YES/NO Prediction Markets Explained

Every prediction market, however complicated it looks, is built out of one component: a YES/NO contract. Understanding that single building block is enough to trade almost anything on any venue.

The contract in one line

A YES contract pays $1 if the stated event happens, and $0 if it does not. A NO contract does exactly the reverse. Their prices should add to about $1, because between them they cover every possibility.

A worked example

Say "Will the central bank cut rates at the next meeting?" has YES at $0.72 and NO at $0.29.

• Buying 100 YES costs $72. If they cut, you receive $100 — a $28 profit. If not, you lose the $72.
• Buying 100 NO costs $29. If they hold, you receive $100 — a $71 profit. If they cut, you lose the $29.
• The extra cent in $0.72 + $0.29 is the spread the venue and the market makers keep.

Selling is just buying the other side

You do not need a short-selling facility. If you hold YES and want out, you sell your YES contracts to another trader. If you have no position and think the event will not happen, you buy NO. Many interfaces show both as one toggle, which hides how simple it is underneath.

Multi-outcome markets

A five-candidate election is five YES contracts sharing one event. The prices should sum to roughly 100%, and buying NO on a candidate is economically the same as buying a slice of every other candidate. Some venues let you do that directly; others leave you to construct it.

Scalar and bracket markets

Questions about a number — inflation, a price level, a total — are usually split into brackets: "under 2.5%", "2.5–3.0%", "over 3.0%". Each bracket is its own YES/NO contract and, again, they should add to about 100%.

Reading the rules like a trader

• What exactly counts as the event? Announced, confirmed, or reported by a specific source?
• What is the deadline, in which time zone?
• What happens if the event is cancelled, postponed or ambiguous? Void, NO, or a delayed resolution?
• Which source has the final word, and is there a dispute window?

Most avoidable losses in prediction markets are not bad forecasts. They are traders who were right about the world and wrong about the wording.

Risk, plainly

Your maximum loss is what you paid, and it is a total loss — there is no partial refund for being nearly right. A 90-cent contract looks safe and loses 90 cents one time in ten. Size accordingly.