FOMO and information hygiene for traders – how information selection shapes decisions
When you see a rapid price move or someone else’s profit, it is easy to feel that an opportunity is passing you by. That stimulus does not tell you whether your method’s conditions are met, and a single published result does not show every attempt or loss. This article explains how to distinguish a planned entry from a decision made under pressure, why visible results can distort comparisons, and how to organise information flow. Instead of a universal list of prohibited sources, you will learn how to assign each source a function, define windows for using it and record your own decision conditions. This comes before the reset covered in the previous article: it concerns the information that reaches you before an order is placed.

Follow the claim upstream before counting confirmations
Several messages can contain the same piece of information. To assess them, ask where each claim came from and what, if anything, it adds. Counting the messages without tracing their origin can make repetition look like corroboration. The audit concerns the sources, not whether the reader felt pressure.
Imagine a hypothetical release at 14:00:00 UTC. An automated headline quotes it at 14:00:01. A blog repeats that headline at 14:00:04. A group post shares a screenshot of the blog at 14:00:06. At 14:00:08 your phone displays all three messages. In this invented chain, they trace to one original release. The screenshot and repost add distribution paths; they do not add independent observations of the release value.
| Message | Declared upstream source |
|---|---|
| Automated headline | Original release |
| Blog post | Automated headline |
| Group screenshot | Blog post |
Retain the original document, reference period and version where available. A later correction must remain distinguishable from the earlier version. A second author repeating the first is not the same evidence as a separate measurement.
Keep the source test separate from the entry test
Suppose the plan written before the release permits a hypothetical long entry only after the original value is verified and only while the observed Ask is no higher than 1.1040. The example demonstrates a rule check; it does not propose a profitable news strategy. At 14:00:08 the observed Ask is 1.1048, with one pip defined as 0.0001.
The snapshot is 8 pips above the stated ceiling. Even if the original release value is verified later, that does not make the earlier price observation satisfy the rule. Source verification and price eligibility are separate conditions. Neither a headline’s confidence nor its number of reposts changes the written ceiling.
An observed Ask is also not a guarantee of an available fill for any quantity. Preserve the quote timestamp, quantity assumptions and actual execution record if an order is sent. A decision can fail the stated entry condition without its subsequent trade losing money. Conversely, a later favourable move does not rewrite which condition was met when the decision was made.
Make a changed rule visible instead of editing the old record
At 14:00:11 the original release is checked. At 14:00:12 the trader changes the ceiling to 1.1050 after seeing the price. Save that as a new version with its own time and rationale. Do not overwrite the earlier ceiling and then describe an earlier request as compliant with it.
A method may legitimately permit updates when new information arrives. Whether this update is allowed depends on the update rule that existed beforehand. The example does not ban revision; it makes the distinction between a permitted revision and an undocumented exception checkable. An entry after a price move is not automatically a process breach.
The record should separate the source message, the rule version used, the observed conditions, the request and the eventual result. These fields show what could be checked before an action. They do not identify an emotion from an order timestamp or establish that a notification caused a particular trade.
Give each information source a job you can inspect
For this workflow, the original document verifies a value, the quote record checks a price condition and the order history verifies execution. Commentary may raise a question to investigate; it does not automatically replace any of those records. Choose the source role before an urgent message arrives, and keep unresolved claims marked as unresolved.
Barber and Odean’s research examined attention and stock-buying behaviour in investor records. It did not test this forex workflow or prove that turning off notifications improves profit. Our chain is an original illustration of evidence tracking, not a measurement of a psychological effect.
During a later review, look for duplicated origins, missing timestamps and rule versions changed without a retained earlier copy. Those are concrete information-process problems to repair. The full lesson develops selective publication, comparisons with other traders and the difference between a planned entry and a decision that departs from earlier criteria.
Three exercises with worked answers
The exercise data are hypothetical. Choose an answer, then check the calculation.
From a message chain to a deliberate information workflow
The detailed lesson retains its original publication-filter calculation, illustrations, process examples and research sources. Use it to examine what reaches your screen and what each source can actually establish.
1. What you cannot see in someone else's result – the denominator problem
Let us start with something that is not yet psychology, only arithmetic. When you see a published result, you see the numerator: one account, one week, one screenshot. For that numerator to mean anything, you need a denominator – the number of everyone who could have shown a result, and the number of those who did not. The denominator is not on the screen and usually cannot be reconstructed. This is not the fault of any particular service or any particular author. That is how every set behaves when the person publishing decides whether to publish.
