Behavioural Finance for Traders: A Diagnostic for Telling Bias From Everything Else

By Sofia Harchich | Trading Psychologist & Behavioural Finance Writer | thewealthmirror.com

Behavioural finance for traders is usually delivered as a glossary, and a glossary is close to useless at four in the afternoon with a position open.

Twenty biases, each with a definition and a short example. The reader recognises most of them, finishes the article better informed, and behaves identically the following week. The gap sits in the format rather than the content: a list tells you what exists in the world and says nothing about what is running in your account right now.

This article is built the other way round. It starts from a mistake you actually made and works backwards, and the first job is the one every glossary skips: establishing whether the mistake was a bias at all. Most trading errors attributed to cognitive bias turn out to be three other things, and each of the three responds to a completely different intervention.

Test one: bias, or a knowledge gap?

Take the specific decision that went wrong. Then answer one question honestly: at the moment of the decision, did you know what the correct action was? A trader who is profitable on demo but not live has usually answered it already, because the decisions were made correctly on demo, where nothing was at stake.

If the answer is no — the setup was outside anything you had defined, the instrument behaved in a way you had not studied, the position size was wrong because you had never worked out the arithmetic — this is not a bias. It is a gap, and no amount of behavioural work will close it.

The distinction matters because the two produce identical-looking outcomes and opposite remedies. A trader treating a knowledge gap as a bias will journal about their psychology for six months while the actual missing piece stays missing. A trader treating a bias as a knowledge gap will read a fourth book on market structure and repeat the behaviour on page one of the fifth.

The discriminating detail is what happened before the decision. Bias operates on information you had. If the information was absent, start with the information.

If you knew, continue.

Test two: where behavioural finance for traders stops describing you

Now the harder separation, and the one behavioural finance for traders almost never addresses, because behavioural finance grew out of experiments conducted on calm people at desks.

Log one word before each session for ten sessions — settled, activated, or flat — and mark which sessions contained the error.

If the errors cluster on activated or flat days and are absent on settled ones, you are not looking at a stable cognitive tendency. You are looking at what happens to an otherwise sound decision process when the physiology underneath it changes. The bias literature describes tendencies that show up reliably in resting subjects; what you have is a capacity that degrades under load.

This changes the intervention completely. A bias is addressed by structuring the decision — written criteria, pre-commitment, a checklist. A state problem is addressed physiologically, before the session, and adding more decision structure to it simply produces a more elaborate procedure that gets abandoned at the same point.

The tell is inconsistency. A trader who exhibits a textbook bias three days a week and not the other two does not have a partial bias. They have a decision process that holds in one state and fails in another, and the inconsistency itself is the useful reading rather than noise to be averaged out.

If the errors appear regardless of state, continue.

Test three: bias, or an incentive problem?

The last of the three, and the least discussed anywhere.

Ask what the behaviour produced other than the financial outcome. Cutting winners early produces a reliable sense of competence and a shorter period of uncertainty. Holding a loser preserves the possibility of being right. Overtrading generates evidence of effort. Checking positions at midnight reduces the discomfort of not knowing.

If the behaviour reliably delivers something you want, you are making a trade rather than a reasoning error — comfort now against expectancy later — and it will survive every correction aimed at your reasoning, because your reasoning was never the problem.

A behaviour that pays you in something other than money is an unexamined transaction rather than a bias.

The remedy here is neither structural nor physiological. It is to name the payment explicitly, in writing, and then decide whether you want to keep making the trade. Some traders decide they do, which is a legitimate answer and an enormous improvement on not knowing.

What is left is bias, and it sorts into three families

Anything surviving all three tests is a genuine cognitive tendency, and the useful move is to sort it by which of the five layers it operates on rather than by its name.

Family one: reference-point biases. How the situation is framed determines the decision. Prospect theory is the foundational work here, and it covers loss aversion, the disposition effect, and most of what happens around exits. These live on the behavioural layer and respond to changing what is displayed — the reference point is set by your platform far more often than by your psychology.

Family two: evidence-handling biases. How incoming information is weighted. Confirmation bias, recency, availability. These live at the decision step of the sequence and respond to pre-commitment: the invalidation condition written before entry, when the weighting machinery is not yet under load.

Family three: self-attribution biases. How outcomes are assigned to skill or circumstance. Overconfidence, hindsight, the self-serving pattern. These live on the core-belief layer and respond to almost nothing structural, because the belief being protected is not about markets. They need a second scoreboard rather than a better rule.

Three families, three intervention types. A trader who knows which family they are in has narrowed the field from twenty options to about four.

