Why Is Polymarket a Scam? Our Investigation and Experience with the Platform

Is Polymarket a scam or just a dangerous tool? In this article, I use recent cases to explain why the platform’s market-resolution system is designed in a way that leaves the average trader losing almost every time.

From May 9 to 11, 2026, a three-day “ceasefire” was in effect between Ukraine and russia, at the latter’s repeated request, so the kremlin could hold its May 9 parade in moscow. On May 12, one day after it ended, the russians launched more than 200 drones at Kyiv, Dnipro, and Kharkiv. The war continued as before. Yet on Polymarket, the “Russia x Ukraine ceasefire by May 31, 2026” market, with $141 million in volume, resolved to “Yes”. This happened even though the market’s own rules explicitly excluded humanitarian pauses and any informal agreements.

This is a fundamental problem with the way the platform works. In this article, I explain why Polymarket, with $26 billion in quarterly volume, handed market resolution to the small UMA protocol with a market capitalization of $95 million. I also explain how whales make money from this, why there is nothing you can do about it, and why Ukraine blocked Polymarket back in December 2025.

To put it briefly: I do not recommend treating Polymarket as a serious tool. Details below.

Polymarket: its scale and why it matters

Polymarket is currently the world’s largest decentralized prediction market. A few figures to show the scale:

• $21.5 billion in volume in 2025 and $26.2 billion in the first quarter of 2026 alone.
• $10.57 billion in March 2026 — the first month in which volume exceeded $10 billion.
• Single-day record: $425 million (February 2026), higher than on the day of Trump’s election in 2024.
• $8 billion valuation after a $2 billion investment from ICE/NYSE (October 2025).
• 95 million transactions in 2025 and approximately 600,000 active traders.

Polymarket operates on the Polygon blockchain, and all bets are placed in USDC. One point is critical to this discussion: Polymarket is a non-custodial service. The platform does not hold user funds and does not make decisions about event resolution. This is a deliberate architecture. It emerged after the CFTC issued Polymarket a cease-and-desist order in January 2022 and forced it to pay a $1.4 million fine for operating as an illegal derivatives exchange. The team redesigned the system so that it formally controls nothing: trading runs through smart contracts, while outcomes are decided through the external UMA oracle. Legally, this means: “we are not an exchange; we are just an interface”.

For users, this means something else: when you bet on “Yes”, Polymarket is not the project deciding whether you won. Another project does that. Now let us look at how it works.

What UMA is and where this oracle came from

Before examining the mechanics, we need some context. UMA (Universal Market Access) is a separate project founded in 2018 by Hart Lambur, a former Goldman Sachs trader. UMA was originally built as a platform for synthetic assets and derivatives, where users could create tokenized versions of Apple shares or the S&P 500 index. The synthetic-assets idea did not take off (Synthetix did it better and earlier), so UMA pivoted to the main product that remained alive: an oracle for real-world data.

Today, UMA Optimistic Oracle serves more than Polymarket. Its clients include Across (a cross-chain bridge), Sherlock (smart-contract insurance), and other DeFi projects. But Polymarket is its largest and most controversial client. Looking at the statistics, the overwhelming majority of disputes in UMA come from Polymarket markets, because this is the only use case in which outcomes are subjective and politically charged.

Polymarket and UMA are legally two different companies, but they are connected. In 2024, Polymarket and UMA announced a partnership with EigenLayer to develop a “next-generation” oracle based on restaking. Polymarket officially calls UMA its “trusted oracle”, while UMA officially calls Polymarket its “largest client”. In other words, they are formally independent partners, but without Polymarket, UMA would lose much of its reason to exist. This creates another hidden asymmetry of interests, which I will return to below.

How market resolution works through UMA Optimistic Oracle

UMA’s architecture is called Optimistic Oracle (OO), and it has two levels.

