要聞列表耶魯研究:Polymarket 3% 高手吃下 30% 利潤,67% 散戶承擔全部虧損
動區 BlockTempo2026-04-27 00:55:32

耶魯研究:Polymarket 3% 高手吃下 30% 利潤,67% 散戶承擔全部虧損

AI 影響分析Grok 分析中...
📄完整原文· 由 trafilatura 自動擷取Gemini 翻譯2455 字
Researchers from London Business School and Yale University analyzed over $13.7 billion in trading volume on Polymarket and found that the so-called "wisdom of the crowd" is merely a game for a small number of experts: only 3.14% of accounts dominate most price discovery, and together with market makers, they take away more than 30% of the profits; meanwhile, 67% of losing accounts bear all the losses on the platform. At the same time, the CFTC filed the first-ever insider trading lawsuit involving event contracts, accusing an active-duty U.S. Army Sergeant Major of placing massive bets to profit over $400,000 before classified military operations were exposed. (Previous coverage: Polymarket insider caught! Israeli soldier used classified intelligence to bet, winning 6 out of 6, sued for making $150,000) (Background supplement: Can we track the next Polymarket insider trader? Definitely, and the barrier to entry is low) The most fascinating selling point of prediction markets is that "millions of people bet collectively, and wisdom naturally emerges"—but this story might have been wrong from the very beginning. A latest working paper released by researchers from London Business School (LBS) and Yale University directly points out: the price accuracy of Polymarket comes overwhelmingly from a tiny group of informed elites, rather than true crowd wisdom. This conclusion is a direct hit to the entire prediction market narrative. Four scholars—Roberto Gomez-Cram, Yunhan Guo, and Howard Kung (all from LBS) and Theis Ingerslev Jensen (Yale University)—systematically analyzed Polymarket’s complete transaction records from 2023 to 2025 in a working paper titled "Prediction Market Accuracy: Crowd Wisdom or Informed Minority?" released on SSRN on April 20, 2026: 98,906 events, 210,322 markets, approximately $13.76 billion in trading volume, and 1.72 million accounts. The paper's title itself is the answer: Does the accuracy of prediction markets come from crowd wisdom or an informed minority? The research results are sobering: only 3.14% of accounts can be classified as "expert winners"—accounts whose order flow consistently predicts short-term price fluctuations and final outcomes. This group, combined with market makers, accounts for less than 3.5% of total accounts but captures over 30% of all profits on the platform. On the other end, losing accounts classified by the research team as "unlucky or unskilled" make up 67% of the total, and this group bears all the net losses on the platform. The remaining accounts fall in between, essentially breaking even. In other words: the operational logic of Polymarket is more like a mechanism where a few informed individuals harvest the majority of retail investors, rather than an information aggregation platform as defined in textbooks. Even more concerning is the paper's systematic flagging of suspected insider trading accounts. The research team identified 1,950 accounts with specific behavioral patterns: opened shortly before a single major event and went dormant immediately after the event settled. The price-moving power of these accounts is about 7 to 12 times that of typical skilled traders—but because they are highly concentrated on isolated events, they do not significantly improve Polymarket's overall prediction accuracy. A specific case study in the paper is particularly striking: three accounts were opened between December 27, 2024, and January 3, 2025, and subsequently focused on betting on the "Yes" contract for "Will Venezuelan President Nicolás Maduro step down before January 31, 2026?" These accounts held heavy positions before the U.S. military intelligence was exposed, ultimately profiting over $630,000 in total. The research team pointed out in the paper that these account behaviors exhibit clear characteristics of information asymmetry—precise betting, perfect timing, and disappearing after the fact. Shortly after the paper was released, real-world judicial proceedings provided a stunning corroboration of the academic analysis. On Thursday, April 24, 2026, the U.S. Commodity Futures Trading Commission (CFTC) officially filed the first-ever insider trading lawsuit involving event contracts, with the defendant being active-duty U.S. Army Sergeant Major Gannon Ken Van Dyke (North Carolina). The CFTC stated in a press release that before the classified intelligence of the U.S. military raid on Maduro-related targets, codenamed "Operation Absolute Resolve," was exposed, Van Dyke purchased over 436,000 "Yes" contracts for "Maduro to step down before January 31, 2026" on Polymarket between
資料狀態✓ 已擷取全文閱讀原文(動區 BlockTempo)
🔍歷史類似事件· 關鍵字 + 標的比對6 則
💡 目前用關鍵字 + 標的比對(MVP)· 之後會升級為 embedding 語意搜尋
原始資訊
ID:d863ff4ee9
來源:動區 BlockTempo
發佈:2026-04-27 00:55:32
分類:zh_news · 導出分類 zh
標的:未指定
社群投票:+0 /0 · ⭐ 0 重要 · 💬 0 留言
耶魯研究:Polymarket 3% 高手吃下 30% 利潤,67% 散戶承擔全部虧損 | Feel.Trading