It is easiest to see this in a calculation that contains no skill by assumption.
Assumptions (illustrative values; these are not market data): 1,000 accounts · 5 sessions · probability that a single session ends in profit equal to 0.5 · sessions independent · no edge, a pure coin toss.
The probability that one particular account closes all five sessions in profit is calculated by compounding, that is by multiplication: 0.55 = 1/32 = 3.125%. The additive calculation – 5 × 50% = 250% – produces a meaningless result, and that is exactly why it is worth putting side by side. The method is not a technical detail; it decides whether the number means what it appears to mean.
| Profitable sessions out of five | Formula | Probability | Expected number of accounts out of 1,000 |
|---|---|---|---|
| 5 of 5 | 1/32 | 3.125% | 31.25 |
| 4 of 5 | 5/32 | 15.625% | 156.25 |
| 3 of 5 | 10/32 | 31.25% | 312.5 |
| 2 of 5 | 10/32 | 31.25% | 312.5 |
| 1 of 5 | 5/32 | 15.625% | 156.25 |
| 0 of 5 | 1/32 | 3.125% | 31.25 |
| Total | 32/32 | 100% | 1,000 |
The table has been checked in three relations: every cell against its own formula (the number of combinations divided by 32), the sum of the components against the whole (32/32 and 1,000 accounts), and every cell against its neighbour in the same row (number of accounts = probability × 1,000). The fractional values in the last column have been left unrounded – after rounding the total would stop adding up, and it is precisely the total that proves here that nothing dropped out.
Now the most important part. Assume a hard filter: only someone who closed all five sessions in profit publishes a result. Among the published posts the share of the "perfect week" is 100%. In the population those posts come from – 3.125%. The difference can be given in two ways and both are worth seeing next to each other: additively it is 96.875 percentage points, in terms of compounding it is a 32-fold overstatement of the frequency. The number 32 is not a coincidence – it is exactly 25 from the calculation above.
The hard filter is unrealistic, though, because people also publish weaker weeks. So take a mild one: everyone who had 4 of 5 or 5 of 5 publishes. That gives 156.25 + 31.25 = 187.5 publishers, or 18.75% of the population. Among their posts the perfect week accounts for 31.25 / 187.5 = 16.667%, against 3.125% in the population. Additively: a difference of 13.54 percentage points. Compounding: a 5.33-fold overstatement. Nobody lied, nobody faked a screenshot, no result is false – and the picture is still shifted more than fivefold.
What this calculation does not say. It does not say that people who publish have no skill – the coin-toss assumption was made in order to see the filter effect in isolation; describing anyone's results was not the point here. Nor does it say what the true distribution of results among retail traders is; that cannot be established from visible publications, and that is exactly the heart of the matter. It says one thing: a publication filter produces an overstated picture even when every result shown is true. Why a long losing streak sits within the mathematical norm and why it is not a signal of failure – that is the subject of the article on trading under uncertainty.
2. FOMO is visible in the record, not in the feeling
FOMO stands for fear of missing out. In 2013, Przybylski and colleagues developed a measurement scale and investigated its correlates[1]; they did not coin the term itself. The paper comprised three studies: an international online sample (N = 1,013), a nationally representative British sample of working-age adults (N = 2,079), and 87 first-year university students. It concerned well-being and social media use, not financial decisions. We can use the concept to describe pressure, but applying the findings to trading is an extrapolation rather than a result of that research.
What is needed, then, is a definition that can be checked in documents. The feeling that something is getting away is not in itself a mistake, and no procedure can be built on it: it does not appear in the order history, and reconstructed after the fact it is already distorted by the outcome.
Working definition. This article examines cases in which pressure to avoid missing an opportunity accompanies a departure from criteria established before the stimulus. The trace is a difference between the earlier plan and execution. Not every change in a criterion is FOMO: an update allowed by the method may be valid. A record of a breach alone also does not prove which emotion caused it.