Level 1: optimistic assertion. When a market closes, anyone can propose an outcome (for example, “Yes” to the question “Was there a ceasefire?”). To do this, they must post a bond, usually $750 USDC on Polymarket. A two-hour challenge period then begins. If nobody disputes the proposal, the outcome is considered correct, and the proposer receives the bond back plus a small reward. According to UMA statistics, only around 1.5% of proposals are disputed. This is the fast path, and it covers the overwhelming majority of markets.

Level 2: Data Verification Mechanism (DVM). If someone disputes the proposal (also posting $750 USDC), voting begins. This is where the interesting details start.

Only UMA token holders who have staked their tokens can vote. Voting takes place in two stages:

1. Commit phase (24 hours): voters publish a hash of their vote. The actual vote is hidden so that nobody can copy another participant’s decision.

2. Reveal phase (24 hours): voters reveal their votes together with the salt used to generate the hash.

Voting power is proportional to the number of staked UMA tokens. A supermajority of more than 65% is required to make a decision. Those who vote with the majority receive rewards, while those in the minority are slashed (around 0.1% of their stake). The slashing is minor, but it creates a strong psychological effect: smaller voters are afraid of being wrong and vote “with the crowd” rather than according to reality.

Once voting ends, the result is returned to the Polymarket smart contract, the market closes, and payouts go to the winners. That is it. As Polymarket’s documentation states directly: “Once finalized by UMA, outcomes are immutable”. Polymarket cannot change the result, even if it wants to.

Now for the key question: who are these voters?

UMA’s structural problem: the numbers do not add up

This is where the circus begins. Let us look at the numbers.

Let that sink in. The platform’s monthly volume is more than 100 times the market capitalization of the project that resolves all disputes on it. When the Zelenskyy suit controversy erupted in July 2025 (more on that below), a single market with $242 million in volume was decided by people holding tokens worth $95 million. Literally, it can be cheaper to buy a share of the voters than to lose the market.

In March 2025, analyst Folke Hermansen examined voting concentration and found that several of the largest whales control more than 50% of active voting power in an average vote. Data from high-profile cases shows that the top 10 stakers account for around 30% of average voting activity — the fate of most markets is decided by a handful of addresses. For a system marketed as decentralized, this is effectively an oligopoly disguised with language about “truth-seeking”.

Separately, these whales often trade on Polymarket. They take a $1 million position, then vote in UMA in favor of their own outcome. The conflict of interest is built directly into the system’s architecture. UMA argues that honesty is more profitable because discrediting the system would cause the token price to fall. That is a reasonable theory, but it breaks down in specific cases where a one-off profit on a $7 million market exceeds the expected losses from a falling token price.

And one more thing: only UMA holders can vote. If you enter Polymarket, put down $50,000, and are clearly right on the facts, you still have no influence over the resolution. Completely different people vote — people who may not trade on the market at all or, more often, trade on the opposite side. This is the fundamental asymmetry behind the entire problem.

Recent case: Russia x Ukraine ceasefire by May 31, 2026 (~$141 million, May 2026)

The most recent and most revealing case. Let us break it down in detail.

Background. Polymarket had been running a series of markets titled “Russia x Ukraine ceasefire by 2026” for some time. The largest had deadlines of May 31, 2026, and June 30, 2026. The market rules were very clear, and this is not my summary but a direct quote:

“This market will resolve to “Yes” if there is an official ceasefire agreement, defined as a publicly announced and mutually agreed halt in military engagement, between Russia and Ukraine. Only ceasefires which constitute a general pause in the conflict will qualify. Ceasefires which only apply to energy infrastructure, the Black Sea, or other similar agreements will not qualify. Any form of informal agreement will not be considered an official ceasefire. Humanitarian pauses will not count toward the resolution of this market.”

The key point: humanitarian pauses do not count, informal agreements do not count either, and the ceasefire must be a “general pause in the conflict” — an actual halt to military activity across the entire front.

What happened. On May 8, 2026, Trump announced a three-day ceasefire for May 9–11, timed to the May 9 parade in moscow. Both sides described it as a short-term pause, and a 1,000-for-1,000 prisoner exchange was also announced. There was no “general halt to military activity”: it was a symbolic temporary pause.

