> For the complete documentation index, see [llms.txt](https://polymarketguide.gitbook.io/polymarketguide/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://polymarketguide.gitbook.io/polymarketguide/rules/precedents.md). # Precedents **Note:** This page is an archive of selected precedents based on past UMA Oracle resolutions. Interpretations may change over time, and some entries may reflect practices or reasoning from a particular period rather than current practice. Some precedents have since been incorporated into Polymarket's written rules and are no longer maintained here. Always start with the **market's resolution criteria**, which define the explicit terms of the market. The precedents below document recurring interpretations and implicit rules that may help with cases not fully covered by the written criteria. They are an educational reference and do not override current Polymarket rules or clarifications.
UMA Oracle Philosophy **Early Resolution and Consensus** UMA voters may favor early resolution when an outcome is overwhelmingly clear and further delay is unlikely to change the result. For example, when a market is already trading near 99.8% and supported by a strong consensus of credible reporting, voters may accept resolution without waiting for formal confirmation. **Waiting for Finalized Data** UMA voters may prefer finalized data from the listed resolution source even when preliminary information appears reliable. This can be important when results remain subject to review or revision, such as NASCAR results. UMA voters may also wait for the listed source even when revisions are unlikely. For example, voters have waited for Federal Register publication of executive orders even when the same information was already available through other official sources. Strict adherence to finalized data can also lead to disputes based on timing or technicalities rather than disagreement over the underlying outcome. **Proposer Due Diligence** UMA voters may expect proposers to understand how a resolution source publishes and revises its data, including practices that are not explicitly stated in the market's resolution criteria. *Examples*: * NASCAR results may appear immediately after a race but remain subject to review, protest, or penalties under NASCAR's own policies. Even if the market rules do not distinguish between preliminary and final results, a proposal based on preliminary results may be disputed for being too early. * In *Megaquake before August?*, a proposal was submitted shortly after USGS reported a qualifying earthquake. However, USGS routinely revises seismic data after initial publication. Although the reported magnitude was accurate when proposed, the source's revision practices provided grounds to dispute whether the proposal was made too early. **The Spirit of the Market** UMA voters may consider the spirit and intent of a market when the written criteria leave room for interpretation. This can include how a reasonable participant would have understood the market when it was created. The spirit can be restrictive or inclusive. For example, if a market asks whether someone ate a sandwich, participants may reasonably understand this to exclude a hot dog even if some definitions classify it as a sandwich. Conversely, a market asking whether someone gives another person a nickname may be interpreted broadly if the rules do not establish a specific formula for what qualifies. The spirit of a market can itself become a point of debate. Participants may disagree over what traders understood the market to mean when it was created or whether later evidence fits that original understanding. Explicit resolution criteria, definitions, and clarifications remain the primary guide.
Market Intention Markets may resolve according to the clear intent of the Polymarket interface when the written criteria leave room for interpretation. Context, how traders understood the market when it was created, and prior related resolutions may help establish that intent. *Examples*: * *Will MicroStrategy purchase Bitcoin?* — After MicroStrategy rebranded as Strategy, references to "MicroStrategy" were still understood to refer to the same company. The name change did not alter the intended subject of the market. * *Will Elon leave DOGE?* — A previous market, *Will Elon join DOGE?*, resolved to Yes, treating Elon Musk as having joined the Department of Government Efficiency. Even if someone could argue that he never formally "joined," that earlier resolution established how the follow-up market should be understood. * *US recession in 2025?* — The market required an announcement from the NBER, reflecting the market's intent to use the recognized formal designation of a U.S. recession rather than alternative definitions. Market intent is most useful when the rules are ambiguous or incomplete. Explicit resolution criteria, definitions, and clarifications still take priority.
Rule-Defined Terms When the resolution criteria define a term or event, that definition determines what qualifies. The resolution source provides evidence that something occurred, but does not override the definition in the rules. A resolution source's description or classification of something does not necessarily make it qualify if it does not meet the definition in the rules. *Examples*: * *Who will Trump publicly insult by August 31?* — The rules define what qualifies as an insult and distinguish personal or professional disparagement from criticism of specific actions, policies, or decisions. A statement described by credible reporting as insulting would not necessarily qualify if it does not meet this definition. * *How many transgenders become Federal Deputy in Brazil Elections?* — The rules classified a candidate as "transgender" only if they declared a transgender gender identity visible on the official Brazilian candidature portal at the specified time. Self-descriptions, rumors, behavior, later changes, and sexual orientation were not considered. * *Kai and Speed finish their Minecraft marathon by...?* — The rules defined when the "marathon" concluded and explicitly stated that completing the Minecraft challenge did not determine resolution. Finishing the challenge therefore did not necessarily mean the marathon had ended. In other markets, the resolution criteria may explicitly rely on the resolution source's own classification or measurement. In these cases, the source's classification determines what qualifies because the rules make it part of the resolution criteria. *Examples*: * *Will a hurricane make landfall in the US in August?* — The rules rely on NHC or CPHC classifications to determine whether a storm qualifies as a hurricane and whether a qualifying landfall occurred. * *Where will a 6.0+ earthquake occur by end of August?* — The rules rely on USGS data, including the preferred magnitude and the country to which the earthquake is attributed, to determine whether the event qualifies.
Typos and Minor Errors Minor errors generally do not affect resolution when the intended meaning of the Polymarket interface is clear. This can include incorrect years, misspelled names, or small wording errors in supporting materials. *Examples*: * *Super Bowl Championship 2026* — The UMA proposal incorrectly referenced 2025, while the Polymarket interface consistently used 2026. The market resolved according to the interface. * *Zohran Mamdani wins the Democratic primary for NYC Mayor* — Mamdani's name was misspelled, but the intended subject was clear. * *"However a consensus of credible reopening may also be used"* — A wording error in the resolution note did not change its clear intended meaning. When the intended meaning is clear, minor textual errors generally do not change the resolution.
False Equivalences False equivalences often arise when participants argue that a condition has been met by treating a related event, phrase, or term as equivalent to what the market actually requires. These arguments generally appear at the event, phrase, or word level. We will take a look at two markets where these types of reasoning appear. ***Will Iran close the Strait of Hormuz before July?***\ This market required Iran to halt or severely restrict international maritime traffic through the Strait of Hormuz. ***Will Zelenskyy wear a suit before July?***\ This market required Volodymyr Zelenskyy to be photographed or videotaped wearing a suit. *** **Event-Level False Equivalence** *Misinterpreting what kind of real-world event qualifies.* **Example:** Treating a slowdown as equivalent to a closure.\ **Focus:** Misjudging the scale, nature, or completeness of the event. In the *Strait of Hormuz* market, some participants argued that a large drop in maritime traffic caused by Iranian GPS jamming satisfied the market. However, a reduction in traffic is not necessarily equivalent to Iran halting or severely restricting maritime movement. *** **Phrase-Level False Equivalence** *Interpreting a phrase more broadly than the resolution criteria support.* **Example:** Treating any major disruption as meeting the phrase "severely restricts."\ **Focus:** Stretching a phrase beyond the threshold required by the market. In the *Strait of Hormuz* market, participants disagreed over what level of disruption qualified as "severely restricts." A significant effect on traffic did not automatically establish that the specific threshold in the rules had been met. *** **Word-Level False Equivalence** *Treating related but non-equivalent terms as interchangeable.* **Example:** Treating a suit jacket as equivalent to a suit.\ **Focus:** Substituting a related term for the specific term required by the market. In the *Zelenskyy suit* market, a suit referred to a Western-style matching jacket and trousers. A *suit* is not categorically the same as a *suit jacket*, *jacket*, or another related term. Foreign-language terms such as *костюм*, *traje*, or *completo* may qualify when the context clearly refers to a Western-style suit. These terms can have broader or narrower meanings, so a direct translation alone may not be sufficient.
Resolution by Necessity Some markets hinge on a single key event. In these situations, participants may argue that a borderline or ambiguous event must be considered because there may be no clearer or more definitive occurrence before the deadline. If the potential qualifying event is ignored, the market may resolve to No by default even though something arguably satisfying the condition already occurred. This does not mean an ambiguous event automatically qualifies. The argument is that the event deserves closer consideration because it may be the only plausible qualifying instance available before resolution. *Examples*: * *Will anyone audibly fart during the Digital Asset Summit?* — If an unclear sound that might be a fart is heard, some participants may argue that it qualifies because a more obvious example is unlikely to occur before the deadline. Others may argue that without clearer evidence, such as visual cues or an unmistakable sound, it does not satisfy the criteria. * *Biden falls asleep during debate?* — If footage shows Biden with his eyes closed or his head nodding, some participants may argue that it is the only moment that plausibly qualifies. Others may reject it if the evidence does not clearly establish that he actually fell asleep.
