Trading Journal and Review: What to Record and What to Look For

A journal is the only way to find out what you actually do, as opposed to what you believe you do. The gap between those two is usually where the money goes.

Trading Journal and Review at a glance

What it is and why it works

Every trader has a story about how they trade. It is usually a description of their plan rather than their behaviour, and the difference is invisible from the inside because memory is selective in a very specific way: it keeps the trades that confirm the story and discards the ones that do not. Ask a struggling trader what is going wrong and you will get a theory. Ask for their records and you will get an answer.

A journal closes that gap by capturing what happened before the outcome could colour it. The essential trick is writing the reason for the trade at the moment of entry. Once a trade wins, its reason feels sound; once it loses, the same reason feels like a lapse. Only a note written beforehand survives that revision, and it is the reason a journal reveals things a broker statement never can.

A broker statement tells you what you did. A journal tells you why, in what conditions, feeling what, at what size, and that is where the patterns are. The most common findings are unflattering and immediately actionable: most of the losses came from one instrument, or one hour of the day, or from trades taken within twenty minutes of a previous loss, or from the setups the trader would describe as their B option.

Then there is the review, which is what converts a record into an improvement. Recording without reviewing is a diary. Reviewing on a fixed schedule (weekly for behaviour, monthly for statistics) forces you to look at the whole sample rather than the last trade, which is the only vantage point from which a pattern is visible at all.

How to trade it, step by step

1. Record the mechanical facts for every trade, without exception. Date and time, instrument, direction, entry price, stop price, target, lot size, the money at risk, exit price, and the result in R. Losing trades and the ones you would rather forget are the most important rows in the file; a journal containing only your good trades is worse than no journal, because it produces confident conclusions from a filtered sample.
2. Write the reason for entry before you know the outcome. One or two sentences naming the setup and what has to happen for the idea to be wrong. “Pullback into the 4H range low with a rejection candle; wrong if it closes below the low.” This single field does more work than everything else combined, because it is the only part of the record that cannot be rewritten by hindsight.
3. Take a screenshot at entry and another at exit. Mark the entry, stop and target on the entry chart. Months later the numbers will mean nothing on their own, and the pictures will show you instantly that you were, say, consistently buying into resistance. Store them with the trade ID so the row and the image stay connected.
4. Add a small set of tags you can filter on later. Setup type, session, whether it was in your plan, whether it was taken within an hour of a loss, and how you felt in one word. Keep the list short and fixed, five or six tags used consistently are far more useful than twenty used sporadically, because the analysis depends on being able to group trades.
5. Review weekly at a fixed time, looking at behaviour rather than profit. Set aside half an hour on the same day each week. Ask three questions: which trades were outside my plan, which mistakes repeated from last week, and what was the single largest loss and why. Write one sentence naming the one thing to change next week, only one, or nothing will change.
6. Review monthly for statistics. Calculate win rate, average win and average loss in R, expectancy, maximum drawdown and longest losing streak; the method is on the risk-reward and expectancy page. Compare each figure with the previous month and investigate what moved rather than the headline profit, which is the least informative number in the file.
7. Segment the results by your tags and look for the concentration. Split expectancy by instrument, by session, by setup type and by in-plan versus out-of-plan. This is where journals pay for themselves: it is extremely common for one instrument, one hour or one setup to account for most of the losses, and cutting it is a change you can make immediately.
8. Check your risk consistency explicitly. Put the money risked on every trade in a column and look at the spread. If your largest risk is much more than one and a half times your smallest, you are sizing by feel, which invalidates every R-based statistic you have calculated. See position sizing for the fix.
9. Keep a short record of the trades you did not take. Setups that met your rules but were skipped, and why. This catches the most invisible problem in trading, hesitation after losses, which never appears in a record of executed trades and can cost as much as any bad trade you did take.

Size every one of those entries with the position size calculator and check the trade is worth taking with the risk/reward calculator before you commit.

The conditions it needs

Entries made at the time, not reconstructed later

A journal written up at the weekend is a record of your memory, and memory has already tidied the evidence. Entry reasons in particular have to be written before the result is known, otherwise every winning trade acquires a good reason and every loser a bad one. Two minutes at the moment of the trade is worth more than an hour of reconstruction.

Complete coverage, including the trades you regret

Selective recording produces confident conclusions from a biased sample: the same survivorship problem that ruins backtests. The impulsive trade taken out of boredom is precisely the row that explains your results, and it is the one most likely to go unrecorded.

A fixed review schedule

Reviewing only after a bad run means you analyse in a poor emotional state and study an unrepresentative sample. A fixed weekly slot means you also review after good weeks, which is when overconfidence and rising position sizes show up. The schedule matters more than the depth of any individual session.

A small, consistent set of fields

Journals fail from ambition. An elaborate template with thirty columns gets abandoned within a fortnight, while ten fields filled in reliably for a year produce a genuinely valuable dataset. Consistency is what makes grouping and comparison possible, and inconsistent tagging makes analysis impossible regardless of how much was recorded.

When it fails

• Recording results without reasons. A list of profits and losses is a broker statement with extra steps. Without the reason for entry and the conditions at the time, there is nothing to diagnose; you can see that you lost, but not what to change. The reason field is the difference between data and information.
• Journalling only losses. Reviewing only losers teaches you that losing trades exist. The comparison is what carries information: what was different about the winners, and did the losses come from a different setup, session or emotional state. Without both sides there is no contrast to learn from.
• Writing entries after the fact. Hindsight rewrites motives with complete conviction. A trade taken on impulse becomes, three days later, a considered decision that went wrong, and the impulsive pattern stays invisible because every instance was recorded as reasoned.
• Reviewing the last trade instead of the sample. Individual trades contain almost no information, because any single result is dominated by chance. Traders who overhaul their method after two losses are responding to noise and destroying whatever consistency they had. Patterns exist across dozens of trades or not at all.
• Building a system too elaborate to maintain. Thirty fields, screenshots at four timeframes and a written essay per trade lasts about two weeks. The journal that gets used is the one that takes two minutes, and a simple spreadsheet completed for six months beats sophisticated software abandoned in March.
• Reviewing without producing a change. A review that ends in a general resolution to be more disciplined changes nothing. It has to end in one specific, checkable rule (no trades after two losses in a session, no positions in the last hour, no setups outside the written plan) and next week’s review has to check whether the rule was kept.

For different levels of experience

If you are brand new

Start with a spreadsheet and eight columns: date, instrument, direction, entry, stop, size, result in R, and the reason you took it. That is enough. Add the reason before you know the outcome, because that is the column that will eventually tell you something.

Take a screenshot of the chart at entry with your stop and target marked. It takes five seconds and, months later, the pictures show you patterns that the numbers cannot, that you keep buying at the top of the day’s range, or entering just before a session ends.

Expect the first fifty trades to feel like a chore with no payoff, because a journal needs a sample before it can say anything. Around trade fifty it starts to answer questions you could not otherwise ask: which setup actually works for you, which hour costs you money, and whether your losses are bigger than you intended. Pair it with a written trading plan, since the journal is largely a record of whether the plan was followed.

If your results are inconsistent

The intermediate trader usually keeps a journal and does not review it, which is the same as not keeping one. Recording is the easy half; the value is entirely in the analysis, and it is the half that gets skipped because it is uncomfortable.

Run the segmentation and be prepared for the answer. Split expectancy by instrument, by session, by setup and by in-plan versus out-of-plan. The typical finding is stark: one instrument or one setup is responsible for most of the losses, and out-of-plan trades have a clearly negative expectancy. That gives you a change you can make on Monday, without learning anything new about markets.

Also track the two behavioural fields most journals omit, time since the previous trade, and whether the previous trade was a loss. Revenge trading is the most expensive habit at this level and it hides in plain sight, because each individual instance looks like an ordinary trade. It only becomes visible when you group by what came before it.

If you are experienced

At a professional level the journal is a research dataset and should be structured as one: one row per trade, one column per attribute, with tags applied from a controlled vocabulary so that grouping is reliable. Store the market context alongside the trade (realised volatility, session, spread at entry, distance to the day’s range extremes, whether a scheduled release fell inside the holding period) because those covariates are what let you test whether an apparent edge is genuine or a regime artefact.

The most valuable analysis is usually attribution rather than aggregate performance. Decompose the return into setup, execution and management: what the signal was worth if traded mechanically, what entry and exit timing added or destroyed, and what discretionary intervention did. Traders frequently discover a positive signal with negative execution alpha, which is a completely different problem from an absent edge and calls for automation rather than a new method.

Treat sample size with the same discipline you would apply to a backtest. Segmenting a 200-trade record into six buckets leaves cells too small to support a conclusion, and multiple comparisons across many tags will manufacture apparent effects. State the hypothesis before slicing the data, and treat anything found by inspection as something to test forward rather than to act on.

Risk management for this strategy

A journal is a risk control in its own right, because it is the only mechanism that reliably detects a drift in your risk. Position size, stop discipline and trade frequency all creep gradually, and none of them are noticeable from inside a single day; they are only visible as a trend across a column of numbers.

