Two Polymarket wallets that didn’t exist before this week have staked roughly $62,000 on a single, audacious idea: that the prediction market will be forced to reverse itself — and pay out more than $13.8 million in the process. They don’t look like insiders. They look like something stranger: traders betting on a contradiction Polymarket wrote into its own rules.
At the center sits a deceptively simple question — did MicroStrategy sell any Bitcoin by May 31, 2026? — and a market worth tens of millions of dollars that still can’t agree on the answer, even though the company itself says yes.
KEY FACTS AT A GLANCE
- The market: Polymarket’s “MicroStrategy sells any Bitcoin by May 31, 2026?”
- What happened: Strategy sold 32 BTC (~$2.5M) between May 26 and 31, disclosed via an SEC 8-K on June 1
- The catch: the sale fell inside the window, but the disclosure landed about 8 hours after the deadline
- The status: disputed and headed to a UMA token-holder vote; “Yes” has fallen from ~81% to under 1%
- The whales: two new wallets staked ~$62K on “Yes” for a combined ~$13.8M potential payout
- As of June 3, 2026: still unresolved
The Bet on a Contradiction
The two wallets are near-mirror images. The first, “kahanetzadak,” was created just days ago, has traded exactly one market in its life, and holds nearly 10 million “Yes” shares bought for about $34,279. If the market resolves “Yes,” that position pays out roughly $9.96 million — a return near 290 times the stake. The second, “jezfan,” fits the same template: a brand-new, single-market wallet that put down about $27,972 for a potential $3.82 million, or roughly 137 times.
One detail reframes everything: both wallets bought their “Yes” shares on June 2 — after the dispute had already erupted and the price had cratered below a penny. These aren’t traders caught on the wrong side of a surprise ruling. They walked in with eyes open and bought the disputed lottery ticket at a discount. On-chain investigators have exposed genuine insider activity on Polymarket before, but neither of these wallets left a traceable funding trail, and neither has surfaced in the mainstream coverage, which has fixated on other traders.
So why would anyone pay real money for an outcome the market has priced at less than 1%? The answer is buried in what Strategy actually did — and, more importantly, when it told anyone.
What Strategy Actually Did
On June 1, Strategy — the Michael Saylor company formerly known as MicroStrategy — filed an SEC Form 8-K disclosing that it had sold 32 Bitcoin for about $2.5 million, at an average of $77,135 per coin. It was the company’s first Bitcoin sale since 2022, a symbolic crack in Saylor’s relentless “never sell” gospel — yet financially trivial, just 0.0038% of its 843,706-BTC hoard. The reason wasn’t distress or tax-loss harvesting (the coins were sold at a small gain); it was to fund dividends on the company’s preferred stock. Saylor had even telegraphed the move weeks earlier. Polymarket, for its part, has become a venue where Bitcoin bets move tens of millions of dollars a day, so a contested $50-million market drew a crowd almost instantly.
“We will probably sell some bitcoin to pay a dividend, just to inoculate the market and send the message that we did it.”
— Michael Saylor, on Strategy’s Q1 2026 earnings call
The crucial fact is the timing. The sale transactions ran from May 26 through May 31, and the filing itself timestamps the holdings “as of May 31, 4:00 p.m. ET” — comfortably inside the market’s window. But the 8-K that proved it didn’t reach the SEC’s wire until the morning of June 1, roughly eight hours after the market’s 11:59 p.m. ET cutoff. The event landed on time. The paperwork did not.
The Rule That Wasn’t There
Read the market’s own rule and the “Yes” case looks airtight. It resolves “Yes” “if MicroStrategy sells any of its Bitcoin by 11:59 PM ET” on May 31. Strategy sold Bitcoin by then. Case closed — or so thousands of traders assumed when they drove “Yes” to roughly 81%.
Then Polymarket posted a clarification to the oracle that settles the market, and the price collapsed to under a penny:
“No information from MSTR, on-chain data, or consensus of credible reporting confirmed that MicroStrategy sold Bitcoin within the market’s timeframe. Confirmation achieved outside of the market’s time frame does not qualify.”
— Polymarket resolution bulletin
Critics call this a retroactive rule change, and they have a point. Nowhere in UMA’s resolution rulebook — the standard that actually governs these markets — is there a written requirement that an event be publicly confirmed by the deadline, only that it happen by then. Polymarket effectively bolted a “disclosed-by-the-deadline” clause onto a rule that asked about the sale itself. The two readings split the entire dispute:
TWO READINGS OF THE SAME RULE
The “Yes” case — did it happen?
- The rule asks whether Strategy sold “by 11:59 PM ET” — and it did, on May 31.
