Binary, multi-outcome and range markets are the three main ways a prediction market can frame a question: a binary market asks one yes-or-no question, a multi-outcome market lists several outcomes where only one can win, and a range market splits a number into bands. Every one of them is built from the same basic unit, the event contract, which pays $1 if its outcome happens and $0 if it does not. What changes from one type to the next is how many contracts can pay out and, as a result, how you should read the prices you see on screen.

Key takeaways

  • A binary market asks one yes/no question, and the Yes and No prices together equal $1.
  • A multi-outcome market lists several outcomes for the same event, and only one of them can win.
  • A range market turns a number into outcomes: either separate bands (buckets) or "above X" thresholds.
  • The same price, say a hypothetical 30 cents, means something different in each type, so always check what exactly the contract pays on.
  • The bid-ask spread and the amount of liquidity limit how precise any displayed price really is.

What Is an Event Contract?

An event contract is a contract tied to a real-world question that settles at $1 if the answer goes its way and at $0 if it does not. Its price, which moves between a few cents and just under a dollar, is set by people trading with each other, and many traders read it as a rough probability estimate.

Not every question has two answers. "Will it rain tomorrow?" is yes-or-no, "Who will win the championship?" has many answers, and "How high will a number be at month-end?" has a whole range. So binary markets cover one yes/no question, multi-outcome markets a list of competing answers, and range markets the possible values of a number. New to the topic? Start with how prediction markets work or the short definition of a prediction market.

As of October 2026, market structures and rules change, so check the rules of each market before trading.

Binary Markets: One Question, Yes or No

The binary market is the simplest and most common form of prediction market. According to Kalshi, each market has a YES contract and a NO contract covering both possible results. When the question is settled, the right side pays $1 per contract and the other side pays nothing.

Because exactly one side wins, the two prices are linked. On Kalshi, contracts are priced between $0.01 and $0.99, and a YES and its matching NO always add up to $1: if YES costs $0.60, NO costs $0.40. On Polymarket, shares trade between $0.00 and $1.00, and the platform describes the price as an implied probability, so a share at $0.50 is read as a 50% chance. Polymarket does not set these prices itself. They come out of a central limit order book, where buyers and sellers post orders, and the first price of a market appears when a Yes order and a No order that add up to $1.00 are matched, for example Yes at $0.60 and No at $0.40.

Reading a binary price as a probability

For illustration, a Yes share at 62 cents is usually read as "about a 62% chance this happens." That is a useful shortcut, but it is an estimate, not a fact: it depends on liquidity, fees and who happens to be trading, and a small market can move a lot when more people join.

HYPOTHETICAL example (not a real market): imagine a binary market asking whether a city will get more than one inch of snow on a given day. The Yes share trades at 62 cents, so No is roughly 38 cents.

  • If you bought one Yes share at 62 cents and it snows more than an inch, the share settles at $1, so the difference is 38 cents before any fees.
  • If it does not, the share settles at $0 and the 62 cents are lost.

The price shows how the crowd currently balances those two scenarios, not what will happen. For more, see how to read Polymarket odds and prices and the short explainer on implied probability.

Multi-Outcome Markets: Several Outcomes, One Winner

Many questions have more than two answers: candidates in an election, nominees for an award, teams in a tournament. A multi-outcome market groups these answers under one event. Each outcome usually works like its own small binary market with a Yes and a No, but at the end only one outcome resolves Yes.

Mutually Exclusive Events: Why Only One Outcome Can Win

In a prediction market, mutually exclusive events are outcomes that cannot happen together: if one of them is true, all the others must be false. "Candidate A wins" and "Candidate B wins" are mutually exclusive in a single-winner race, because there can only be one winner. That is the defining feature of a multi-outcome market.

This affects prices. If exactly one outcome pays $1 and the rest pay $0, holding one Yes share in every outcome returns exactly $1. In an efficient market, the Yes prices of all outcomes should add up to about 100%.

How Polymarket Handles Multi-Outcome Events

Polymarket's documentation describes a mechanism called "negative risk" for events with several mutually exclusive outcomes. The idea follows from the logic above: if only one outcome can win, then betting that one outcome will not happen is economically close to betting that one of the others will. Polymarket lets a No share in one market of the event be converted into one Yes share in every other market of the same event. Its own example: someone holding No on "Other" can convert it into Yes on each named candidate.

Some events are created before every outcome is known, such as a race new candidates may still enter. For these, Polymarket uses augmented negative risk, which separates:

  • Named outcomes, listed at creation or clearly added later.
  • Placeholder outcomes, reserved for outcomes not yet named.
  • An explicit "Other", covering everything not named.

