The first number you see before making an estimate quietly biases your final answer, even when that number is random.

In classic experiments by psychologists Amos Tversky and Daniel Kahneman, participants spun a wheel rigged to land on either a low or high number, then were asked to estimate an unrelated quantity, like the percentage of African countries in the United Nations. People who saw the higher number on the wheel consistently gave higher estimates than those who saw the lower number, even though the wheel’s result had nothing to do with the question.

This tendency, called anchoring, has been replicated in negotiations, real-estate pricing, courtroom sentencing, and everyday shopping decisions, and it holds even when people are explicitly told the anchor number is random or irrelevant. It’s one of the most robust findings in the study of judgment and decision-making, and it’s part of why an initial asking price in a negotiation tends to pull the entire conversation toward it.