Wisdom of Crowds

mental-model

A crowd can outperform its members when judgments are genuinely independent and errors point in different directions. Diversity improves the average by cancelling mistakes; shared biases make the same averaging process preserve or amplify them.

Imagine collecting a large pile of guesses and discovering that every person was wrong—but the average was sensible. The apparent contradiction disappears once you notice that the guesses missed in different directions.

Accuracy emerges from uncorrelated mistakes

Aggregation does not magically turn ignorance into knowledge. It works when people bring different backgrounds, ideologies, information and ways of reasoning, so one person’s overestimate can offset another’s underestimate. The useful signal survives while opposing errors cancel.

Independence is therefore as important as headcount. A heterogeneous crowd produces a spread of mistakes that an average can reduce. When members copy one another or inherit the same assumptions, their errors become correlated: adding more voices then adds repetitions, not information. This is why the mechanism is better understood through the Diversity Prediction Theorem than through the slogan that crowds are simply wise.

E1 E2

Where it shows up

Different ideologies, balanced judgment

People shaped by different backgrounds carry different inner biases. When those biases pull in opposing directions, combining their judgments can leave behind a more sensible collective view.

Wrong guesses, useful average

A collection of individually inaccurate guesses can still converge on a better estimate because the misses are distributed on both sides rather than concentrated in one direction.

A crowd cannot average away a shared blind spot

The model breaks when diversity is cosmetic or judgments are not independent. If participants share the same information, ideology or social cue, they may all err in the same direction; averaging then launders a common bias into apparent consensus. Group conformism can make this worse by suppressing the disagreement the mechanism needs.

E1 E2

Collect judgments before people confer

Tomorrow, when you need a group estimate or forecast, ask each person to submit an answer and brief rationale privately before discussion begins. Compare the independent responses first, then aggregate them; otherwise the earliest confident voice may turn a potentially wise crowd into one repeated opinion.

Episodes that teach this

The Madness of Crowds - Now Online - Future IQ 5,535 views
With different backgrounds and ideologies, people's inner biases "cancel each other out" and what remains is "usually a sensible wise thing."
The Wisdom of the Crowd Effect: Theory, Experiment, Examples and Explanation · explained at 3:53 2,169 views
"if you take enough guesses from different people all of them are wrong but they are wrong in different ways and their errors cancel out"