The Hidden Liquidity Problem in Prediction Markets

On Polymarket, the price you see is rarely the price you can get — and that gap is where most traders silently lose their edge.

The interface looks clean. A binary question, two prices, a probability that updates in real time. Yes at 63 cents. No at 37 cents. It feels like a market the way a stock chart feels like a market. Visible, continuous, executable.

It isn’t.

The Quoted Price Is Not The Tradeable Price

The number on the screen is the best bid or best offer. It is not the price at which size can be moved. It is the price at which a single small order can be filled, often by a market maker who is quoting tight on the top of book and much wider underneath.

Click on the order book and the structure becomes visible. The first level might hold a few hundred dollars. The second level is three cents away. The third is six cents away. By the time you have absorbed a thousand dollars of size, you are no longer buying at 63. You are buying at an average closer to 65 or 66, depending on the market.

This is not a defect. It is the reality of any market with finite participation. But the interface does not show it until you are already moving size, which is why most traders only notice it once they have crossed the spread, paid the slippage, and started wondering why their thesis was correct and their P&L was not.

Why Spreads Are Wide

Prediction markets attract relatively low capital compared to the events they cover. A market on a national election might trade hundreds of millions over its lifetime, but at any given moment, the depth available within one or two cents of the mid-price is often a small fraction of that.

There are reasons for this. Market makers face binary settlement risk. The contract resolves to one or zero, and any inventory carried into resolution is fully exposed. To compensate, market makers widen their quotes and reduce their depth. This is not greed. It is risk pricing.

The thinner the market, the wider the protective buffer. The wider the buffer, the harder it is for a directional trader to enter at attractive prices and exit at attractive prices. Both sides of the round trip pay the spread.

For markets that trade for months, the cumulative spread cost can exceed the actual edge in the thesis. A trader who is structurally correct about the underlying probability can still lose money because the market never offered the price they needed at the size they wanted.

Time As A Position

Illiquidity does something subtle to a position. It removes the option to exit cleanly. Once the trade is on, the question is no longer whether the thesis is right. The question is how long the trader can hold it.

In a liquid market, the holding period is a choice. In an illiquid one, it is a constraint. The trader cannot exit at a fair price, so they wait. They wait for new participants to arrive, for news to move the contract, for the order book to thicken. They wait for the market to come to them, because they cannot afford to chase it.

This is where time compounds without decision. The trader has not made a new choice. They have not adjusted their thesis. They are simply still in the position because the alternative was paying through the spread to leave. Each hour the position remains open accumulates exposure, opportunity cost, and psychological weight. Holding becomes a position in itself.

A trader who plans for a two-week hold often ends up in a three-month one. Not because the thesis changed, but because the exit did not exist on the timeline they assumed.

Where The Gaps Are Most Visible

The liquidity problem is most observable in markets with niche or specialized resolution criteria. A general election market with widespread attention will have tighter quotes and more participants. A market on a specific regulatory ruling, a corporate event, or a less-followed sports outcome will trade with wide spreads and shallow depth for most of its lifecycle.

This is where you can trade prediction markets on Polymarket and observe the structural friction directly. The order book is public. The depth is visible. The trader can quantify, before entering, exactly how much of a haircut they will take to acquire a meaningful position and how much they will pay to leave it.

What looks like a market with a 2-cent spread at the top often has a 6-cent realized spread for any position of consequence. The interface does not lie. It just shows the best case, and the trader’s mental model fills in the assumption that the best case is the normal case.

It is not the normal case. It is the smallest case.

Resolution Risk And The Exit Window

Every prediction market has a resolution date. As that date approaches, two things happen at once. The probability distribution tightens, because the outcome becomes more certain. And the liquidity profile changes, because participants begin closing positions and market makers reduce their inventory exposure.

In the final hours before resolution, the spread can compress dramatically if the outcome is clear. Or it can widen dramatically if it remains contested. The trader who waited too long to exit can find that the exit window they assumed would exist has narrowed faster than they expected.

This is a different kind of liquidity risk. It is not the static depth of the order book. It is the dynamic shrinking of the exit window as the contract approaches settlement. A position that was tradeable at fair prices three weeks before resolution may be untradeable at fair prices three days before.

The trader who built a thesis without modeling the exit cost has built half a trade.

What Edge Actually Requires

In equity markets, edge is often framed as predictive accuracy. The trader who correctly forecasts more outcomes than the market does, profits over time. The mechanics of execution are assumed to be a solved problem.

In prediction markets, this assumption breaks down. The trader who is more accurate than the market still has to acquire exposure to that accuracy at a price that leaves room for the spread, the depth cost, the holding period, and the exit. If any one of these structural costs is mispriced in the planning phase, the realized return on the thesis collapses.

Edge in prediction markets is therefore a composite. Part of it is forecasting. Part of it is execution. Part of it is patience to wait for liquidity to appear at acceptable prices rather than crossing the spread out of impatience. A trader who has solved the forecasting problem but not the execution problem will produce a trade log full of correct calls and uncorrelated P&L.

The Quiet Cost Of Crossing

Every time a trader crosses the spread, they pay a structural tax. In a tight market, the tax is small. In a wide one, it is the dominant cost of the trade.

A trader who crosses on entry and crosses on exit pays the spread twice. If the spread is six cents on a contract priced near 50, the round trip costs roughly twelve cents of the dollar of resolution value. That means the trader needs the contract to move more than twelve cents in their favor just to break even on the execution costs, before accounting for the thesis being correct at all.

Most theses do not have that much room. The structural cost consumes the edge.

The traders who consistently extract value from prediction markets are not the ones with the strongest convictions. They are the ones who refuse to cross when crossing is expensive. They post limits. They wait. They accept the possibility of not getting filled. They treat the spread as the most important number on the screen and the quoted price as a suggestion.

The Visible And The Real

Prediction markets are a useful study in the difference between a market that looks tradeable and a market that is tradeable in the sense a directional trader needs.

The interface is visible. The depth is visible. The spread is visible. None of it is hidden, in the sense of being concealed. But it is hidden in the sense of being routinely ignored by traders who focus on the probability number and the binary thesis and skip the structural mechanics underneath.

The edge is not in being right about the world. The edge is in being able to express that rightness at prices that survive the round trip.

A trader who internalizes this stops seeing prediction markets as a pure information game. They start seeing them as a market microstructure problem with an information layer sitting on top. The information layer is where the conversation happens. The microstructure layer is where the money is made or lost.

The price you see is the price the market offers to a tiny order. Every other order pays a different price, and that difference is the actual cost of having a view.

More from SwapHunt

Long-form observations on structure, behavior, and timing.

Trade prediction markets: Polymarket — Probability-driven markets on real-world events.

Ebooks:

📘 Quiet Edges — On tempo, structure, and optionality

📗 Reading the Market, Not the News — On structure, behavior, and second-order effects

📙 When Not to Trade — On decision-making under uncertainty

Follow @SwapHunt for daily observations.

This content is for educational purposes only. Not financial advice.

---

The Hidden Liquidity Problem in Prediction Markets was originally published in Coinmonks on Medium, where people are continuing the conversation by highlighting and responding to this story.