The War-Betting Investigation Has a Serious Evidence Problem

60 Minutes has identified a genuine danger: people with privileged military information can turn that access into money, and prediction markets can create incentives to compromise military security and even threaten journalists. But its most dramatic claims deserve considerably more scrutiny than the broadcast gives them.

The central problem is the distinction between suspicious trading and proven insider trading. A remarkable winning record, a cluster of connected accounts or a perfectly timed financial transaction can justify an investigation. None, by itself, identifies the person responsible or establishes what that person knew.

The 98% figure is not a 98% win rate

The most dramatic statistic in the broadcast does not mean what most viewers would understand it to mean.

On September 22, blockchain analytics company Bitquery independently reconstructed the nine accounts' transactions from Polymarket's public blockchain records. It confirmed their combined profits of approximately $2.477 million, almost exactly matching Bubblemaps' reported $2.474 million. But it found that only 86 of 101 settled positions won: an 85% success rate by number of bets. The reported 98.2% measures the proportion of money associated with winning positions, because the losing positions were comparatively small. Even the larger military bets, above $1,000, won 75 out of 83 times, or 90%. Bitquery's reconstruction.

This is not an insignificant distinction. An 85% success rate is still extraordinary for wagers frequently priced as unlikely outcomes. But calling a money-weighted result a 98% win rate makes the performance sound almost infallible. The broadcast never explains the difference.

The accounts also lost bets on earlier proposed US strike dates in February before correctly predicting February 28. That matters because the story's lottery-winning analogy obscures the actual sequence of successes and failures.

The $2.4 million is real. The 98% headline needs qualification.

Nine accounts do not automatically make one insider network

The more important discovery in Bitquery's investigation concerns the word connected.

Bubblemaps described the nine accounts as a connected group. Bitquery confirmed their profits and found striking similarities: the accounts used the same cryptocurrency bridge, traded in the same markets during similar periods and withdrew funds through similar routes.

But Bitquery could not establish common ownership.

It found no direct transfers between any two of the nine accounts. Five accounts sent money into three shared Binance deposit addresses, but those addresses had also received funds from dozens of unrelated senders. They could belong to financial intermediaries serving multiple customers.

Bitquery explicitly concluded that the financial connections were insufficient to prove that one person controlled the accounts. The blockchain also cannot reveal who held the private keys or what information the traders possessed. Bitquery's account-linkage analysis.

This does not establish that the accounts were unrelated. Their synchronized behaviour remains suspicious. But there is a substantial difference between identifying similar trading behaviour and establishing a coordinated group of insiders.

60 Minutes presents the connection as an established fact without showing viewers the underlying wallet relationships.

That is precisely the evidence needed to move from an interesting pattern to an investigative finding.

The 52% statistic needs its missing methodology

Michelle Kendler-Kretsch's research at the Anti-Corruption Data Collective provides some of the strongest evidence in the programme. Its April 2026 report found that 51.8% of qualifying longshot bets on military and defence markets were placed on winning outcomes.

But the definition matters.

ACDC counted a longshot bet when a single wallet accumulated at least $2,500 in purchases, at a weighted average price of 35 cents or less, within one hour. This is a specialised research measure, not a count of every individual military wager.

The study examined 66,425 qualifying longshot bets across 15,110 markets and 10,944 wallets. Its military and defence finding represented approximately $9.3 million in wagers. ACDC's original research.

The distinction is important because the broadcast presents the 52% figure alongside a 7% sports comparison as an apples-to-apples result, without giving viewers the underlying sample definitions or comparison period.

ACDC's own report shows that sports longshot success changed from 5.2% historically to 9.4% in the first ten weeks of 2026. The comparison therefore needs a clearly stated time window.

More importantly, the researchers themselves acknowledge that unusually successful longshot betting can result from chance, market mispricing or structural biases. Their methodology adds wallet-level concentration and the timing of bets to strengthen the identification of potential insider activity.

That qualification is essential. The research identifies serious information asymmetries and possible insider trading. It does not establish the identity or guilt of every successful trader.

Interestingly, ACDC's subsequent August research identified 152 particularly successful military-market wallets, which it called Orcas. It also found that larger traders and automated accounts followed these wallets, amplifying their positions. The potential security problem is therefore broader than nine accounts, but copying a successful trader does not itself prove access to classified information. ACDC's August investigation.

The $800 million oil transaction is not an $80 million insider payday

The March 23 oil trades deserve investigation. The timing was unusual, and the subsequent price movement was substantial.

But 60 Minutes makes a significant leap when it describes more than $800 million as money staked on falling oil prices and then suggests that the resulting profit could have reached $80 million.