Such a definition has one practical advantage: it makes FOMO checkable. The table below lists the changes that leave a trace and – equally important – what each trace does not yet prove.
| What is visible in the record | How to measure it | What it does not yet prove |
|---|---|---|
| Entry without a satisfied condition | The "entry condition" field empty, or filled in after the order was placed | A method may be changed – check whether the change is written down and predates the entry |
| Entry price far from the level written down | The difference between the order price and the level from preparation | Slippage and a price gap are not the same thing as chasing – first check whether the level was written down at all |
| Very short time from stimulus to order | The timestamp of the stimulus and the timestamp of the order | A fast decision can be planned; if the condition was ready, time alone is not an accusation |
| A skipped checklist step | A missing tick while the order of the remaining fields is preserved | A checklist can be badly designed – a dead checklist is a separate problem, described in the article on procedures |
| An instrument outside your own list | The instrument name against the list from preparation | Extending the list is allowed if it was written down before the session |
| Position size outside the rule | The value against the rule from the plan | The rule comes from risk management and may be changed – what counts is when it was changed |
| Entry on someone else's call without your own condition | The "decision source" field in the record | Someone else's information can be a valid input to your own condition – see section 7 |
The last column is not hedging. Labelling a behaviour "FOMO" after the fact does not prove that pressure caused it; it is still a label fitted to the outcome. Before you attach it, check the simpler explanations: market conditions at that moment – a widened spread, a price gap, elevated volatility – the instrument, the time of day, and finally an ordinary mistake in operating the platform. The psychological explanation joins that list last.
It is also worth separating FOMO from something that looks similar and is in fact correct behaviour: reacting to fast-moving information. If a data release changes the picture and your method provides for reacting to such a change, acting fast is executing the method and there is no reason to read it as a symptom of pressure. The difference lies in the same place as before: whether the condition for reacting existed earlier, or came into being during the move.
3. Anatomy of an episode: from stimulus to order
An episode has a repeatable course and – more usefully – leaves a different kind of trace at each point. Four links, and at each one what can be observed and where it can be broken.
1. The stimulus. A move on the chart, a price alert, a headline, somebody else's published result. A stimulus is neutral in itself and cannot be banned – some stimuli are simply necessary to the method. What matters is that it arrives at a moment when you have no ready condition for that instrument. Trace: the time and the source. This is the only field that cannot be reconstructed later.
2. The narrative. A sentence appears that closes the gap: "this is the opportunity", "the move is only just starting", "it will not happen twice". The narrative replaces the missing condition. Since there is no condition, a story takes its place. Calling this a defect of thinking explains nothing here. Trace: the content of the justification, provided it was written before the order. A justification written after the order describes a different state of knowledge.
3. Time pressure. The decision horizon may shrink to “now” while the move continues. Recording two timestamps — when the stimulus was noticed and when the order was submitted — lets you measure the interval between them. That interval does not itself measure psychological pressure or rule out a valid rapid response. Without both fields, you cannot reconstruct reaction time precisely; other traces, such as complete conditions or the timing of the justification, can still be checked.
4. Entry without a plan. The order is placed and the justification comes after it. You will recognise this by one thing: the justification contains information that did not yet exist at the moment of the decision.
Chasing price is a special case of the fourth link and is worth defining separately, because the name tends to be used too broadly. Chasing means successive entries at prices ever further from a level that was written down. Without a level written down there is nothing to chase – there is an ordinary entry after a move, and whether that is a mistake is settled differently (section 4). The trace is numerical and beyond dispute: the difference between the entry price and the level from preparation, across successive attempts. If it grows, you have chasing. If there was no level, you have a lack of preparation – a different problem and a different repair.
Where to intervene follows from the layout of the chain. The later, the more expensive: at the fourth link the only thing standing against the move is a decision made under pressure, that is the weakest safeguard available – why that is so is explained by the article on procedures as a substitute for self-control. The cheapest intervention sits between the first and the second link, that is in the rules of information flow, which are set before the stimulus arrives. That is the subject of section 7. The transition from state to order itself is covered separately by the article on emotions and impulsivity.
4. Entering after a move that is not FOMO
This is the place for the counterexample, without which the rest of the article would be irrefutable and therefore useless. Entering after a move is not in itself a symptom of anything. The two sequences below are an illustration; they describe nobody's experience and deliberately carry no dates, amounts or names.
Common to both: the same instrument, the same move, the same time, the same entry price, the same result. On the chart they are indistinguishable.
