Confirmation, Recency, Anchoring and Hindsight Bias in Market Decisions
Recognise four information biases and use a written bias-check sheet before and after trades.
What you will learn
- How confirmation bias, recency bias, anchoring and hindsight bias appear in trading.
- How to calculate what an anchor really demands from the market.
- How to judge a short losing run against your strategy's normal behaviour.
- How a prediction log protects you from hindsight bias.
Explanation
A cognitive bias is a systematic error in thinking: not a random mistake, but one that repeats in the same direction. The four biases in this lesson change which information you notice and how you judge it.
Confirmation bias
Confirmation bias is the tendency to search for, notice and remember information that supports what you already believe. A trader who is long reads bullish comments, sees every green candle as proof and explains away warning signs. Counter-measure: before entry, write down the invalidation, the specific price or event that would prove your idea wrong, and actively list arguments against the trade.
Recency bias
Recency bias means giving too much weight to the most recent events. After three losses, a sound strategy "feels broken"; after three wins, a weak one "feels great". Counter-measure: compare a recent run with what your tested statistics say is normal (you saw in the risk management course how losing streaks follow from the win rate).
Anchoring
Anchoring is the habit of relying too much on a first number you saw. In trading, typical anchors are your entry price, an old all-time high or an analyst's price target. The market does not know your entry price. Counter-measure: calculate what the anchor demands, and decide based on current structure.
Hindsight bias
Hindsight bias is the "I knew it all along" feeling after an outcome is known. It makes past moves look obvious, which creates false confidence and unfair self-criticism ("How could I miss that?"). Counter-measure: a prediction log, where you write your expectation and confidence before the event, so you can later compare honestly.
Worked example
Hypothetical example: a trader uses a bias-check sheet on three situations. Fees, spreads and taxes are excluded.
1. Anchoring
She bought a share at €50. It now trades at €40, and her plan's stop was €44, already broken. She wants to "wait until it gets back to €50".
- Required rise = (€50 − €40) ÷ €40 = €10 ÷ €40 = 25%.
- Question: "Would I buy at €40 today with a target of €50?" Her chart shows a downtrend with lower highs; the answer is no.
- Conclusion: the €50 is an anchor, not an analysis. She exits per plan.
2. Recency
Her strategy has a tested win rate of 45%, so the loss rate is 55%. She has just lost three trades in a row and wants to change the rules.
- Probability that any three given trades in a row all lose = 0.55 × 0.55 × 0.55 ≈ 0.166, about 1 in 6.
- Conclusion: a three-loss run is a normal event for this strategy, not evidence that it stopped working. She keeps the rules and reviews after the planned sample of trades.
3. Confirmation
Before a long trade she writes three reasons for and is forced to write three reasons against:
| For the trade | Against the trade |
|---|---|
| Higher-timeframe uptrend | Price is close to a major resistance zone |
| Pullback to a former support level | Important economic report in two hours |
| Momentum turning up | Volume on the pullback was rising, not falling |
She defines invalidation: "A daily close below €38.50 proves the idea wrong." Because two of the counter-arguments are significant, she halves her planned risk from 1% to 0.5% of the account, as her plan allows for B-quality setups.
Common mistake
Practice activity
On paper: (a) You bought at €80; the price is now €60. Calculate the rise needed to get back to €80. (b) Write one sentence for a prediction log about an upcoming event, including your expected outcome and your confidence in percent.
Show solution
(a) Required rise = (€80 − €60) ÷ €60 = €20 ÷ €60 ≈ 33.3%. The fall was 25%, but the recovery needs 33.3%, because the base is now smaller.
(b) Example: "Date: Monday. I expect the index to stay inside last week's range until Friday. Confidence: 60%. Reason: no major data releases this week." On Friday, record what happened and whether the reason was right.
Limitations: the three-loss probability assumes trades are independent and the tested win rate is still valid; real markets can change, which is why reviews happen after a planned sample size.
Key takeaways
- Confirmation bias filters information; write the invalidation and the counter-arguments before entry.
- Recency bias overrates the last few trades; compare runs with your tested statistics.
- Anchors like the entry price mean nothing to the market; calculate what they demand and ask the "fresh entry" question.
- A prediction log written in advance is the best protection against hindsight bias.
Next step
Next, you will look at biases that inflate your belief in your own skill: overconfidence, the illusion of control and the hot-hand trap.
Self-check
Answer the questions, then check your answers. You can retry as often as you like.
Automatic self-check for your own revision. Your answers are checked in your browser and are not seen or assessed by us. It is not an exam and not a qualification.
Your progress
“Complete” only means you marked the lesson as done. It is not a test of your understanding. Progress and notes are saved only in this browser on this device — not in an account.