Polymarket's guidance is to trade only on named outcomes and to ignore placeholder outcomes until they are named or until the market resolves. If the eventual winner was never named, the market resolves to "Other." For readers, the practical lesson is to always check which outcomes are actually named and what "Other" covers. For more on how the final answer is decided, see our guide on how Polymarket markets resolve.

Do the Prices Add Up to 100%?

In theory, yes; in practice, rarely exactly. Real prices can drift above or below 100% because each outcome has its own bid-ask spread, fees make tiny corrections not worth the effort, and less popular outcomes often have thin liquidity.

Compare this with a sportsbook, which builds its margin, called the vigorish or "vig," into its odds, so its implied probabilities add up to more than 100%. So-called no vig odds are the same prices with that margin removed. Prediction-market prices are set by traders, not a bookmaker, but can still show a small excess or gap.

HYPOTHETICAL example (not a real market): a made-up event lists three named outcomes and "Other." The Yes prices are 55, 30 and 12 cents for the named outcomes and 6 cents for "Other."

  • Total: 55 + 30 + 12 + 6 = 103 cents, so the prices add up to 103%.
  • The extra 3% does not mean the market expects more than one winner. It reflects spreads, fees and uneven liquidity across outcomes.
  • To compare outcomes more fairly, some readers divide each price by the total (for example, 55 / 103 ≈ 53%). This is a rough normalization, not an official figure.

Range Markets: Outcomes That Are Bands of a Number

Some questions are about a number: a temperature, an economic figure or an interest-rate decision. Generic "fed rate odds" questions are a typical example. A range market turns that number into a set of contracts that each cover part of the possible values.

According to Kalshi's API documentation, markets are either binary (settling Yes or No) or scalar (settling to a continuous number), and a "strike type" describes how settlement is evaluated: greater, greater-or-equal, less, less-or-equal, "between" (a value inside a range with a floor and a cap), plus functional, custom and structured types. This describes what the system supports, not what is offered to US users at any moment, so check each market's rules. See our Kalshi review for more.

Buckets vs. "Above" Thresholds

Range markets usually come in two shapes, and mixing them up is one of the most common reading errors.

Buckets ("between"). Each contract covers one band of values, such as "between 2.0 and 2.5." Bands do not overlap, so exactly one bucket resolves Yes. This is what many people mean by a Kalshi bracket: a ladder of side-by-side brackets that together cover the whole range. Because the buckets are mutually exclusive, their prices should add up to about 100%, just like a multi-outcome market.

Thresholds ("above"). Each contract asks whether the number will be above a certain level, such as "above 2.0," "above 2.5" and "above 3.0." These are nested, not exclusive: if the highest threshold pays, every lower threshold pays too. Their prices fall as the threshold rises, and they do not add up to 100%. A 55% price on "above 2.5" is the chance of being anywhere above 2.5, not the chance of landing in one particular band.

According to one third-party analysis published by Turbine in 2026, threshold contracts traded between 21 and 52 times the volume of the matching bucket markets in that author's data. That is one data point from one source, not a rule, but it suggests many traders work with thresholds and then do the math to figure out the bands.

A Range Market Is a Probability Distribution

The most useful way to read a range market is as a whole. In a prediction market, a probability distribution is the full set of prices across all the bands of a number, showing where the crowd thinks the value is most likely to land: tall where traders see likely values, low at the extremes.

HYPOTHETICAL example (not a real market or forecast): imagine a made-up numeric question with these "above" thresholds:

  • Above 2.0: 90%
  • Above 2.5: 55%
  • Above 3.0: 15%

By subtracting neighbouring thresholds, you can rebuild the buckets:

  • 2.0 or less: 100% − 90% = 10%
  • Between 2.0 and 2.5: 90% − 55% = 35%
  • Between 2.5 and 3.0: 55% − 15% = 40%
  • Above 3.0: 15%

The four buckets add up to 100%, and the most likely band is 2.5 to 3.0. Thresholds and buckets describe the same distribution from two angles; real markets are messier because each contract has its own spread. Scalar contracts that settle directly to a number also exist as a market type in Kalshi's API; this guide does not assume how many are offered.

