Explainer · PicksByOdds

Why a 70-cent YES isn't a 70% win: decoding prediction market liquidity

A 70-cent YES only reads as a 70 percent chance when the order book behind it is deep, tight and recent; here is how spread, depth, volume and stale trades bend the number, and what to check before trusting it.

A YES contract showing 70 cents looks like a 70 percent chance of collecting a dollar. Sometimes it is close to that. Often the number on the screen is a midpoint, a stale last trade, or the footprint of one large order, and the price you can actually get is different. Liquidity is the gap between the number and the trade, and reading it is what separates a price you can act on from a price you can only look at.

What "70 cents" is actually showing you

A prediction market contract pays $1 if the event resolves YES and $0 if it resolves NO, so its price lives between 0 and 100 cents and reads as a probability. That part is solid.

What is less solid is which price you are looking at. Every contract has an order book: a list of bids (what buyers will pay) and asks (what sellers want). The displayed figure is usually one of three things: the last trade, the midpoint between best bid and best ask, or the best ask. In a busy market these sit within a cent or two of each other and the distinction barely matters. In a quiet one they can be far apart, and "70 cents" may describe a trade that happened hours ago at a level nobody is currently offering.

Why it is not a 70 percent win even at face value

Suppose the price really is 70 cents and you buy one YES. If the event happens you collect $1, a 30 cent gain. If it does not, you lose the 70 cents. For that bet to break even over many repetitions, the event has to happen at least 70 percent of the time. To make money, it has to happen more often than that.

So a YES bought at 70 cents pays off only if the true chance is above 70 percent, which is a different claim from "a 70 percent win." Any trading fee the venue charges comes out of the 30 cent payout and pushes the break-even higher still. Buying at the market price with no view of your own has an expected return of about zero before fees and slightly negative after them.

How thin books distort the number

Liquidity means how much you can buy or sell without moving the price. Three things go wrong when it is scarce:

  • The spread widens. In a deep market the best bid and best ask might be 69 and 71. In a thin one they might be 62 and 78. The "70" midpoint is a fiction in the second case: buying costs 78, selling gets 62, and no one is trading at 70.
  • Single orders move the price. If there are only a few contracts offered at each level, one buyer who wants a hundred contracts walks the price up through the book and leaves the displayed price well above where the next trader would value it. The biggest movers page often includes moves of this kind, where a small dollar amount produced a large jump.
  • Prices go stale. With no trades for a day, the last-trade figure reflects yesterday's information. News may have arrived and nobody has repriced yet.

None of these change the true probability of the event. They change how well the number tracks it.

Three ways to read the same 70 cents

Deep book. Bids and asks stacked within a cent or two of 70, steady trading on both sides, and a large total volume. Here 70 is a real consensus and you can trade near it in size. A move from 70 to 72 means people changed their minds.

Thin book. A wide spread with a few contracts at each level. The 70 might be one trader's position rather than a shared belief. Do not treat it as 70 percent; treat it as "somewhere between the bid and the ask," which may be a very wide range.

Shocked book. News hits, and the first traders to react are the fastest rather than the best informed. In a thin market the price can overshoot in either direction before slower, larger money comes in to correct it. A price seen within minutes of a headline is provisional.

What to check before trusting a price

  1. Spread. Look at the best bid and best ask, not just the headline number. If they are more than a few cents apart, the midpoint is an estimate rather than a price.
  2. Depth. How many contracts sit at the best bid and ask, and how quickly does the book thin out beyond them? Depth tells you how much you could trade before you become the reason the price moved.
  3. Volume and open interest. Total dollars traded and dollars currently held in positions are two figures every market page on this site shows. Both indicate how many people have real money behind the current level. The highest volume list is a quick way to find contracts whose prices carry the most information.
  4. Recency. When was the last trade? A price set an hour ago in an active market and a price set three days ago in a dead one are not the same kind of evidence.

The practical rule

Read a price as a probability only after you have read the book behind it. A deep, tight, recently traded 70 is close to a 70 percent estimate, and buying it is a bet that the true chance is higher. A wide, thin, stale 70 is a rough range with a 70 in the middle, and the only number that matters is what you would actually pay to get filled.

The contract pays a dollar or nothing. The market's estimate of how often it pays is what the price is trying to tell you. Liquidity determines how clearly it can say it.

Related reading: what implied probability actually means covers the other reasons implied and actual probability drift apart.

More explainers

Related reading

Get new posts in your inbox. No spam, unsubscribe anytime.