You'll hear that a market priced at 70¢ means "a 70% chance." It's a useful shortcut, but it's not literally true. Several forces push the price away from the real probability — and knowing them is how you find mispriced markets. Educational only, not financial advice.
In an ideal, frictionless market, a Yes share trading at 70¢ should mean traders collectively believe there's about a 70% chance of Yes. That's the "wisdom of crowds" idea, and as a rough gauge it works. The problems start when you treat it as exact.
If a platform takes a cut on trades or withdrawals, rational traders price that in — nudging Yes prices down and No prices up so the pair no longer sums cleanly to 100¢. Any fee means the "probability" you read is already shaded by the house's take.
A share that resolves in two years ties up your cash the whole time. Traders demand a discount for that, so long-dated markets can sit below the "true" probability simply because the money could be earning elsewhere. The further away resolution is, the more the price reflects impatience, not just odds.
People overpay for cheap longshots (the 5¢ lottery-ticket thrill) and underpay for near-certainties. This well-documented favorite-longshot bias means extreme prices are the least reliable: a 3¢ market is often really a 1% event, and a 97¢ market is often more certain than it looks.
In a market with a wide spread and little depth, the "price" might just be where one stale order sits. A single trade can move it 8¢ with no new information. A price is only a good probability estimate when there's real money and real depth behind it — see reading the order book.
Prices also drift because related markets don't always agree. If "Candidate wins the nomination" trades higher than "Candidate wins the presidency," but the second requires the first, the pair can be logically inconsistent — occasionally enough to arbitrage. More often the lesson is subtler: check whether a market's price lines up with its own sub-questions and with polling or betting elsewhere. Big disagreements between correlated markets are a flag that at least one price is wrong.
Treat the price as a starting estimate of probability, then adjust: widen your uncertainty on long-dated, low-liquidity, or extreme-priced markets. The edge in prediction markets comes from spotting where the price has drifted from the real probability for one of these structural reasons — not from assuming it's already correct.
To know if a price is off, you need an independent estimate to compare it against — otherwise you're just trusting the number you're trying to judge. Start from a base rate (how often this kind of thing happens historically), adjust for the specifics, and only then look at the market. If your reasoned estimate is 60% and the market says 70¢, you've found a potential edge; if you can't beat the crowd's reasoning, the price is probably fine and you should pass.
Before you read a market price as a probability, sanity-check it: