On Prediction

Curated content for builders, investors & researchers

Prediction Market Concepts

84 concepts across 6 clusters: Oracle & Resolution, Liquidity & Trading, Information Theory, Mechanism Design, Governance & Decisions, Business & Platforms.

adverse selection: Risk that counterparties trade because they have superior information, causing losses for the market maker.
AI agents: Autonomous software systems that trade on prediction markets, potentially improving liquidity and accuracy.
arbitrage: Exploiting price differences for equivalent outcomes across markets or platforms for risk-free profit.
attention markets: Markets that trade on what topics or events will capture public attention and discourse.
batched auctions: Collecting orders over a time window and clearing them simultaneously at a single price to reduce manipulation.
bid-ask spread: The difference between the highest price a buyer will pay and the lowest price a seller will accept, representing the cost of immediacy for traders.
binary contracts: Prediction market contracts that resolve to exactly one of two values, typically $1 (yes) or $0 (no), based on whether a specified event occurs. The binary structure eliminates ambiguity at resolution and enables direct probabilistic interpretation of prices.
bonding trades: A prediction market strategy of betting on near-certain outcomes (>95% probability) to earn a small but reliable yield, analogous to holding a zero-coupon bond to maturity. The 'yield' is the gap between the market price and the $1 payout, but carries catastrophic tail risk if the improbable outcome occurs.
Brier score: A proper scoring rule that measures probabilistic forecast accuracy as the mean squared difference between predicted probabilities and binary outcomes; lower scores indicate better calibration.
calibration: When stated probabilities match empirical frequencies—e.g., events given 70% odds happen 70% of the time.
combinatorial prediction markets: Prediction markets that allow participants to place bets on conditional events and Boolean combinations of base events, not just individual outcomes. By enabling richer betting vocabularies, they can elicit joint probability distributions over related events rather than only marginal probabilities.
conditional tokens: Tokens representing positions contingent on specific outcomes, enabling composable prediction market positions.
continuous double auction: A trading mechanism where buy and sell orders are matched immediately as they arrive, with price-time priority determining execution order.
corruption value multiple: The ratio of economic value at stake outside a market versus the cost of corrupting its oracle.
covariance markets: Markets that trade on the correlation between events, allowing positions on how outcomes relate to each other.
credibility markets: Markets where participants stake on the accuracy of claims or sources, building continuously updated trust scores rather than static reputations. Treats trustworthiness as a tradeable probability over time.
cross-platform arbitrage: Trading the same event across different prediction market platforms to profit from price discrepancies.
cross-subsidization: Using revenue from commercially profitable prediction markets to fund liquidity incentives on socially valuable but self-sustaining markets, mirroring how newspaper ads subsidize investigative journalism.
decision markets: Prediction markets explicitly designed to inform specific decisions by forecasting outcomes of each option.
demand markets: Markets that capture consumer purchase intent as a probability signal before production, replacing surveys and focus groups with skin-in-the-game demand forecasting.
dispute resolution: Mechanisms for challenging and adjudicating contested market outcomes when participants disagree on results.
distribution markets: Markets that trade on full probability distributions rather than single binary yes/no outcomes.
election markets: Prediction markets focused on political elections, often the highest-volume and most visible markets.
endogeneity: When the existence or visibility of a prediction market influences the very outcome it aims to forecast, undermining its reliability as a neutral signal. Particularly relevant in political and social markets where public price movements can shift behavior.
event contracts: Standardized binary contracts on specific real-world events, the core trading unit of prediction markets.
execution quality: How favorably a trader transacts relative to the prevailing market price. In prediction markets, execution quality separates profitable from unprofitable traders: automated participants systematically obtain better prices than manual traders, and this edge outweighs directional forecasting skill as a determinant of returns.
federal preemption: The legal doctrine under which federal law displaces state regulation. In prediction markets, the question is whether the CFTC's exclusive jurisdiction over swaps under the Commodity Exchange Act preempts state gaming laws from regulating exchange-traded event contracts.
forecasting accuracy: Measuring how well predicted probabilities match actual outcome frequencies over many events.
futarchy: A governance system where policy decisions are made based on which option prediction markets forecast will produce the best outcomes.
gap risk: Exposure to sudden large price jumps in binary markets when new information arrives between trades.
hedging: Taking offsetting positions to reduce exposure to adverse price movements or uncertain outcomes.
hyperstition markets: Markets where collective belief in an outcome actively contributes to making that outcome happen.
impact markets: Markets that trade on the expected social impact of projects, funding those predicted to be most effective.
implied correlation: The degree to which related prediction market outcomes should move together based on economic logic, creating trading opportunities when markets price correlated events inconsistently.
incentive compatibility: Designing mechanisms where truthful participation is each player's optimal strategy regardless of others' actions.
info finance: Using financial market mechanisms to elicit, aggregate, and trade on informational signals about future events.
information aggregation: The process by which markets combine dispersed private knowledge into a single consensus price signal.
information asymmetry: When some traders possess materially better knowledge about likely outcomes than other participants.
insider trading: Trading on material non-public information (MNPI) obtained through a breach of a confidentiality duty or implied promise. In prediction markets, this arises when participants trade on privileged access to information about upcoming events, creating adverse selection for other traders.
Kelly criterion: A formula for optimal bet sizing that maximizes long-run growth by balancing edge against risk of ruin.
liquidity fragmentation: The dispersion of trading capital across multiple independent order books when a single question is split into many binary contracts. In prediction markets, this creates ghost markets where tail outcomes receive little or no volume, reducing the information captured by the market structure.
liquidity provision: Supplying capital so traders can buy and sell positions without excessive price impact or delays.
