Los riesgos estructurales y mecanismos de explotación de los mercados predictivos: un estudio de caso de Polymarket

Prediction markets have emerged as a novel mechanism for aggregating information and forecasting future events, leveraging the "wisdom of crowds" to derive probabilistic outcomes. Platforms like Polymarket, which reached a valuation of nearly $9 billion by 2025, exemplify this trend. However, beneath their veneer of democratized forecasting lies a complex interplay of behavioral finance biases, structural risks, and exploitative mechanics. This analysis examines Polymarket's design through the lens of behavioral finance and market efficiency, revealing how psychological biases and institutional conflicts undermine both user trust and market integrity.

Behavioral Finance Biases in Prediction Markets

Behavioral finance has long challenged the Efficient Market Hypothesis (EMH), which assumes rational actors and fully priced information. In prediction markets, however, biases such as anchoring, overconfidence, herding, and loss aversion distort outcomes. For instance, a 2024 study found that retail investors on platforms like Polymarket often anchor to short-term price data rather than long-term fundamentals, leading to mispriced contracts. Overconfidence is equally pervasive: 2025 research highlighted its role in Nepal's emerging markets, where investors overestimated their ability to time events, a pattern mirrored in Polymarket's speculative bets.

Herd behavior further exacerbates volatility. Social media-driven trading, such as the 25% volume surge on Polymarket following viral election memes, demonstrates how collective action without due diligence creates speculative bubbles. Loss aversion, meanwhile, causes users to cling to losing positions, skewing liquidity and price discovery. These biases, amplified by AI-driven nudging tools, create a feedback loop where prediction markets become less about accurate forecasting and more about psychological manipulation.

Market Efficiency and Structural Paradoxes

The EMH posits that markets efficiently incorporate all available information, but prediction markets like Polymarket reveal inherent paradoxes. While contracts on events such as the EU AI Act enforcement timeline aggregate diverse opinions, their efficiency is contingent on liquidity and participation. A contract with only $0.75 liquidity, for example, may reflect a 75% probability but face severe slippage during rapid price shifts. This fragility is compounded by the fact that 85% of Polymarket traders ended 2025 with negative balances, underscoring the platform's inability to sustain profitability in a model reliant on user losses.

Polymarket's Exploitative Mechanics

Polymarket's structural design introduces exploitative mechanics that prioritize platform revenue over market integrity. In 2024, the platform hired an in-house trading team to trade directly against users-a move critics liken to a traditional sportsbook. This creates a conflict of interest, as the platform leverages data asymmetry to manipulate pricing decisions in its favor. Additionally, Polymarket's liquidity incentives, which dropped from $10 million in 2024 to $0.025 per $100 traded in 2025, highlight unsustainable practices.

The platform's reliance on off-chain oracles for settlement further introduces trust-based risks. Ambiguous event definitions, such as those in cultural or political contracts, can lead to disputes and settlement uncertainty. Meanwhile, a 40% probability of a major platform ban within the next year could reduce liquidity by 27-30%, eroding $2-3 billion in annual activity.

Systemic Risks and Investor Implications

The interplay of behavioral biases and structural flaws creates systemic risks for both users and the broader market. For instance, Polymarket's novelty segments-accounting for 40% of trading volume-exhibit thin liquidity and wide bid-ask spreads (4.8% vs. 2.5% in sports contracts), making them prime targets for manipulation. Social media-driven herding behavior amplifies this vulnerability, as traders follow trends without independent analysis.

Investors must also contend with the platform's decentralized structure, which, while fostering innovation, lacks robust oversight. The ability to shift liquidity to blockchain DEXs offers a mitigant but does little to address the root issue: a system designed to profit from user irrationality.

Conclusion

Polymarket's rise as a $9 billion prediction market underscores the allure of democratized forecasting. Yet, its structural risks-rooted in behavioral finance biases and exploitative mechanics-highlight a critical tension between market efficiency and institutional self-interest. For investors, the lesson is clear: prediction markets are not immune to the psychological pitfalls that plague traditional finance. As regulatory scrutiny intensifies and user trust wanes, platforms like Polymarket must reconcile their design with the principles of transparency and fairness-or risk becoming the next cautionary tale in the behavioral finance playbook.