Official Sources Some markets require confirmation from an official or authoritative source before resolution. When the rules specify such a source, other reporting generally does not replace it unless the market explicitly allows another source or a consensus of credible reporting. **Elections** Markets requiring certified results must wait for the relevant certification. Early calls or projections do not substitute for certification unless the rules allow them. *Example*: *Has Georgia certified the vote?* **Economic Indicators** Markets tied to CPI, GDP, employment, or similar data generally rely on the official release from the listed agency or source rather than estimates, leaks, or third-party summaries. *Example*: *Will the June CPI exceed 3.5% year-over-year?* **Government Announcements** Markets involving government decisions may require formal documentation from the Federal Register, Federal Reserve, White House, or another named official source. Reporting or speculation does not qualify if the required official action has not been documented. *Example*: *Will the Fed raise interest rates at the July 2025 meeting?* **Corporate Results** Markets focused on company performance may require the official release from the company's investor relations site or another listed source. Rumors, analyst commentary, or early leaks do not qualify. *Example*: *Tesla deliveries this quarter?* **Interviews / Earnings Calls** Markets based on interviews, press conferences, earnings calls, or similar events may require the actual audio or video recording. Summaries or third-party reporting do not replace the event itself unless the rules allow them. *Example*: *What will Elon Musk say during Tesla Q1 2025 earnings call?* **Sports** Sports markets may require confirmation from the official governing body or event operator, especially where results remain subject to review, penalties, or correction. *Example*: *F1 Constructors Champion*
Source Data and Timing **Time Inclusion** Markets generally include the final minute listed. For example, if a market ends at 11:59 PM, events occurring during the full 11:59 PM minute may still qualify. The resolution criteria take priority when they specify an exact cutoff time or define the market window differently. **Time Zones** The applicable time zone depends on the market's resolution criteria, event, and resolution source. Markets may use ET, UTC, local time, or another specified time zone. When no time zone is specified, the coverage of the resolution source may help determine when the market window has fully ended. For example, a source covering events across multiple U.S. time zones may require waiting until the deadline has passed in the latest applicable time zone. *Examples*: * *Another commercial airline evacuation before March?* — A "No" proposal was submitted shortly after 12:00 AM ET on March 1. Because the FAA reports incidents in local time across the United States, it was still February 28 in Hawaii, where a qualifying evacuation could still have occurred. Resolution could not confirm No until February 28 had ended in the latest applicable time zone. * *How many different countries will Israel strike in 2026?* — Although the market covers multiple countries, it uses ET for its start and end times. * *Where will a 6.0+ earthquake occur by end of August?* — Although the market covers multiple countries, it uses UTC for the earthquake's origin time and reporting deadline. When a market specifies a time zone, that time zone governs regardless of where the event occurs. **Clearly Incorrect Data** Initial data from an official source can be disregarded when it is clearly incorrect, even if it appears first. *Example*: If a price feed briefly displays an obvious glitch, such as BTC at $1, that value would not be used for resolution. Similarly, if there is strong evidence that a source was hacked, manipulated, or otherwise displaying incorrect data, that data may be disregarded even if it appears to come from the named source. **Preliminary vs. Final Information** When a market calls for final or official data, early or unofficial versions from the source are not sufficient unless the market explicitly allows them. When the rules do not specify whether preliminary or final data should be used, UMA voters may prefer finalized data, particularly when the source routinely revises its initial figures. However, this may depend on the circumstances and whether the initial data remains valid under the source's normal revision process. Some markets instead specify that the initial release will be used for resolution. In these cases, later revisions generally do not affect the outcome unless the rules provide an exception, such as a correction to clearly erroneous data. *Examples*: * *China Annual GDP Growth 2026* — Uses data from the initial GDP release and disregards later revisions. * *Will Anthropic's valuation hit $925B by June 30?* — Uses initially published NPM data and disregards later revisions unless they correct clearly erroneous data. Some markets treat revisions asymmetrically: a later revision can establish that a qualifying result occurred, but cannot invalidate a previously published data point that already qualified. *Examples*: * *Conservatives flip Liberals for Canada Seats Polls in 2026?