Spread Capture

Earning the bid-ask spread through limit orders in binary contract markets

What is spread capture?

Spread capture — also known as market making — is the practice of placing limit orders on both the buy and sell side of a market, earning the difference (the spread) when both orders fill. It is one of the oldest trading strategies, dating back to the specialist system on the New York Stock Exchange.

In a prediction market context: if a contract has a best bid of 85 cents and a best ask of 91 cents, a trader could post a buy limit at 86 cents and a sell limit at 90 cents. If both fill, the trader earns 4 cents per contract, minus fees.

How it works in practice

A spread capture strategy on Kalshi typically involves:

1. Identify wide-spread markets — Look for markets where the bid-ask spread is wider than exchange fees, indicating there is room to profit by providing liquidity.
2. Place limit orders inside the spread — Post a buy order above the current best bid and a sell order below the current best ask, tightening the spread and offering a better price to other participants.
3. Manage inventory — If one side fills but not the other, the trader has directional exposure. Managing this inventory risk is the central challenge of market making.
4. Repeat — Once both sides fill (or the position is closed), the process restarts.

Academic foundations

The Glosten-Milgrom model

Glosten and Milgrom (1985) developed the foundational model of why bid-ask spreads exist. In their framework, a market maker faces two types of counterparties: informed traders (who have better information about the true value) and noise traders (who trade for other reasons). The spread compensates the market maker for the losses they expect from informed traders.

Inventory risk

Ho and Stoll (1981) modeled how a dealer adjusts bid and ask prices based on their current inventory. A market maker who accumulates a large position on one side faces the risk of a price move against them. Their model shows that optimal spreads widen as inventory risk increases.

Optimal market making

Avellaneda and Stoikov (2008) developed a continuous-time model for optimal market making in a limit order book. Their framework balances the revenue from capturing spreads against the risk of adverse price moves and inventory accumulation. The optimal strategy involves skewing quotes based on current position and time horizon.

Binary contract specifics

Spread capture in prediction markets differs from equity markets in several important ways:

• Terminal settlement — Prediction contracts settle at 0 or 100 cents. Unlike stocks, where prices fluctuate continuously, a binary contract has a known endpoint. This creates discontinuous risk — a position can go from 90 cents to 0 cents in an instant at settlement.
• No short selling mechanics — On Kalshi, selling NO is buying YES and vice versa. The relationship between YES and NO prices means a market maker can always express both sides, but the mechanics differ from equity market making.
• Fee structure — Exchange fees on Kalshi are charged per trade. If the captured spread is 4 cents but round-trip fees are 2-3 cents, the net capture is only 1-2 cents — a narrow margin for error.
• No obligation to quote — Unlike designated market makers on traditional exchanges, Kalshi traders have no obligation to maintain continuous quotes. They can choose which markets to provide liquidity in and when.

When spread capture works

The strategy tends to be more viable in specific conditions:

• Low-information markets — Markets where no participants have a strong informational edge, so the spread compensates for liquidity risk rather than adverse selection.
• Stable prices — Markets where the true probability is relatively stable (not rapidly shifting due to incoming information), reducing the risk of being picked off.
• Sufficient volume — Enough trading activity that both sides of the order can reasonably expect to fill within a useful time window.

Key considerations

• Asymmetric payoffs — Spread capture earns small amounts per round-trip but faces the risk of being stuck with a position that goes to 0 or 100. One settlement against you can erase many successful captures.
• Requires active management — Unlike a buy-and-hold directional strategy, spread capture requires monitoring open orders, managing fills, and adjusting quotes.
• Capital efficiency — Capital is tied up in resting orders. If orders don't fill, the capital earns nothing.
• Competition — If other participants adopt the same strategy, spreads tighten and the opportunity diminishes. Market making is inherently competitive.