On April 7th, Trump announced a two-week ceasefire between the US and Iran.
On April 21st, the day before the ceasefire was set to expire, he issued a statement on Truth Social, announcing an indefinite extension of the ceasefire.

Subsequently, Reuters, AP, BBC, Al Jazeera, and The Wall Street Journal all reported the extension of the ceasefire. Iran’s foreign minister tweeted to acknowledge the decision to extend the ceasefire.

In the real world, the ceasefire has continued under strict lockdown conditions.

But on Polymarket, the market asking “Will the US-Iran ceasefire be extended before April 22nd” currently has a 0.1% probability of “Yes.”

In other words, the whole world knows the ceasefire has been extended, but the largest prediction market in the world believes it has not been extended.

Markets with contentious outcomes often see bets placed with expectations of huge profits at this time: some traders wager one dollar, or even hundreds of dollars, trying to create a myth of getting rich quickly.

Within the past 24 hours, an account bought $100k worth of “Yes” shares, with a potential payout exceeding $50 million.

Officially intervening to change the rules, has the ceasefire really been extended?

This controversy was embedded in the rules from the start.

Polymarket defines “ceasefire extension” as: both US and Iran must make clear, public statements, or there must be a “consensus of overwhelmingly credible media reports.” The US’s public statement came from Trump himself on Truth Social.

The problem lies with Iran. Iran’s official statement used the word “acknowledged,” not the rule-required “mutually agreed.”

It is this wording that caused the dispute over the $150 million trading volume in this market: “Yes” holders believe Trump’s statement plus the global media consensus constitute “overwhelming consensus”; “No” holders argue Iran did not directly confirm in its own name, so the condition is not met.

On April 24th, Polymarket’s official team directly intervened, adding a note to the market page: as of 23:59 on April 22nd, there is no condition that meets “Yes” for the ceasefire extension.

With official backing, market sentiment suddenly shifted, and the probability plummeted below 1%.

Following that, many skilled traders familiar with the rules and trading strategies began to act: under the premise of official qualification, buying “No” has become an almost zero-risk, high-return investment.

Among these, the third-largest “No” holder is an account called NotBakerMcKenzie, which wagered about $8.5 million. Baker McKenzie is a top global law firm based in Chicago, providing compliance legal services for prediction markets like Polymarket, with deep knowledge of oracle settlement mechanisms and platform rules.

Interpreting the rules as a law firm and placing real money bets seems like a declaration to all traders about the final settlement direction of this market.

However, the top “Yes” holder, Pedro, clearly disagrees: Polymarket’s official statement can only serve as a reference for settlement, and the final result is decided solely by votes from UMA’s decentralized oracle. As long as UMA token stakers vote “Yes,” what the official says doesn’t matter.

This is exactly what Pedro is betting on: compared to the potential $50 million payout, risking $100k on a possible oracle vote surprise is very cost-effective.

Valuable Oracle Attack Worth $50 Million

Pedro’s Polymarket account homepage has a link to a website he created for his own token—$pedros-coin. Although the site is filled with semi-finished architecture and rough design, the token’s rules are quite eye-catching.

$pedros-coin cannot be purchased through conventional meme rules; the only way to acquire it is through action: watch a live stream to earn 1 coin per hour, post content on social media to earn 20 coins each— all tightly bound to online dissemination.

The value of this token depends entirely on the probability of “Yes” in the “ceasefire extension” market. If the probability is 100%, each token is worth $1; if the final settlement is “No,” the token becomes worthless.

Pedro’s holdings in this market also serve as a perfect payout guarantee: he holds 50 million “Yes” shares, with potential payouts exceeding $50 million. Only if he wins can he get paid.

Putting these features together, the logic of this design becomes clear: Pedro uses $pedros-coin to bind the interests of hundreds of people to his “Yes” position, encouraging them to keep voicing opinions online, aiming to create enough public pressure before oracle voting to make as many stakers believe the market should settle as “Yes.”

From a narrative perspective, this mobilization mechanism has a strange Web3 spirit—Pedro invests over $100k of his own money, leading retail investors to unite through tokens earned by action, fighting against the large holders’ capital-driven finality, with their betting direction aligning with a real, existing ceasefire.

But in Pedro’s Discord channel, another layer emerges, making this matter less pure.

The Foolish Pedro and the Market-Manipulating Oracle Whale

On the morning of April 30th, a user named Euan posted a message in Pedro’s Discord: “As you can see, I own the richest UMA wallet. Willing to accept bribes to manipulate the vote to ‘Yes.’ DM open.”

Next to the message, two screenshots were attached: one showing 2.9 million UMA tokens held, and another of an account named borntoolate.eth.

The 2.9 million UMA tokens account for about 16.4% of the current total staked 17.71 million.

This single screenshot is already impressive, and for long-time Polymarket players, the name borntoolate carries weight comparable to that 16.4%.

In March 2025, the Polymarket market “Will Ukraine agree to Trump’s mineral resource deal before the end of March” was settled as “Yes” by oracle voting, despite ongoing negotiations and no formal signing.

This was the infamous Polymarket oracle attack, orchestrated by borntoolate. With relatively low overall voting participation, borntoolate used large holdings of UMA tokens to push the settlement result forcibly toward “Yes,” contrary to the facts.

The core assumption of UMA’s security model is “attack costs exceed attack benefits”—attackers need to buy enough UMA tokens to control voting, and this cost should surpass any gains from the attack. But the total market cap of UMA is only about $40 million.

We cannot verify whether Euan is truly borntoolate himself. But if this round of settlement results again flip, Pedro’s $100k wager could net over $50 million.

So far, the results of UMA oracle voting seem quite clear: among publicly available votes, “No” votes exceed 10.27 million, while “Yes” votes are only 25.

The only uncertainty in this round’s vote is the ownership of about 8.69 million UMA tokens whose votes are still private. If more than 2.33 million of these vote “No,” the market will be considered to have reached consensus, settling as “No.” If not, the vote is invalid, and the dispute continues to the next round—exactly what Pedro is waiting for.

As of writing, Pedro is still actively buying “Yes.”

Click to learn about the job openings at BlockBeats.

Join the official BlockBeats community:
Telegram Subscription Group: https://t.me/theblockbeats
Telegram Discussion Group: https://t.me/BlockBeats_App
Twitter Official Account: https://twitter.com/BlockBeatsAsia