* — A later revision showing a qualifying projection can count if published within the market's timeframe, but it cannot invalidate an earlier data point that already resolved the market. * *Will a hurricane make landfall in the US in August?* — A later correction can establish that a qualifying landfall occurred, while a later downgrade does not invalidate an earlier qualifying advisory when the rules preserve that initial determination. **Methodology Changes** Some markets continue to use the named resolution source even if that source changes how it calculates the relevant metric. *Examples*: * *Trump approval rating on August 21?* — Changes to Silver Bulletin's methodology do not affect resolution as long as the specified approval rating remains available. * *Conservatives flip Liberals for Canada Seats Polls in 2026?* — The market uses 338Canada's published seat projections regardless of methodological changes, confidence intervals, or projections from other sources. **Proposals Submitted Too Early** A proposal may be rejected as Too Early (P4) if the required data is not yet available. However, if a proposal is made when sufficient valid data exists to satisfy the resolution criteria, it may be accepted even if contradictory information emerges later. All relevant data must be considered, but if the proposal was valid at the time it was submitted, later developments do not invalidate it. On the other hand, if no proposal is submitted before contradictory data emerges, then the initial data may no longer be sufficient. In that case, the market must be evaluated using the full set of information, including the new conflicting evidence. Later confirmation also does not retroactively validate a proposal that was too early when submitted. Voter preference for waiting is not enough to justify a P4 dispute if the resolution criteria have already been met.
Event Deadline vs. Data Availability Deadline An event may occur within the market window but appear on the named resolution source later. The event deadline determines when the event must occur, while the data availability deadline determines when the confirming data must appear. The confirming data may appear after the event deadline if the rules provide a grace period or delayed publication window. Resolution should follow the source and timing specified in the market criteria rather than outside reporting alone. **Earthquakes** *Example*: In *Earthquake 7.0 or above before August?*, a qualifying earthquake had to occur within the market window, but USGS could report it afterward. The market could remain open until the specified data availability deadline for the event to appear on the resolution source. If it still had not appeared by then, the rules allowed other credible sources to be considered. **Revision Window** Some markets also include a revision window after qualifying data first appears. *Example*: If a qualifying earthquake is registered before the deadline, the market may remain open for an additional period to allow USGS to revise the recorded magnitude. Resolution then uses the latest available data after that revision window, even if the window extends beyond the market's event deadline. **Executive Orders** *Example*: Executive order markets may give the Federal Register until 12:00 PM ET the following day to publish an order signed before the event deadline. If the order does not appear by the data availability deadline, the market may resolve to No even if the order was signed. The key distinction is that an event can occur on time while the evidence required for resolution becomes available later.
When to Wait for the Next Data Point Some markets cannot resolve until the next data point has been published. The next data point may be used to confirm that the data relevant to the market is finalized. *Examples*: * **Daily temperature markets:** The next day's first data point may finalize the previous day's temperature data, after which later revisions are no longer considered. * **Approval rating markets:** The rating for a specified date may not be considered finalized until the next data point is published.
Early Resolution Some markets include an explicit early resolution clause, such as allowing a consensus of credible reporting to suffice before official confirmation. Other markets do not. An outcome being highly likely or effectively certain does not necessarily mean the market can resolve early. *Example*: In *Who will win Dem nomination for NYC Mayor?*, the rules allowed an overwhelming consensus of credible reporting to be used instead of waiting for the first official announcement of the results. For an early proposal to be valid, the required consensus or other qualifying condition must exist at the time of proposal. Later confirmation does not retroactively make an earlier proposal valid. Additionally, if it becomes materially impossible for a candidate or outcome to proceed, such as through concession or missing a required filing deadline, this may support early resolution if it is consistent with the market's rules and the available reporting. A consensus of credible reporting can function as an early-resolution mechanism when the rules allow it, letting a market resolve before the official source, certification, or final data is available. If the market requires the official source or final data and does not provide a consensus-based alternative, the market generally should not resolve early.
