REPORTED: facts stated in press coverage, with a link to the source. I did not verify these independently.
• INFERENCE: my own reasoning about why a resolution was contested.
• UNCERTAIN / OPEN: places where sources conflict or I could not check the claim.

About "exact criteria". The quoted rules below are the description text Polymarket serves today, copied verbatim with typos kept (for example "is is" and "arial"). Two limits apply:

1. Polymarket sometimes adds "Additional context" or clarification notes to the market page. These notes are not in the API description field and only render in the page's JavaScript, so I could not retrieve them. Where press coverage quotes one, I mark it as REPORTED.
2. The API returns the current text. If Polymarket edited a description after launch, the original wording may differ.

How I ranked them (INFERENCE / judgment). No official "controversy" metric exists. I combined four things: (a) money at stake, (b) how many UMA dispute rounds the market went through, (c) how much mainstream or crypto-press coverage there was, and (d) consequences such as a lawsuit, a Polymarket override, or oracle reform. Someone weighing these differently could reasonably order the list differently.

---

Historical summary from the original report (before IMD comparison)

#	Market (exact question)	Resolved	Volume (API)	UMA history (API)	Core controversy
1	Will Zelenskyy wear a suit before July?	No	$242.2M	5× proposed→disputed	Widely reported "suit" at NATO; oracle said No
2	MicroStrategy sells any Bitcoin by May 31, 2026?	No	$375.8M	2× disputed	8-K showed 32 BTC sold May 26–31; lawsuit filed
3	US x Iran ceasefire extended by April 22, 2026?	No	$203.6M	3× disputed	Trump announced an extension; Iran's confirmation contested
4	US x Iran permanent peace deal by June 15, 2026?	Yes	$177.4M	2× disputed	Was a 60-day interim MOU a "permanent" deal?
5	Ukraine agrees to Trump mineral deal before April?	Yes	$7.1M	(not recorded)	UMA whale; no deal existed; Polymarket called it "unprecedented"
6	Will the U.S. invade Venezuela by January 31, 2026?	No	$8.4M	proposed (no dispute)	Maduro captured in a U.S. raid, but ruled "not an invasion"
7	Will Nicolas Maduro Win the 2024 Venezuela presidential election? (and Edmundo González market)	Maduro No / González Yes	$2.6M / $1.7M (event $6.15M)	(not recorded)	Official primary source said Maduro won
8	Was Barron involved in $DJT?	No	$1.14M	(not recorded)	Polymarket publicly said UMA was wrong and refunded
9	Trump declassifies UFO files in 2025?	Yes	$16.7M	2× disputed	No formal declassification; AARO video deemed enough
10	Gold missing from Fort Knox?	No	$3.5M	(not recorded)	Closed months early as No; manipulation alleged

Honorable mention: "Israel x Hezbollah Ceasefire extended by April 26, 2026?" resolved Yes ($20.9M, 2× disputed), even though its rules require confirmation from "Hezbollah". It is covered at the end.

---

1. "Suitgate": Will Zelenskyy wear a suit before July?

FACT (primary, API):

• Slug will-zelenskyy-wear-a-suit-before-july.
• Opened 2025-05-22; end date 2025-06-30; closed 2025-07-09.
• Volume $242,231,180. Final outcome No.
• umaResolutionStatuses: proposed, disputed, proposed, disputed, proposed, disputed, proposed, disputed, proposed, disputed.

Exact rules:

This market will resolve to "Yes" if Volodymyr Zelenskyy is is photographed or videotaped wearing a suit between May 22 and June 30, 2025 ET. Otherwise, this market will resolve to "No".

For this market to resolve to "Yes" the images or video must be taken and released within this market's timeframe. The images or video must be authentic, not the result of artificial intelligence or video editing.

The resolution source will be a consensus of credible reporting.

REPORTED:

• Zelenskyy appeared at the NATO summit on June 24, 2025, in a black jacket and trousers that many outlets described as a suit.
• UMA finalized "No", citing insufficient "credible reporting consensus" (Decrypt, 2025-07-09; CoinDesk, 2025-07-07).
• Traders alleged that UMA whales with "No" positions decided the vote. UMA said it saw no evidence of foul play.
• Defenders pointed to a May precedent, where a similar outfit was judged not to be a suit.
• Menswear writer "Derek Guy" weighed in on whether it was a suit (The Defiant; CryptoRank).

INFERENCE: The rules never define "suit" and rely only on "a consensus of credible reporting". That left the question of what counts as a suit to UMA token holders.

