The Trading Journal And Post-Mortem Template We Actually Use

A trading journal is the single highest-leverage tool a discretionary trader owns and almost nobody uses one that works. The standard failure mode is a Notion page with thirty fields and screenshots that never get filled in past week two. T

A trading journal is the single highest-leverage tool a discretionary trader owns and almost nobody uses one that works. The standard failure mode is a Notion page with thirty fields and screenshots that never get filled in past week two. The second failure mode is a P&L spreadsheet pretending to be a journal. The third is a Discord channel of celebratory posts. None of those produce the thing a journal is supposed to produce: enough structured data, captured consistently, that you can run an honest post-mortem and find what is actually killing your edge. The template is short on purpose, the post-mortem questions are fixed on purpose, and the cadence is non-negotiable on purpose.

What a journal is for

A journal exists to do exactly one thing: produce a record honest enough that your future self can find patterns your current self cannot see. That is it. Not motivation. Not record-keeping for tax. Not content for social media. Pattern discovery, full stop.

Once you accept that framing, almost every "should I add this field?" question answers itself. If the field will not feed a pattern you can run a query against in three months, it does not belong in the journal. Most journals fail because they collect things that feel important in the moment and produce nothing queryable downstream.

The twelve fields and nothing else

Here is the exact field set Tradoki cohorts use. We have iterated on this for months. Adding fields almost always hurt completion. Removing fields almost always hurt analysis. Twelve is what survived.

1. Date and time of entry. Including the timezone you trade in.
2. Instrument. Symbol and asset class. EURUSD spot is a different beast from a EURUSD CFD with overnight financing.
3. Setup name. From your written playbook. If a trade does not have a setup name from the playbook it is not a trade, it is a punt, and that should be visible in the journal.
4. Timeframe. The execution timeframe and the higher timeframe you are aligning to.
5. Entry price and rule. Not just the price — the rule that produced it. "Pullback to 20EMA in trend confirmed by candle close" is a rule. "Looked like a buy" is not.
6. Invalidation price. Where this idea is wrong. The level at which you know the setup failed, not where your stop happens to sit.
7. Stop and target. With explicit risk-to-reward in R-multiples.
8. Position size and planned R. In whatever R unit your risk framework uses.
9. Pre-trade screenshot. Annotated chart of what you saw before clicking buy.
10. Pre-trade rationale paragraph. Three to five sentences, written before entering. This is the field everyone wants to skip and skipping it is the single biggest predictor of journal failure.
11. Emotional state, one word, fixed scale. Calm / focused / anxious / impatient / tilted. Reuse the same scale every time. Single-word entries make this analysable across hundreds of trades.
12. Outcome. Closed price, R achieved, post-trade screenshot, and a one-line "what actually happened versus the plan."

That is the entire template. There is no field for "lessons learned" because lessons live in the weekly post-mortem, not in per-trade entries. Mixing the two means neither cadence works properly.

The four post-mortem questions

The journal records. The post-mortem analyses. They are different artefacts on different cadences and conflating them is why most traders never extract anything from their own data.

The post-mortem is weekly. Sit down with the last week of entries, sorted by setup name. For each setup that occurred more than once, run these four questions:

Question one: did I follow my own rules? Not "was I profitable" — did I follow my own rules. Count the trades that match the playbook entry rule exactly versus the trades that do not. The ratio is your discipline score for the week. It is the most important number in the post-mortem and it is the one most traders avoid computing because it is uncomfortable.

Question two: when I followed the rules, what was the average R? This is your edge per setup, conditional on execution. If a setup has positive R when followed cleanly, you have something. If a setup has negative R even when followed cleanly, you have a backtest problem, not a discipline problem.

Question three: when I broke the rules, what was the average R? Almost always negative. Sometimes catastrophically so. Putting a number on it removes the temptation to argue with yourself about it.

Question four: what is the recurring failure mode? Tilted entries on Mondays? Position sizing creep on Thursdays after losses? Skipping invalidation on instruments you "feel good about"? Find one pattern per week and write it down.