Does this qualify as a “humanitarian pause”, which the rules explicitly exclude? Yes. Is it an “informal agreement” without an official treaty? Also yes. Is it a “general pause in the conflict”? Clearly not: on May 12, one day after the “ceasefire” ended, the russians launched more than 200 drones at Ukrainian cities. The war continued without any pause.

How it was resolved. The “Russia x Ukraine ceasefire by May 31, 2026” market, with approximately $141 million in volume, resolved to “Yes”. Trump announced the ceasefire on social media, and the Polymarket team hid the markets, allegedly for technical reasons. While the markets were hidden, someone proposed a “Yes” outcome in UMA. Because the markets were inaccessible on Polymarket, traders did not see it, and within hours all “Russia x Ukraine ceasefire by 2026” markets resolved to “Yes”. UMA and the Polymarket team did nothing: the proposal passed, and nobody disputed it. As a result, traders who followed the rules, bet on “No” because the war had not actually ended, and assumed the rules meant something were left with losses.

They did not. Whales holding “Yes” positions pushed the outcome through UMA, and Polymarket did nothing. A reminder: “Once finalized by UMA, outcomes are immutable”. Polymarket officially has no authority to reverse the result.

If you open this market now, the comments are full of complaints posted almost every day: endless “refund?” and “scam”. There will be no refund. It is the same pattern we will see below with the Zelenskyy suit and Ukraine mineral deal cases: the Polymarket team says “this is not a market failure”, “we are working with UMA on improvements” — and returns nothing.

Why this is systemic. This case exposes the core problem. It does not matter how clearly the rules are written. If UMA voters vote against those rules, the UMA decision wins. In other words, the market rules shown to you before you place a bet have no legal authority. The only thing with real authority is a vote by holders of 20 million staked tokens concentrated in the hands of a few people. And those people can vote in their own interest because they are also betting on the market.

Voting dynamics. I reviewed the comments and discussions in the UMA Discord channels #evidence-rationale and #voting-discussion, where disputes are theoretically discussed. There were attempts to challenge the outcome on this market, but they did not cover anything close to the full volume of open positions. Who will post a $750 USDC bond and risk losing it to protect someone else’s position? The dispute therefore lost, and the original “Yes” proposal was accepted. The result was a situation in which formally clear market rules diverged from the actual resolution, and none of the “No” traders received compensation.

Here is a representative reaction from Twitter. The Ukraine Battle Map account wrote: “Polymarket Traders who put $14.5 million worth of bets on a ceasefire between Russia and Ukraine are complaining after the market resolved to YES due to the 3-day ceasefire for May 9th parade. Even though the war will continue after it’s over”. This was not a fringe thread. It was a public controversy seen by millions.

This is not an isolated case: other manipulations

Russia x Ukraine ceasefire is the hottest case, but it is far from the only one. Let us briefly look at other examples to confirm that this is a pattern, not an accident.

Ukraine mineral deal with Trump (March 2025, $7 million)

Market: “Will Ukraine agree to a minerals deal with Trump before April?”. Rules: official confirmation from the governments of the United States and Ukraine.

March 24–25, 2025 Trump publicly said he “expected the deal to be signed soon”. There was no official confirmation, and no deal had been signed. Yet the market price for “Yes” jumped from 9% to 100% in 24 hours. How? An anonymous whale using the address “BornTooLate.eth”, who had accumulated 1.3 million UMA and became one of the top five voters, plus another whale with 5 million UMA distributed across three wallets and accounting for 25% of the total weight, pushed through a “Yes” resolution.

A pool of more than $7 million was affected, and traders who bet on “No” lost everything. The crypto community called it “the most controversial incident in the history of the service”. The Polymarket team wrote in Discord that this was “not a market failure” and there would be no refund: “We’re committed to building the future of prediction markets… we will build up systems”. The system was never built. Two months later, the next controversy arrived.

Zelenskyy suit ($237 million, July 2025)

The highest-profile case. Market: “Will Zelenskyy be photographed wearing a suit between May 22 and June 30, 2025?”. The resolution was supposed to depend on the “consensus of credible reporting”.