UNCERTAIN:

• Decrypt gives the window as "March 22" to June 30. The API says May 22, and I treat the API as authoritative.
• Press reports give different dates for when UMA's "No" became final (July 1 vs. July 8). The API's closedTime is 2025-07-09 00:30 UTC.

2. MicroStrategy sells any Bitcoin by May 31, 2026?

FACT (primary, API):

• Slug microstrategy-sells-any-bitcoin-by-may-31-2026, in event "MicroStrategy sells any Bitcoin by ___ ?" (event volume $407.2M).
• Market volume $375,813,105. Final outcome No.
• UMA history: proposed, disputed, proposed, disputed. Closed 2026-06-04.
• The sibling market "…by June 30, 2026?" resolved Yes.

Exact rules (API description):

This market will resolve to "Yes" if MicroStrategy sells any of its Bitcoin by 11:59 PM ET on the date specified in the title. Otherwise, this market will resolve to "No".

The primary resolution source for this market will be information from MSTR and on-chain data, however a consensus of credible reporting will also be used.

REPORTED:

• Strategy's Form 8-K (disclosed June 1) reported selling 32 BTC between May 26 and May 31.
• Polymarket's review ended June 3 with "No", after a UMA vote.
• Two Yes holders, William Wood and Thomas Bush, sued Polymarket, CEO Shayne Coplan and CMO Matthew Modabber in New York Supreme Court on July 3, 2026. Their claims include breach of contract, bad faith, unjust enrichment and NY GBL deceptive practices.
• The plaintiffs allege that Polymarket added "clarifying language" that "effectively required public confirmation by the May 31 deadline rather than merely a sale by that date" (The Block, 2026-07-07).
• Polymarket's reported position was that "Confirmation achieved outside of the market's time frame does not qualify" (Yahoo Finance).

INFERENCE: The description served by the API contains no requirement that confirmation happen within the window. The dispute is about the gap between that base text and a clarification that sits outside it.

OPEN:

• I could not retrieve the clarification text or its timestamp.
• I found no public outcome for the lawsuit as of the research date.

3. US x Iran ceasefire extended by April 22, 2026?

FACT (primary, API):

• Slug us-x-iran-ceasefire-extended-by-april-22-2026.
• Volume $203,621,392. Final outcome No.
• UMA history: 3× proposed→disputed. Closed 2026-05-01.

Exact rules:

This market will resolve to "Yes" if there is an official extension of the two-week ceasefire agreement between the United States and Iran announced on April 7, 2026, defined as a publicly announced and mutually agreed extension to the halt in direct military engagement between the United States and Iran, by the specified date, 11:59 PM ET. Otherwise, this market will resolve to "No".

Both extensions of the April 7 ceasefire and new agreements for an extended ceasefire will qualify, even if a brief period occurs during which there is no formal ceasefire in effect after the expiration of the April 7 ceasefire.

If a qualifying agreement is officially reached before the resolution date, this market will resolve to "Yes," regardless of whether the ceasefire extension ultimately takes effect.

An extension of the ceasefire agreement requires clear public confirmation from both the United States government and the government of Iran that they have agreed to halt military hostilities against one another for longer than the initially agreed two-week period, or for an official extension of the ceasefire agreement in place to be otherwise confirmed by an overwhelming consensus of media reporting.

Any form of informal understanding, backchannel communication, de-escalation, or unilateral pause in hostilities without a confirmed agreement on a qualifying extension will not qualify. Similarly, newly agreed-upon humanitarian pauses, limited operational pauses, or temporary tactical stand-downs will not qualify.

A newly agreed-upon broader peace deal will qualify if it includes a qualifying extension of the ceasefire agreement/halt in military hostilities. Agreements that outline future negotiations or de-escalation measures, but do not explicitly commit to extending the ceasefire, will not qualify.

This market's resolution will be based on official statements from the United States government and the government of Iran. However, an overwhelming consensus of credible media reporting confirming that an official ceasefire extension agreement has been reached will suffice.

REPORTED:

• On April 21, Trump announced on Truth Social an "indefinite extension at Pakistan's request".
• Pakistan's PM Shehbaz Sharif confirmed the extension on X, and the UN Secretary-General welcomed it.
• The CoinCentral article (2026-04-28) reports that Reuters, AP, BBC and WSJ covered the extension, but it cites no direct Iranian confirmation (CoinCentral).

