Prediction markets have a well-earned reputation for being fairly accurate: contracts priced at 70 cents tend to win about 70% of the time. This relationship between price and outcome frequency holds across thousands of markets and even outperforms many expert forecasts and traditional polls, making these markets powerful tools for gauging probabilities.
However, that 70% figure is more of a useful guideline than an exact truth. Studies reveal consistent biases in how contract prices translate to actual probabilities. One familiar distortion is the favorite-longshot bias: cheap longshot contracts end up winning less often than their prices would suggest, while favorites win slightly more frequently. Investors betting on longshots typically face negative expected returns due to this skew.
Another subtle but impactful factor is capital lock-up. When a contract promises a $1 payout months down the line, its value today isn’t just about the probability but also the fact that money is tied up and not earning interest. This drags down prices for longer-dated contracts relative to their “fair” probability-based value.
What Distorts Probability Mapping?
Calibration isn’t uniform across categories and timeframes. Political prediction markets, for example, often show prices clustered closer to 50% in long-term bets. This likely happens because opposing partisan bets cancel each other out rather than providing new informative signals. Fees, bid-ask spreads, and the mechanics of market-making also chip away at actual returns, especially when contract prices trade in very small increments. also resolution risk looms behind every contract, as a correctly predicted outcome might still fail to pay due to administrative issues.
Underlying the rapid creation and trading of prediction contracts in the U.S. is a regulatory nuance buried in the Dodd-Frank Act. A simple filing allows exchanges to list new markets without prior approval, though the Commodity Futures Trading Commission recently opened a rulemaking process to clarify the legal definitions that have long been vague. These three small words in legislation shape the entire ecosystem.
Understanding prediction market prices requires embracing that prices reflect a market-clearing figure shaped by capital constraints and trading frictions, not a perfect probability number downloaded from an oracle. This insight helps explain why traders who treat prices as literal probabilities often lose money in predictable ways. For a deeper get into how this impacts crypto and financial markets, see our coverage on the expected growth of prediction markets and how AI could fuel crypto demand.
Disclaimer: This article is for informational purposes only and is not financial advice.
