LMSR stands for Logarithmic Market Scoring Rule. Robin Hanson developed it as an automated market maker for prediction markets. The formula quotes a price for every outcome, accepts trades without waiting for another trader, and moves each price as shares are bought or sold. One liquidity setting controls how sharply prices react.
Traders meet LMSR in smaller forecasting markets, play-money markets, and market designs where continuous liquidity matters more than a conventional order book. It turns every trade into both a position and a probability update.
How does LMSR work?
LMSR keeps a running count of shares sold for every possible outcome. It converts those counts into prices with an exponential formula. When a trader buys one outcome, its share count rises, its quoted price rises, and the prices of the other mutually exclusive outcomes fall. The outcome prices continue to sum to one.
For a two-outcome market, Cultivate Labs writes the price of outcome one as `e^(q1/b) / (e^(q1/b) + e^(q2/b))`. The `q` values are outstanding shares. The `b` value controls liquidity. Traders do not need to solve the equation by hand, but they should understand what moves the quote.
The market maker charges the difference between its cost function before and after a trade. A larger order moves farther along that curve, so later shares can cost more than earlier shares in the same order. The mechanism always posts a quote. Execution does not depend on a matching limit order from another person.
What does the LMSR liquidity parameter do?
The LMSR liquidity parameter, written as `b`, sets the market's sensitivity to each trade. A smaller value makes prices react sharply to a small purchase. A larger value makes the quote more resistant, so a trader must buy more shares to produce the same probability change. Market operators choose this setting when they launch a market.
Cultivate Labs published a practical example using the same ten-share purchase across markets with different liquidity settings. The quoted probability moved by about 1 percentage point in one market, 4 points in another, and 26 points in the third. The order size stayed fixed. The liquidity parameter changed the response.
Traders should read an LMSR price together with the liquidity setting and recent activity. A 60 percent quote in a thin, reactive market may represent far less committed capital than the same quote in a deeper market. Price alone hides how hard it was to move.
Why do prediction markets use LMSR?
Prediction markets use LMSR when the operator wants every participant to receive a quote at any time. A conventional continuous double auction needs a buyer and seller whose prices cross. A thin market can show a wide spread or no executable order. LMSR supplies the other side through a formula, even before a crowd arrives.
Hanson's 2002 paper connects market scoring rules with group probability estimates. Each trader pays to move the public estimate and can profit if the final outcome rewards that move. The design lets later traders correct earlier prices while preserving a continuous public forecast.
That makes LMSR useful for internal forecasting programs and small communities. Large real-money venues often use order books instead. If you need the broader mechanics first, read what is a prediction market.
What are the limits of LMSR?
LMSR guarantees a quote, but the operator still chooses the liquidity setting and funds the market maker's potential loss. A setting that is too small lets one modest trade swing the forecast. A setting that is too large can make informed traders pay too much to correct a stale price. The formula guarantees availability. People still supply the judgment.
A trader can also mistake mechanical liquidity for genuine agreement. The quote reflects the share inventory and the configured curve. It does not reveal how many independent traders contributed information, how concentrated the positions are, or whether one participant caused the move.
Order-book venues expose bids, asks, spreads, and depth directly. Traders can study those mechanics in the Kalshi order book guide. The prediction market arbitrage guide explains how prices can diverge across venues even when contracts appear to cover the same event.
How should a trader read an LMSR price?
A trader should treat an LMSR price as the market maker's current implied probability, then inspect what it would cost to move that price. Check the liquidity parameter, the size of the proposed order, and the resulting quote after slippage. Those three values reveal more than the displayed probability alone.
Run the same test on the exit. LMSR keeps liquidity available, yet a position can still be expensive to unwind after the curve moves. Estimate the full round trip before sizing the trade. Event contracts can settle at zero, and continuous liquidity does not remove outcome risk.
Kairos is built for traders using live order books across Polymarket, Kalshi, and Predict.fun. Sub-second data, one cross-venue book, and real execution. Open the Kairos terminal.
Sources
Robin Hanson's paper provides the original market-scoring-rule design. Cultivate Labs supplies the two-outcome price equation, trade-cost calculation, and practical examples of the liquidity parameter. These sources explain LMSR mechanics and the operator's design choices. Individual markets still control their own implementation, currency, liquidity setting, and settlement rules.