On June 24, Zelenskyy arrived at the NATO summit in The Hague wearing a black blazer, matching trousers, and a shirt. More than 40 global media outlets (Reuters, AP, BBC, Wired, and others) described it as a suit. Menswear expert Derek Guy told Wired: “This falls under the technical definition of a suit”. Polymarket Intel itself (a community account associated with the platform) tweeted “President Zelenskyy in a suit last night” on June 25. It then quietly changed its bio to “community ran”.

On July 1, UMA voted that it was “not a suit”, and the result was “No”. Why? Because there was “insufficient credible reporting consensus”. Yet 40 headlines from major outlets obviously constitute a consensus. There had been a similar market in May, which also resolved to “No”, and UMA voters cited that precedent.

A few figures for context:

• Market volume: $237–242 million.
• UMA’s total market capitalization at the time: approximately $95 million.
• Dispute duration: 9 days with several rounds of challenges.

Menswear influencer Derek Guy separately bet $3.6 million on “No”, apparently to earn a potential $72,000. An anonymous X user, Atlantislq, wrote: “This is the biggest scam in PM history. Despite all the evidence, despite dozens of articles from major media outlets that are more than sufficient for consensus, they still sided with the UMA manipulators and gave them all the money”.

Once again, Polymarket did nothing. There was no refund. UMA co-founder Hart Lambur told Wired: “There is no evidence of UMA manipulation”. Formally, that is true: there is no direct evidence of collusion. But there is a structural reality: 10 people hold 30% of the votes, and when they want one outcome, that outcome wins.

Fort Knox gold, Israel-Lebanon, US-Iran ceasefire, and others

These are only three examples among many that appear regularly:

• Fort Knox gold (March 2025, $3.5 million): a market on whether gold had disappeared from the US Fort Knox depository resolved to “No” under questionable circumstances. In his thread, Folke Hermansen claimed that during the two-hour challenge period, the dispute button simply “disappeared” from the interface, making it impossible to challenge the proposal.
• Israel x Hezbollah ceasefire: the market was renamed “Israel x Lebanon ceasefire” while trading was already underway, changing the resolution conditions. Users had bet on “Hezbollah”, while the market was resolved based on “Lebanon” — fundamentally different entities.
• US x Iran ceasefire (April 2026): a trader with $4,000 on “No” lost the position because the market resolved to “Yes” based on a temporary pause that should not have counted under the rules.
• Barron Trump meme coin: the oracle had to decide whether Trump’s son was involved in a pump-and-dump scheme, even though the alleged involvement could not be verified through public sources.
• OceanGate submarine “found”: a disputed market on whether the OceanGate submarine wreckage had been found, with disagreement over what should count as “found”.
• Venezuela elections: a controversial resolution of the Venezuelan election, in which one side formally declared a winner while alternative data suggested otherwise.

Polymarket’s Trustpilot rating is 1.3/5 based on 519 reviews. Most complaints fall into the same category: “the rules are not enforced”, “whales decide, not reality”, “whales buy UMA votes and change the outcome”. You could say that Trustpilot is generally a place where people complain. But when the complaints are this consistent and match what independent analysts are writing, they cannot be ignored.

Folke Hermansen’s analysis: why the system breaks

In March 2025, independent analyst Folke Hermansen published a detailed thread about manipulation on Polymarket. It is worth examining separately. His key claims:

1. Concentration of power: several of the largest whales control more than 50% of active voting power, while the top 10 stakers account for around 30% of average voting activity. This is confirmed on-chain concentration, not a theoretical risk.

2. Insiders write the rules: according to Hermansen, insiders write market rules, take positions, and then coordinate with verifiers who approve proposals. “Polymarket is revealing itself to be a totally fraudulent platform. Insiders write rules, place bets, and coordinate with verifiers to rig markets and scam their own customers for millions daily”, he wrote directly.