INFERENCE:

• The likely basis for "No" is the clause that Iran must also confirm ("mutually agreed", "clear public confirmation from both").
• Yes holders relied on the "overwhelming consensus of media reporting" fallback.
• Iran's unilateral silence versus the U.S. announcement is the crux.

UNCERTAIN: I did not find a primary statement from Polymarket or UMA explaining the final "No".

4. US x Iran permanent peace deal by June 15, 2026?

FACT (primary, API):

• Slug us-x-iran-permanent-peace-deal-by-june-15-2026-734-856-129.
• Volume $177,370,818. Final outcome Yes.
• UMA history: proposed, disputed, proposed, disputed. Closed 2026-06-18. Event total $478.9M.

Exact rules:

This market will resolve to "Yes" if Iran and the United states agree to a permanent peace deal by the specified date, 11:59 PM ET. Otherwise, this market will resolve to "No".

A permanent peace deal refers to any agreement which explicitly indicates that military hostilities between the United States and Iran have ended or will permanently cease, or uses equivalent language clearly signaling a lasting end to military hostilities between the United States and Iran. Agreements that are explicitly temporary or which do not include a definitive agreement to end military hostilities between the US and Iran on a lasting basis (e.g. a temporary extension of the two-week ceasefire agreement announced on April 7, 2026), will not qualify.

A qualifying agreement will be considered to have been established if either of the following conditions are met:

• The United States and Iran each sign or formally adopt a written agreement (e.g. a treaty or multi-point agreement) which meets the above criteria.

• Both the governments of the United States and Iran provide clear public confirmation that a qualifying agreement has been definitively established. Negotiations, statements of progress, or other statements which do not constitute a definitive announcement that a qualifying agreement has been reached will not count.

The primary resolution source for this market will be official information from the governments of the United States and Iran; however, a consensus of credible reporting may also be used.

REPORTED:

• Pakistan announced a deal including "immediate and permanent termination of military operations on all fronts".
• Its details showed an interim 60-day framework, with a memorandum of understanding to follow (The Next Web, 2026-06-15).
• Citing Bloomberg, TNW says nine wallets hold more than half the UMA tokens used in dispute votes, and more than 60% of active UMA voters have Polymarket accounts.
• Calcalist describes traders contesting whether the MOU met the "permanent" definition, given continued attacks (Calcalist/CTech).

INFERENCE: The text says "permanent end", but the substance was transitional. The rules key on explicit language ("explicitly indicates … will permanently cease"), which favors Yes on the letter and No on the spirit.

UNCERTAIN: Sources disagree on when the MOU was signed (June 14 vs. June 17), which matters for a June 15 deadline. I could not resolve this.

5. Ukraine agrees to Trump mineral deal before April?

FACT (primary, API):

• Slug ukraine-agrees-to-give-trump-rare-earth-metals-before-april.
• Opened 2025-02-03; end date 2025-03-31; closed 2025-03-25.
• Volume $7,079,947. Final outcome Yes.
• The API has no recorded UMA history for this older market.

Exact rules:

This market will resolve to "Yes" if the United States and Ukraine agree to any deal between February 2 and March 31, 2025, 11:59 PM ET, that explicitly involves Ukrainian rare earth elements. Otherwise this market will resolve to "No".

This includes, but is not limited to, agreements related to the exchange of Ukrainian rare earths for U.S. aid (military or civilian), partnerships involving rare earth metals, future rights to rare earth resources, mining rights, or any other form of cooperation related to rare earth elements.

An announcement of a deal will qualify regardless of if/when the deal is enacted.

The resolution source for this market will be official information from the governments of the US and Ukraine.

REPORTED:

• The price went from about 9% to 100% between March 24 and 25. One holder with about 5M UMA across three accounts cast roughly 25% of the votes.
• Polymarket called it "unprecedented", said it was not a "market failure", and issued no refunds (The Block; CoinDesk, 2025-03-27).
• The actual US–Ukraine minerals agreement was signed at the end of April 2025, after the window closed.
• The episode is cited as a driver of UMIP-189 / MOOV2, which introduced whitelisted proposers in August 2025 (web3isgoinggreat).

INFERENCE: The broad "any deal … any other form of cooperation" language gave some cover for Yes. But no official US or Ukraine agreement existed by March 25, and the rules name those governments as the resolution source.

6. Will the U.S. invade Venezuela by January 31, 2026?

FACT (primary, API):

• Slug will-the-us-invade-venezuela-by-january-31-2026, in event "Will the U.S. invade Venezuela by...?".
• Volume $8,368,551. Final outcome No.
• UMA history: proposed only, with no formal dispute recorded. Closed 2026-02-01.

