Index and spread markets compare two sides of a scoreboard or index. Proposition bets—props—zoom into micro-outcomes: a debate mention, a cabinet shortlist name, first scorer, earnings beat, whether a bill passes before a date. They are often implemented as binaries or small categoricals, but liquidity, wording, and parent-event correlation dominate the trade.
What makes a prop different from a headline market
Macro binaries on “who wins” or “Fed cuts” usually attract depth and tight spreads. Props sit inside a larger story: the game still has to be played, the debate still has to happen, the nominee still has to survive the path. Rulebooks run longer and quirkier—what counts as a “mention,” which debate, what happens if the event is cancelled.
Retail flow loves props because narratives are fun; professionals respect them because spreads and disputes eat edge faster than on marquee contracts.
Taxonomy in plain language
Occurrence props ask whether something happens at all—keyword in a speech, rate cut count at least three. Timing props add a calendar: bill passes before August 1. Count props threshold a total. Personnel props are categorical slices on a shortlist. Performance props frame surprise versus consensus on a print or earnings.
Many props are conditional in story even when the title omits “if”: the chance a candidate says “border wall” in Debate 2 is tiny if the debate is cancelled, even if the contract does not say “given debate occurs” in the headline.
Estimating probability on a thin edge
Suppose you assign 98% that the debate happens, 30% that the phrase is spoken given it happens, and 29.4% joint. A YES price of 28¢ looks roughly fair before fees. If the ask is 32¢ and the bid 26¢, buying at the ask may be negative expected value despite a good story—the spread is the market’s tax on props.
Require a larger edge buffer on props than on liquid binaries; a 3¢ theoretical edge can vanish in one round trip.
Correlation on one night
Five props on the same debate—mention topics, interruption counts, zinger lines—move together when the event starts. Treat them as one cluster for bankroll caps, not five independent positions. Conjunction fallacy appears when traders buy a basket of “likely” props that all require the same campaign path; joint probability is lower than the product of marginals when shocks are shared.
Parlay-style bundles on the same evening embed positive correlation; fair joint probability is often higher than the product of legs.
Personnel slates and the NO leg
“Treasury Secretary is Person P” at 18¢ in a six-name menu might mean your 12% true chance is below market—negative edge on YES. The NO leg at 84¢ is often overlooked; renormalize the whole slate when a name is withdrawn, as you would on a categorical primary board.
When the shortlist shrinks, dead names should trade near zero—if they do not, check whether the market is voided or still live per rules.
Resolution traps unique to props
Does “mentions” mean spoken aloud or chyron text? National debate versus town hall? If the game is postponed, does the market void, refund, or settle NO? Historical disputes on retail political platforms often hinged on English, not on math—treat rules as half the edge.
Regulatory props on economic releases may be cleaner than culture props; dispute rate scales with ambiguity.
News jumps and calendar discipline
Debate mention props spike at open and decay; injury news gaps player props. Pre-define entries on a calendar instead of chasing the first tick after a viral clip—by then the spread has widened and joint probabilities shifted.
Manipulation on thin props is cheaper than moving a national binary—fade spikes only when rules and expected value agree, not because “it feels too high.”
Minimum edge on niche contracts
Many traders demand 5¢ plus fees on props versus 3¢ on liquid binaries because half the spread can erase a thin model edge. If joint probability is 29.4% and the ask is 32¢, pass—even when the narrative feels right.
Scoring and calibration on props
Low-base-rate props punish overconfidence in Brier and log scores. Logging props separately from macro binaries keeps calibration honest—you may be sharp on Fed cuts and terrible on debate mentions.
Cross-checking nested products
“Mentions border policy” is not “border bill passes.” Arb requires identical events. Compare props to related scalars and win markets only after a rule diff.
Cancellation and void
If the debate is canceled, mention props may void, refund, or settle NO—three different economic outcomes. Parent cancellation is the first line in your prop checklist.
Why organizations list props
Props attract flow on cultural moments and micro policy beats; they also generate dispute risk. For traders, props are where rules and spreads matter more than clever narratives. For platforms, props are engagement with higher operational load.
Core concepts to remember
Classify occurrence, timing, count, or personnel. Estimate joint paths with parent events. Require wider edge buffers. Cap cluster risk per evening. Read mention and cancellation definitions before size.
Culture versus policy props
Culture props can move on memes; policy props often move on schedules and leaks. Liquidity and dispute rates differ. The same trading math applies; the rulebook and cluster caps should differ.
Kalshi, Polymarket, and retail caps
Regulated macro props may have tighter spreads than meme props on global platforms. Academic caps widen spreads everywhere on niche politics. Match your size to the venue’s friction, not to your confidence alone.
Common mistakes
Sizing props like marquee binaries. Ignoring cancellation of the parent event. Buying correlated mention tickets as diversification. Chasing the first tick after news without spread math.
What comes next
Props are micro-outcomes with implicit paths. Conditional markets make the path explicit.
Next: Conditional Markets: If X Happens, Then Y?