Common beginner mistakes

Four traps, one theme

Beginners everywhere look at the same chart and make the same behavioural mistakes. The local difference is aggressive CFD marketing, messaging channels, and confusing supervision with a safe broker. This lesson does not rank companies — it organises process mistakes. And it does so in the only useful way: every mistake has a trace that is visible in the journal.

Error catalogue: cost in R and trace in the journal

The table collects the mistakes that recur most often in retail journals. The last column is the most important: it is a signal countable in a spreadsheet, not an impression after the session. Amounts in the cost column are illustrative.

How to detect a mistake instead of feeling it

Most traders do not know which mistake they make most often, because they have never counted it. The table below turns four traps into four measurable indicators — each calculated from the journal after 20-30 transactions:

The first indicator is the most interesting, because it shows something you cannot see in yourself but which is visible in the numbers: whether after a loss you actually increase position size. Most people claim they do not. Most spreadsheets say otherwise.

Revenge and I will make it back in 10 minutes

After a loss, the brain wants an immediate repair. In practice you get a worse fill, a wider spread, and a decision outside the checklist. Playbook rule: after the daily limit, the platform may stay open, clicking may not. See the psychology lesson and the session playbook.

The concrete mechanism that works better than a resolution: a break of defined length after every loss, not only after the daily limit. If your next transaction after a loss is opened within a few minutes, it is not analysis — it is a reaction. Record in the journal the minute the previous position closed and the next one opened; mark an interval below 15 minutes after a loss as a reaction, not a setup.

Mechanism: why size grows exactly after a loss

A loss triggers two processes: loss aversion, which makes the same amount hurt more, and the need for closure — an open minus is an unfinished state. Double size shortens the path to removing it, so revenge rarely looks like a decision and always looks like an opportunity.

Calculation: one doubled position and eight correct trades

Take eight plan-compliant transactions, each with 1R of risk:

• five profitable at +1R give +5R,
• three losing at -1R give -3R,
• net result: 5 - 3 = +2R.

That is +0.25R per transaction, because 2 / 8 = 0.25. After the third loss the trader opens a position with double size and gives up a full stop. That single transaction costs 2R instead of 1R, so the result falls to 2 - 2 = 0R — the gain from eight correct transactions disappears in one click. Rebuilding 2R at an edge of 0.25R requires 2 / 0.25 = 8 transactions. Check: 5 + 3 = 8, 5 - 3 = 2R, 2 / 8 = 0.25R, 2 / 0.25 = 8.

Overleverage because the broker allows it

Leverage is a credit limit, not a size target. Size follows from the distance to invalidation and the monetary limit. Compare a local or EU broker with an offshore one: higher leverage often means faster ruin, not more edge.

The mechanism is arithmetic, not psychological: with higher leverage, the same stop in points allows a larger lot, so the same market move costs more. The broker does not give you an edge — it gives you a higher ceiling on the size of your mistake. More in the lesson Leverage and margin.

Signals without a process

A ready-made entry without size, stop, and daily limit is someone else’s risk on your account. If a community discusses setups substantively — fine, as learning the language of the market. If it says enter now / TP1 TP2 without your checklist — that is not a curriculum.

One more layer that is rarely discussed: a Telegram signal has an unknown history. You see the result, you do not see the number of transactions that built it. Even honestly published profits without a denominator (how many entries, what size, what drawdown) say nothing about a strategy. Before you treat any channel as study material, check whether it publishes all entries or only selected ones.

Spread, slippage, news

The market at the moment of a macro release is often more expensive than a backtest assumed. A limit order may not be filled. Rule: smaller size or no position around high-risk events, regardless of how certain the direction seems. See also market versus limit.

No exit plan, that is, I will exit when I feel it

A condition that cannot be written before entry is not a rule — it is an intention. I will exit when I feel it does not say what has to happen for the transaction to end, nor how much you lose when it does not end. An exit plan is the invalidation level, the way the position is managed, and the maximum time it is held.

In the journal, a missing plan is visible immediately: the exit level column is empty or filled in after the fact. You cannot calculate the swing up and down, so you cannot distinguish a good exit from a lucky one. You cannot calculate R, and without R there is no expectancy, no drawdown, and no answer to whether a transaction was a mistake or bad luck.

When a stop may move

A stop is not a slider for fine-tuning the result, but a boundary beyond which the thesis is false. One criterion settles it: whether after the move the maximum loss on that position is smaller than in the plan. If so, the stop narrows. If not, it is a mistake, regardless of the justification.