Sequence A. In the preparation before the session a sentence was written down: "if price moves above level X and returns to it, I enter – provided that Y occurs; if Y does not occur, I do not enter". The move started without the trader, price came back to X, Y occurred, the entry followed. Entering after the move is here the execution of a written record.
Sequence B. There was no preparation. The move was the stimulus, the condition formed while watching the move, the justification was added after the order was placed. On the chart it looks like A.
The difference is single and checkable: the existence of a record made before the decision, with a timestamp earlier than the stimulus. Not the outcome. Never the outcome. If A ends in a loss, it is still the execution of a plan and there is nothing in it to repair. If B ends in a profit, it is still an entry without a condition – and that is the didactically most dangerous case, because profit reinforces a behaviour that has no basis. Separating the quality of a decision from the quality of an outcome is the subject of the article on outcome versus the decision process.
A third sequence: the deliberate "no entry". Y did not occur, the trader did not enter, price went on without them. In terms of outcome this looks like a missed opportunity. In terms of process it is a correct result of the procedure – exactly as correct as an entry. The trouble is that a "no entry" disappears from the record by default: it is not in the orders, it is not in the statistics, and after a week all that remains in memory are the moves that got away. That is why it needs a trace of its own: a short entry with the time, the instrument, the level and a note on which condition was not met.
This is, incidentally, the same denominator problem, only on your side. Without a record of "no entries" your own picture of yourself is filtered too: the numerator remains, that is the entries, and the denominator disappears, that is all the situations in which the procedure worked and precisely for that reason nothing happened. You are then comparing other people's results, selected by someone else's filter, with your own memory, selected by your own. It would be hard to find a worse pairing.
5. Comparisons and social proof – what is documented
The tendency to judge yourself by comparison with others was described by Festinger in 1954[2]. It has to be said straight away what that work is and what it is not: it is a theoretical text, with no sample, no measurement and no effect size. It may not be cited as proof. One observation may be taken from it, and in trading it can be checked: social comparisons move in where an objective yardstick is missing.
In trading that condition is met in its strong version. Your own short-term result is not fit to serve as a running measure of quality, because it is largely noise – a week of results does not distinguish a good process from a bad one. What remains is a yardstick that does not work, and an empty space into which somebody else's visible result moves without difficulty. That is why the comparison does not disappear once a person learns that it is meaningless: as long as there is no record of your own process, there is nothing to replace it with.
The comparison itself answers a question about position: how am I doing relative to others. Social proof answers a different question – what counts as the right thing to do – and derives the answer from the number of people who have already done it. In a transaction record you recognise it by one behaviour: the entry condition is treated as satisfied because a few visible people entered the same instrument or showed a result from it. The count then takes the place of the condition, even though none of those posts brings your method any information that your condition uses.
How can you recognise a pattern in your own record? For an illustrative set of thirty entries, record whether the instrument appeared in your pre-session preparation and whether it had appeared in your sources beforehand. These groups can overlap: a planned instrument can also be popular. Compare them with the condition’s timestamp and any plan breaches. The review may suggest a hypothesis about visibility, but one group being larger does not establish a cause. Thirty is an example of a review window, not a reliability threshold; do not reconstruct unknown exposure to stimuli from memory.
It also has to be said what social proof in this sense is not. It is not the use of somebody else's information that passes through your condition and is let through or rejected by it; more on that below in this section. It is not your entry coinciding with somebody else's, if your condition was written down earlier – two independent processes may point at the same instrument and nothing follows from that. Nor, finally, is it the popularity of an instrument in itself: elevated turnover can be a market fact, and a method has every right to use it as data with a defined role. That does not turn the fact into a recommendation.
I have no source for this concept in trading and I say so plainly: I have found nowhere a measurement linking the number of visible entries to the decisions of retail traders, and I invoke nothing of the kind here. The description above therefore keeps to what you can check in your own account export and your own entries. One thing, however, follows directly from the first section: the assessment of social proof is also affected by the selection of what is visible. You are counting those who showed. Those who entered and wrote nothing, and those who entered and closed at a loss, do not enter that count. The premise "since so many people are doing it" is therefore a numerator without a denominator – exactly the one at stake in the binomial calculation, with the difference that entries are counted instead of published results.
Among behavioural research, the work closest to the topic is Barber and Odean's 2008 paper on the role of attention[3]. The data come from real accounts of individual investors in the United States, from the stock market. The finding: individual investors are net buyers of assets that attracted attention on a given day – through unusually high turnover, an extreme price change from the previous day, or presence in the news. The direction repeats across several independent datasets inside the paper itself, and that is its strong point.
