Cognitive Biases in Trading Decisions

Summary

Confirmation bias, anchoring and loss aversion offer ways to examine decisions without assuming a universal emotional response.

Research and individual decisions

Behavioural research studies patterns in how people evaluate uncertain outcomes. Those patterns can help frame journal questions, but they do not diagnose an individual or guarantee better financial results. No level of intelligence or discipline removes market uncertainty.

Confirmation bias

Confirmation bias describes giving more attention or weight to evidence that supports an existing belief. In a practice journal, record the original rationale and at least one observation that would challenge it. Compare how you treated supporting and conflicting information without assuming the exercise removes the bias.

Anchoring

An initial number can influence a later judgement even when it provides little information about current value. A purchase price may become such an anchor. For a worksheet, compare the reasons for keeping a hypothetical position with the reasons for choosing it today; also record costs and constraints. That comparison is a reflection prompt, not an automatic instruction to sell.

Loss aversion and the two-times shorthand

Loss aversion describes losses receiving greater weight than comparable gains relative to a reference point. Tversky and Kahneman's 1992 model estimated a loss-aversion parameter from experimental choices. The familiar 'about twice' shorthand is not a universal measurement of emotional pain. Later studies estimate different magnitudes across people, tasks and methods, and a model parameter should not be treated as an individual trading rule.

A journal exercise

Write down your expected reaction to a hypothetical $100 gain and $100 loss, then describe how the reaction might change with the starting balance or reference point. This exercise observes your reasoning; it does not measure a validated loss-aversion coefficient. A written plan can make decisions easier to review, but it cannot guarantee an exit price or prevent loss. TradeHQ does not execute automatic stop orders.

Recency bias and sample limits

Recency bias can mean giving disproportionate weight to recent observations. A short winning or losing run does not establish a reliable strategy. Compare recent and longer journal periods while noting that market conditions may have changed. No fixed trade count or rolling win-rate window eliminates bias or proves an edge.

(Educational only — not financial advice. TradeHQ offers $100,000 virtual cash for supported spot instruments. Options contracts, Greeks, futures margin and settlement are conceptual exercises; the simulator does not execute them.)

Key takeaways

• Record supporting and conflicting evidence when reviewing a practice decision.
• A purchase-price anchor is a reflection prompt, not an automatic exit rule.
• Loss aversion varies; the two-times shorthand is not a universal emotional ratio.
• Journal samples and written plans cannot guarantee returns or execution.

Check your understanding

• Loss aversion describes: Options: Identical reactions from everyone; Giving losses more weight than comparable gains relative to a reference point; Preferring losses to gains; Avoiding all trades. Correct answer: Giving losses more weight than comparable gains relative to a reference point. Why: The concept concerns asymmetric evaluation around a reference point. Its magnitude varies and is not a universal two-times emotional ratio.
• Which example illustrates a possible purchase-price anchor? Options: Treating the entry price as the only reason to keep a position; Recording transaction costs; Comparing company revenues; Reading conflicting evidence. Correct answer: Treating the entry price as the only reason to keep a position. Why: An entry price can anchor judgement without establishing current value or dictating whether a position should be kept.
• Which journal exercise examines confirmation bias? Options: Record only supporting headlines; Record both supporting and conflicting evidence; Assume a profitable result proves the reasoning; Ignore uncertainty. Correct answer: Record both supporting and conflicting evidence. Why: Comparing how both kinds of evidence were treated can help examine reasoning; it does not guarantee the bias disappears.
• Recency bias can lead someone to: Options: Give all observations equal weight; Over-weight the last few outcomes; Ignore all recent evidence; Know the next market move. Correct answer: Over-weight the last few outcomes. Why: Recent outcomes can receive disproportionate weight, while neither a short nor a fixed-size sample proves an edge.

Sources

• Tversky & Kahneman — Cumulative Prospect Theory (1992) (https://doi.org/10.1007/BF00122574)
• Walasek, Mullett & Stewart — A Meta-analysis of Loss Aversion in Risky Contexts (2024) (https://wrap.warwick.ac.uk/id/eprint/185745/13/1-s2.0-S0167487024000485-main.pdf)

Practise this lesson

Use $100,000 in virtual cash to practise supported spot instruments and general habits such as recording a trade thesis. Options contracts, Greeks, futures margin and daily settlement are conceptual exercises: calculate their examples in a worksheet, because TradeHQ does not execute derivatives contracts. Educational simulation only — not financial advice.