Once the knowledge gap and the incentive problem are ruled out, what remains is bias acting inside the instability loop — the distortion between a market event and the decision it produces, which is the only place a bias can cost money.

The measurement that makes this behavioural finance rather than a reading list

Every one of the interventions above needs a way of telling whether it worked, and the same two numbers do the job for all of them.

Log, per trade: how early you identified the pattern, scored 0 to 5, and how often you acted against it in the same session, scored 0 to 5.

The first number rises quickly once you start looking. Within three weeks most traders can name their tendency, describe its trigger and predict when it will return. The second frequently does not move for months, and the distance between them — the awareness–action gap — is the only figure in this article that reports on your actual progress.

This is where behavioural finance for traders parts company with behavioural finance in general. Knowing the taxonomy improves the first number, which is not the number that costs you money. What closes the second is structure, physiology or an honest look at the payment — which is why the three tests come before the glossary rather than after it. Before paying for a behavioural finance course, the useful question is which of the two numbers it is built to move.

Running the three tests on one real mistake

The sequence is quicker in practice than it reads, and running it once on a real error is worth more than a second pass through the theory.

Take an actual trade. Say a position cut for a small profit that then ran a considerable distance further, and say it happened three times last month.

Test one. At the moment of the exit, did you know the correct action? If your plan specified a target and you exited before it, yes — you knew. If the plan never specified where to exit winners, the answer is no, and this is a gap rather than a bias. That single question resolves a surprising proportion of cases, because a great deal of what gets attributed to psychology is an undefined exit.

Test two. Mark those three sessions against your pre-session state word. If all three were activated days and your other twelve exits were on settled days, the decision process is intact and degrades under load. Behavioural finance for traders has essentially nothing to offer here; the intervention is physiological and belongs before the open.

Test three. What did cutting early deliver? Almost always: certainty, and the end of an uncomfortable interval. That is a genuine payment, received reliably, and it explains why the behaviour survived three repetitions and a stern review.

Only if all three tests come back clean does the glossary become relevant — and then the family sorting narrows it further, in this case to the reference-point family, where the remedy is a display change rather than a resolution.

Worth noticing how much has been ruled out by four minutes of questions. What began as I have a psychological problem with cutting winners early has become one of four specific things, and only one of those four is psychological in the sense the phrase implies.

That is what behavioural finance for traders is actually for. Not a taxonomy of the ways minds go wrong, but a method for finding out which of a small number of things is happening in your account this month.

Why the glossary approach persists

Worth being fair to it, because the standard treatment is not stupid and the researchers behind it were doing something else entirely.

The biases were catalogued in controlled experiments designed to isolate reasoning from everything around it — no money at risk, no physiological load, no identity attached to the answer. Under those conditions the tendencies show up cleanly, which is exactly what makes the findings robust and exactly what makes them incomplete for a trader.

A person at a desk with size on is not the subject in that experiment. They have all three of the things the experiment removed, and behavioural finance for traders has to account for them or it becomes a very well-evidenced description of a situation nobody is in.

None of that invalidates the research. It means the research is one of three inputs rather than the whole account, and the three tests above are how you find out which input you need.

What this does to the reading you have already done

Most traders arriving here have read the standard material and have found it accurate and inert — recognisable, true, and somehow not usable on a Tuesday.

That is not a failure of comprehension, and it is worth saying plainly, because the usual conclusion is that the reader is insufficiently disciplined to apply what they understand. The material was assembled to explain a class of reasoning error, and what most traders bring to it is a mixture of one reasoning error, one knowledge gap, one state problem and one unexamined transaction. A tool built for one of the four returns nothing on the other three, and the reader concludes the fault is theirs.

Sorting the four is most of the work. What remains after sorting is smaller, more specific, and considerably more tractable than the original problem felt.

Where to go from the result

• Reference-point family: why traders hold losing positions and sunk cost fallacy in trading
• Evidence-handling family: confirmation bias in technical analysis and recency bias trading
• Self-attribution family: why intelligent investors lose money
• State problem rather than bias: what is trading psychology

If the three tests did not resolve cleanly — which happens, and usually means two things are running at once — the pattern quiz identifies which emotional pattern your behaviour under pressure fits, and that usually indicates which of the two to treat first.

---

About the Author

Sofia Harchich is a Trading Psychologist and Behavioural Finance Writer with a Master’s in Psychology. She works at the intersection of Jungian shadow work, neuroscience, and market behaviour — helping traders understand the psychology driving their decisions, not just the strategy.

Read more at thewealthmirror.com/about