Original | Odaily Planet Daily (@OdailyChina)

Author | Azuma (@azuma_eth)

Polymarket is once again embroiled in a controversy regarding fairness.

The incident originated from the prediction market "Will the U.S. invade Venezuela by…?" On January 4, Polymarket issued a clarification stating that "the previous action of the U.S. capturing Venezuelan President Maduro does not meet the definition of invasion." This statement caused a sharp decline in the YES shares for the prediction of a U.S. invasion by January 31, directly affecting the actual interests of many users.

• Odaily Note: The image shows the price trend of the YES shares for the January 31 prediction, with the turning point being the time when Polymarket issued the clarification.

This is not the first time Polymarket has faced similar controversies. Last year, we mentioned similar cases in the articles "Polymarket Faces Oracle Manipulation Attack, Large Players Can Use Voting Power to 'Reverse Black and White'?" and "Polymarket Again in Truth Controversy: What Zelensky Wears Will Determine the Ownership of $140 Million," and briefly analyzed Polymarket's result determination logic.

In the discussion of this incident, we found that many readers are aware that Polymarket relies on the oracle protocol UMA for its determinations, but they are unclear about how this process works. Therefore, Odaily has published this article to analyze its determination mechanism and explore the potential gray areas that may lead to disputes over results.

Predetermined Rules and Clarifications

First, any prediction market on Polymarket will have predetermined rules written at the time of its launch, which will clearly define the conditions for determining the outcome and the effective date, as well as how to judge in various unforeseen circumstances.

Taking the prediction market "Will the U.S. invade Venezuela by…?" as an example, the text below the Rules section is the predetermined rule for this market. The conditions for judgment and the effective date are as follows: if the U.S. launches a military offensive aimed at controlling any part of Venezuelan territory between November 3, 2025, and January 31, 2026 (11:59 PM Eastern Time), it will be judged as YES; otherwise, it will be judged as NO.

However, even with various unforeseen circumstances anticipated, sometimes developments can still be unexpected. For instance, in this incident, no one could have predicted that a president could be so suddenly captured by another power. Therefore, in very rare cases, for unforeseen circumstances that were not anticipated at the time of market creation, Polymarket will personally issue clarifications to further explain the rules—whether to clarify this step is not entirely decided unilaterally by Polymarket; users can actively request clarification in the #market-review channel on Polymarket Discord if they have concerns.

Sharp-eyed friends may have noticed that below the Rules section, there is a lighter font section labeled Additional context, with an updated date (the predetermined rules were published on December 18 of last year, while this content was updated on January 4). This is actually the content of Polymarket's clarification, which states: "This market involves U.S. military actions aimed at establishing control. President Trump stated he would 'manage' Venezuela while referring to ongoing negotiations with the Venezuelan government, but this statement alone is not sufficient to classify the 'capture and extraction' mission aimed at capturing Maduro as an invasion."

In simple terms, Polymarket does not consider the U.S. capture of Maduro to be defined as an invasion of Venezuela, and therefore does not support determining the outcome as YES based on this.

We will not discuss whether Polymarket's clarification is reasonable; what is more important to note is that the effective period of this prediction market (January 31) has not yet ended, meaning the final determination process has not yet begun. The reason for emphasizing this point is to remind that all current disputes essentially stem from ambiguities in the rules, unrelated to the determination phase; and to indicate that this dispute has not yet been resolved, and the current losses for users are actually unrealized losses—everything will need to wait for the final determination to conclude before a result is reached.

So how is the final determination process executed?

Determination Procedure: Results Are Proposed by People

For any prediction market on Polymarket, a result must be proposed by someone during the final determination process. Using the previous prediction market as an example, the proposal result window is actually located in the Propose resolution section below the Rules.

Of course, not everyone can casually propose results; UMA and Polymarket have designed two restrictions regarding economic incentives and whitelist requirements.

Economic incentives mean that proposing a result requires submitting a certain amount of USDC as a deposit (usually 750 USDC, some markets may be higher). After submission, it enters a questioning window (usually 2 hours); if no one questions it during this period, the result is deemed valid and will be used as the basis for the final determination of the prediction market, and it will not be changed again. The proposer can receive a certain reward (usually 5 USDC); conversely, it will enter a dispute phase, and the proposer faces the risk of losing their deposit (which will be explained in detail later). In simple terms, if someone proposes a result just to cause trouble, the risk will far outweigh the reward.

• Odaily Note: By clicking into the Propose resolution section of the prediction market, you can see the deposit requirements and reward amounts for proposing results.

The whitelist restriction means that Polymarket initially allowed anyone to propose results for determinations, but later, to improve determination efficiency, a whitelist maintained in collaboration with Risk Labs was introduced last August, and only whitelisted addresses are allowed to propose results. There are three ways to enter the whitelist: joining the Risk Labs team, joining the Polymarket team, or submitting more than 20 proposals with an accuracy rate exceeding 95% in the past three months—any address can check through this contract; initially, there were only 40 addresses, but the number has significantly increased since then.