Futures-market volume measures contracts traded. It does not, by itself, establish whether traders were opening short positions, closing existing positions, buying contracts or selling them. CME Group explicitly explains this limitation in its educational material on futures trading. CME Group's explanation.

The Wall Street Journal subsequently examined trading records and confirmed that more than $800 million in oil futures changed hands around the announcement. It identified several firms with profitable trades that day, including Qube Research & Technologies, which gained approximately $5 million, and Forza Fund, which gained approximately $10 million on adjusted trading results.

But the results were mixed. Jane Street gained approximately $19 million on its adjusted daily crude-futures trading, while Jump Trading lost approximately $15 million. These figures concern trading across the day, not necessarily the specific pre-announcement burst. None of the firms was accused of wrongdoing.

Crucially, some firms attributed their activity to a Semafor headline published at 6:50 a.m., reporting that the White House was considering an exit from the Iran war. That is a potentially important public-information explanation which investigators must test against the actual trading records. Wall Street Journal investigation, May 20, 2026.

The CFTC is investigating the transactions. That is appropriate.

But David Covel's estimate that someone could have made $80 million is not an established profit figure. The programme does not show the trader identities, individual positions, opening and closing prices or calculations necessary to establish that amount.

The distinction between trading volume, a directional position and realised profit is not financial pedantry. It is the difference between an alarming chart and evidence of a crime.

The journalist's threats were real, but the betting dispute was more complicated

Emanuel Fabian's experience is among the strongest human stories in the investigation. He received threats against himself and his family, including an explicit warning that the sender would spend money to finish him if he caused a betting loss.

There is no legitimate defence for threatening a journalist to influence his reporting.

However, 60 Minutes omits important details about the actual market dispute.

The March 10 Polymarket contract did not simply ask whether an explosion occurred in Israel. Its rules specifically excluded missiles that had been intercepted, even if debris reached Israeli territory. They also required confirmation of an eligible strike within 48 hours.

After Fabian reported the incident near Beit Shemesh, other users argued that the explosion could have involved interceptor debris rather than an un-intercepted missile. Fabian maintained that the available footage and information from Israeli military sources supported his reporting.

The Washington Post examined the dispute and explained that Polymarket's resolution process involved voting by holders of its cryptocurrency token. The journalist's report was important evidence, but it was not itself the market's final settlement mechanism. The Washington Post, March 17, 2026.

There is also an unresolved difference in how the market's financial exposure is described.

Fabian's Times of Israel account said more than $14 million had been wagered on March 10. He subsequently told 60 Minutes that the figure had risen to approximately $22 million by the time the market closed.

The Washington Post's separate analysis estimated that approximately $200,000 was actually at stake at the end of March 10, with another $7 million wagered as the dispute developed over subsequent days.

Those figures may describe different measures of trading activity and outstanding exposure. They should not be treated as interchangeable.

The broadcast repeats the $22 million figure without explaining precisely what it represents. That is an avoidable omission in a story whose central argument depends on the financial incentives created by prediction markets.

A more direct insider-trading case was already public

Perhaps the most consequential omission concerns Israel.

Before the 60 Minutes broadcast, Israeli authorities had already indicted an Israeli Air Force reservist and a civilian for allegedly using classified military information to place Polymarket bets.

The subsequent court disclosures described a remarkably direct arrangement. The reservist, an Air Force major, allegedly received confidential information about the timing of the June 2025 operation against Iran and passed it to his civilian associate.

The civilian placed the wagers using multiple newly created accounts. The pair allegedly split approximately $128,000 from their initial successful bet, with their total profits from subsequent military-related wagers reaching more than $150,000.

Prosecutors also alleged that the arrangement continued around a subsequent Israeli operation in Yemen. The Guardian published detailed reporting on the indictment on May 6, eleven days before the 60 Minutes broadcast. The Guardian, May 6, 2026.

These remain allegations, not convictions. The defendants' lawyers disputed the case, and their pleas had not been publicly disclosed in the Guardian's reporting.

Nevertheless, this was a much more direct example of the alleged mechanism: a military officer with privileged information communicating operational timing to a civilian who placed bets and shared the proceeds.

Rob Schwartz describes the Van Dyke case as unprecedented, saying there was nothing comparable. If he meant unprecedented in the United States, that distinction should have been made clear. The Israeli case demonstrates that the underlying conduct was already being prosecuted elsewhere.

Leaving it out deprived viewers of an important international comparison and a more concrete account of how classified information could reach a betting account.

