The Prediction Markets Glossary: Every Term You Need, in Plain English

Prediction market terms come from three worlds at once: exchange trading (order book, maker, taker), probability (implied probability, expected value, longshot bias), and federal derivatives regulation (DCM, self-certification, CFTC). This glossary defines the ones you’ll actually run into — grouped by theme, alphabetical within each group, with real numbers where a number makes the definition click. If you’re brand new, start with our explainer on what a prediction market is, then use this page as the reference you keep coming back to.

Every term on this list is a different way of asking the same question: what does the price say, and should you trust it?

Contracts, outcomes, and resolution

Event contract

The basic unit of a prediction market: a contract tied to a yes-or-no question about a future event, which pays a fixed amount — usually $1 — if the event happens and nothing if it doesn’t. “Will the Fed cut rates in September?” is a question; the event contract is the thing you actually buy and sell, priced between 1¢ and 99¢. On regulated US exchanges, “event contract” is also the formal regulatory term for these products.

Outcome

One possible answer to a market’s question. A binary market has two outcomes, Yes and No. A multi-outcome market — “Who wins the nomination?” — has one contract per candidate, and the Yes prices across all of them should sum to roughly $1, since exactly one outcome occurs.

Resolution

The moment a market’s question gets an official answer. Once a market resolves, trading stops and the contracts on the winning side become worth exactly $1 while the losing side becomes worth zero. Resolution is a determination of fact; settlement (below) is the payment that follows.

Resolution criteria

The fine print that defines precisely what counts as Yes. Good criteria remove all judgment: a market on “Will inflation be mentioned at the Fed press conference?” will specify which transcript governs and whether “inflationary” counts for “inflation.” Reading the resolution criteria before trading is the closest thing prediction markets have to a golden rule — the market resolves on the written rules, not on what you assumed the question meant.

Resolution source

The specific authority the rules name for settling the question — the Bureau of Labor Statistics release for a CPI market, the official transcript for a mention market, the Associated Press call for an election market. If the named source is ambiguous or never publishes, the rules say what happens next; if they don’t, expect a dispute.

Settlement

The payout step after resolution: holders of winning contracts are credited $1 each, holders of losing contracts get nothing, and the positions disappear from your account. On regulated exchanges settlement runs through a clearinghouse (see DCO below), which is why the winner gets paid even if the person on the other side of the trade vanishes.

Yes/No contracts

The two sides of a binary market. They are mirror images: buying No at 30¢ is economically identical to selling Yes at 70¢, because Yes and No always sum to $1. Platforms display both mostly for convenience — “buy No” is easier to think about than “short Yes.”

The order book: how trading actually works

Ask

The lowest price at which anyone is currently willing to sell a contract. If the ask on Yes is 63¢, you can buy immediately at 63¢. Also called the offer.

Bid

The highest price at which anyone is currently willing to buy. If the bid is 61¢, you can sell immediately at 61¢. The gap between bid and ask is the spread.

Depth

How many contracts are available at each price level, not just the best one. A market showing 40 contracts at the 63¢ ask and thousands more at 64–66¢ is deep; one where a 200-contract order would blow through five price levels is shallow. Depth is what separates a headline price you can actually trade at from one you can’t.

Limit order

An order that names your price and waits: “buy Yes at 60¢ or better.” It only executes if someone comes to meet you, which means you might never be filled — but you never pay more than you said. Resting limit orders are what fill the order book.

Liquidity

How easily you can trade meaningful size without moving the price. Liquid markets have tight spreads and real depth; illiquid ones might show a last-traded price of 55¢ but a bid-ask of 45¢/65¢, which tells you the “price” is closer to a rumor than a fact. Thin liquidity is the single biggest practical difference between a headline market and an obscure one.

Maker and taker

A maker places a resting limit order that adds liquidity to the book; a taker executes against an existing order and removes it. Exchanges commonly charge takers more than makers because makers are providing the service of a tradeable market. As of August 2026, Kalshi’s fee schedule charges takers roughly 0.07 × P × (1 − P) per contract — about 1.75¢ on a 50¢ contract — with maker fees at about a quarter of that where they apply.

Market maker

A trader — often a professional firm — who continuously quotes both a bid and an ask, earning the spread in exchange for always being willing to trade. Market makers are why you can buy a contract at 2 a.m. on a market nobody is talking about. Some exchanges pay or rebate designated market makers to keep quotes tight.

Market order

An order that executes immediately at the best available price, whatever that is. Fast but blind: in a thin market, a large market order can walk up the book and fill at much worse prices than the quote you saw. In most prediction markets, a limit order priced at the ask gets you the same fill with a ceiling on the damage.

Order book

The live list of all resting buy and sell orders at every price, best prices on top. Prediction market exchanges run central limit order books — the same market structure as a stock exchange — which is the core mechanical difference between an exchange and a sportsbook, where you can only accept the house’s posted price.

Position

What you currently hold: 500 Yes contracts in the September Fed market is a position. Your exposure is the gap between what you paid and the $1-or-zero outcomes — 500 contracts bought at 40¢ risk $200 to win $300.

Slippage

The difference between the price you expected and the price you got, usually because your order was bigger than the depth at the best quote. Quoted at 63¢, filled at an average of 65¢ — that 2¢ is slippage, and in a market that only pays out $1, it’s a real tax on the trade.

