Psychology

What is Anchoring Bias?

Reviewed by Karthikeyan Anbarasan (Founder, TradeLyser) · Updated 19 Oct 2026 · Editorial policy

DefinitionAnchoring bias is leaning too heavily on the first number you saw, such as your entry price, yesterday’s close or a price target, when deciding what to do next. A trader who bought an illustrative stock at ₹1,480 and refuses to sell below breakeven after the reason for the trade has failed is anchored to ₹1,480, a number the market neither knows nor cares about.

How anchoring works

Amos Tversky and Daniel Kahneman described anchoring in their work on judgement under uncertainty: people start from an initial number and adjust away from it too little, even when the number has little to do with the question. In trading the anchor is usually a price you saw or paid, and the decision it distorts is when to exit, add or enter.

AnchorHow it shows upBetter reference
Your entry priceHolding a loser "until it comes back"; moving the stop below the planThe planned stop and current structure
Yesterday’s close or highCalling a gap-up stock "too expensive" and skipping a valid longToday’s opening range and VWAP
52-week high or IPO priceAveraging down because "it was ₹2,000 last year"The current trend and your risk per trade
The day’s first moveRefusing a valid short because the morning was bullishThe setup rules as they stand now
An analyst or social media targetHolding past your exit because the target is "only 3% away"Your written exit and R-multiple

A worked example

Illustrative, before charges. A trader buys 200 shares of a large-cap NSE stock at ₹1,480 with a stop at ₹1,450, so the planned risk is ₹30 × 200 = ₹6,000, or 1R. The stock reaches ₹1,450 and the trader cancels the stop, waiting for "at least breakeven". It falls to ₹1,395, then bounces to ₹1,430, where the trader sells in relief.

Planned loss at the stop
−₹6,000
Open loss at ₹1,395
−₹17,000
Actual exit at ₹1,430
−₹10,000
Cost of the anchor
−₹4,000

At the low the open loss was ₹85 × 200 = ₹17,000, or 2.83R. The exit at ₹1,430 lost ₹50 × 200 = ₹10,000, or 1.67R, which is ₹4,000 more than the plan allowed. Breakeven at ₹1,480 was never reached, and the decision to hold was driven by that number rather than by anything on the chart. The bounce to ₹1,430 also felt like relief, because it was closer to the anchor than the low had been, so the trader sold there. Measured against the plan, it was simply a 1.67R loss. Anchoring works in the other direction too: a trader who sold this stock at ₹1,450 may refuse to buy it back at ₹1,470 on a fresh valid setup, because ₹1,470 now feels expensive compared with the price they sold at.

Spotting it in your journal

  • Stops moved further away after entry, especially just before they would have been hit.
  • Exits at exactly the entry price after a long hold, logged as "scratched".
  • Averaging down on a position that has already passed its planned stop.
  • Notes that mention a past price: "it was ₹1,520 last week", "it opened higher than this".
  • Losses larger than 1R on trades that had a stop in the plan.
  1. Before entry, write the stop, the target and one sentence on what would make the trade wrong.
  2. Track open positions in R from the planned stop, not in rupees from the entry price.
  3. When tempted to hold or add, ask: if I had no position, would I enter here with this stop? If not, exit.
  4. Each Friday, filter for trades where the stop was moved and compare their average R with the rest.
Rule of thumb
Any loss larger than 1.2R on a trade with a written stop deserves a one-line note on why. If the answer mentions the entry price, it was anchoring.

Anchoring vs sunk cost

Anchoring bias

A reference number, such as the entry price or an old high, distorts what you think a fair price or exit is.
Typical note: "it should get back to ₹1,480"

Sunk cost fallacy

Money or effort already spent keeps you in a position, because exiting would make the loss feel final.
Typical note: "I have already lost too much to sell now"
For F&O traders in India
The premium you paid is a strong anchor. An option bought at ₹180 that falls to ₹90 is often held "until it comes back", while time decay keeps working against it, most of all in the last sessions before expiry. Base option exits on the underlying level written in your plan, not on the premium you paid.

Reference guide

ContextValueReading
Exit decided by the planned stop and the setupUnanchoredThe entry price sets the size of the loss, not whether to hold
Holding a loser until it gets back to breakevenAnchored to entryAsk whether you would buy it now at this price with this stop; if not, the anchor is doing the holding
Target taken from an old high or an analyst numberAnchored targetBase targets on current structure and R, and write the source of each target in the plan

Questions

What is anchoring bias in trading?

It is giving too much weight to a reference number, usually your entry price, a previous close or a target, when deciding what to do now. It shows up as holding losers for breakeven, moving stops and skipping valid trades because price "already moved".

What is an example of anchoring bias?

Buying at ₹1,480 with a stop at ₹1,450, then cancelling the stop to wait for ₹1,480 again. The plan allowed a ₹6,000 loss on 200 shares; waiting for the anchor can turn it into ₹10,000 or more.

How do I avoid anchoring bias?

Decide the stop and target before entry, measure open trades in R rather than rupees from entry, and ask whether you would take the trade fresh at the current price. Review every trade where the stop was moved.

Is using support and resistance a form of anchoring?

Not by itself. Levels that many traders watch can matter because of how orders cluster around them. It becomes anchoring when a personal number, like your own entry, overrides what the plan and the current structure say.

What is the difference between anchoring and confirmation bias?

Anchoring fixes on a number. Confirmation bias filters information to support a view. They often work together: the anchor sets the view and confirmation bias collects reasons to keep it.

Related terms