- The 8-K timestamps the sale “as of May 31, 4:00 p.m. ET,” inside the window.
- When wording is ambiguous, it should favor the trader who took the plain-language bet.
The “No” case — was it knowable?
- No filing or credible report confirmed the sale before the deadline.
- Polymarket says confirmation after the window “does not qualify.”
- A market should resolve on what was provable at the cutoff, not in hindsight.
Same Filing, Opposite Verdicts
If the “confirmation timing” argument still feels defensible, here’s the fact that turns it inside out. Strategy’s single June 1 filing didn’t only feed the May 31 market — it also settled the sibling markets asking the very same question for June 30 and December 31, 2026. Both of those resolved cleanly to “Yes.” The identical sale, proven by the identical document, counts for two of the three markets and is being rejected only for the one where the disclosure happened to arrive a few hours late.
How Polymarket Actually Decides
Polymarket doesn’t resolve its own markets. It outsources the verdict to UMA, a crypto “optimistic oracle” that runs in three stages:
STEP 1: PROPOSE
Someone posts a ~$750 bond proposing “Yes” or “No.” If no one challenges it, the result finalizes in about two hours.
STEP 2: DISPUTE
Anyone can challenge with a matching bond. Here, two “No” proposals were filed — and both were disputed.
STEP 3: TOKEN VOTE
UMA token-holders settle it through a 48-to-96-hour vote. That’s where this market sits right now.
This market has already torn through the first two stages — two separate “No” proposals were filed, and both were disputed — so it now heads to a token-holder vote. That is where the real risk lives. Roughly nine wallets control about half of UMA’s voting power, and voters can hold positions in the very markets they are adjudicating. Precedent offers little comfort, because it points in every direction at once:
| Past dispute | How it resolved | Why it matters here |
|---|---|---|
| Zelenskyy “wears a suit” (~$237M) | Resolved “No” | UMA decided broad “reporting consensus” wasn’t enough — strict wording won. |
| Ukraine minerals deal | Resolved “Yes” (contested) | A whale’s votes swung a result many observers called flat-out wrong. |
| Barron Trump token market | Polymarket overrode UMA | Proof the platform can contradict its own oracle and refund traders. |
In other words, the outcome may turn less on what the rule says than on who shows up to vote — which is exactly why regulators are already circling prediction markets and why integrity questions keep dogging the sector.
So, Do They Know Something?
Back to our two whales. The honest answer is: probably not in the way the question implies. By the time kahanetzadak and jezfan bought in on June 2, the sale was already public — there was no secret left to know. What they purchased wasn’t inside information; it was a cheap, asymmetric wager that Polymarket’s own contradiction gets resolved in favor of the plain-language rule, or that the swelling outrage forces a “Yes.” At under a penny a share, you risk a little to make 100-to-290 times your money. You don’t need a tip. You need the contradiction to be noticed.
That’s the uncomfortable lesson of this episode: over a long enough horizon, the most valuable edge in a prediction market may not be knowing the future — it may be reading the fine print more closely than the house that wrote it. If you trade these markets yourself, our crypto tax calculator can help you keep your wins and losses straight at filing time.
DEVELOPING STORY
As of June 3, 2026, this market remains unresolved and in UMA’s dispute queue. A token-holder vote — typically 48 to 96 hours — will determine the final outcome. The positions, prices, and odds described here reflect the state of the market at the time of writing.
KEY TAKEAWAYS
- The sale was real but tiny — 32 BTC (~$2.5M), Strategy’s first sale since 2022, to fund preferred-stock dividends.
- Timing is the entire fight — the sale happened by May 31, but the SEC disclosure didn’t arrive until June 1.
- The “disclosed-by-deadline” rule isn’t written — UMA’s rulebook requires the event to happen by the date, not to be confirmed by it.
- The same filing settled the June and December markets “Yes” — which makes the May 31 “No” look incoherent.
- The whales aren’t insiders — they bought after the collapse, arbitraging a contradiction for a 100-to-290× payout.
- A handful of UMA voters will decide it — not Polymarket, and not the rule’s plain text.
Sources
- Form 8-K — Bitcoin sale disclosure — U.S. Securities and Exchange Commission
- MicroStrategy sells any Bitcoin by May 31, 2026? — market page — Polymarket
- Trader profile — “kahanetzadak” — Polymarket
- Trader profile — “jezfan” — Polymarket
- How markets are resolved — Polymarket Documentation
- How UMA’s Optimistic Oracle works — UMA Protocol
- UMIP-107 — YES_OR_NO_QUERY resolution rules — UMA Improvement Proposals
