Joachim Andersen

Accelerated Learning

Anchoring Bias: Why the First Number Sticks

An antique brass ship anchor and chain on a dark walnut desk at night, the chain wrapped around a single ivory die, with small brass weights and a brass balance scale in warm golden light
The first number we see holds our judgment in place.


Anchoring Bias is the tendency to rely too heavily on the first piece of information I see when I make a judgment. Once a number or an idea is in my head, it becomes the starting point, and every estimate I make afterward is pulled toward it, even when that first number is arbitrary or obviously wrong.

I notice it most when I am buying or selling something without a clear market price. The first figure that is mentioned quietly sets the range for the whole conversation, and I often only realize later how much it shaped what I thought was reasonable.

The essence of Anchoring Bias is to ask: “Where did my starting point come from, and would I land somewhere else if I had started from a different number?”


Origin: Anchoring and Adjustment

The idea was introduced by the psychologists Amos Tversky and Daniel Kahneman in their 1974 paper Judgment under Uncertainty: Heuristics and Biases. They described anchoring and adjustment as one of three mental shortcuts we use when we have to estimate something under uncertainty: we start from an initial value and adjust from it until the answer feels right. The problem is that the adjustment is usually too small.

In one of their experiments, people had five seconds to estimate the product of the numbers one through eight. Those who saw the sequence as 1 × 2 × 3 × 4 × 5 × 6 × 7 × 8 gave a median estimate of 512, while those who saw it as 8 × 7 × 6 × 5 × 4 × 3 × 2 × 1 gave a median estimate of 2,250. The correct answer is 40,320. In another, a wheel of fortune rigged to stop on either 10 or 65 was spun before people estimated the percentage of African countries in the United Nations. Those who saw 10 guessed 25% on average, while those who saw 65 guessed 45%. The wheel had nothing to do with the question, yet it moved the answers by 20% points.


Key Concepts

  • The Anchor
    The first number or reference point we encounter, whether it’s a price, a forecast, a demand or a random figure that happens to be in front of us.

  • Insufficient Adjustment
    We usually move away from the anchor in the right direction, but we stop too early, as soon as the answer seems plausible.

  • Selective Accessibility
    Considering an anchor makes information that fits it easier to recall, so we end up building a case for an answer close to it.

  • Even Absurd Anchors Work
    In a 1997 study by Fritz Strack and Thomas Mussweiler, people asked whether Gandhi died before or after the age of 9 guessed 50 on average, while those asked about the age of 140 guessed 67.

  • Precision Matters
    A precise number like 4,985 looks better informed than a round number like 5,000, so I tend to adjust less from it.


Some Examples

  • The Expert House Valuers
    In a 1987 study, Gregory Northcraft and Margaret Neale let students and experienced real estate agents tour the same house and gave them the same information, except for the listing price, which varied between groups. The listing price pulled both groups’ estimates of the house’s value in its direction. Yet when asked afterward, the agents denied that the listing price had influenced them.

  • Sentences Decided by Dice
    In a 2006 study by Birte Englich, Thomas Mussweiler and Fritz Strack, young lawyers with their first experience as judges read the same shoplifting case and then threw a pair of dice to determine the prosecutor’s sentencing demand. The dice were loaded to show either 3 or 9. Those who rolled 9 gave an average sentence of 7.81 months on probation, while those who rolled 3 gave 5.28 months. They knew the number was random, and it still shaped their judgment.

  • The First Offer Sets the Stage
    In a 2001 study of negotiations, Adam Galinsky and Thomas Mussweiler found that whichever side made the first offer tended to get the better deal, because the final agreement was strongly linked to that opening number. The effect weakened when the other side deliberately focused on the first party’s alternatives, or on their own ideal outcome.

  • The Odd-Looking Asking Price
    In a series of studies published in 2008, Chris Janiszewski and Dan Uy found that precise anchors are stickier than round ones. People who read an asking price for a beach house of $799,800 stayed closer to it in their own estimates than people who read $800,000, because a precise number makes us adjust in smaller steps.


Applying Anchoring Bias

Researchers have found that one of the few things that reliably weakens anchoring is to deliberately think of reasons why the anchor might be wrong, a strategy Mussweiler, Strack and Tim Pfeiffer called considering the opposite. This is the simple process I follow when a number matters:

  1. Make My Own Estimate First
    Before I look at a price, a forecast or someone else’s proposal, I write down my own estimate and how I got there.

  2. Name the Anchor
    I ask where the first number came from, who chose it and why, and whether it has anything to do with the real value.

  3. Argue Against It
    I list reasons why the anchor might be too high or too low, which makes the information that does not fit it easier to recall.

  4. Use Independent Reference Points
    I look up comparable prices, base rates or several independent estimates, so that no single number gets to set the range.

  5. Consider Setting the First Number
    In a negotiation, I prepare well enough that I can make a reasonable first offer myself, rather than letting the other side anchor the conversation.


Pitfalls and Limitations

  • Knowing about the bias is not enough. In a 1996 study, Timothy Wilson and colleagues found that warning people about anchoring and offering them incentives to be accurate did little to remove its effect.

  • Expertise does not protect me. As the studies of real estate agents and judges show, people with experience are anchored too, and they may be less likely to notice it.

  • Not every first number is a bad anchor. When the starting point is a good estimate, such as a well-researched market price or a reliable base rate, adjusting from it is sensible. The problem is relying on it without asking where it came from.

  • Anchors come in many forms. Last year’s budget, a previous salary, a price shown as reduced from a higher amount or an early forecast can all quietly set the range for what follows, which is why I try to notice them before I start adjusting.


Related Mental Models

Anchoring Bias reminds me that my judgments often begin long before I start thinking, with whatever number happened to arrive first. By making my own estimate before I see anyone else’s, asking where each starting point came from and deliberately arguing against it, I give myself a better chance of landing on the answer the facts support. For more on how first impressions and context shape our choices, see my summary of Predictably Irrational and my guide to Effective Decision-Making.



To learn more about mental models and philosophy,
check out: Wise: How to Attain a Deep Understanding of the World, History, Our Place in It, and How to Live Well



📚 Sources and References