The wisdom of crowds: why a market can out-forecast an expert
How prediction markets turn thousands of scattered opinions into a single number that often beats the individual expert — and where the effect breaks down.
Outcomer Team · Jul 12, 2026
Ask one person to guess a hard number and they will usually be off. Ask a thousand people, throw away nothing, and average the answers — and the result is often startlingly close to the truth. That effect is the engine behind every prediction market. Understanding it helps you read a price for what it is: not a guess, but an aggregate.
The ox that started it
In 1906, the statistician Francis Galton watched a weight-guessing contest at a country fair in Plymouth. Around 800 people paid to guess the weight of a dressed ox; about 787 tickets were usable. Galton expected the crowd to be hopeless. Instead, when he averaged every guess, the crowd's estimate came to 1,197 pounds — and the ox actually weighed 1,198. The average of ordinary fair-goers, most of whom knew little about cattle, landed within a single pound of the truth. Galton published the result in Nature in 1907, and it has been the go-to illustration of collective accuracy ever since.
The lesson was not that crowds are wise in some mystical way. It was that individual errors tend to cancel. Some people guess too high, some too low, and if those mistakes are roughly independent, they wash out in the average, leaving the signal behind.
From guesses to prices
A prediction market is Galton's experiment run continuously, with money instead of paper tickets. Every trade is a person staking something on their view. Buy Yes and you push the price up; sell and you push it down. The market price settles where buyers and sellers balance — which is another way of saying it settles on the crowd's aggregate estimate of the probability.
Money sharpens the effect in two ways. First, it filters for conviction: someone who is genuinely confident will stake more, so informed views carry more weight than idle ones. Second, it rewards being right and punishes being wrong, so people who consistently misjudge lose their stake and their influence shrinks over time. A poll asks everyone equally; a market lets the people with an edge press it.
If you want the mechanics of how a single price encodes a probability, we walk through it in reading the odds. And for the basics of what these markets are, start with what is a prediction market.
Why it beats a lone expert
An individual expert has deep knowledge of one slice of a problem and blind spots everywhere else. A market pools thousands of partial views — the local reporter who heard something, the statistician who ran the numbers, the person who simply lives where the event is happening. No single participant sees the whole picture, but the aggregate can.
This is why markets frequently match or beat expert panels and polls on questions with clear, verifiable outcomes. The price updates the instant new information arrives, because someone always has an incentive to trade on it before the crowd catches up.
The three conditions it needs
Wisdom of crowds is not automatic. It relies on three things, and when any of them fails, the crowd can be as wrong as one person — sometimes more.
Independence. Errors have to be roughly uncorrelated. If everyone is reading the same viral headline and copying the same take, the mistakes stop cancelling and start compounding.
Diversity. The crowd needs different information and different angles. A room of identical specialists is really just one opinion repeated.
A real incentive to be right. People have to be rewarded for accuracy, not for cheering. That is exactly what a stake provides — and why a market with actual skin in the game tends to be sharper than a free online poll.
When these hold, the price is a genuinely useful forecast. When they break — during a panic, a rumour cascade, or a thin market with only a few traders — it can drift badly. Reading a market well means knowing which situation you are in.
Seeing it for yourself
The fastest way to build intuition for this is to become one of the guesses. Watch a market where you have a view, place a small position, and see how the price moves as others arrive with theirs. You will start to feel the difference between a deep, liquid market that has absorbed a lot of information and a thin one that is really just a handful of opinions.
You can do all of that on Outcomer with virtual money — no risk, just the chance to watch the crowd think in real time and test whether your read is better than the average. That average, as Galton found, is a tougher opponent than it looks.