Dispute Phase: Economic Interest Game

In the previous section, we mentioned that if the proposed result does not receive any objections during the questioning window, it will be deemed valid, which is the final outcome for the vast majority of prediction markets. However, in very rare cases, if there are objections, how should a determination be made?

First, it should be noted that, just like proposing a result, objections cannot be made casually—the objector must pay an equivalent amount of USDC as a deposit (usually still 750 USDC) to confront the proposer, meaning both parties must put up equal stakes for discussion. However, unlike the proposer, the objector does not need to provide a complete result; they only need to point out an error in the proposer's result.

Once an objection is confirmed, the UMA community will debate it, which generally lasts 24 to 48 hours (voting will take place the next day, with at least 24 hours reserved for discussion each time). Anyone wishing to provide evidence for the discussion can give their opinions in the #evidence-rationale and #voting-discussion channels on the UMA Discord server.

After the debate concludes, UMA token holders will vote on the matter (this process takes about another 48 hours), and the following four outcomes may occur:

• Proposer wins: The proposer retrieves their deposit, plus half of the objector's deposit as a reward. The objector loses their deposit.
• Objector wins: The objector retrieves their deposit, plus half of the proposer's deposit as a reward. The proposer loses their deposit.
• Too early: This result applies to proposals where the relevant event has not yet been settled, such as ongoing sports match results. The objector will receive a refund, plus half of the proposer's deposit as a reward. The proposer loses their deposit.
• Tie (50:50): The rarest situation, in which the objector will recover their deposit and receive half of the proposer's deposit as a reward. The proposer loses their deposit.

There are two points to note in the above voting.

First, in the four potential outcomes, the objector has three scenarios in which they can profit, while the proposer has only one—this is an intentional design by UMA, aimed at increasing the accuracy of proposed results through the imbalance of risks and rewards between both parties, as the objector only needs to point out one flaw to win, so the proposer must provide results that are as accurate and standard as possible.

The second point is that the governance voting power of UMA has absolute authority over the final outcome. In other words, the billion-dollar prediction market created by Polymarket is fundamentally supported by a protocol with a fully diluted valuation (FDV) of only 100 million dollars.

Exploring the Gray Areas

Combining the analysis of Polymarket's determination process with a review of historical real dispute cases, it is not difficult to find that there are certain gray areas that can lead to disputes, whether in the rule setting and clarification stages during the operation of the prediction market or in the final determination process.

First, in the rule setting and clarification stages, the essence of the ambiguity includes the fact that textual rules sometimes cannot cover real-world variables, and the same text can often be interpreted in different ways. For example, in last year's "Did Zelensky Wear a Suit?" incident, the rules did not explicitly mention whether "military-style suits count as suits." Although Polymarket clarified that "reliable reports have not confirmed whether Zelensky wore a suit," it did not explain what constitutes "reliable reports." Such ambiguities are bound to cause disputes.

If Polymarket, as the platform, could maintain neutrality, it might not provoke public outrage every time, but the situation is rarely that ideal. Polymarket's operating entity is located in the United States, which makes it difficult to achieve complete neutrality on all issues involving geopolitics due to the regulatory environment and political context it faces. For instance, in the case of "Will the U.S. invade Venezuela?", when it comes to military and diplomatic actions involving the U.S. itself, the interpretation of the rules tends to lean towards a more conservative "demilitarized expression." This is not unreasonable, but ultimately, it is the users who suffer losses.

As for the determination process, the source of ambiguity directly points to the potential for fraud in UMA voting. Although UMA has designed a reward and punishment mechanism to constrain proposal behavior and improve the accuracy of results, this mechanism can only constrain economic interests within its system. When external interests come into play, there theoretically remains a space for wrongdoing. This is not baseless suspicion; in last year's "Ukrainian Rare Earth" incident, large players in UMA manipulated voting power to reverse the truth, resulting in a misjudgment of 7 million dollars in bets.

The existence of these ambiguities is the root of frequent questioning of Polymarket's fairness and is a structural issue that the prediction market needs to address. In fact, any prediction market involving complex real-world events will inevitably face the following threefold dilemma—first, real events themselves often cannot be clearly dichotomized; geopolitical issues, military actions, and diplomatic games are inherently filled with gray areas; second, rules must be expressed in language, but language naturally has interpretive space; third, once a determination mechanism introduces human or governance participation, interest games become unavoidable.

From the user's perspective, you may need to realize early on—that in the prediction market, what you are betting on is not "what will happen in the world," but rather "how the rules will ultimately be interpreted."

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。