Prediction Markets // Article

Why a Prediction Market at 70% Isn't Exactly a 70% Chance

SPUNK13  ·  7 min read  ·  Updated Aug 2026
In this article
  1. The clean version of the claim
  2. Fees and the vig shift the number
  3. Time value: your money is locked up
  4. Risk and the favorite-longshot bias
  5. Thin liquidity distorts the read
  6. Correlated markets and arbitrage
  7. So how should you use the price?
  8. How to find your own estimate
  9. A checklist for reading a price

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.

The clean version of the claim

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.

Fees and the vig shift the number

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.

Time value: your money is locked up

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.

Risk and the favorite-longshot bias

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.

Thin liquidity distorts the read

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.

Correlated markets and arbitrage

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.

So how should you use the price?

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.

How to find your own estimate

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.

A checklist for reading a price

Before you read a market price as a probability, sanity-check it:

// FAQ
Does a prediction market price equal the true probability?
Roughly, but not exactly. Fees, time value, risk preferences (the favorite-longshot bias), and thin liquidity all push the price away from the real probability, especially on long-dated or extreme-priced markets.
Why are prediction market longshots usually overpriced?
Because of the favorite-longshot bias: people overpay for cheap, high-payout 'lottery ticket' outcomes and underpay for near-certainties, so very low prices tend to overstate the real chance of the event.
Track prediction markets on 13.markets
Live odds, platform comparisons, and beginner-friendly market analysis.
Explore 13.markets

More Articles

How to Read a Prediction Market Order Book as a BeginnerWhat Happens When a Prediction Market Outcome Is Disputed