Long-shot wagers of $2,500 or more placed at odds of 35 percent or less on Polymarket had an average win rate of around 52 percent in markets on military and defense actions, according to analysis by the Anti-Corruption Data Collective.
Anti-Corruption Data Collective's analysis
The Anti-Corruption Data Collective, a non-profit research and advocacy group, examined betting on Polymarket and reported that “insider trading is rife on Polymarket.” Their definition for a long-shot bet was explicit: wagers of $2,500 or more placed when the market assessed the chance at 35 percent or less. Using that threshold, the group calculated win rates across different topical categories of markets on the platform.
Long-shot bets on military and defense actions
Within markets tied to military and defense actions, these long-shot bets recorded an average win rate of about 52 percent. That outcome implies that wagers meeting the long-shot criteria in those markets succeeded roughly half the time—far above what would be expected from random guessing at the stated odds and markedly different from other topical areas on the platform, as the Collective’s numbers show.
Comparative performance: politics-focused and all markets
The Collective contrasted the 52 percent win rate for long-shot military/defense wagers with two benchmarks drawn from Polymarket more broadly: a 25 percent win rate across all politics-focused markets, and a 14 percent win rate across all markets on the platform. Those three figures—52 percent, 25 percent, and 14 percent—are presented side-by-side by the Collective to underline the disparity in outcomes between military/defense markets and other categories.
What this means for Polymarket operators, policymakers, and the public
- Polymarket operators: The data places questions about market integrity and the platform’s exposure to information asymmetries squarely on the table. The pronounced outperformance of large, low-odds wagers in military and defense markets is the pattern the Collective highlights.
- Policymakers and regulators: The Collective’s analysis and its accompanying commentary — which calls the legal status “absolutely insane” in light of the apparent distortion — position legality and potential oversight as central concerns to be considered.
- The general public and bettors: For those who participate in or follow prediction markets, the numbers suggest that some markets may be influenced by nonpublic information or actors with advantaged access, changing the risk-reward calculus for large wagers.
Conclusion
The Anti-Corruption Data Collective’s narrow, numeric finding is stark: large, low-odds wagers on Polymarket’s military and defense markets have won at an average rate of about 52 percent, compared with 25 percent in politics-focused markets and 14 percent across the platform. Presented alongside the Collective’s blunt assessment of legality and comparison to insider betting in sports, the data raises a single, concrete question for those who run, regulate, or use these markets: how should a platform respond when specific categories of bets outperform the broader marketplace by such wide margins?
