What is a prediction market?
A prediction market is a place where people bet real money on a question about the future, and the price of the bet becomes the crowd’s live estimate of how likely it is. Here is how that works, piece by piece.
A prediction market sells you a claim on an answer, not a thing. Its price is the crowd’s live probability, and a share pays $1 if you’re right and $0 if you’re wrong, which is exactly the terrain an AI fund can work with research alone.
Most markets sell you a thing, a share of a company, an ounce of gold. A prediction market sells you a claim on an answer. That one difference changes everything about how it works, and it is why an AI fund can operate here with nothing but research and discipline.
01One market, one question
Every market on a venue like Polymarket is a single question the future will settle: Will Barcelona win Saturday’s match? Will this candidate win the election? Will this coin close above $100,000 by July? The question comes with written resolution rules, exactly what counts as Yes, decided by what source, on what date. No opinions, no quarterly spin: the answer arrives, and the market closes.
02The price is a probability
Each market trades between 0¢ and 100¢. That price is the crowd’s current estimate of the chance the answer is Yes. Barcelona at 62¢ means the crowd collectively believes: about a 62% chance. Every new bet nudges the number, so the price is a live, money-backed forecast, often sharper than any single pundit, because everyone in it pays for being wrong.
| You see | It means |
|---|---|
| 62¢ | Crowd says ~62% chance of Yes |
| 95¢ | Near-certainty, little left to win |
| 8¢ | Longshot, cheap, usually for a reason |
| 50¢ | A coin-flip, or a market that hasn’t made up its mind |
03Shares: $1 if right, $0 if wrong
When you bet, you buy shares of one side. Yes or No, at the current price. At resolution, every share of the correct answer pays exactly $1; every share of the wrong one pays $0. Buy Yes at 62¢ and Barcelona wins: each share returns $1, a 38¢ gain. Barcelona loses: the share is worth nothing. You can also sell any time before resolution at whatever the market then pays, a skill of its own.
04Where the opportunity lives
If prices were always exactly right, there would be nothing to do here. But crowds overreact to headlines, sleep on dull facts, and lag the news. Whenever the price drifts away from the real chance, the market is mispriced, and buying the underpriced side is how a disciplined fund gets paid. That gap has a name: edge.
A price is the crowd’s opinion with money behind it, usually good, never gospel. The whole craft is knowing when it’s wrong.
That is the entire machine: a question, a price that is a probability, shares that settle at $1 or $0, and a crowd that is mostly, but not always, right. Everything your fund does starts from these four pieces. For the shortest version, the FAQ answers it in four sentences.
What to remember
- Each market is one yes/no question the future will settle.
- The price is the probability 62¢ means the crowd sees a 62% chance.
- A share is worth $1 if right, $0 if wrong; the gap between price and true odds is the opening.
- No inventory, no timing tricks, just mispriced answers, which is where your fund earns.