Anchoring Bias, explained.
Anchoring bias is the influence of an initial number or reference point on a later estimate, including when the starting value is weak or irrelevant evidence.
Why it happens
An anchor can shape which values feel plausible and where adjustment begins. Moving away from it is not enough if the adjustment remains insufficient. Relevant starting information may be useful; arbitrary numbers deserve a different treatment.
Judgment can remain influenced by an initial value even when better evidence is available. This demonstration makes the adjustment rule visible; it does not measure your susceptibility.
Read the result
Change the anchor while holding the independent evidence fixed. The model makes anchoring influence explicit through its controls; it does not measure your susceptibility or prove that every estimate follows the displayed formula.
A worked example
Estimating a job's duration
Someone suggests the work should take four hours before anyone examines the requirements.
The group adjusts upward to six hours, but an independent breakdown supports twelve.
Generate an estimate from the actual tasks before comparing it with the initial suggestion.
OPTIONAL DEEPER DETAILGo deeper: inside the model
Inside this model
Comparable items have a fixed evidence-based estimate of 80 units. The toy estimate is anchor + adjustment × (80 − anchor). At full adjustment the anchor has no effect; at zero adjustment the estimate equals the asking price.
Where this idea is useful
A practical use
When evaluating a used bicycle, gather comparable sales before considering the seller's asking price. This gives your estimate an independent starting point.
A common misconception
“Any use of a reference number is a bias.”
A relevant, well-supported reference may be informative. Bias arises when the starting value receives more influence than the evidence justifies.
What this explanation leaves out
- A linear adjustment rule is an illustration, not a fitted psychological law. Some anchors contain useful information and responses differ across people and contexts.
How can I reduce anchoring in an estimate?
Build an independent estimate, consider a plausible range, and examine reasons the initial value could be wrong. These practices improve the comparison; they do not guarantee immunity to anchoring.
Which number arrived first, and what evidence supports it?
Associated thinkers
Further reading
Explore the original research or the teaching reference behind this experiment.