Wisdom of crowds
Updated November 10, 2026
Wisdom of crowds is the hypothesis that combining many traders’ views can sometimes beat individual polls or pundits, because money at stake penalizes being wrong. It is not a law; thin markets and correlated errors can fail it.
A poll shows 52% support; a liquid election contract trades 58¢ for weeks. Traders with private models and hedging needs may price incumbency risk differently than survey respondents. Combined market forecasts can sometimes track outcomes better than polling, but sometimes is not always.
- Support in the poll
- 52%
- Liquid election contract, for weeks
- 58¢
Why combining views can help
If many people each estimate the same thing and their errors are independent, some guess too high and some too low, and the errors partly cancel when the estimates are combined. The combined figure can then sit closer to the truth than most of the individual guesses.
A market combines views through the price. A poll records what a sample says and costs respondents nothing either way. A contract price records what traders were willing to transact at, and being wrong costs whoever holds the position. Anyone who thinks the quote is off can trade against it, and the price can update within seconds of new information instead of waiting for the next field period.
When the crowd gets it wrong
The cancellation only works if the errors are independent. Where traders draw on the same few sources, one mistake gets counted many times and the price can look settled without much diversity of judgment behind it. Herding, anchoring on the visible quote or on other traders instead of forming an independent estimate, pushes in the same direction.
Thin markets are the other common failure. On a contract with little volume, the quote may reflect a handful of traders, and a single order can move it several cents. The 58¢ contract in the example is described as liquid and stable for weeks; a thinly traded contract at the same price would carry much less information. Unclear settlement wording adds noise too, because part of the price can reflect disagreement about what the question means. Market efficiency covers these conditions in more detail.
FAQ
Is the wisdom of crowds real?
The mechanism is real: independent errors partly cancel, and a market adds a cost for being wrong. Whether it produces a good estimate on a given contract depends on how much independent judgment and volume sit behind the price.
What is the difference between the wisdom of crowds and herding?
The wisdom of crowds depends on traders forming their own estimates. Herding is the opposite: traders follow the visible price or each other, so one view gets repeated rather than combined with others.
Are prediction markets more accurate than polls?
They measure different things, and neither is right every time. Comparing them takes scoring both against outcomes over many questions, using measures such as calibration and the Brier score.