3. Slashing fear creates herding: “UMA, in theory, is a neutral third-party blockchain protocol which incentivizes truth-seeking. In reality, it incentivizes crowding towards whatever other people are voting for”. In other words, smaller voters fear slashing and vote with the majority even when the majority is clearly wrong. As a result, more than 95% of votes end “almost unanimously”, because nobody wants to be in the minority.

4. Bond barriers: the $750 USDC bond required to dispute a proposal is a barrier for ordinary users. Insiders holding millions can easily post several bonds. An ordinary trader with a $100 position will not risk $750 to challenge someone else’s proposal.

5. Anonymity protects manipulators: UMA voting is anonymous (a commit-reveal scheme), and this is not a bug but a feature. Nobody knows how others voted until the votes are revealed. But it also means that tracking who is pushing a specific outcome is difficult. If you want to determine that 25% of votes came through three affiliated wallets, you need to conduct an on-chain analysis after the fact, and very few people do that.

The UMA team publicly rejects allegations of manipulation. UMA co-founder Hart Lambur told Wired in July 2025: “There is no evidence of UMA manipulation”. Formally, that is true. There is no direct evidence of collusion between insiders because such evidence is nearly impossible to obtain: all votes are anonymous. But the structural reality remains: 10 people hold 30% of the votes in an average vote, and that concentration alone makes the system vulnerable, regardless of whether anyone acts maliciously on a particular day.

How whales profit from manipulation: the economics of the scam

To understand why this is not an accident, we need to examine the economics of manipulation from a whale’s perspective. Let us consider a hypothetical but realistic scheme using the Russia-Ukraine ceasefire market as an example.

A whale with 5 million UMA tokens (worth approximately $25 million at the current price) looks at a market with around $141 million in volume. The price of “No” falls to 5–10% during the panic after Trump’s announcement. The whale buys $1 million worth of “No” positions for $50,000–100,000 because the price is low. When the time comes to resolve the market, the whale simply votes with its stake in favor of “No”, convinces a few smaller voters to do the same (for example, through the UMA Discord, where the discussion theoretically takes place), and collects a clean $1 million because “No” wins.

There are two elegant parts to this scheme. First, the risks are minimal: a $750 USDC bond and potential slashing of 0.1% of the stake ($25,000 in our example) against a potential $1 million payout. That is a 40x asymmetric bet with limited downside. Second, the manipulation is not technically illegal because nobody violates UMA rules. Everyone acts within a protocol that allows large stakers to vote however they want.

Looking at the history, we see the same pattern again and again. In the mineral deal case, a whale bought “Yes” positions at a low price (8–9%), then pushed “Yes” through UMA and captured the difference. In the Zelenskyy suit case, as far as we can reconstruct, whales bought “No” positions cheaply when most people believed there was a suit and then voted for “No”. In both cases, the whale’s behavior is economically rational, and it is possible precisely because UMA’s architecture allows market participants to vote on markets in which they hold positions.

For context: compared with traditional financial markets, this is similar to allowing major shareholders of a company to vote on how its quarterly report should be interpreted. Traditional markets impose liability for that kind of conflict of interest. In prediction markets, it is legal under the label of “decentralization”.

Why Polymarket users are powerless against UMA

The structural problem is that a Polymarket user has no influence over UMA. They are different projects. Even if you deposit $100 million on Polymarket, a UMA whale with a $1 million stake has more power and authority than you do.

Imagine this. You enter Polymarket, read the market rules (the same clear rules about a ceasefire), and bet $10,000 on “No”. The market closes and, contrary to the rules, resolves to “Yes”. You lose the entire $10,000. What can you do?

1. Challenge the proposal during the two-hour challenge period? You can, by posting $750 USDC. But you will most likely lose the vote and the bond as well, because UMA stakers with more tokens will say you are wrong. And unless you spend all day in the UMA Discord and monitor the market 24/7, you may not even see the moment when the proposal appears.

2. Vote in the DVM? You cannot. You need to own staked UMA tokens. If all you have is USDC on Polymarket, you have no voting power.