After entry, price makes a new high and the stop travels below the new low, above the entry price — risk falls to zero, so the decision is justified. When price moves towards the stop and you move it from 20 to 40 pips to give the market room, risk grows from 1R to 2R. The new level must make sense in the structure, because a stop in a random place increases the chance of being taken out by noise.

Overtrading: the cost you only see in R

Capital of 50,000, risk of 1R is 1% of capital, that is 500. The stop is 25 pips, so a pip is worth 500 / 25 = 20. Round-trip cost is a spread of 1.5 pips plus a commission of 0.5 pips, together 2 pips, that is 40. We divide: 40 / 500 = 0.08, and the check along the second route gives 2 / 25 = 0.08.

With 60 transactions, costs alone come to 60 x 0.08R = 4.8R. A gross edge of 0.12R gives 7.2R, and net 2.4R — costs eat 4.8 / 7.2 = about 67% of the edge. Only a transaction with a gross edge above 0.08R is therefore worthwhile. Twenty entries at 0.05R give gross 1.0R, cost 1.6R, and net -0.6R; check: 20 x (0.05 - 0.08) = -0.6R.

One transaction says nothing

The result of a single transaction is the sum of decision and randomness. At a deviation of 1.2R it is noise. A decisive measurement begins only with a sample of about 62 transactions: at an edge of 0.3R the standard error formula gives (1.96 x 1.2 / 0.3)^2 = 7.84^2 ≈ 61.5. Check: 1.96 x 1.2 = 2.352; 2.352 / 0.3 = 7.84; 7.84^2 = 61.47.

Change the system only once the sample counts about 62 transactions; a change after 5 transactions is a reaction to noise. Three losses in a row at a 40% hit rate occur with probability 0.6^3 = 0.216, that is once every five blocks. After one transaction, assess only the quality of the decision; the full calculation is described in the lesson Expectancy and drawdown.

Error journal — 5 sessions

Keep a transaction journal and an error journal separately. The second has one column that does all the work — whether the violation was deliberate:

The deliberate column is the most important in the whole lesson. A deliberate violation is a process problem — it must be fixed. An undeliberate one is a knowledge gap — you need to learn more. If you do not separate the two, the only conclusion from 20 transactions will be I must be more disciplined, and that is not a conclusion, it is an emotion.

How to tell a process error from bad luck

A practical rule you can apply immediately:

• Bad luck — a transaction compliant with the plan: size calculated from the stop, invalidation recorded before entry, exit according to the rules. A loss on such a transaction says nothing about you or the system.
• Process error — a deliberate violation of the checklist, the daily limit, or size. Even if the transaction came out positive, it is a mistake — and that is the most dangerous category, because a win teaches a bad habit. Learn more if the violation was undeliberate.

Consequence: you assess a transaction by the quality of the decision, not by the result. A winning transaction opened without a stop is worse than a losing transaction compliant with the plan.

Measurable exercise

For 5 demo sessions use the checklist and keep an error journal. Calculate three indicators: % of transactions outside the checklist, % of deliberate violations, and average size after a loss divided by average size after a win. Educational goal: the first two ≤ 10% (better 0), and the third as close to 1.0 as possible. After about 62 transactions, repeat the measurement. Thinking about live before that makes no sense — see demo versus live.

Common questions

Are signals always a scam?

Not always. The educational problem is copying entries without your own risk, size, and invalidation. The test for a channel is simple: does it publish all entries, or only selected ones? Without a denominator, a result means nothing.

How do I tell a process error from bad luck?

Bad luck happens on plan-compliant transactions — size from the stop, invalidation recorded before entry. A process error is a deliberate violation of the checklist, the daily limit, or size. Watch out for wins after a mistake: they reward a bad habit and are harder to spot than losses.

How do I know revenge trading is my most common mistake?

Calculate the average size of transactions opened after a loss and divide it by the average size after a win. If the result exceeds about 1.3, you are increasing the position in reaction to a loss — regardless of what you think about yourself. Also check the time interval: an entry a few minutes after the previous position closed is a reaction, not a setup.

How do I check whether costs are eating my edge?

Divide round-trip cost by the value of 1R. With a 25-pip stop and a 2-pip cost, that comes to 0.08R, that is 8% of R. A gross edge must be higher than that number. Over 60 transactions, a cost of 0.08R gives 4.8R, so at an edge of 0.12R net remains 2.4R — costs take two thirds of the result.

Can I trade without a journal if the result is positive?

You can, but then you do not know where the positive comes from. Without a journal you cannot calculate the R distribution, the average cost, or the interval between transactions. A positive result without measurement describes favorable volatility rather than your method, and usually does not hold once that changes.

Is this investment advice?

No. This is educational material about risk hygiene. Trading involves the risk of losing capital.