Investigations are not convictions

There is an important development that strengthens the programme's central concern.

In September, WIRED obtained CFTC voting records through a Freedom of Information Act request. They showed that chairman Michael Selig approved an investigation order concerning Iran-related Polymarket event contracts at the end of May, approximately two weeks after the 60 Minutes broadcast.

The order authorised the enforcement division to investigate, obtain testimony and seek documents. It did not establish that the nine accounts were operated by insiders or that anyone had committed a crime. WIRED, September 11, 2026.

As of Bitquery's September 22 investigation, no charges in the Iran-account matter had been made public.

The distinction matters. A regulator opening an investigation is evidence that a matter deserves scrutiny. It is not confirmation of the allegations that prompted it.

And the broader concern is not imaginary. The United States has charged Van Dyke, while Israel has prosecuted a separate alleged military-insider betting arrangement. ACDC's research also identifies a wider population of unusually successful wallets.

There is enough here to justify serious investigative work.

The verdict: A real scandal deserves better journalism

I do not dispute the danger of betting on military secrets. Nor do I dispute that some prediction-market accounts have produced extraordinary profits under circumstances that deserve investigation.

What I dispute is the confidence with which 60 Minutes moves between different kinds of evidence.

• The nine accounts' $2.4 million profit is independently reproducible, but their 98% figure is money-weighted, not a 98% success rate by number of bets.
• Their common control has not been established.
• ACDC's 52% result concerns a carefully defined subset of longshot wagers, not all military betting.
• The $800 million oil figure describes trading activity, while the suggested $80 million profit remains unsubstantiated.
• Fabian's threats are real, but the broadcast omits material details about the contract's resolution rules and financial exposure.
• A direct Israeli insider-betting prosecution was already publicly documented but absent from the programme.

These are not minor technical objections. They determine whether viewers are being shown a suspicious pattern, an identifiable insider network or an established criminal operation.

Those are three different conclusions.

60 Minutes had the ingredients for an exceptional investigation. Its mistake was allowing the most dramatic interpretation of the evidence to become the dominant impression before showing viewers enough of the evidence itself.

War betting deserves scrutiny. Classified information must not become a private source of profit. But investigative journalism must be just as disciplined about its own numbers and conclusions as it expects financial markets to be about theirs.

References

1. Prediction market bets on war defy odds — 60 Minutes, CBS News, May 17, 2026. The original broadcast and transcript establish the claims, interviews and figures being examined.

2. CFTC Polymarket Probes: The Iran and Biden Pardon Bets Traced — Bitquery Research, September 22, 2026. Reconstructs the nine accounts' transactions and profits, corrects the 98% win-rate interpretation and examines the limits of the account-linkage evidence.

3. Insider Risks in Polymarket Political Markets — Anti-Corruption Data Collective, April 30, 2026. Provides the original statistical methodology, longshot definition, military-market results and explicit qualifications concerning the identification of insider trading.

4. Classifying Insider Trading Risk: Analyzing the Longshot Betting Ecosystem on Polymarket — Anti-Corruption Data Collective, August 20, 2026. Identifies 152 unusually successful military-market wallets and documents how larger traders and automated accounts follow their positions.

5. What Is Volume? — CME Group. Explains why futures trading volume alone cannot establish whether traders are opening or closing positions, or buying or selling.

6. Flurry of Suspicious Oil Trades Worth $800 Million Triggers Regulatory Probe — The Wall Street Journal, May 20, 2026. Provides subsequent trading-record analysis, named firms, adjusted daily gains and losses, and the alternative explanation involving a public Semafor headline.

7. A Journalist Reported a Missile Strike. Then Came the Death Threats — The Washington Post, March 17, 2026. Examines the threats against Fabian, the contract's settlement conditions and the difference between money at stake and subsequent trading activity.

8. Gamblers Trying to Win a Bet on Polymarket Are Vowing to Kill Me If I Don't Rewrite an Iran Missile Story — The Times of Israel, March 16, 2026. Provides Fabian's firsthand account, the threats, the attempted inducement and his explanation of the underlying market.

9. Keep Quiet 'So We Don't Go to Jail': The Israelis Charged with Bribery After Suspicious Bets Placed on Iran Strikes — The Guardian, May 6, 2026. Examines the Israeli indictment, the alleged military-to-civilian information transfer, cryptocurrency accounts and division of betting proceeds.

10. The US Government Launched 3 Previously Unreported Investigations of Polymarket Trades — WIRED, September 11, 2026. Uses CFTC records obtained through FOIA to establish that a formal investigation into Iran-related Polymarket contracts was authorised after the broadcast.