Author: Alexander Lin, Crypto KOL
Compiled by: Felix, PANews
Opinions on prediction markets have always been mixed. Some see them as a disruptive new infrastructure capable of disrupting traditional institutions, while others believe they are unlikely to become a true part of mainstream finance. Recently, crypto KOL Alexander Lin published an article pointing out 23 flaws in prediction markets, the details of which are as follows.
1. Low capital efficiency
Prediction markets require full collateralization and prohibit the use of leverage. Compared to the 5-10% notional value margin requirement of perpetual contracts (Perps), prediction markets are 10 to 20 times less capital efficient. This does not even take into account the zero return on locked capital and the inability to cross margin calls across positions.
2. Capital turnover rate is structurally disrupted.
Because capital is locked up throughout the contract period, resulting in a binary outcome, capital turnover is structurally disrupted. Positions are voided (expired) immediately after contract settlement, thus eliminating balance sheet efficiency and preventing market maker assets from compounding. The same funds would generate a much higher turnover (5-10 times) if used for perpetual account trading during the same period: inventory is recycled, positions are rolled over, and hedging operations continue.
3. LP inventory has fundamental flaws.
At settlement, half of the assets in the liquidity pool are destined to go to zero. For example, spot liquidity pools will be rebalanced among assets with remaining value; but for prediction markets, there is neither rebalancing nor residual value, only the "binary collapse" of the losers.
4. Lack of natural hedgers
Unlike commodities, interest rates, or foreign exchange, prediction markets lack a "natural hedge" that provides inverse liquidity. No entity or trader has a natural economic need to stand on the opposite side of event risk. Market makers face pure adverse selection, lacking structured counterparties. This is a fundamental obstacle to scalability.
5. Adverse selection intensifies as settlement approaches.
As the market nears settlement, adverse selection intensifies. Traders with an advantage or more accurate information are able to buy the winners at better prices from those who are still pricing based on outdated prior information. This attrition is structural and worsens over time.
6. The Starting Dilemma: The Structural Liquidity Trap
The lack of liquidity in new markets discourages informed traders from entering (to avoid losses due to slippage); and as long as prices remain inaccurate, no more traders will emerge. Long-tail markets often fail before they even begin, and no amount of subsidies can solve this problem.
7. Lack of an endogenous demand cycle
Every dollar of trading volume relies on external attention (such as elections, news, and sporting events), with no support between events. In contrast, perpetual contracts create an internal flywheel: trading generates funding rates, funding rates create arbitrage opportunities, and arbitrage brings in more funds.
8. Disconnected from institutional asset allocation
Predicting markets is not linked to risk premiums, portfolio returns, or factor exposure. Institutional capital lacks a systematic framework for scaling or risk management of these positions. These markets do not conform to any standard portfolio construction language or strategy, and therefore cannot be truly scaled.
9. Liquidity is reset to zero at each settlement.
Liquidity is reset to zero after each settlement, requiring rebuilding from scratch. The open interest (OI) and depth that accumulate over time in perpetual contracts are structurally impossible to achieve in prediction markets.
10. Subsidy-driven False Prosperity
Subsidies are the only reason bid-ask spreads haven't permanently spiraled out of control. Once the incentives stop, market liquidity collapses. "Bribed" liquidity is inherently a corrupt and short-sighted market structure.
11. The contradiction between transaction volume and information quality
Platforms profit from trading volume (e.g., "We need gambling volume!") rather than accuracy, while regulators need predictive utility to justify these platforms' existence. This trade-off leads to unsatisfactory product/feature decisions.
12. Accuracy becomes an illusion
In high-attention markets, marginal participants without informational advantages simply follow public consensus, causing prices to reflect what people "already believe," rather than pricing in fragmented signals. Accuracy becomes an illusion.
13. Unrestricted market creation is rife with noise.
When listing requires no cost, liquidity and attention are dispersed across thousands of markets. The drivers of growth and the drivers of selection are directly opposed.
14. Problem design can be used as an attack tool.
The person who writes the questions controls the criteria for judging the final results. There is neither a neutral drafting process nor an incentive mechanism to ensure the accuracy of the questions, and there is no recourse if someone takes advantage of the loopholes.
15. Oracle Risks
Decentralized oracles determine the truth based on token weights. When the market capitalization of an oracle is less than the value of the funds it guarantees (locked), manipulation becomes a rational transaction. Centralized settlement, on the other hand, faces the risk of its operators being captured or becoming ineffective.
16. Inflated nominal trading volume
Trading volume is reported without price adjustments. A $1 trade at $0.90 is entirely different from a $1 trade at $0.50. The actual amount of risk transferred is exaggerated by an order of magnitude, but everyone is citing that inflated figure.
17. Reflexivity after scaling up
When the prediction market is large enough, a high probability (e.g., >90%) prediction will itself change the behavior of relevant participants. This logic of "discovering the truth" has structural limitations.
18. Cross-platform credibility risks
If the same event yields different settlement results on different platforms, then the entire industry appears unreliable. Credibility is shared, and discrepancies between different platforms generally lead to negative expectations.
19. Market Manipulation
Traders can secure their positions in the prediction market (secondary market) by manipulating real-world events (primary market). Currently, there are no effective position limits or regulatory implementations in place.
20. Risk of manipulation
Because there are no position limits and limited regulatory enforcement against manipulation, this means that a single wallet can leverage a shallow market and use this volatility to make contrarian trades without any consequences (no one can hold them accountable). This problem is particularly severe on Polymarket compared to Kalshi.
21. Lack of complex financial instruments
There is no term structure, conditional instructions, or composability. The entire derivatives toolkit is completely absent, except for a single binary outcome, which prevents professional institutions from entering the market.
22. Fragmented regulation
As regulations tighten, differences between federal and state levels will force liquidity fragmentation. When the market is segmented into pools of different participants, price discovery collapses.
23. The Innovator's Dilemma
Existing giants have no incentive to redesign their architecture. If trading volumes continue to grow and regulatory moats continue to form, any architectural changes will become increasingly expensive. This is a classic innovator's dilemma.
Related reading: Polymarket vs Kalshi: Who is the king of prediction markets?