"I had been trading for three years and I had never once written down whether I followed my own rules. The week I started counting was the week I realised I was running two strategies, not one — the written one and the tilted one. Only one of them had positive expectancy."

— Tradoki cohort feedback, week six

What we deliberately excluded and why

There are fields we tested and dropped. Worth saying which, because the absence might look like an oversight.

No P&L in dollars. Only R-multiples. Dollar P&L screws with risk perception week to week as account size changes. R-multiples normalise across position sizes and across accounts.

No long emotional journal entries. One word, fixed scale. We tested paragraph-style emotional entries and within four weeks they decayed into either therapy text or copy-pasted boilerplate. Neither is analytically useful.

No "market sentiment" field. It correlates with nothing actionable in our cohort data so far and it adds friction to the entry. A field that does not produce a pattern is a field that should not be in the template.

No "lessons learned" per-trade. Lessons go in the weekly post-mortem. Per-trade lessons are usually post-hoc rationalisation written in the heat of the close. They are noise pretending to be signal.

No screenshots of every timeframe. Two screenshots — execution timeframe pre-trade and execution timeframe post-trade. More screenshots, in our testing, did not improve review quality and significantly hurt completion.

We are open to revisiting any of these if cohort data gives us reason to. None of them have so far.

The Tradoki eight-week syllabus introduces the journal in week one and the post-mortem cadence in week three. The risk-of-ruin pillar guide defines the R-multiple system the journal references. The ninety-day practice plan is what students run on top of this template after the cohort ends.

How to start tomorrow without the cohort

You do not need a cohort to start. You need a notebook, an honest hour on Saturday morning, and the discipline to write the pre-trade paragraph before you click buy. That is the entire activation cost.

Start with the twelve fields. Do not customise. Do not add. Do not skip the rationale paragraph because you are in a hurry — if you are in a hurry, the answer is to take fewer trades, not skip the field. After two weeks you will have enough data to run the first post-mortem. After eight weeks you will know more about your own trading than ninety percent of retail traders ever know.

The reason we build cohorts around this template, despite it being free in the sense that anyone can copy it, is that knowing the template and running the template are completely different problems. Most traders know they should journal. The cohort is the structure that gets them to actually do it. If you can run it solo, run it solo. If you cannot, that is what the cohort is for.

What the dataset gives you after ninety days

If you run the template honestly for three months, you will have somewhere between thirty and two hundred entries depending on your cadence. That is enough to compute, with reasonable confidence:

• Your discipline score (rule-followed trades / total trades) per setup
• Your conditional edge (R-multiple when rules followed) per setup
• Your tilt cost (R-multiple when rules broken) per setup
• The day-of-week and emotional-state combinations correlated with your worst trades

That is the analytical foundation almost no retail trader has. It is the foundation we believe is required before any conversation about scaling size, adding instruments, or running a more aggressive strategy can be had honestly. The asset-selection framework and the future-of-retail-trading essay both lean on the assumption that a working journal exists. This article is the template that makes that assumption real.

● FAQ

What fields belong in a trading journal?

Setup name, instrument, timeframe, entry rule, invalidation level, target, position size, planned R, actual R, screenshot of pre-trade chart, screenshot of post-trade chart, and a one-paragraph rationale written before entering. Anything else is optional and most additions hurt completion rates.

How long should a journal entry take?

Three to six minutes pre-trade, ninety seconds post-trade for the close, and ten to fifteen minutes for the weekly review. If it takes longer the template is wrong, not the trader.

Why pre-trade and post-trade screenshots?

Because hindsight rewrites every chart. The pre-trade screenshot is the only honest record of what you actually saw. The post-trade screenshot, with the move played out, is what your future self will compare against to learn pattern recognition.

What goes in a post-mortem versus the journal?

The journal is per-trade and behavioural. The post-mortem is weekly or monthly and pattern-level. The journal asks what happened. The post-mortem asks what is recurring. Different cadences, different questions.

Should journals track emotions?

One word, on a scale you reuse. Not a paragraph. Long emotional entries decay into therapy and stop being analytical. A reusable scale lets you correlate emotional state with execution quality across hundreds of trades.