3. File a complaint with Polymarket? You can, but the outcome is predictable: the Polymarket team ignores your complaints, says “this is not a market failure”, “we do not have the authority to change UMA’s decision”, and “thank you for your feedback”.

4. Contact a regulator? Polymarket is non-custodial, legally “not an exchange”, and operates offshore. There is nowhere to file a complaint with a Ukrainian regulator, especially considering that Polymarket is officially banned in Ukraine (more on this below).

So even if you have one million dollars on the platform, your actual voting power in the resolution system is zero. Decisions about your money are made in a system in which you are not a participant. You simply contribute to a treasury from which funds are distributed based on other people’s decisions.

For comparison, Kalshi (Polymarket’s main competitor in the United States) resolves markets itself, with external audits and clear appeal procedures. If you believe Kalshi made a mistake, you can formally challenge the decision and have it reviewed. Polymarket has no such mechanism at all: UMA decisions are final.

Why Polymarket changes nothing: this is not a bug, but the architecture

This is where my expert opinion begins. I do not believe that the Polymarket team is unaware of the problem. It is discussed on Twitter every two or three months, every major controversy generates hundreds of publications in crypto media, and Atlantislq and Hermansen have produced very precise analyses. The team is not blind.

So why do they change nothing? Because the “non-custodial + external oracle” architecture is their legal shield.

Let us look at the timeline. In January 2022, the CFTC issued Polymarket a cease-and-desist order and forced it to pay a $1.4 million fine for operating as an illegal derivatives exchange. The platform stopped serving users in the United States and moved offshore. Then, when the CFTC granted Polymarket a no-action letter in January 2026 and allowed it to return to the US market, ICE/NYSE invested $2 billion at an $8 billion valuation. In January 2026 (announced on January 7), Polymarket signed an exclusive partnership with Dow Jones. Polymarket forecasts are now displayed on the Wall Street Journal, MarketWatch, Barron’s, and Investor’s Business Daily.

In other words, the platform gained legitimacy as a “source of truth”. But that legitimacy depends on the fact that Polymarket formally makes no decisions. If it begins intervening in resolutions, reversing UMA decisions, and paying compensation, it becomes a conventional exchange with all the consequences: KYC everywhere, a full regulatory burden, and responsibility for every decision. For now, the company can say “it was not us, it was UMA”, and legally, that works.

Polymarket is developing a new oracle with EigenLayer (using restaking) and UMA. It was announced in 2024, but there has still been no actual launch. The platform FAQ even contains the phrase: “We anticipate rolling out a new rewards and oracle-resolution system later this year”. “Later this year” is already running through its second “this year”.

My assessment: they know about the problem, but changing the architecture is not in their interest. The current architecture works: $26 billion per quarter, an $8 billion valuation, a Dow Jones partnership, and investment from NYSE. As long as the money keeps flowing, the game is worth it. If UMA breaks individual markets, that is a user problem. The team is paid for volume, not for fair markets.

Could this change? Yes. If the CFTC or another major regulator looks at the next $200+ million controversy and decides that it constitutes market manipulation, Polymarket will have a very unpleasant conversation. Or major traders may begin moving to Kalshi, where this specific problem does not exist. So far, neither has happened.

Ukraine: blocked since December 10, 2025

For a Ukrainian audience, this is important context. Polymarket has been officially blocked in Ukraine since December 10, 2025.

Details: The National Commission for the State Regulation of Electronic Communications (NCEC) adopted Resolution No. 695, requiring internet service providers in Ukraine to restrict access to polymarket.com. The reason was that Polymarket provides gambling services without a Ukrainian license. The initiative came from PlayCity, Ukraine’s gambling regulator, which had already blocked more than 4,500 illegal gambling websites.

The justification also noted that Polymarket allowed bets on events related to the russian invasion. By December 2025, approximately 240 markets related to the war in Ukraine had already closed on the platform, with total volume exceeding $270 million. Another 120 active markets had volume of more than $140 million. The platform used data from the Ukrainian OSINT project DeepState without permission, while the Institute for the Study of War (ISW) criticized the use of changes on its map as the basis for bets on the occupation of specific Ukrainian cities.