"Yes" Proposals After Deadline "Yes" proposals can generally be submitted once a qualifying event occurs, while "No" proposals often require waiting until the market deadline. In some disputes, however, a "Yes" proposal is submitted after the deadline based on an event that occurred much earlier. When this happens, the market's behavior around the alleged event can provide useful context. If the event was publicly known, no "Yes" proposal was submitted at the time, and the market continued trading at a low probability afterward, this may weigh against a later argument that the event clearly satisfied the resolution criteria. *Example*: In *Will Tesla launch a driverless Robotaxi service before July?*, a "Yes" proposal submitted after the deadline relied on an event that had occurred weeks earlier. After that event, the market had traded around 10%, suggesting that traders at the time generally did not view it as clearly satisfying the market. Market price and proposal history do not determine resolution on their own. However, they can provide useful evidence of how traders understood the event at the time.
Concurrent or Linked Markets Related markets may resolve from the same underlying event when they cover the same event, use overlapping criteria, or represent different outcomes of the same event. If one related market resolves without challenge while another is disputed over the same issue, UMA may look to the accepted resolution for consistency. This is especially relevant when the dispute concerns timing, a shared definition, or related outcomes. **Single-Winner Markets** In single-winner markets, only one outcome can ultimately resolve to Yes. This can create consistency issues when some contracts resolve while others remain disputed. *Example*: In *Highest grossing movie in 2026?*, each movie is represented by a separate contract, but only one can ultimately be the highest grossing. If some losing contracts resolve to No without being disputed as too early while another contract is disputed on the same timing grounds, those accepted resolutions can weigh against the Too Early argument. All of the contracts depend on the same underlying ranking being sufficiently final. If the same timing issue applies to every related contract, a disputer may need to challenge all of them rather than only one. **Identical Markets Across Timeframes** Some markets ask the same question across different deadlines and use identical or substantially similar resolution criteria. *Examples*: * *US-China trade deal before June?* * *Trump trade deal before June?* * *Trump trade deal before July?* If the same qualifying event falls within multiple market windows and the relevant criteria are identical, it may qualify across multiple markets. Resolutions in one timeframe may also help inform how related versions are interpreted. **Shared Definitions** A definition established through a dispute, prior resolution, or Polymarket clarification may carry over to other versions of the same market when the wording and resolution criteria are identical. *Example*: In disputes over *Will Zelenskyy wear a suit before July?*, UMA accepted a particular interpretation of what qualified as a suit. That interpretation was then applied to other versions of the market with the same criteria.
Rolled Votes When a vote fails to reach UMA's required consensus and rolls into another round, voting patterns may change as the underlying event develops or voters reconsider whether the original proposal was too early. Historically, some votes split between P4 (Too Early) and P1/P2 have shifted toward the substantive outcome in later rounds once the event became clearer. *Examples*: * *Major cyberattack on Iran in June?* * *Fordow nuclear facility destroyed before July?* In both cases, early voting included substantial support for P4 and P2, but the markets ultimately resolved P2 after later voting rounds. Large UMA stakers can also materially affect the result of a rolled vote if they change their position between rounds. For this reason, an early vote distribution does not necessarily predict the final resolution.
Sports **Games & Competitions** Sports markets generally resolve according to the official result from the governing body or event operator specified in the resolution criteria. Other sources may be used when the rules explicitly allow them. A game or competition ending does not always mean its result is final. Results may remain subject to review, penalties, protests, or other official changes before they are confirmed. *Examples*: * *Super Bowl Champion 2025* — Resolution source: NFL. * *NBA Finals: Will any player score 40+ points?* — Resolution source: NBA. * *Europa League Final: Player of the Match* — Resolution source: Europa League. Some sports markets also specify how later revisions are treated. Some football and soccer rules explicitly state that revisions to officially declared final scores made after market resolution are not considered. *Example*: *Eintracht Frankfurt vs. Hull City* explicitly excluded revisions to officially declared final scores made after market resolution. **Transfer & Roster Markets** Markets involving player transfers, team changes, or roster movements may rely on official announcements from the player, team, club, or other parties specified in the resolution criteria. Reporting from sports media does not replace those sources unless the rules allow a consensus of credible reporting. *Examples*: * *Will Neymar leave Al-Hilal?* — Resolution sources: Neymar, Al-Hilal, or the signing club. * *Will Jordan Henderson leave Saudi Arabia this transfer window?* — Resolution sources: Al-Ettifaq or the signing club. * *Which Soccer Players Will Sign With New Clubs?* — Resolution sources include the clubs involved or a consensus of credible reporting. **Event Participation & Hosting** Markets about where an event takes place, when it occurs, or who participates generally follow the official sources and any consensus-based alternatives specified in the resolution criteria. *Examples*: * *Will Novak Djokovic play in the 2022 French Open?* — Resolution sources include the French Open and credible media sources. * *Will Novak Djokovic win a gold medal in Paris?* — Resolution sources include the International Olympic Committee or a consensus of credible reporting. * *Will the 2021 Tokyo Olympics take place?* — Resolution sources include the International Olympic Committee, the official Olympics site, and the Markets Integrity Committee.
