The anchor
Lesson 6 of 31 in the course “Brain Hacks”. Free, with animations and a quiz.
A wheel that knows nothing
New level: thinking bugs. First up: the anchor. In a famous study, students spun a wheel of fortune that was rigged to stop only at 10 or 65. Then they were asked: what percentage of UN member countries are in Africa? The wheel obviously has nothing to do with it. But those who got 10 answered 25% on average, and those who got 65 answered 45%. Any number you see before making an estimate pulls your answer towards it, like an anchor holds a boat.
Even the pros
Maybe it only happens to students? In a study with German judges who had more than 15 years on the bench on average, they read a shoplifting case. Before deciding, the judges rolled dice rigged to land on 3 or 9. Those who rolled 9 gave 8 months on average; those who rolled 3 gave 5 months. Experience doesn’t protect you from anchors, because anchors act on autopilot, not on knowledge. That’s why it matters to know about them in advance.
Why the anchor holds
Two mechanisms are at work. The first is the pilot. It starts from the anchor and moves in the right direction, but stops as soon as the answer feels “about right”. That’s almost always too close to the anchor. The second is autopilot. The question “Was Gandhi older than 114 when he died?” makes your brain search for images of a very old Gandhi. Related thoughts pop up, just like in the association machine, and the estimate goes up. In the end, both pilots work for the anchor.
Anchors in stores
A crossed-out price, “was $200, now $99”, is a classic anchor. Your brain compares $99 with $200 and celebrates the discount, but the real question is different: are these sneakers worth $99 to you? Same with an expensive dish on a menu that makes everything else seem reasonable. It works on pros too: in one study real estate agents insisted the listing price didn’t affect them, yet their valuations of a house moved with it. An anchor works even when you’re sure it doesn’t.
The anchor in crypto
Danya sees a coin: a month ago it was $10, now it’s $3. First thought: “−70%, cheap, gotta buy!” But $10 is just an anchor, a past price at the peak of the hype. It says nothing about what the coin is worth now. Maybe $10 was wildly inflated and $3 is still too much. The question “is it cheap?” always needs a follow-up: “compared to what?” Compare with what the project actually does, not with the prettiest number on the chart.
The anchor in haggling
Danya is selling his old console. If he says first, “yours for 120”, the haggling will circle around 120. If the buyer says first, “I’ll give you 50”, it will circle around 50. So when you know what something is worth, name the price first. And if someone throws an outrageous anchor at you, don’t haggle from it: that only locks the number in. Better to say plainly the offer doesn’t work and name your price with a reason. That way you don’t let their anchor settle in your head.
Hack: argue with the anchor
Just knowing about anchors isn’t enough: the effect stays even when people are warned. What helps is an active move. When you see a number before a decision, deliberately come up with three reasons why it’s wrong: why this price is too high, why this forecast is too bold. In experiments, this “think the opposite” move noticeably weakened the anchor. And ask yourself: “What number would I give if I’d never seen this one?” That way the pilot starts from your own reasoning, not someone else’s figure.
Try it: your own price
For your next three purchases that cost more than a coffee, do this. Before you look at the price tag, write in your notes how much the item is worth to you. Then compare with the price. Danya did this with headphones, “50% off” sneakers and a game. Two out of three cost more than his honest estimate, even though the sneaker discount looked huge. The number you wrote down becomes your anchor, and it’s harder for someone else’s to drag you.
Check yourself
- A student’s wheel stopped at 65, and he guessed a higher share of African countries in the UN than those who got 10. Why?
Correct answer: An anchor: even a random number pulls the estimate. Yes. The wheel knows nothing about the UN, but 65 became an anchor: the pilot didn’t adjust far enough, and autopilot supplied reasons in its favour. - A buyer messages Danya: “I’ll give 30 for your console”, though they sell for 110. What’s best?
Correct answer: Reject it and name his own price with a reason. Right. Haggling from an outrageous anchor locks it in. A clear no plus your own price with a reason brings the frame back to real value. - A coin fell from $10 to $3. A friend says: “That’s −70%, it’s cheap!” What’s wrong?
Correct answer: $10 is an anchor; a past price says nothing about value. Right. “Cheap” only makes sense compared to what the project is worth. The hype-peak price might have been wildly inflated. - Which move actually weakens an anchor?
Correct answer: Come up with 3 reasons the number is wrong. Yes. “Think the opposite” makes the pilot look for arguments against the anchor, and in experiments the effect noticeably weakened.