Spread

The gap between the best bid and the best ask. A 61¢/63¢ market has a 2¢ spread. The spread is a round-trip cost: buy at the ask, sell at the bid, and you lose the spread even if the probability never moved. Tight spreads are the signature of a healthy market.

Prices, probability, and edge

Arbitrage

Locking in a riskless profit from inconsistent prices. The clean version: if Yes trades at 40¢ on one venue and No at 55¢ on another for the same event with the same resolution rules, buying both costs 95¢ and pays $1 no matter what happens. In practice fees, withdrawal frictions, and subtly different resolution criteria eat most apparent arbitrages — the “same” question resolving differently on two platforms is a classic way an arb turns into a loss.

Edge

The gap between your probability estimate and the market’s. If you believe an event is 60% likely and Yes trades at 50¢, you believe you have a 10-point edge. Whether you actually do is the entire question — markets aggregate a lot of information, and most traders who think they see an edge are looking at the market’s information from behind.

Expected value (EV)

The average outcome of a trade if you could repeat it many times: probability of winning × payout, minus probability of losing × cost. Buy Yes at 30¢ on an event you think is 40% likely, and your EV per contract is (0.40 × 70¢) − (0.60 × 30¢) = +10¢. Positive EV doesn’t mean you’ll win this time; it means the price is wrong if your probability is right.

Hedging

Taking a market position to offset a risk you already have, rather than to speculate. A farmer buying Yes on “major drought declared” or an importer trading a tariff market is using the contract as insurance — one of the oldest economic justifications for event markets, and one regulators weigh when deciding what may list.

Implied probability

The probability reading you get straight from the price: a Yes contract at 27¢ implies the market collectively puts the chance of the event at about 27%. This is the whole trick of prediction markets — the price is the forecast. It’s an aggregate estimate, not an oracle: implied probabilities move with news, and 27% events happen about 27% of the time in a well-calibrated market, which means “the market said it wouldn’t happen” is never quite what the market said.

Longshot bias

The well-documented tendency for low-probability outcomes to be overpriced and near-certainties to be slightly underpriced. A contract “worth” 2¢ often trades at 4¢ because small prices feel like cheap lottery tickets. Inherited from decades of horse-racing data, it shows up in event markets too — it’s the standard caution against reading extreme tail prices too literally in either direction.

The regulatory layer

CFTC

The Commodity Futures Trading Commission — the federal agency that regulates derivatives exchanges in the United States, event contract exchanges included. The CFTC designates exchanges, oversees what they list, and can review or challenge specific contracts. As of August 2026 it is also actively rethinking the category: the agency proposed a new framework for event contracts in June 2026, and its staff issued an advisory in July 2026 tightening how exchanges may self-certify whole series of contracts at once.

DCM (designated contract market)

The CFTC’s license category for a full derivatives exchange — the same designation the CME holds. Kalshi operates as a DCM, which is what makes its event contracts federally regulated products rather than bets. DCM status carries obligations: rulebooks, market surveillance, and listing procedures set out in CFTC regulations.

DCO (derivatives clearing organization)

The regulated clearinghouse that stands between every buyer and seller, holds the collateral, and guarantees settlement. When you win $1 on a resolved contract, the DCO is the entity making sure that dollar arrives. It’s the plumbing that lets strangers trade against each other without counterparty risk.

KYC (know your customer)

The identity-verification requirements — legal name, date of birth, Social Security number for US users — that regulated exchanges must impose before you can trade. KYC is a practical dividing line between regulated venues and offshore ones, and it’s why “anonymous” and “CFTC-regulated” don’t coexist.

Self-certification

The process under CFTC Regulation 40.2 by which a DCM lists a new contract by certifying it complies with the law, rather than waiting for advance approval — file it, and it can trade, unless the CFTC later objects. Self-certification is how event contract listings have moved so fast. It’s also where current regulatory friction concentrates: the CFTC’s July 2026 staff advisory says broad templates covering unspecified future contract variations shouldn’t be filed as a single self-certification.

Market types you’ll encounter

Mention market

A contract on whether a specific word or phrase gets said in a defined setting — “Will the Fed chair say ‘inflation’ at the press conference?” Resolution keys to a named official transcript, with the rules specifying which word forms count. As of August 2026 mention markets are under a cloud: NPR reported in August 2026 that federal regulators are probing them, with critics arguing the speaker can influence the outcome — an insider-knowledge problem baked into the market design. The platforms dispute that characterization; nothing had been prohibited as of this writing.

Multi-outcome market

A market with more than two possible answers — an election with five candidates, a “which month?” question — structured as a set of linked binary contracts, one per outcome. Because exactly one outcome wins, the Yes prices should sum to about $1; when they sum to meaningfully more or less, that gap is either fees, thin liquidity, or an arbitrage.

Parlay / combo

A single position bundling multiple contracts where every leg must resolve Yes to pay out. Kalshi calls these combos; sportsbooks call the equivalent a parlay. Priced correctly, a combo should cost about the product of its legs’ probabilities — two independent 50% legs ≈ 25¢ — but correlation between legs and dealer margin complicate that math. As of August 2026, combos are among the fastest-growing products on US event exchanges and a focus of the regulatory debate over how prediction markets differ from sports betting.

Scalar / range market

A market on a number rather than a yes/no — “How many rate cuts in 2026?” — typically implemented as a ladder of binary contracts (“exactly 0,” “exactly 1,” “2 or more”). The whole ladder read together gives you the market’s full probability distribution over the number, which is often more informative than any single contract in it.