A binary contract is the simplest tradable forecast: one event, two outcomes, a fixed payout at settlement. Almost every headline market you see on regulated U.S. venues, crypto platforms, or academic exchanges is built from this YES/NO pair. Understanding binaries is the key to reading every other product structure in this module.
What you are actually buying
When you buy YES, you are purchasing a claim that pays $1 per share if the stated event occurs, and $0 if it does not. Buying NO is the mirror image: it pays $1 if the event does not occur. You pay a price today—often quoted in cents on the dollar—and your maximum loss is typically what you paid in; your upside on a winning share is $1 minus that price, before fees.
A YES price of 62¢ is usually read as an implied 62% chance the event resolves YES, as discussed in the foundations module on price and probability. That mapping is only as good as the liquidity you can trade at and the clarity of the resolution rules. The mid on a wide book is not your fill; the ask to buy YES is.
The complement identity
In a well-functioning binary on a single venue, YES and NO should be nearly perfect complements:
YES price + NO price ≈ $1.00
If both sides are expensive—say YES at 64¢ and NO at 37¢, summing to 101¢—you are paying a small overround, similar to vig on a sportsbook. If the sum is below $1, sophisticated traders may buy both legs to lock a near-riskless box, subject to fees and identical settlement rules. Persistent gaps across venues often mean different resolution text, not free money.
When YES looks expensive, check the NO ask. The cheaper risk expression is often the overlooked leg.
A minimal payoff walkthrough
Suppose you buy 1,000 YES shares at 41¢ on “the Fed cuts 25 basis points at the March meeting.” You pay $410 now. If the cut happens, you receive $1,000 at settlement—a profit of $590 before fees. If it does not, you receive $0 and lose the $410 you paid.
Break-even thinking is the bridge between trading and forecasting: at 41¢, you need the event to happen more than 41% of the time in expectation to justify the YES buy, ignoring fees and risk preference. If your model says 48%, the raw edge per share is about 7¢; a two-cent round-trip fee might leave +5¢—still positive, but size should respect variance and headline risk, as in the chapters on expected value and Kelly sizing.
How binaries trade in practice
On a central limit order book, you walk the ladder when size exceeds depth at the best ask. On an automated market maker, the pool price is a starting point; a 5,000-share buy moves the curve. Comparing a crypto pool quote to a regulated best ask without matching size and fees is a common mistake when hunting cross-venue gaps.
You can hold to settlement if your edge is in the final outcome, or sell before expiry if news reprices the contract. Buying YES at 40¢ and selling later at 52¢ when debate shifts fair value captures profit without waiting for the meeting. Your forecast log should reflect whether you treat the exit as an updated belief or a pure trade.
Resolution is part of the instrument
Binary payoff is only $1 or $0 after the rulebook event occurs. Cutoff timezone, official data source, and exact wording—“incumbent wins” versus “Party X wins”—can split otherwise similar headlines into different contracts. Always read the Rules tab before sizing; the same tweet can mean YES on one market and irrelevant noise on another.
Historical retail platforms sometimes used platform-specific withdrawal or candidate rules that differed from national books—arb without rule parity is betting on clerical errors.
Portfolio thinking beyond one ticker
Ten YES positions on the same narrative cluster—same party, same macro day—are not ten independent bets. Correlation and joint probability bounds apply: a poll shock can move “wins Pennsylvania” and “wins Michigan” together. Size binaries as clusters, not isolated legs on each screen line.
Fréchet-style thinking helps: given only marginal probabilities on two events, the joint probability of both living in a coherent world is bounded. You cannot treat ten correlated YES tickets as ten unrelated coin flips.
Platform flavor without changing the math
Regulated U.S. event contracts on order-book venues emphasize published fee schedules and CLOB depth on marquee macro and politics. Crypto-native YES tokens may settle in stablecoin with AMM-first liquidity on long tails. Academic-style shops with per-market share caps can leave YES+NO sums distorted longer because arb size is capped.
The economics are the same; friction schedules differ. Expected value must use your venue’s executable prices.
Holding to expiry versus trading out
Holding makes sense when your edge is in the final outcome and fees favor patience. Trading out makes sense when news moves fair value before the event—your edge was timing, not only terminal probability. Boxing YES and NO when both legs are cheap relative to $1 is a structural theme from arbitrage chapters, but only when settlement rules match.
Expected value in one paragraph
If your model says 48% and you pay 41¢ for YES, expected value per share is roughly 7¢ before fees—the same p minus executable price identity from the probability module. If fees cost 2¢, net edge is 5¢. Positive expected value is not a command to max size; variance near 50¢ contracts is high, and correlated clusters mean one election night can swamp many small edges.
Reading binaries on a chart
A single price is a snapshot; the path matters. A move from 40¢ to 55¢ in an hour signals marginal traders repriced the event—whether or not you know the headline yet. Journalists cite “market odds” because level and speed both carry information. Your trade thesis should say whether you disagree with the level, the trend, or both.
Structural complements and boxes
When YES plus NO sums below $1 on asks, buying both legs returns more than you paid at settlement if rules are identical—a theme from cross-contract arbitrage. When the sum is above $1, opening riskless profit requires shorting or NO legs, not only buying YES. Fees and position limits determine whether the gap is tradable.
Who is on the other side
Retail traders, political junkies, crypto natives, hedge funds, and market makers share the book. The price you see is the marginal trade among whoever showed up under these rules today. A $500 market on an obscure tweet is entertainment, not a scientific consensus—efficiency is conditional on liquidity and participant pool.
Core concepts to remember
Binary means $1 or $0 after clear resolution. YES and NO are complements near $1 on one venue. Edge is your p minus executable price. Resolution text is part of the instrument. Correlated binaries are one cluster.
Common misconceptions
Treating the mid as your fill on a wide retail book overstates edge. Assuming YES + NO = 1 across platforms with different fees and position caps invites false arbitrage. Buying YES on “Candidate wins” and YES on “Party wins” is diversification in name only when both rise and fall on the same poll wave.
Another mistake is treating every platform quote as comparable. The same headline event on two venues can trade at different levels because resolution text, fees, and who may trade differ. “The market says” is shorthand for this pool, these rules, this moment.
What comes next in this module
Binary contracts are the atomic unit. The chapters ahead in Event Contracts & Product Structures stack them into slates, numbers, spreads, props, conditionals, trees, bundles, and finally the clocks and courts that turn prices into cash.
Next: Multi-Outcome Contracts (Categorical Markets)