Exact rules:

This market will resolve to "Yes" if the United States commences a military offensive intended to establish control over any portion of Venezuela between November 3, 2025, and January 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to "No".

For the purposes of this market, land de facto controlled by Venezuela or the United States as of September 6, 2025, 12:00 PM ET, will be considered the sovereign territory of that country.

The resolution source for this market will be a consensus of credible sources.

REPORTED:

• U.S. forces captured Nicolás Maduro in Caracas on Jan 3, 2026.
• Polymarket's added "Additional Context" said that Trump's statement that the U.S. will "run" Venezuela "does not alone qualify the snatch-and-extract mission to capture Maduro as an invasion" (DeFi Rate, 2026-01-07; Forbes, 2026-01-07).
• Traders called the ruling "Polyscam" (Futurism).
• Separately, insider-trading allegations surfaced around the "Maduro out" markets.

INFERENCE: This is controversial for the opposite reason to the others: Polymarket's own context note, rather than a whale vote, steered the outcome. The fight was over "intended to establish control".

7. Venezuela 2024 presidential election

FACT (primary, API):

• Event venezuela-election-winner, $6,152,241 total.
• "Will Nicolas Maduro Win the 2024 Venezuela presidential election?" resolved No ($2.59M).
• "Will Edmundo González win the 2024 Venezuela presidential election?" resolved Yes ($1.73M).
• Both closed on 2024-08-06.

Exact rules (Maduro market; González is identical except for the name):

The 2024 Venezuela presidential election is scheduled to take place on July 28, 2024.

This market will resolve to "Yes" if Nicolás Maduro wins. Otherwise, this market will resolve to "No."

This market includes any potential second round. If the result of this election isn't known by December 31, 2024, 11:59 PM ET, the market will resolve to "No."

In the case of a two-round election, if this candidate is eliminated before the second round this market may immediately resolve to "No".

The primary resolution source for this market will be official information from Venezuela, however a consensus of credible reporting will also suffice.

REPORTED:

• The CNE declared Maduro the winner with 51.2%.
• The U.S. Secretary of State said González "won the most votes" (CRS).
• Critics argued UMA voters ignored the primary source (Bitcoin.com, 2024-08-09; LessWrong).

INFERENCE: On a literal reading, "primary … official information from Venezuela" favors Maduro. The resolution relied on the secondary "credible reporting" clause, which turned a factual question into a legitimacy judgment.

8. Was Barron involved in $DJT?

FACT (primary, API):

• Slug was-barron-involved-in-djt.
• Opened 2024-06-21; end date 2024-06-23; closed 2024-06-26.
• Volume $1,143,639. Final on-chain outcome No.
• A sibling market, "Did Barron Trump launch $DJT?" ($2.08M), with a stricter "definitive evidence" standard, also resolved No.

Exact rules:

This market will resolve to "Yes" if a preponderance of evidence suggests that Barron Trump was involved in the creation of the Solana token $DJT. Otherwise this market will resolve to "No".

Determination as to whether Barron was involved in the creation of $DJT will be made by this market's decentralized resolver, UMA, and will take into account all available evidence as of 12 PM ET, June 23.

REPORTED:

• After UMA voted No, Polymarket posted "We firmly believe that UMA got this resolution wrong" (The Block, 2024-06-27).
• Polymarket then said it was "conclusive" Barron was involved "in some way" and refunded Yes holders without changing the on-chain outcome (crypto.news).

INFERENCE: This is the clearest documented case of Polymarket publicly overriding its own oracle. That makes it controversial in both directions: some objected to the oracle's ruling, others to the override.

9. Trump declassifies UFO files in 2025?

FACT (primary, API):

• Slug trump-declassifies-ufo-files-in-2025.
• Opened 2025-04-17; closed 2025-12-10.
• Volume $16,680,241. Final outcome Yes.
• UMA history: proposed, disputed, proposed, disputed.

Exact rules:

This market will resolve to "Yes" if the Trump administration declassifies any previously classified files pertaining to extraterrestrial life and/or unexplained arial phenomena by December 31, 2025, 11:59 PM ET. Otherwise, this market will resolve to "No".

Announcements of declassifications that are not implemented within this market's timeframe will not count.

The primary resolution source for declassification will be official information from the government of the United States, however a consensus of credible reporting will also be used.

REPORTED:

• The trigger was a Pentagon AARO release of a military sensor video. AARO itself called the object "unremarkable".
• Critics said no formal declassification had happened and called the mechanism "proof-of-whales" (Yahoo Finance; Bitget/CryptoSlate syndication).
• Polymarket's own newsletter argued the release "quite possibly" counted (The Oracle by Polymarket).

