What is anchoring bias?
- Anchoring bias makes you lean too hard on the first number you see, even when that number is random or meaningless.
- Psychologists Daniel Kahneman and Amos Tversky proved the effect in 1974 with a rigged roulette wheel that shifted average guesses by 20 percentage points.
- In investing, a past-price anchor distorts buy, hold, or sell decisions, making you hold losers and sell winners too early.
- Knowing about the bias does not stop it; even experts and cash rewards fail to remove the pull.
- The best defense is the consider-the-opposite method: force yourself to argue against your first number before you act.
What is anchoring bias?
Anchoring bias is the human tendency to rely too heavily on the first information you see. That first number becomes your anchor to a reference price, an arbitrary reference, even when it is pure chance.
You get stuck on a number that has no real meaning. Then you adjust your thinking around it, but never enough. The result? Bad calls on prices, trades, and deals.
Where does anchoring bias come from?
Kahneman Tversky, the psychologists whose names became shorthand for the effect, coined it in 1974. They asked people to guess the share of African nations in the UN. A rigged wheel landed on 10 or 65. The 10 group guessed 25 percent. The 65 group guessed 45 percent.
Their second test used math. People had to multiply 1 through 8 in five seconds. Those who started with small numbers guessed 512. Those who started with big numbers guessed 2,250. The real answer was 40,320.
Why anchoring works
Anchoring and adjustment drives it. You grab the first number and nudge your answer toward it, never far enough. The anchor can be external, a listing price or a first offer; internal, the price you already paid. Both pull you back to the first number.
How anchoring bias shows up in real life
Dan Ariely found that the last two digits of your Social Security number can set your bids. Participants wrote them down, then bid on wine and gadgets. High digits meant 60 to 120 percent higher offers, with zero link to value. A high listing price does the same to buyers.
The pull is not only numeric. Anchoring reaches physical judgments like length, weight, and volume, so a heavy first object changes how heavy the next one feels. Even absurd anchors work. Students asked if Gandhi died before age 9 guessed 50 on average. Students asked before 140 guessed 67.
How sellers weaponize your anchors
Menus and price lists turn the bias against you. A restaurant shows its costliest dish first, and the cheaper plates look fair by comparison. That top price is your anchor, and you judge every other dish against it.
Decoy pricing leans on the same trick. A bundle's parts sold separately make the set meal look like a steal. Retailers set those parts high, and a precise anchor like $799,800 drags your estimate lower than a round $800,000 ever would.
Negotiation runs the same play. Whoever names the first number sets the range, so the side that goes first often lands closer to its price. A salary talk works the same way; the opening figure becomes the anchor. Sometimes you hand yourself the anchor.
How anchoring bias distorts your investing
Here is where it hurts you most. A past-price anchor distorts buy, hold, or sell. You bought at 100. It drops to 70. You refuse to sell because 100 is your number. That anchor costs you real money.
App-based investors show the same trap. Your first stock purchase price becomes your anchor. Start small and you stay small. The study found small first buys led to less wealth over time. You anchor to what you paid, not what the company is worth, so you hold losers too long.
Why you can't just avoid it
Knowing the bias does not stop it. Studies warned people the anchor would contaminate their answers. They still moved toward it. Even cash rewards failed to fix the pull, and no amount of warning removed it.
Experts are not safe. Real estate agents denied being influenced by listing prices. The data showed they were just as biased as students. Judges, doctors, and analysts all show the same pattern.
The anchor does not fade with time. A week after the anchor was set, people still answered to it, as if it were still fresh. The first figure keeps pulling even after you know it was random. Understanding can blunt it, but it rarely dissolves on its own.
How anchoring bias spreads through groups
Anchoring is not just a solo habit. When one board member names a price first, the rest adjust to it instead of naming their own, so that figure becomes the shared anchor for the group. The social pull compounds the error once it is on the table.
Why machines inherit your anchors
Automated systems do not escape the bias, because the humans using them still carry it. A business intelligence tool that surfaces one plausible number can shape a forecast, and the software offers no cure for it. A system that filters a bad anchor can still leave you inside its reach.
Algorithms are trained on data that already carries human bias. If the examples that built the model lean on a skewed first value, the machine reproduces the anchor. A model can inherit the very misjudgment it was built to correct, so automation amplifies anchoring instead of removing it.
How to beat anchoring bias
The best fix is the consider-the-opposite method. Before you decide, force yourself to argue the other side. Ask what the price would be if the anchor never existed. Write it down before you act.
Set your own numbers first. Decide a fair price before you look at any listing. Decide a sell point before you buy. That way your anchor to a reference price is yours, not someone else's. Get a second opinion before you commit.