Liquidity and Slippage in Prediction Markets: Read the Book Before You Trade

Liquidity in a prediction market is not the same as volume. It is the amount you can trade at or near the quoted price without moving it. Slippage is the difference between the price you saw and the average price you actually got. A market can trade millions of dollars in a day and still be thin enough that a $5,000 order walks the book several cents.

That distinction matters more in prediction markets than in most financial markets because the tick size is coarse. One cent is one percentage point of implied probability. In a 50-cent market, a three-cent slip is a 6% loss on the position before the event has even been decided.

The order book is a ladder, not a number

Every outcome has two sides. Bids are resting buy orders; asks are resting sell orders. The best bid and best ask are the top of the book. Behind them are additional price levels, each with its own size. Polymarket's order book API returns the full ladder, including the minimum order size and tick size for the market.Order book data

The top of the book is what the screen shows. The rest is what you actually trade against. If the best ask is 60 cents for 500 shares, a 500-share buy gets 60 cents. A 3,000-share buy does not.

Suppose the ask side looks like this: 500 shares at 60 cents, 2,000 shares at 61 cents, and 5,000 shares at 62 cents. You buy 3,000 shares. The first 500 cost $300. The next 2,000 cost $1,220. The last 500 cost $310. Total cost is $1,830, an average of 61 cents. The quoted best ask was 60 cents, so your slippage is 1 cent per share, or $30. That is 1.6% of the $1,830 you spent.

If you had assumed the whole order would fill at 60 cents, you would have underestimated the cost by $30. On a 3,000-share position, that is the difference between a trade that needs a 60% win probability and one that needs 61%.

Why thin markets are expensive

A thin market has few resting orders and wide gaps between levels. That produces three costs at once. The spread is wider, so you pay more to enter and receive less to exit. The depth is lower, so your own order moves the price. And the book can empty quickly when news arrives, leaving you unable to exit at anything close to the last mark.

Prediction markets are especially exposed to the third problem because resolution is binary. A market can trade for weeks with modest liquidity and then see a court ruling, a data release or an election call that changes the fair value by tens of cents in minutes. Market makers pull quotes when they cannot price the risk; Polymarket's market-making guidance treats stale-quote cancellation and inventory limits as core controls.Market making The traders left holding positions discover that the exit was never as deep as it looked.

Polymarket's documentation is explicit that the order book has no trading size limit but that large orders may move the price significantly, and it tells users to check depth before trading in size.Prices and order books That is a fair description of an exchange: there is no house maximum stake, but there is a market maximum that changes by the second.

How to measure liquidity

Start with the quoted spread. A one-cent spread on a 50-cent market is tight. A five-cent spread is wide. Convert it to a percentage of the midpoint so you can compare markets at different price levels: a 2-cent spread is 4% at 50 cents but 40% at 5 cents.

Then look at depth, not just the top level. Add up the shares available within one cent and two cents of the touch. Ask how much it would cost to buy or sell your intended size, using the full ladder. That number, not the quoted price, is your expected execution cost.

Volume is a weaker signal. High volume can mean genuine activity, but it can also reflect market-maker churn, incentive programs or wash trading. Depth and the cost to trade your size are better signals, though displayed orders can still be canceled. A market with $2 million of daily volume and $500 of resting depth is not liquid for a $10,000 order.

Finally, look at the number of trades and the frequency of book updates. A market with a few large trades per day behaves differently from one with continuous two-sided quoting. The latter is more likely to absorb an order without a gap.

Practical execution rules

1. Use limit orders. On Polymarket, every order is technically a limit order; a "market order" is just a limit order priced to execute immediately. That means you can set a worst-case price instead of accepting whatever the book offers.Prices and order books

1. Check the full book before sizing. If the depth within one cent is less than your order, either reduce the order or accept that you are paying up.

1. Break large orders into clips. A series of smaller limit orders gives the book time to replenish and reduces the signal you send to other traders. It also reduces the chance of filling the worst levels in one sweep.

1. Join the bid instead of crossing the ask when you are not in a hurry. You give up immediacy but save the spread. On a 4-cent spread, that is 8% of a 50-cent position.

1. Avoid trading immediately before known catalysts unless you intend to take event risk. Liquidity is usually thinnest when uncertainty is highest.

1. Account for fees in the execution price. A fill that looks 1 cent better can be worse after taker fees. Polymarket's fee formula depends on price, so the same share count costs more at 50 cents than at 20 cents.Fee schedule

1. Do not assume displayed depth is real. Orders can be canceled before you reach them, especially in thin markets. A quote is a signal, not a contract, until it trades.

The liquidity trap

The most dangerous version of illiquidity is the one that appears after you buy. A position can be marked at 70 cents on the screen while the realizable bid is 64 cents. If you need to exit, you discover the difference. This is why risk management in prediction markets has to use the bid, not the midpoint, when estimating portfolio value.

Liquidity also interacts with resolution. A market that is liquid today may become untradeable once the outcome is known but before redemption. Your capital is locked for the settlement process, which can take hours or days. Plan for that delay rather than assuming you can redeploy the cash immediately.

None of this means small markets are untradeable. It means the trade has to be sized to the book. A $500 order in a thin market can be perfectly reasonable; a $50,000 order in the same market is a different trade with a different risk profile.

A simple checklist

Before any order, write down four numbers: the best bid, the best ask, the depth within one cent, and the average fill price for your size. If the average fill is more than a cent away from the touch, you are paying for liquidity. Decide whether the edge still exists after that cost. If it does not, the market is telling you the trade is too small to matter or too large for the book.

Liquidity is not a background condition. It is the price of admission, and in prediction markets it can change faster than the probability.