Who Sets Prediction Market Prices?
Nobody does — and that is the point. Prices on a prediction market emerge from thousands of independent traders buying and selling against each other. The platform matches trades; it does not pick the numbers. Here is how the process works.
Prices Are Collective Opinions — With Money on the Line
A prediction market contract pays $1 if an event happens and $0 if it does not. The current price — say 63¢ — is what traders are willing to pay for that contract right now. Since the maximum payout is $1, a price of 63¢ implies the market collectively estimates a 63% probability that the event will occur.
That price is not set by an algorithm, an editor, or the exchange itself. It is the result of continuous negotiation between buyers and sellers. When more people believe the probability is higher than 63%, they buy — pushing the price up. When more believe it is lower, they sell — pushing the price down.
This mechanism is called price discovery: the market "discovers" the consensus probability through real trades from real participants who have skin in the game.
Key concept: the order book
Behind every prediction market is an order book — a live list of all open buy and sell offers. A buyer might offer 61¢ for "Yes." A seller might ask 64¢. When a buyer and seller agree on a price, the trade executes and the order book updates. The mid-point of the best bid and best ask is the "market price" you see displayed.
Three Types of Participants — Each Moves Prices Differently
Not everyone who trades a prediction market is trading for the same reason. Understanding who is on the other side of your trade explains why prices move the way they do.
Market Makers
Role: Provide liquidity by posting both buy and sell orders simultaneously
How they move prices: They anchor the initial price range and keep spreads tight. If informed traders consistently buy one side, market makers adjust their quotes to reflect the new consensus.
Example: A market maker might quote 48¢–52¢ on an election market. If traders keep hitting the 52¢ offer, the market maker shifts the quote to 51¢–55¢ to avoid losing money to better-informed counterparties.
Informed Traders
Role: Trade based on private research, models, or edge — the sharpest signal in the market
How they move prices: Every trade they make is a vote. When informed traders push heavily to one side, prices move toward their belief. Over time, their aggregate activity drives prices toward true probabilities.
Example: A trader who reads 50 state-level polls and builds a regression model might buy "Yes" at 44¢ because their model says the true probability is 62%. Their purchases move the price toward that level.
Uninformed / Retail Traders
Role: Trade on opinions, news, or vibes — introduce noise but also volume
How they move prices: Large coordinated retail flows can temporarily move prices away from true probabilities, creating opportunities for informed traders to correct them. Over a long event window, retail noise tends to wash out.
Example: After a viral news story, retail traders might pile into "Yes" on a political market, pushing it briefly to 78¢. Informed traders often sell into this spike if they believe the true probability is lower.
How Does a New Market Start — Before Anyone Has Traded?
When a prediction market is first listed, it has no price history and no established consensus. Here is the typical sequence of events:
Platform lists the market
The exchange creates a contract with defined terms: what event it covers, how it resolves, and when. It does not set the probability — it opens the order book.
Market makers seed initial quotes
Professional liquidity providers post initial bid and ask orders, often near 50¢ on uncertain events. These quotes are tentative — they expect to adjust quickly once they see how others are trading.
Informed traders arrive first
Traders who have done research often see a new market as an opportunity to buy at a mispriced level before the crowd adjusts it. Their early trades are the first strong signal about true probability.
Price discovery begins in earnest
As volume builds, the bid-ask spread typically tightens and the price converges toward a stable level. The more liquid the market, the faster this convergence happens.
Ongoing updates as information arrives
Every poll release, news event, or statement from key parties triggers new waves of trading and price adjustment. The market is always recalibrating.
What Actually Moves the Price After It Is Set?
Once a market is established, prices shift for four main reasons.
New information
A poll releases, a court rules, a company announces earnings. Informed traders incorporate new data faster than the rest of the market and adjust their positions. Prices reflect this almost immediately on liquid markets.
Order imbalance
When many more traders want to buy than sell (or vice versa), market makers widen their quotes and shift prices to rebalance the order book. The "pressure" of unmatched orders moves the mid-price.
Liquidity changes
If large participants withdraw from the market — reducing the depth of the order book — small trades can move prices more dramatically. Thin markets are more volatile, not necessarily more informed.
Time decay and resolution proximity
As an event approaches its resolution date, prices naturally converge toward 0 or 100 as uncertainty resolves. A market at 60¢ in January may be at 95¢ in October on the same underlying event — purely from the passage of time and accumulated evidence.
What Does the Platform Actually Do?
On a CFTC-regulated prediction market exchange, the platform plays a specific and limited role:
- Operates the order-matching engine that pairs buyers and sellers
- Lists contracts with defined terms and resolution rules
- Enforces rules against manipulation and ensures fair access
- Charges a transaction fee on matched trades (typically a percentage of the notional value)
- Does not take positions — it does not bet on outcomes
The platform profits whether "Yes" or "No" wins — its revenue comes from trading activity, not from picking the right side. This is structurally different from a sportsbook, where the house sets lines to profit from bettors' expected losses.
Why Do Prices Differ Between Platforms?
Each exchange is a separate order book with different participants and liquidity. A contract trading at 61¢ on one platform and 64¢ on another does not mean one platform is more accurate — it means their respective crowds have not yet arbitraged the gap away.
Contract wording can also legitimately cause price differences. "Will [candidate] win the election?" might resolve differently if one platform uses AP calls and another uses electoral vote certification. Read the resolution rules before comparing prices across platforms.
Why prediction markets disagree on the same event →Frequently Asked Questions
Does the platform set prediction market prices?
No. On CFTC-regulated exchanges like Kalshi and Polymarket's DCM, the platform operates an order-matching engine but does not set prices. Prices emerge from bids and offers submitted by independent traders. The platform earns a fee on matched trades, not from taking positions.
Who decides the starting price of a new market?
Typically the platform will list a market at 50¢ (equal probability) or at a seed price based on market maker quotes. Within minutes or hours, informed traders and market makers adjust prices toward what they believe is the correct probability.
Can one trader move prediction market prices?
Yes, on thin markets. If an order book has limited depth, a single large trade can shift the price significantly. On deep, liquid markets (like major election markets), moving the price requires much larger capital because market makers and arbitrageurs will push back against outlier prices.
Why do prices sometimes jump suddenly without news?
Large block trades, an informed trader acting on private research, or coordinated activity can all move prices without public news. On CFTC-regulated exchanges, platforms are required to monitor for manipulation — but informed trading (trading on research or skill) is not manipulation and is permitted.
Why do Kalshi and Polymarket show different prices for the same event?
Each platform has a separate order book with different participants, market makers, liquidity depth, and contract wording. A price difference of a few cents is normal and often reflects liquidity gaps, not a fundamental disagreement about probability. Large persistent differences may signal an arbitrage opportunity.
Are prediction market prices always accurate?
No — but they tend to be more accurate than polls, pundit forecasts, and most individual models, especially over thousands of markets. On low-liquidity markets or very long time horizons, prices can stay away from true probabilities for extended periods. Accuracy improves with liquidity and time pressure.
Keep Learning
Why Prediction Markets Move
What causes prices to shift over days, weeks, and months — not just in real time.
Why Platforms Disagree on the Same Event
Contract wording, liquidity, and oracle differences that explain cross-platform price gaps.
Model vs. Market: Why They Diverge
Statistical models and prediction markets both forecast outcomes — but they measure different things.