LMSR (logarithmic market scoring rule): An automated market maker that prices trades using a logarithmic cost function, guaranteeing bounded loss.
long-tail markets: Prediction markets on niche or specialized questions that individually attract thin liquidity but collectively generate significant informational value, analogous to how long-tail content drives aggregate engagement on platforms like YouTube.
longshot bias: The tendency for low-probability outcomes to be systematically overpriced in prediction markets.
LOX (log-odds excess lateness): A metric measuring how much a forecaster's probability updates lag behind optimal Bayesian updating.
market making: Continuously quoting bid and ask prices to facilitate trading, earning the spread while managing inventory risk.
market manipulation: Deliberately trading to distort prices away from true probabilities for strategic or financial gain.
market scoring rules: A class of automated market maker mechanisms that subsidize trade by penalizing a market maker according to a proper scoring rule. Traders profit by moving prices closer to their beliefs, ensuring the market maker absorbs losses in exchange for eliciting honest probability estimates. LMSR is the most widely used instance.
market surveillance: Systematic monitoring of trading activity by exchanges or regulators to detect insider trading, manipulation, and other abusive practices. In prediction markets, surveillance is complicated by pseudonymous blockchain wallets and the absence of traditional issuer-based disclosure obligations.
minimum viable liquidity: The threshold level of trading volume a prediction market needs to attract informed traders who can correct mispricing, calculated as Cost of Expertise divided by Price Gap. Beyond this threshold, additional liquidity does not improve forecast accuracy.
multi-outcome markets: Prediction markets where an event is divided into multiple mutually exclusive outcome ranges (e.g., price brackets), each tradeable. Contrasts with binary markets that only ask yes/no questions.
network effects: When a platform becomes more valuable to each user as more participants join and trade on it.
no-loss prediction markets: Designs where participants risk only potential yield, not principal, lowering the barrier to entry.
noise decomposition: Separating observed price movements into components attributable to genuine new information versus random microstructure noise, used to assess how much of a market's short-term variance reflects real signal.
nowcasting: Using prediction market prices as real-time proxies for economic indicators that are officially reported with a delay, such as inflation or employment figures. Enables faster decision-making by treating continuously updated market odds as a high-frequency data source.
opportunity markets: Markets for discovering and pricing unexploited opportunities across domains using crowd forecasting.
oracle design: Systems that feed real-world outcomes into prediction markets, determining how events are verified and settled.
order book: A list of outstanding buy and sell orders at various prices, showing available liquidity at each level.
orderflow arbitrage: Profiting from temporary price dislocations caused by large orders that push the market away from fair value before it mean-reverts.
parimutuel markets: A betting structure where all wagers pool together and payouts are split proportionally among winners after the event resolves. Parimutuel markets bootstrap liquidity without market makers, making them useful for niche or long-tail prediction markets, but face challenges around locked positions, timing, and real-time price readability.
parlays: Multi-leg bets combining outcomes across several events, offering higher payoffs at lower probability.
peer prediction: Methods for eliciting honest subjective reports by comparing respondents against each other statistically.
platform competition: How prediction market platforms differentiate, compete for users, and defend market share.
position collateralization: Using prediction market positions as collateral to borrow capital, unlocking liquidity without selling. This addresses the capital lock-up problem in long-dated markets and enables composability with the broader DeFi ecosystem.
price discovery: The process through which trading activity reveals the fair value or true probability of an event.
probability infrastructure: The concept of prediction market mechanisms as a general-purpose layer that embeds live probability signals into decision surfaces beyond trading, such as attention, credibility, and demand.
proper scoring rules: Incentive-compatible functions that reward forecasters most when they report their true beliefs honestly.
reflexivity: When market prices influence the very outcomes they predict, creating feedback loops between beliefs and reality.
regulatory arbitrage: Exploiting differences in prediction market regulations across jurisdictions to offer otherwise restricted products.
regulatory classification: The legal categorization of prediction market contracts as gambling, financial derivatives, or a distinct product class, which determines which regulatory regime applies and what compliance obligations operators face.
relative value trading: A strategy that bets on the relationship between two correlated prediction market prices rather than the direction of either one, profiting when a mispriced spread converges.
resolution criteria: Pre-defined rules specifying exactly what outcome counts as a win, loss, or void for a market.
retail flow: Trading activity from non-professional participants, generally considered less informed and more predictable.
self-resolving markets: Markets whose resolution depends on on-chain or automatically verifiable data, needing no human oracle.
semantic tick size: The minimum price increment in a prediction market that doubles as a narrative unit: because contracts resolve at $0 or $1, each one-cent move is universally read as a one-percentage-point probability revision, making microstructure noise appear informative.
superforecasting: Techniques and traits of forecasters who consistently outperform base rates and prediction markets.
temporal arbitrage: Profiting from swings in probability estimates as a prediction market converges toward its binary resolution, trading volatility rather than the final outcome.
time arbitrage: Profiting from a delay between when information becomes known and when the prediction market price adjusts to reflect it.
toxic flow: Order flow from informed traders that systematically moves against the market maker's positions.
UMA protocol: Optimistic oracle system where outcomes are assumed correct unless disputed, using economic bonds for security.
wash trading: Executing offsetting buy and sell transactions to inflate apparent trading volume without taking real market risk. On prediction markets, wash trading can artificially boost platform metrics and distort liquidity signals, and is federally prohibited under the Commodity Exchange Act.
wisdom of crowds: The phenomenon where aggregated group estimates often outperform individual experts in forecasting accuracy.
yes bias: The systematic tendency for YES shares in prediction markets to be overpriced relative to true probabilities, potentially explaining what was previously attributed to favorite-longshot bias.