Prediction MarketsSeptember 7, 2026

What Is a Market Maker? How Liquidity Gets Made

AK

Austin Kennedy

6 min read

A market maker is a firm or trader that continuously posts prices to buy and sell an asset. Those standing bids and asks give other traders someone to trade with. In prediction markets, market makers quote outcome contracts, manage the positions created by fills, and carry the risk when prices move against their inventory.

What does a market maker do?

A market maker keeps orders on both sides of an order book. The bid states the highest price the maker will pay. The ask states the lowest price the maker will accept. A trader who wants immediate execution can sell into the bid or buy at the ask.

The SEC's Investor.gov glossary defines a market maker as a firm that stands ready to buy or sell at publicly quoted prices. Polymarket's market-making documentation applies the same mechanism to outcome contracts: continuous bids and asks deepen the book, tighten the spread, and let other traders execute.

Without resting quotes, a buyer has to wait for a seller who wants the opposite trade at the same moment. Market makers bridge that timing gap by committing capital before the next trader arrives.

How do market makers make money?

Market makers seek to earn the spread between their bid and ask. A maker quoting 47 cents bid and 49 cents ask can buy at 47 and sell at 49, collecting 2 cents if both orders fill and the contract's fair value does not move first. Makers collect that difference in exchange for supplying capital and taking execution risk.

Market makers do not earn a guaranteed profit from the spread. One side can fill while the other sits untouched. New information can then move fair value against the position before the maker exits. Polymarket also documents liquidity rewards and maker rebates for qualifying activity, but each program has its own eligibility and payout rules.

Serious market makers measure realized spread after adverse selection, fees, rebates, and hedging costs. The displayed spread is the starting number. The fills decide the result.

How does market making work in prediction markets?

Prediction market makers quote contracts that settle according to a stated outcome. A YES contract priced at 52 cents and its complementary NO side encode the same market view from opposite directions. Polymarket's documentation gives the direct equivalence: buying NO at 48 cents has the same economics as selling YES at 52 cents.

The maker estimates fair value, places bids and asks around it, and updates those orders when news, order flow, or inventory changes. Order book depth shows how much size rests at each price. A tight top-of-book spread can still hide thin liquidity one level deeper.

Binary contracts add settlement risk to the usual trading problem. The market's resolution source and rules determine the final payout. A maker must price the event, read the rulebook, and keep enough capital or outcome tokens available to support open orders.

What risks do market makers carry?

A maker takes inventory risk with every fill. If traders keep buying one side, the maker accumulates the opposite exposure. The maker can adjust prices or order sizes to attract offsetting flow, reduce the position elsewhere, or stop quoting.

  • Adverse selection: a faster or better-informed trader executes against a stale quote.
  • Inventory risk: one-sided fills leave the maker exposed to the event outcome.
  • Operational risk: delayed data, rejected orders, or a failed cancel can leave an old price live.

Polymarket tells market makers to cancel stale quotes, track real-time order updates, validate prices against the book midpoint, set size limits, and maintain a kill switch. Those controls exist because market making turns small execution failures into live positions. Traders also need a sizing rule. Our prediction market position sizing guide covers that decision.

Why do market makers need speed and one book?

Market makers compete on the freshness of their prices. A quote based on old data gives faster traders a free option to trade only when it benefits them. Real-time market data, fast cancels, batch order entry, and current fill updates keep the maker's view aligned with the venue.

Traders provide prediction market liquidity across Kalshi, Polymarket, and Predict.fun. A trader watching one venue can miss a better price or a move that already happened elsewhere. Kairos aggregates the venues into one book with sub-second data, global best bid and ask, and low-latency execution. Read the prediction market terminal guide for the full execution stack.

Quote the market with the full book in view. Trade Kalshi, Polymarket, and Predict.fun on Kairos. See you in the order books.

Sources

Frequently asked questions

A market maker keeps prices available for buyers and sellers by posting bids and asks. Other traders can execute against those orders immediately, while the market maker manages the resulting inventory and price risk.
A broker routes or arranges a customer's trade. A market maker trades as the counterparty using its own quoted prices and capital. One firm can perform both roles in different parts of its business, but the functions are distinct.
Market makers seek to capture the difference between their bid and ask, called the spread. Some venues also offer qualifying makers rebates or liquidity rewards. Inventory losses, stale quotes, fees, and hedging costs can exceed that income.
A matching engine can pair compatible orders, but it cannot create an order on either side. Market makers commit capital and post resting bids and asks before another trader arrives, giving the matching engine prices to execute.
A prediction market maker quotes buy and sell prices for outcome contracts such as YES and NO. The maker updates those prices as information, order flow, inventory, and the time remaining before resolution change.
Yes. Market makers can lose when prices move against accumulated inventory, informed traders hit stale quotes, or operational failures leave old orders exposed. The spread compensates them for risk. It does not remove it.

Related reading