Comparing the Four Structures

TypeQuestion it asksHow many outcomes can winWhat the price meansCommon trap
BinaryWill this one thing happen?One side: Yes or NoRough chance the answer is YesTreating the price as a certainty instead of an estimate
Multi-outcomeWhich of these outcomes will happen?Exactly one outcomeRough chance that this specific outcome winsExpecting prices to add up to exactly 100%, or ignoring what "Other" covers
Range (bucket)Which band will the number land in?Exactly one bucketRough chance the value falls inside this bandForgetting where each band starts and ends
Range ("above" threshold)Will the number be above this level?Every threshold below the final valueRough chance the value is above this levelReading a threshold price as the chance of a single band

How the Price Reading Changes From One Type to the Next

The number on screen looks the same in every market, but its meaning shifts:

  • In a binary market, the Yes price and the No price are two views of the same question. Look at one, and you know roughly where the other sits.
  • In a multi-outcome market, a price only makes sense next to the others. In a hypothetical case, 30 cents on one candidate means something different if the leader is at 40 cents or at 65 cents.
  • In a bucket market, each price covers a narrow band. A low price on every bucket does not mean the market expects nothing; it may just mean many bands share the probability.
  • In a threshold market, each price covers everything above a line. Subtract neighbouring thresholds to see the chance of a specific band.
  • In every type, many people loosely call these numbers "Polymarket odds" or "Kalshi odds," but they are prices first. Converting them into percentages is a reading, not a promise.

The Bid-Ask Spread: Why the Price You See Is a Midpoint

The bid-ask spread is the gap between the highest price someone is willing to pay (the bid) and the lowest price someone is willing to sell at (the ask). The midpoint is the value halfway between them. The number shown on a market page is often not the last trade, but a summary of this spread.

According to Polymarket's help center, the price displayed for a market is the midpoint of the bid-ask spread. If the spread is wider than $0.10, the displayed price switches to the last traded price instead. Polymarket gives this example: with a bid of 34 cents and an ask of 40 cents, the midpoint is 37 cents, displayed as 37%. If you wanted to buy right away, though, you would pay closer to 40 cents, and if you wanted to sell right away, you would receive closer to 34 cents. Large orders can also move the price, because they use up the orders sitting at the best levels.

A thin market, with few orders, usually has a wide spread. That signals uncertainty or low interest, and it means the displayed price is less precise. This matters most in multi-outcome and range markets, where less popular outcomes or extreme buckets can have very wide spreads even when the main outcomes are tight. The glossary entries for liquidity, spread and market maker explain who supplies these orders and why it matters. Fees are a separate cost; our fee calculator shows how they work.

Common Mistakes When Reading Market Types

  • Ignoring the exact wording. "Above 2.5" and "between 2.5 and 3.0" are different contracts. Always read the title and the rules, including the resolution source and the date.
  • Adding threshold prices together. Nested "above" contracts do not add up to 100%. Only mutually exclusive outcomes, such as buckets or named candidates, should come close.
  • Overlooking "Other" and placeholders. In events that are still open to new outcomes, an unnamed winner can resolve the market to "Other."
  • Trusting prices in thin markets. A price on an outcome with a wide spread and little volume can be far from where a real trade would happen.
  • Reading a price as a forecast. A 62% price is an estimate shaped by liquidity, fees and who is trading. It is not a guarantee, and it is not advice.

Key terms

  • Event contract: a contract tied to a real-world question that pays $1 if its outcome happens and $0 if it does not. Its price is set by traders.
  • Binary market: a market with one yes-or-no question, where the Yes and No prices together equal $1.
  • Multi-outcome market: a market that groups several mutually exclusive outcomes of one event, where only one outcome resolves Yes.
  • Range market: a market that splits a number into outcomes, either as non-overlapping buckets or as nested "above" thresholds.
  • Bid-ask spread: the gap between the best buy price and the best sell price. A wide spread usually means the market is thin or uncertain.

Want to see the three market types in action?

Browsing live markets is a calm way to recognize binary, multi-outcome and range structures and to notice how their prices behave. You do not need to trade to learn from them. If you decide to go further, start with our beginner guide on how to trade on Polymarket.

Sources

  • Kalshi, Working with Event Contracts — https://news.kalshi.com/p/what-are-event-contracts
  • Kalshi API documentation, Get Markets — https://docs.kalshi.com/api-reference/market/get-markets
  • Polymarket documentation, Negative Risk Markets — https://docs.polymarket.com/developers/neg-risk/overview
  • Polymarket Help Center, How Are Prices Calculated? — https://help.polymarket.com/en/articles/13364488-how-are-prices-calculated
  • Polymarket documentation, Prices & Orderbook — https://docs.polymarket.com/concepts/prices-orderbook
  • Wikipedia, Prediction market — https://en.wikipedia.org/wiki/Prediction_market
  • Turbine, Automating Strike Ladders on Kalshi — https://www.turbinefi.com/blog/automating-strike-ladders-kalshi-2026