INFERENCE: "Any previously classified files" is a low bar if any released video counts. The dispute was whether that AARO clip was previously classified and whether "the Trump administration declassifies" happened at all.

10. Gold missing from Fort Knox?

FACT (primary, API):

• Slug gold-missing-from-fort-knox.
• Opened 2025-02-17; end date 2025-06-30; but closed 2025-03-11.
• Volume $3,518,425. Final outcome No.

Exact rules:

According to the U.S. Government, Fort Knox holds 147,341,858.38 Troy Ounces of Gold (see: https://www.fiscal.treasury.gov/reports-statements/gold-report/21-02.html)

This market will resolve to "Yes" if it is confirmed that Fort Knox holds less than 147,300,000 troy ounces of gold between February 17, 2025, and June 30, 2025, 11:59 PM ET. Otherwise this market will resolve to "No".

If gold is removed from Fort Knox after February 17, 2025, this will not qualify for a "Yes" resolution. If an audit confirms that there is no gold missing from Fort Knox, or that it otherwise contains 147,300,000 troy ounces of gold or more, this market will resolve to "No".

The resolution source for this market will be official information from the U.S. Government, including the Department of Government Efficiency.

REPORTED:

• During the Ukraine-minerals fallout, a critic alleged that manipulators resolved this market "No" early in March, "stealing $3.5 million" (Mitrade, 2025-03-26).
• A trader on X complained it was resolved on a "super dumb technicality" (X post).

INFERENCE: The early "No" (3.5 months before the deadline) presumably relied on some official statement that the gold was all there, treated as "an audit confirms". Whether that statement was really an audit is the likely crux.

UNCERTAIN, and the weakest entry:

• A 2026 explainer (PolyScope) says the dispute hinged on the word "allow". That word does not appear in this market's rules, so the explainer may be describing a different Fort Knox market or be wrong.
• I could not confirm the dispute count or the official statement used.

---

Honorable mention: Israel x Hezbollah Ceasefire extended by April 26, 2026?

FACT (primary, API):

• Resolved Yes. Volume $20,894,918. UMA history: 2× disputed.
• The rules require "clear public confirmation from both the Israeli government and Hezbollah … or … an overwhelming consensus of media reporting".

REPORTED: Critics contrasted it with #3: Hezbollah, which is not a government, did not confirm, yet the market resolved Yes, while #3 resolved No (polymarkets.co.il commentary, a lower-reliability secondary site).

INFERENCE: The pairing suggests the "consensus of media reporting" fallback is applied inconsistently.

Cross-cutting patterns (INFERENCE, supported by the cases above)

1. Fallback clauses. Most of these markets have a primary source plus "a consensus of credible reporting" as a fallback (#1, #3, #4, #7, #9, Hezbollah). Disputes cluster on which of the two governs.
2. Undefined key terms. Contested words include "suit", "invade", "permanent", "declassifies" and "involved".
3. Event time vs. confirmation time (#2): did the event happen in the window, or did it have to be confirmed in the window?
4. Oracle concentration.
   • REPORTED: nine wallets hold more than half the voting UMA, and over 60% of active voters have Polymarket accounts (Bloomberg via TNW).
   • REPORTED: the WSJ found the top-10 wallets cast more than half the votes in the average dispute (via Yahoo Finance). I did not read these underlying analyses directly.
5. Off-API clarifications (#2, #6): Polymarket's "Additional context" notes materially changed outcomes but are not part of the base rules text.

Unanswered questions

• What exact clarification text was added to the MicroStrategy May 31 market, and when? This is the central fact in the lawsuit; I could not retrieve it.
• What is the status of Wood & Bush v. Polymarket after July 2026?
• What official source did UMA rely on for the early Fort Knox "No"?
• When was the US–Iran MOU signed relative to June 15, 2026?
• What were the full UMA vote tallies for each dispute? These could be reconstructed from UMA's Data Verification Mechanism records on-chain; I did not do so.
• Were the Ukraine-minerals, Barron and Venezuela-2024 markets formally disputed? The API's umaResolutionStatuses is empty for pre-2025 markets, so the absence of a recorded dispute there is not evidence of no dispute.
• Completeness: according to one secondary site, 2026 alone had more than 1,150 contested markets. Lower-coverage disputes may have been more egregious than some entries here, so this list reflects disputes that received press attention.