Politics **Senate Appointments** A cabinet nominee is considered confirmed once the Senate vote is closed and the result has been announced. Ongoing votes or reported intentions do not qualify. **Senator Votes** Markets based on how a Senator voted generally require the relevant vote to be officially closed. Reported intentions or incomplete tallies are not sufficient. **Elections & Primaries** Markets requiring certified election results cannot resolve until the relevant certification specified in the rules is complete. *Example*: If a market requires statewide certification, county-level certification alone is not sufficient. **Consensus of Credible Reporting** Political markets may allow a consensus of credible reporting to be used before official results or certification, but only when the resolution criteria provide for it. A concession or a mathematically impossible path to victory may support resolution when consistent with the market's rules and the available reporting. **Executive Orders** Markets requiring publication of an executive order may rely on the Federal Register rather than press briefings, announcements, or other evidence that an order was signed. Some executive order markets provide additional time after the event deadline for publication in the Federal Register. If the order does not appear by the specified data availability deadline, the market may resolve to No even if it was signed. **Conferences & Events** Markets tied to political meetings, summits, or similar events may depend on the first qualifying interaction specified by the rules, such as a meeting, handshake, speech, or other formal contact. **Visits** Markets referring to presidential or diplomatic visits generally require the individual to physically enter the territory specified by the resolution criteria. *Example*: In *Will Trump visit China before September?*, entering Chinese airspace alone did not qualify. The rules required Trump to physically enter Chinese land or maritime territory, as confirmed by official information or credible reporting.
Crypto Crypto markets can raise edge cases around prices, token launches, airdrops, and asset classifications. Resolution depends on the market's specific criteria and the evidence required by those criteria. **Crypto Prices** Price markets should follow the threshold, data source, precision, and timing specified in the resolution criteria. Whether a temporary price movement qualifies depends on how the market defines the relevant price observation. **Token Launches** What qualifies as a token "launch" depends on the market's resolution criteria. Older markets sometimes left the term ambiguous, while newer markets may define the launch condition directly. *Examples*: * *Hawk Tuah token by Thanksgiving* — There was disagreement over whether deployment alone meant the token had launched before an official announcement. The market was not disputed, so it did not establish a clear UMA precedent. * *Will Arc launch a token by \_\_\_?* — The rules require the governance token to be actively and publicly transferable and tradable. Announcements alone do not qualify. **Airdrops** Airdrop markets may distinguish between a token being announced, allocated, distributed, claimable, or tradable. Restrictions such as locked or nonswappable tokens may also matter when the rules explicitly exclude them. *Example*: The LayerZero airdrop raised questions about whether additional claim requirements affected whether an airdrop had occurred. Resolution focused on whether tokens were actually distributed and claimable under the market's criteria rather than whether the distribution followed a typical airdrop structure. **Strategic Bitcoin Reserve** Possession of Bitcoin by a government does not necessarily establish a strategic Bitcoin reserve. Resolution may require a formal policy or designation establishing the holdings as a reserve. *Example*: In a previous UMA dispute over a U.S. Strategic Bitcoin Reserve, seized or otherwise government-controlled Bitcoin was not treated as sufficient on its own without a formal designation or policy establishing it as a strategic reserve.