The limitations belong in the same place. The authors explain part of the asymmetry in terms of attention: when buying, an investor chooses among thousands of stocks, whereas selling is largely limited to existing holdings. This forms part of their interpretation rather than a separate criticism disproving the result. Observational data nevertheless cannot assign a cause to every decision. The paper reports multiple comparisons and measures, rather than one effect transferable to a particular entry. It concerns equities, and the cited data do not establish the same mechanism among retail FX traders.
What may be drawn from this cautiously: what is documented is the co-occurrence of an attention-grabbing stimulus with a buying decision among individual investors in the stock market. It is not documented that watching other people's results causes losses in retail forex. I know of no paper that has measured it – with a population, a method and an effect size. If someone claims otherwise, they should provide those three things instead of the phrase "studies show". Until then, what remains is what can be done honestly: describe the behaviour and design a rule.
What remains is the question of what to compare yourself with, given that comparison itself will not disappear. The only pairing in which you know the denominator is your own record against your own record from an earlier period: how many decisions had a condition written down before the stimulus, how many "no entries" were noted, how many times the entry price departed from the level from preparation. These are dull and unphotogenic quantities. They have one advantage, though: you know how many attempts there were in total – which is something you do not know, and never will know, about anybody else's result.
6. The cost of attention and source redundancy
The problem is not information. The problem is the moment at which it arrives, and how much of it is the same content counted several times.
Assumptions (illustrative values): 12 sources checked · 9 of them repeat the same content · 3 bring independent content · 4 minutes per source.
This calculation is additive – these are shares of a single whole and that is why they add up; there are no probabilities here that would have to be multiplied: 9/12 = 75% redundancy, 3/12 = 25% independent content, 75% + 25% = 100%, and 9 + 3 = 12. In time terms it looks like this: 12 × 4 = 48 minutes against 3 × 4 = 12 minutes, a difference of 36 minutes, that is the same 75%. It is worth noting what this calculation says and what it does not. It speaks about time and about the number of repetitions. It says nothing about the quality of decisions and it is not a promise of a better result.
Does excess content impair judgement itself? In three experiments involving American football predictions, Tsai, Klayman and Hastie found that additional information increased confidence more than accuracy[5]. Accuracy did not invariably remain unchanged. Participants were paid, and the study designs also included performance-related rewards, although they were not trading their own capital. The paper reports statistical tests rather than one universal measure of decision deterioration. Participants knew the sport, samples were small, and the research did not study traders. Any conclusion about trading remains a cautious extrapolation.
Harder data concern something else, but point in the same direction. Barber and Odean analysed 66,465 households at a large American discount broker in the years 1991–1996 in the stock market[4]. The highest-turnover households achieved about 11.4% a year net, the average household about 16.4%, the least active about 18.5%, and the market about 17.9%. The result comes from one set of accounts, one broker and one six-year period; the paper says nothing about how this relationship behaves outside that set, and I invoke here no paper that has checked it.
There are three caveats without which that number misleads. It concerns the stock market and households; retail forex is not covered. A substantial part of the difference is transaction costs, so the paper says above all that trading more often costs money – not that an excess of information damages judgement. And nobody assigned households at random to a level of turnover, so this is a comparison of groups that selected themselves.
Taken together, these papers do not establish that more information automatically increases order counts or reduces accuracy among FX traders. They illustrate two different issues: confidence can increase faster than accuracy, and intensive trading incurs costs. Checking repeated content consumes time, as calculated earlier. An information-flow rule should therefore remove unnecessary repetition and protect decision conditions; its effect on your own execution must be assessed separately.
The practical test fits into a single question, asked only after reading the content: does this content change any condition I have written down? There are three answers – it changes one, it repeats what I already know, or it comments. Only the first is information for a decision. The other two can be valuable for other reasons, and that is exactly why they need a place other than the decision window.
7. Information flow designed before the stimulus
A rule established during a move is established by the move. That sentence is the whole justification for this section: since a stimulus changes decision criteria, the criteria have to come into being earlier than the stimulus – together with a rule saying what is allowed to reach you at all, and when.
