Emotional Stages of Trading: How Well Do the Stage Models Hold Up?
By Sofia Harchich | Trading Psychologist & Behavioural Finance Writer | thewealthmirror.com
Emotional stages of trading appear in almost every course on the subject, usually as a numbered sequence with a diagram, and almost never with a source.
Five stages, seven stages, a curve running from optimism through euphoria to despondency and back. The models are memorable, they describe something recognisable, and they are presented with a confidence that their evidential basis does not support.
That matters practically rather than academically. A trader who believes they are at stage three of five is making predictions about what comes next, and acting on them. Four claims, read for what each one carries.
Claim one: the sequence is a validated psychological model
Not supported, and the provenance is worth knowing.
The stage structures circulating in trading education are adapted, mostly informally, from stage models of grief developed in a clinical context in the late 1960s. Those original models were observational, were not proposed as a universal sequence, and have themselves been substantially revised in the decades since — the finding being that grief does not reliably proceed in stages, and that individuals vary enormously.
Applying a contested clinical model, without adaptation, to an unrelated domain, and presenting the result as established psychology, is a considerable stretch. Nothing in the trading literature validates the sequence, and nothing outside it does either.
This is not an argument that the stages describe nothing. It is an argument against the confidence with which they are drawn.
Claim two: the emotional stages of trading run in order
Not supported, and this is the practically damaging part.
The linearity is the load-bearing assumption. Remove it and the model becomes a list of states rather than a map.
What behavioural records actually show is oscillation without a clean direction: a trader can move from confidence to doubt to confidence within a single session, and the ordering across weeks is driven by outcomes rather than by any internal progression. The same trader will occupy several of these states within a fortnight depending on what the market did, and the sequence will differ next fortnight.
The cost of believing in the ordering is specific. It produces the expectation that a bad period will be passed through if endured, which is a reason to change nothing. Emotional stages of trading, believed literally, is one of the more effective arguments for waiting available to a trader who should be adjusting something.
Claim three: knowing which stage you are in helps
Not supported, and the reason generalises.
Identifying a stage produces a label rather than an action. I am in the despondency phase names an experience and prescribes nothing, and the prescriptions usually attached — be patient, this passes — are the same regardless of which stage is named.
There is a further problem. Self-identification of an emotional stage is performed by someone inside it, and it drifts toward whatever the recent outcome suggests. Self-serving bias operates freely here: good stretches get attributed to progression through the model, bad ones to being temporarily at a difficult stage — and both readings preserve the belief that the sequence is running.
A model that accommodates every outcome as confirmation is not producing information.
Claim four: something real is being described
Supported, and this is what survives.
The models persist because they capture something genuine, and dismissing them entirely would be as sloppy as accepting them uncritically.
Three observations underneath the stage structures do hold up.
States vary and the variation is not random. Traders do move between recognisable conditions, and the conditions are describable.
The state affects the decision quality more than the analysis does. This is well supported and it is the useful core.
Extremes are self-reinforcing. Confidence produces size, size produces outcomes, outcomes produce confidence — and the same loop runs downward. This is a real mechanism.
None of that requires a sequence, a diagram, or a number of stages. It requires a way of measuring state, which the models do not provide.
What the stage models describe as a long arc is closer to the instability loop repeating on a short one — a trigger, a state, a behaviour, an outcome and a belief, cycling within a single week rather than progressing over a career.
What replaces the sequence
Three states rather than five or seven, defined physiologically rather than emotionally, and logged rather than identified.
Settled, activated, or flat — one word, recorded before each session, before any chart is opened. Polyvagal theory describes the underlying conditions: a regulated state supporting engagement, a mobilised state supporting fight or flight, and a shutdown state supporting immobilisation.
Four advantages over the stage models, and they are the reasons to make the swap.
- It is recorded rather than inferred, so it does not drift toward the outcome
- It carries no ordering, so no week is expected to lead to any particular next week
- Each state has a different intervention — down-regulation, up-regulation, or nothing — which is more than any stage model provides
- It produces a testable relationship with rule adherence within ten sessions
The fourth is the one that matters. Emotional stages of trading generate no prediction that could fail. A state word logged beside an adherence proportion generates one immediately: either your deviations cluster by state or they do not, and finding out takes a fortnight.
