Guide

What moves prediction-market odds

A prediction-market price is a number between 0 and 100 cents that doubles as a probability: a contract trading at 65 cents implies the market thinks the outcome is about 65 percent likely right now. That price is not set by an editor or a bookmaker. It moves continuously as people buy and sell, and each trade nudges the implied probability up or down. This page explains the mechanisms behind those moves: how fresh information gets priced in, how order flow and liquidity shape the size of a move, why large or informed traders matter, how prices tend to settle as a resolution date approaches, and how arbitrage keeps the same question priced similarly across platforms. Throughout, keep one thing in mind: these prices are crowd-implied probability estimates, not facts, and they can be wrong. Nothing here is financial advice.

New information is the main driver

The single biggest force on a prediction-market price is new information. Markets aggregate what many independent participants believe, and a price represents a money-weighted average of those views. When something changes what people expect, they trade, and the price moves to a new level that reflects the updated consensus.

The kinds of information that move odds are the same kinds that move the underlying story: a poll release, an economic data print (jobs, inflation, GDP), a company's earnings report, a court ruling, an official statement or policy announcement, a candidate dropping out, an injury before a match, or breaking news of any kind. A surprise relative to expectations tends to produce the sharpest move, because the market had already priced in the consensus forecast; what trades is the gap between what was expected and what actually happened.

This is why a market price often reacts within seconds of a headline while a poll average or a pundit takes days to catch up. The price is a live estimate that updates the moment a participant with conviction acts on what they just learned. It is also why the price can be jumpy: an early, thin report can move it, and a later correction can move it back.

Order flow, liquidity, and the size of a move

Every move happens through trading. On most prediction markets, buy and sell orders meet in a central limit order book, matched by price and time. The best available prices to buy and to sell sit slightly apart, and that gap is the bid-ask spread. A trade that lifts the lowest ask, or hits the highest bid, sets a new last price, which is what most people read as "the odds."

How far a given trade moves the price depends on liquidity, specifically market depth: how many contracts are resting at each price level. A deep market has many orders stacked close together, so even a large trade only walks the price a little and the spread stays tight. A thin market has gaps in the book, so a modest order can jump the price several cents and the spread is wide. This is why an identical piece of news can barely budge a heavily traded headline market yet swing a quiet, obscure one sharply.

Market makers smooth this out by continuously posting both buy and sell orders, earning the spread in exchange for supplying liquidity and absorbing temporary imbalances. When they pull back, during volatile moments or in unpopular markets, spreads widen and prices get noisier. A practical takeaway: in a thin market, a price move may reflect one trader's order hitting an empty book rather than a real shift in what the crowd believes.

Informed traders, conviction, and capital at risk

Prediction markets weight opinions by money, not by headcount. A participant who is confident can back that view with more capital, and a position only pays if the outcome actually resolves their way. This structure rewards being right and penalizes being wrong, which gives informed participants an incentive to trade until the price reflects what they know.

That is the core argument for why these prices can be useful: someone who has done the work, or who holds private information, can move the price toward the truth and profit from doing so, while uninformed noise tends to cancel out across many participants. A sudden, sustained move on no public news is sometimes the market reacting to a well-funded trader who knows or believes something the rest of the crowd does not yet.

The same mechanism is also a limit. Thinly traded markets can be moved by a single large order, sentiment and crowd psychology can push prices away from a fair estimate, and a market can simply be wrong when the crowd is collectively mistaken or when few informed participants are present. A price is the crowd's best money-weighted guess at a moment in time, not a guarantee, and it should be read alongside the underlying reporting rather than in place of it.

Time decay, resolution, and how the rules anchor the price

As a market approaches its resolution date, the price tends to converge. Uncertainty narrows, more participants pile in near the deadline, the spread tightens, and the price grinds toward 0 or 100 cents as the outcome becomes clear. Far from resolution, prices are more volatile because there is more time for new information to arrive and reverse the picture; close to resolution, each remaining piece of news matters less and the price is harder to move.

What ultimately pins the price is the resolution rule: the specific, written source and criterion that decides the outcome. Read it before trusting a number, because ambiguous wording is a real source of error. On Polymarket, markets resolve through UMA's optimistic oracle, where a proposed outcome stands unless someone disputes it within a set window, and disputes escalate to a tokenholder vote. Kalshi, a U.S. exchange regulated by the CFTC, resolves contracts against the official source named in each market's rules and clears them through its registered clearinghouse.

Resolution clarity matters for the price too. Markets with clean, verifiable criteria converge smoothly, while markets with vague or contestable rules can see volume dry up and spreads widen even as the real-world uncertainty shrinks, because traders fear a disputed or surprising settlement.

Arbitrage keeps platforms roughly in line

The same question often trades on more than one venue, and arbitrage is the force that keeps those prices from drifting far apart. If a YES contract is cheap on one platform and the matching NO is cheap on another, a trader can buy both, lock in a position that pays out regardless of the outcome, and pocket the difference if the combined cost is below the payout after fees. That buying pressure pushes the two prices back toward each other.

In practice these gaps are usually small and close fast, often within seconds, and much of the activity is run by bots watching many markets at once. Prices for the same event still diverge sometimes, especially for politically charged questions where the two platforms' user bases differ, or where fees, settlement timing, and the exact resolution wording are not identical. Those frictions are exactly why a perfectly riskless gap is rare and why small differences can persist.

For a reader, the lesson is comparative: when two reputable markets agree on a question, that is a stronger signal than either alone; when they disagree, it is usually a clue that the wording differs, the liquidity is thin, or the crowds genuinely see the question differently. As always, specifics go stale, so check a live market for the current price rather than relying on any number quoted here.