Start by assigning each source a function in the decision. This describes when and why you use the information; it does not replace checking its reliability. Labelling sources “good” or “bad” is insufficient to organise their flow. The table below is an example configuration rather than a ready-made recommendation. Its columns may differ for another trading horizon or method.
Note what is missing from this division. There is no "forbidden" category. The third column can be useful – outside the decision window, as material for learning or for testing your own thesis against gaps you cannot see yourself. Cutting yourself off from published data can be an ordinary mistake and has nothing to do with hygiene if your method uses that data: interpreting macroeconomic releases belongs to fundamental analysis, and the calendar itself to the material on trading hours and the calendar. Information hygiene is a rule saying what should reach you, when and for what purpose. Its resemblance to a demand for silence is purely superficial.
The second step is windows. Instead of the question "how many times a day may I look", to which nobody knows a good answer, you establish blocks with different permissions: in the preparation block all three columns are available, in the decision block only the first, in the review block – the first and the third. This too is an example to be reworked for yourself; a sensible number and length of blocks depends on the horizon you trade in, and there is no prescription for it.
The third step is procedures. Each of the following has five fields, because without any one of them it can neither be executed, nor checked, nor refuted.
| Procedure 1 – a move that started without you | |
|---|---|
| Trigger condition | You see a move on an instrument for which today's preparation contains neither a written level nor an entry condition. |
| Action | You do not place an order. You write down three fields: the instrument, the time you noticed it, and the level at which the move was noticed. Then you check whether an earlier written condition exists for that instrument. If it exists and is satisfied – you enter in line with it. If it exists and is not satisfied – you do not enter. If it does not exist – the instrument goes onto the list for the next preparation and stays without an order in today's session. |
| Completion criterion | The procedure ends the moment one of three decisions is written down: I entered in line with the condition / I did not enter, the condition was not satisfied / there was no condition, I am moving it to preparation. As long as the entry is missing, the procedure is still running. |
| Audit trail | An entry with two timestamps – noticing the move and the decision – plus the name of the stimulus source. |
| Failure criterion | Entries saying "no condition, moving it on" nevertheless end systematically in an entry in the same session; or the entries stop being written even though out-of-list entries still happen. In both cases the procedure is not working and needs a harder form, or is badly matched to the way you work. |
| Procedure 2 – the information window | |
|---|---|
| Trigger condition | The start of a block of work with the market: preparation, session or review. |
| Action | You check only the sources permitted for that block, in an order established beforehand. Sources from the third column are not opened in the decision block, because their function is not to change a condition. Whatever value they have reveals itself outside that block. |
| Completion criterion | Exhausting the list of sources assigned to the block, or the end of the block – whichever comes first. |
| Audit trail | A baseline value collected before the first window: the number of entries without a written condition across as many sessions as the later comparison will cover. Without that number the failure criterion cannot be applied, so the procedure starts by counting it, and the first block comes only afterwards. Beyond that, a list of the sources checked with times, kept for the same number of sessions; and for each source an answer to whether it changed any written condition. |
| Failure criterion | After the agreed number of sessions the number of entries without a written condition does not differ from the number before the windows were introduced; or the preparation block grows until it starts eating the time set aside for review. The first means the problem lies elsewhere; the second – that the rule has turned into a ritual. |
| Procedure 3 – somebody else's directional call | |
|---|---|
| Trigger condition | A specific buy or sell call from outside reaches you. |
| Action | The call is not the basis for an order. You record it as an observation with the source name and the time, and then check it against your own condition. If your own condition is satisfied independently of the call – the decision is yours and the entry says so. If it is not – there is no entry. |
| Completion criterion | An entry containing the answer to whether your own condition was satisfied before the call was taken into account. |
| Audit trail | The "decision source" field in the trade record, filled in before the order is placed: your own condition, or somebody else's call. |
| Failure criterion | The "decision source" field gets filled in after the outcome is known, or stays empty in most entries. In that case you have no data and may not pronounce anything about yourself on the basis of it. |

8. What information hygiene does not fix
Before attributing a run of weak entries to information pressure, examine other possible explanations. Do they concern one instrument, a particular time of day or a period of higher volatility? Did the method or transaction costs change beforehand? Consider these factors alongside traces of information triggers because they may co-occur. The absence of a simple explanation does not prove an information effect; it remains a hypothesis to investigate. The limits of applying a label after the event are discussed in the article on cognitive biases.
