How Random Numbers Hijack Your Judgment
Quick question: Is the population of Turkey greater or less than 35 million? Now estimate the actual population.
Whatever number just popped into your head was pulled toward 35 millionâeven though that number was arbitrary. This is the Anchoring Effect: irrelevant numbers bias our estimates, and we can't escape it even when we know it's happening.
Recreate Tversky & Kahneman's famous 1974 study. Spin the wheel, then answer a question. The wheel is riggedâbut your answer will still be influenced by whatever number comes up.
The anchoring effect isn't just a lab curiosityâit shapes major decisions every day:
In their landmark paper "Judgment Under Uncertainty: Heuristics and Biases," Amos Tversky and Daniel Kahneman introduced the anchoring effect alongside the availability and representativeness heuristics. Their wheel of fortune experiment became one of psychology's most cited demonstrations of cognitive bias.
The original experiment was elegantly simple. Participants watched a wheel of fortune spinârigged to land on either 10 or 65. They were asked: "Is the percentage of African nations in the United Nations higher or lower than this number?" Then: "What do you think the actual percentage is?"
The results were striking. When the wheel landed on 10, the average estimate was 25%. When it landed on 65, the average estimate was 45%. A completely random numberâwith no logical connection to African UN membershipâshifted estimates by 20 percentage points.
"People make estimates by starting from an initial value that is adjusted to yield the final answer. Adjustments are typically insufficient."
â Tversky & Kahneman, 1974
Two mechanisms explain the anchoring effect:
One of the most disturbing findings about anchoring is that expertise doesn't protect you. Real estate agents, judges, doctors, and negotiators all fall prey to anchoringâoften while denying they're affected.
In the famous real estate study, professional agents estimated home values after seeing manipulated listing prices. Despite decades of experience, their estimates were pulled toward the anchor just as much as students' estimates. When interviewed afterward, the agents insisted the listing price hadn't influenced them. The data said otherwise.
To prove that even absurd anchors work, researchers asked participants: "Did Gandhi die before or after age 9?" or "Did Gandhi die before or after age 140?" Both are obviously wrongâGandhi died at 78. Yet the group given the high anchor (140) estimated significantly higher ages of death than the low-anchor group. The difference: over 15 years.
Even anchors that are transparently ridiculous still pull your estimates. Knowing about the bias doesn't make you immune.
Research suggests a few strategies:
But here's the uncomfortable truth: even when warned about anchoring, even when paid to avoid it, even when the anchors are random numbers generated by diceâpeople still anchor. It may be one of the most robust biases in all of psychology.
The next time someone throws out a number in a negotiation, asks you to estimate something after showing you a figure, or displays a "was/now" priceâremember: that first number isn't neutral. It's an anchor, and it's already dragging your judgment toward it.