A model that cannot be wrong about you is not telling you anything about you.
What a state column produces that a stage never did
The practical case for the swap is best made by what the record actually returns after a month.
A correlation, or its absence. State word against rule adherence, ten sessions. Either deviations cluster by state or they do not, and both answers are useful. Emotional stages of trading produce no equivalent — there is no observation that would count against being at stage three.
An intervention that differs by reading. Activated calls for down-regulation, flat for up-regulation, settled for nothing at all. The stage models offer the same counsel at every stage, which is to endure it.
An asymmetry most traders have never noticed. The flat state is the one that reads as composure from outside and produces skipped valid entries, undersized positions and a rulebook that quietly gets stricter. No stage model distinguishes it from patience, because both look like inactivity, and the distinction is only available physiologically.
A frequency count. How often you actually arrive settled. Most traders discover the answer is lower than they assumed, which redirects attention to the pre-session hour — a place no stage model points at, because a stage is something you are in rather than something you arrive with.
Four returns from one word logged before the open, against a diagram that generates none of them.
One thing the swap gives up, and it is worth acknowledging. The stage models tell a story, and a story is easier to hold on a bad month than a column of three-letter entries. Traders who make the change sometimes miss the narrative, which is a real cost rather than an imagined one.
The compensation arrives at about week six, when the column has produced a finding. A trader who can say my deviations occur on activated days and my activated days follow poor sleep has something a stage model could never supply — a specific, testable, personal regularity, discovered rather than inherited.
Where the stage models remain useful
One legitimate use survives the audit, and it is worth preserving.
As a normalising device, the models do real work. A trader who recognises their experience in a described sequence learns that the experience is common, which is genuinely valuable — concealment is what makes most of these patterns worse, and a diagram that makes someone feel less singular has earned its place.
The error is treating a normalising device as a diagnostic one. Emotional stages of trading are good at telling you that you are not unusual and poor at telling you what to do on Tuesday, and almost all of the material treats them as though the second followed from the first.
The prediction worth testing
If you want to check the stage model against your own record rather than taking this article’s word for it, there is a clean test.
Stage models predict ordering: that a given state is followed more often by a specific next state than by others. Log your state word daily for a month, then build a simple table of what followed what.
If the transitions are patterned — activated reliably following settled, flat reliably following activated — the ordering claim has support in your case, and it is worth knowing. If the transitions look close to random with respect to yesterday and closely tied to what the market did, the ordering is coming from outcomes rather than from any internal sequence.
Most records show the second. A minority show something patterned, and those traders have discovered a genuinely useful personal regularity that no published stage model could have told them.
What the models are pointing at
Underneath the audit sits the reason these diagrams keep being drawn.
Trading generates a great deal of emotional experience with very little structure to it. Outcomes arrive unpredictably, the feedback is noisy, and nothing about the experience arranges itself into a story. A stage model supplies an arrangement — a shape, a direction, and the implication that difficulty is a phase rather than a permanent condition.
That is a real need and it is not a foolish one. The trouble is only that the arrangement is imposed rather than found, and an imposed structure will generate expectations the record does not support.
What a logged state column offers instead is less satisfying and more accurate: no narrative, no progression, and a testable relationship between how you arrived and what you did. Emotional stages of trading become useful at exactly the point where the sequence is dropped and the measurement is kept.
For the cluster hub, read fear and greed in trading; for the diagnostic that replaces stage identification, emotional trading.
The pattern quiz identifies which emotional pattern your behaviour under pressure fits, and unlike a stage, a pattern does not imply that anything comes next.
One word before the open, for a month
One word, before the open, for a month.
Settled, activated, or flat — recorded before any chart is visible, beside a rule-adherence proportion after the close.
At the month, check whether the transitions are patterned or driven by outcomes. The emotional stages of trading make a claim about ordering; that is the test of it, and it is the only test anyone has proposed for a model that has been drawn on whiteboards for twenty years.
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
However well the stage models hold up, the useful version is the sequence running in you rather than the one in the diagram. Part One of The Stability System has you map yours from your own sessions.