Record the money at risk on every trade and watch the series. A rising average, or an increase in size after losses, is the earliest warning of the behaviour that ends accounts, and it usually shows up weeks before the damage. Also track your open risk across correlated positions and your worst day of each month, since maximum daily loss is the figure that matters most if you ever intend to attempt a prop firm challenge.

Use the record to set concrete limits rather than intentions. If your journal shows that trades taken within thirty minutes of a loss have a negative expectancy, the rule is a mandatory pause, not a promise to be more careful. Rules derived from your own data are the ones you keep, because you have seen what they cost you.

Where Market Structure Pro fits

The hardest field in any journal is the honest description of market conditions. “Choppy” or “trending” written after the fact tends to describe the outcome rather than the conditions, so the most important variable in your dataset ends up being the least reliable one.

Market Structure Pro fixes that by giving you an objective value to record. Its 27 tools resolve into one verdict (TRADE, TRANSITION or NO TRADE) with a confidence percentage, an A/B/C grade and a plain-English explanation of the reasoning. Logging the state and grade with every trade turns market condition into a stable, comparable field, and because the state locks on the closed bar and does not repaint, the value in your journal is exactly what was displayed when you entered.

That makes a specific and valuable analysis possible: expectancy grouped by grade. If your A-grade trades are strongly positive and your C-grade trades are negative, you have found an improvement that requires no new strategy at all: only fewer trades. Traders regularly discover the same thing about trades taken during a NO TRADE read. MSP is decision support: it does not place trades, it is not a signal service and it guarantees nothing. In a journal, its value is that it makes the quality of a setup a number you recorded rather than a feeling you half-remember.

Frequently asked questions

What should I record in a trading journal?

The mechanical facts (date, time, instrument, direction, entry, stop, target, lot size, money at risk, exit and result in R) plus the reason you took the trade, written before you knew the outcome. Add a screenshot of the chart at entry with the stop and target marked. The reason field is the most valuable part, because it is the only element that hindsight cannot rewrite.

Do I really need a trading journal?

If you want to improve rather than simply continue, yes. Without records you are relying on memory, which systematically keeps the trades that support your beliefs and discards the ones that contradict them. A journal is the only way to find out what you actually do, and the gap between that and what you believe you do is usually where the losses are.

How often should I review my trades?

Weekly for behaviour and monthly for statistics. The weekly review looks at which trades broke your plan, what mistakes repeated, and what the largest loss was. The monthly review calculates win rate, average win and loss, expectancy, maximum drawdown and longest losing streak, and compares them with the previous month.

What statistics should a trader track?

Win rate, average win and average loss in R, expectancy per trade, maximum drawdown and longest losing streak. Total profit is the least informative figure, because it says nothing about how the result was produced or whether it is repeatable. Knowing your longest historical losing streak is also what stops you abandoning a working method during an ordinary bad run.

What is the best trading journal app?

The one you will still be using in six months, which for most traders is a spreadsheet. Dedicated journalling software adds automatic import and analytics, which is convenient, but no tool can supply the reason for entry or the screenshot, and those are the parts that carry the information. Start simple and only add complexity once the habit is established.

How many trades before a journal is useful?

Around fifty before behavioural patterns become readable, and a hundred or more before statistics like win rate and expectancy are stable enough to act on. Before that the record is worth keeping but not worth drawing conclusions from, because small samples are dominated by chance. Changing your method on the strength of twenty trades is responding to noise.

How do I find what I am doing wrong in my trading?

Group your recorded trades and compare expectancy across the groups: by instrument, by session, by setup type, and by whether the trade followed your written plan. The losses are usually concentrated rather than spread evenly, and the most common findings are that one instrument or hour accounts for most of the damage, and that out-of-plan trades have a clearly negative expectancy.

Should I journal trades I did not take?

Yes, at least briefly. Recording setups that met your rules but were skipped, and why, exposes hesitation after losses; a problem that never appears in a record of executed trades and that can cost as much as any bad trade you did take. A short note is enough; there is no need to record every setup you considered.

How do I make journalling a habit?

Keep it small enough that it takes two minutes, and attach it to something you already do, such as closing the trade ticket. Elaborate templates are abandoned quickly, while eight fields filled in reliably for a year produce a genuinely useful dataset. Fixing a specific weekly review slot in the calendar matters more than the depth of any individual entry.

Related reading

• Risk-Reward and Expectancy: The statistics your monthly review should be calculating.
• Trading Plan: The journal is largely a record of whether the plan was followed.
• Trading Psychology: Revenge trading and hesitation are only visible in the records.
• Position Sizing: A risk column in the journal is the earliest warning of size creep.
• Backtesting: The same sampling traps apply to your own trade record.