There is also an answer that has to be admitted on equal terms with the others: it may be noise. A dozen or so trades do not distinguish a change in the process from ordinary variance, and no procedure in this article will change that fact. If after a week it seems to you that the information-flow rule "worked", you are probably looking at variance and taking it for the rule working.
Finally, some problems that look like FOMO are not FOMO and are repaired elsewhere. Successive entries after a losing streak, motivated by making back one specific loss, are revenge trading and tilt. The absence of a moment at which the process gets stopped is a matter for reset protocols. Judging your own decisions through the prism of how they ended is outcome versus process. And if checking the market in the evening comes from fatigue and the time of day, the matter belongs to the article on sleep, fatigue and stress.
Finally the simplest thing, which changes more than any list of blocks: one sentence written down before the session. The condition on which you will enter, the condition on which you will not enter, and the time it was written. For the coming week that sentence will do for the whole procedure, because it lets you answer the only question that matters in this article – whether the record was made before the stimulus or after it. How to keep such a record and how to read it later is described in the article on the trading journal. How physical state changes your ability to stick to your own rules is covered by the next article in this section.
FAQ – frequently asked questions
How will I know it was FOMO and not an ordinary entry?
Should I stop reading market news?
Is a "no entry" a wasted opportunity?
I see other people's results and it still affects me. What can I do about it?
Will an information diet improve my results?
Sources and references
- Przybylski, A.K., Murayama, K., DeHaan, C.R., Gladwell, V. (2013). "Motivational, emotional, and behavioral correlates of fear of missing out." Computers in Human Behavior, 29(4), 1841–1848. – Three correlational studies: an international sample N = 1,013, a nationally representative British working-age adult sample N = 2,079, and first-year students N = 87. Development of a FOMO scale and analysis of well-being and social-media-use correlates. Trading and financial outcomes were not studied; applying the concept is an extrapolation. Source.
- Festinger, L. (1954). "A Theory of Social Comparison Processes." Human Relations, 7(2), 117–140. – A theoretical paper, with no sample, no measurement and no effect size. Used solely as the source of the concept of social comparison; it is not proof of anything about traders. Source.
- Barber, B.M., Odean, T. (2008). "All That Glitters: The Effect of Attention and News on the Buying Behavior of Individual and Institutional Investors." The Review of Financial Studies, 21(2), 785–818. – Real accounts of individual investors in the USA, stock market. Finding: individual investors are net buyers of assets that attracted attention on a given day (unusual turnover, an extreme price change from the previous day, presence in the news). The direction repeats across several datasets within the paper. The asymmetry of the choice set between buying and selling is part of the authors’ proposed explanation, rather than a criticism of their study. Stock market; transfer to FX has not been demonstrated. Source.
- Barber, B.M., Odean, T. (2000). "Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors." The Journal of Finance, 55(2), 773–806. – 66,465 households at a large American discount broker, 1991–1996, stock market. The highest-turnover households about 11.4% a year net, the average household about 16.4%, the least active about 18.5%, the market about 17.9%. A substantial part of the difference comes from transaction costs; the level of turnover was not assigned at random, so this is a comparison of groups that selected themselves. Source.
- Tsai, C.I., Klayman, J., Hastie, R. (2008). "Effects of Amount of Information on Judgment Accuracy and Confidence." Organizational Behavior and Human Decision Processes, 107(2), 97–105. – Three experiments with University of Chicago students (N = 30, 20 and 40), forecasting American football results; participants familiar with the sport were paid, and the designs included performance rewards. Additional information increased confidence more than accuracy. This does not establish unchanged accuracy, an absence of all effects, or the phenomenon among FX traders. Source.
Source corrections from the previous version. Six references and the claims based on them were withdrawn. Alter (2017), Irresistible, and Taleb (2007), The Black Swan, are two popular books previously used as evidence for a mechanism. The CFTC warning, Customer Advisory: Beware of Promoters of Forex Trading (2019), was cited without defining a population or effect size. IG Group’s Client Sentiment Data concerned one provider’s clients but lacked a description of the population and method. Tetlock (2005), Expert Political Judgment, was attributed a claim the book did not support. Heimer and Simsek (2019), Should Retail Investors’ Leverage Be Limited?, concerned leverage limits but were cited for a conclusion about crowd behaviour. No substitute references were added to support the withdrawn claims.