Implementation of the block is inconsistent. Some Ukrainian internet providers comply with the Resolution, while others do not. Polymarket can still be accessed through a VPN, but this violates the platform’s Terms of Service (Section 2.1.4), and Polymarket may freeze an account if it detects VPN use. Technically, its geodetection goes beyond a simple IP check. If your account is blocked for VPN use, funds may remain frozen indefinitely during a “compliance review”. Documented cases of such freezes exist.

If you still want to use prediction markets from Ukraine, there are alternatives that are not directly blocked in the country, but each has limitations:

• Kalshi: a regulated US alternative to Polymarket with a CFTC license. Technically unavailable from Ukraine because it requires a US SSN for KYC. In other words, you cannot sign up without US citizenship or residency. Kalshi is also heavily focused on sports (85% of volume), while political and geopolitical markets are a minority.
• Manifold Markets: play-money markets. There is no real money, only an internal currency. No geographical restrictions, and it works from anywhere. Useful if you want to train your intuition about events, but it is not an investment tool.
• Augur, Gnosis Conditional Tokens: fully decentralized prediction markets operating through DEXs. Accessible through a VPN, with low liquidity and a difficult user experience, but they formally work.
• PredictIt: another US service, also with restrictions for Ukraine and position limits ($850 per market). In 2026, PredictIt is also under regulatory pressure.

Honestly, there are few practical options for serious real-money prediction-market betting from Ukraine today. Polymarket is formally accessible through a VPN, but that is a gray area with a risk of losing access to the account, in addition to all the resolution problems described above. If you still use it through a VPN, the minimum precautions are: do not keep large amounts on the platform, withdraw immediately after winning, and avoid subjective political markets.

Conclusion

I do not recommend using Polymarket as a serious betting tool. Here is why, briefly.

The architectural problem has not been fixed and is not being fixed. Market resolution has been delegated to the external UMA oracle, where voting power is concentrated among several whales who often hold positions on the same markets. The Polymarket rules shown before you place a bet have no legal authority over UMA voting. I have shown how this works in practice using three recent cases: mineral deal, Zelenskyy suit, and Russia-Ukraine ceasefire.

The Polymarket team knows about this. It would be impossible not to know when a new $7–237 million controversy spreads across crypto Twitter every two or three months. But changing the architecture is not in the team’s interest: it generates $26 billion in quarterly volume and provides legitimacy as a “source of truth” through Dow Jones, ICE/NYSE, and the CFTC. As long as the money keeps flowing, the system will remain unchanged.

Everything could change. If the CFTC or another major regulator looks at the next $200+ million controversy and classifies it as market manipulation, Polymarket will have a serious problem. If the team finally launches the new oracle with EigenLayer as promised, something may improve. If major traders begin leaving for Kalshi and Polymarket volume starts falling, the team may finally act. But so far, none of this has happened.

This topic is trending, and you will see plenty of posts saying “Polymarket is a scam”, “do not use it”, and so on. Most are superficial. The real problem is that the platform’s architecture is structurally incapable of providing fair resolution for markets. For objective markets (will Bitcoin exceed $100k? who will win the match?), it works normally because there is nothing to interpret. But the moment a question requires interpretation (“is this a suit?”, “is this a ceasefire?”), the system breaks in favor of UMA holders.

In my view, this platform is unsafe for an ordinary trader. Even in markets where the outcome seems 100% obvious, UMA’s influence can produce a different result.

UMA whales can manipulate markets very easily and leave ordinary traders without liquidity.

If you still want to bet on Polymarket, my advice is to choose only the most objective markets with clear binary criteria — sports, prices, and similar topics. Avoid geopolitics, elections, and anything that requires interpretation of events. Do not bet a large amount on a single market, because if the market is rigged, you will not get anything back. And remember that the platform is officially blocked in Ukraine, so even if everything goes well, you are still operating in a gray area.

Using Polymarket today is at your own risk in the most literal legal and financial sense. I do not recommend it, but the choice is yours.