Product Launch Markets about product or app launches often depend on whether the product is publicly accessible rather than merely announced, internally available, or privately tested. **Public Access** A product may qualify as publicly available when access is open to the general public without invitation or individual selection. This can include paid access, open beta programs, or open rolling waitlists when anyone meeting the stated conditions can participate. Regional availability may still qualify as public access when the product is openly available to the general public within that region. *Examples*: * *Will Grok 4 be released by September 30?* — Grok 4 became available to X Premium+ users in Japan before its official announcement. Because access was available to anyone in the region willing to subscribe rather than an invite-only or selected group, it qualified as public access. * *GPT-6 released by...?* — Current rules explicitly require the model to be launched and publicly accessible, including through an open beta or open rolling waitlist. Closed beta or private access does not qualify. **App Store Launches** For markets requiring an app-store launch, the product generally must be publicly available and installable on the specified store and in the specified region. Pre-orders, placeholder listings, or private testing distributions such as TestFlight do not qualify unless the rules state otherwise.
A More Collaborative Oracle The Oracle is fundamentally collaborative. That means involving a broad range of participants from different backgrounds, perspectives, and communities in discussions around market resolution, interpretation, precedent, and edge cases. It should not be limited to a small circle of longtime Oracle participants or core community members. Prior to July 2025, most resolution discussions were shaped by a small group of UMA community members who held disproportionate influence over how outcomes were interpreted. Even as more users joined the Oracle in early 2025, decision-making remained largely concentrated within that core group. In July 2025, Polymarket issued a public statement encouraging a more open and collaborative Oracle, one accessible to all users and not just longtime governance participants. Today, there are more accessible and centralized places for resolution discussion, along with closer integration between Polymarket and UMA. The Polymarket and UMA teams work together regularly, and Polymarket has become more open to suggestions and feedback on the resolution process, market rules, and related issues. Discussions also increasingly bring together Polymarket users, UMA voters, and other participants.
Oracle Inconsistency UMA resolutions are not always perfectly consistent across markets, especially when similar rule language is applied in different contexts. One example is the use of a consensus of credible reporting. In *2025 Chevrolet Detroit Grand Prix Winner*, the rules allowed either official IndyCar information or a consensus of credible reporting, but voters still preferred to wait for official confirmation. In *Who will win Dem nomination for NYC Mayor?*, a similar consensus clause was accepted without waiting for official publication. Both markets allowed credible reporting and still had some possibility that the result could change, yet they were treated differently. These differences can arise from the type of market, the perceived reliability of the reporting, the risk of later changes, or the particular voters involved. Past resolutions are therefore useful guides, but they do not guarantee that the same reasoning will be applied in every future dispute.
Finding Other Precedents This chapter does not cover every precedent. Relevant interpretations may also be found in past disputes or community discussions. Before proposing or disputing, you can search past dispute threads involving similar titles, rules, or resolution questions. You can also request a proposal and provide supporting evidence in the **#proposal-requests** channel on either the Polymarket or UMA Discord. For unfamiliar categories, consider starting with one proposal and waiting for it to resolve before proposing similar requests. Some precedents are not apparent from the rules alone. Placeholder requests are one example. **Placeholder Requests** In negative-risk markets such as *Presidential Election Winner 2028*, Polymarket may initially create markets for named candidates. These have corresponding UMA request titles, such as: * *Will JD Vance win the 2028 Presidential Election?* * *Will Gavin Newsom win the 2028 Presidential Election?* * *Will Alexandria Ocasio-Cortez win the 2028 Presidential Election?* Polymarket may also create placeholder requests, such as: * *Will Person A win the 2028 Presidential Election?* * *Will Person B win the 2028 Presidential Election?* * *Will Person C win the 2028 Presidential Election?* These placeholder requests reserve slots for additional outcomes and are not initially linked to an outcome on the Polymarket interface. For example, if *Presidential Election Winner 2028* launched without an outcome for Jimmy Donaldson, Polymarket could later add him as an outcome linked to the existing request *Will Person A win the 2028 Presidential Election?* The Polymarket interface would display his name, while the request would retain the "Person A" wording. Two issues can arise when proposing these requests: * **Proposing too early** — "No" proposals for unused placeholders submitted before another outcome has resolved to Yes have frequently been disputed, even when it appeared impossible for the placeholder to resolve to Yes. * **Failing to check the Polymarket interface** — A request that refers to "Person A" may already have been assigned to a named outcome, without this being apparent from the request title. Before proposing, check the Polymarket interface to confirm whether it has been assigned. If so, evaluate the request according to that outcome's rules and clarifications.