The trading journal is one of those practices that almost every experienced trader recommends and almost every beginning trader ignores. It sounds like homework. It takes time that could be spent analyzing charts or looking for new opportunities. The connection between writing down what you did and doing better in the future feels indirect and uncertain.
I kept no journal for the first four months of live trading. Then I started keeping a minimal one — just trade results. Then I added process notes. Then I added the passed signal log described in an earlier post. By month ten, the journal had evolved into a structured document that I review and update daily, and I can say with reasonable confidence that it has improved my trading results more than any system parameter change I've made.
This post describes exactly what I track, why each category matters, and what I've found by reviewing the data over time.
Why a Trading Journal Works — The Mechanism
Before getting into the specifics of what I track, it's worth explaining why tracking produces improvement. The mechanism isn't mysterious, but it's easy to underestimate.
Trading without a journal is primarily an emotional experience. You remember the trades that confirm your beliefs about yourself — the ones where you made a good call, the ones where you got unlucky, the ones where you did everything right but the market went against you. You tend to forget the trades that are inconvenient — the ones where you broke your rules and it happened to work out, the ones where you got lucky, the ones where you were slow and missed an opportunity.
A journal replaces selective memory with systematic record. Once you've recorded something, you can measure it. Once you can measure it, you can see whether your intuitions about your own behavior are accurate. In almost every case I know of where a trader has kept a serious journal, the data has contradicted at least one significant belief they had about their own trading.
For me, the most important contradicted belief was about my exit behavior. I believed — sincerely, not as a rationalization — that I was patient with winning trades and decisive with losing ones. The journal showed the opposite: I was exiting winning trades prematurely and holding losing trades slightly longer than my rules required. The rules were being followed approximately, not precisely, and the direction of the deviations was consistent with the disposition effect. Without the journal, I would never have seen this clearly enough to change it.
What My Journal Tracks — The Seven Categories
Category One: Trade Entry Details
For every trade, I record: the stock name and code, the entry time (exact, from the trade confirmation), the entry price, the number of shares, the signal grade at entry, the primary reason for entry (which tier the stock was on, what the participant breakdown showed, any specific context), and the stop-loss level placed at entry.
The entry time is critical. As I described in the post on hindsight bias, reviewing trades without precise timestamps leads to systematic misinterpretation of where in the price sequence the entry occurred. The exact entry time makes it possible to reconstruct the information available at the moment of entry rather than reading the outcome backward.
Category Two: Trade Exit Details
For every trade, I record: the exit time, the exit price, whether the exit was mechanical (stop-loss or take-profit target hit) or discretionary (manual decision before the target), and if discretionary, the specific stated reason.
The discretionary exit reason is the most valuable item in this category. It forces me to articulate, in writing, why I'm exiting before the defined target. This articulation frequently reveals that the reason is not specific and observable — "market feels weak" or "I have enough profit" — which are the kinds of non-reasons that the three exit rules described in the prior post are designed to prevent. Writing it down makes the non-specificity visible in a way that deciding in the moment doesn't.
Category Three: Outcome Calculation
For every trade, I calculate: the gain or loss in absolute terms (KRW), the gain or loss as a percentage of the blended entry price, whether this was better or worse than the defined stop-loss outcome would have been, and whether this was better or worse than the defined take-profit outcome would have been.
The last two calculations are the ones that reveal the most. A trade that exited at a 2% gain looks like a success in isolation. The same trade, measured against a take-profit target that would have yielded 4.8%, reveals that the exit captured less than half the available outcome. Over hundreds of trades, the pattern of how often you're capturing versus leaving available outcome is one of the most useful signals about execution quality.
Category Four: Rule Adherence
For every trade, I record a binary evaluation: did the entry meet all the stated entry criteria, and did the exit follow the stated exit rules? I use a simple three-point scale: compliant (followed rules as written), minor deviation (followed rules in spirit but not exactly), or significant deviation (rules were substantially overridden).
After twelve months of tracking this, the correlation between rule compliance and outcome is clear: significant deviations from entry rules produce worse average outcomes than compliant entries. Minor deviations are roughly neutral. The same pattern holds for exits. This sounds obvious when stated, but seeing it numerically — where a specific percentage deterioration in average outcome is associated with specific categories of rule deviation — makes it concrete in a way that general awareness doesn't.
Category Five: Market Context at Entry
For every trade, I record: the KOSPI direction on the day (up/flat/down, roughly), the sector trend for the stock's sector, the overall market volatility assessment (low/medium/high based on average ATR readings), and any notable market events on the day (earnings releases, macro announcements, unusual sector activity).
This category exists because the same signal quality grade means different things in different market contexts. An A-grade signal on a day when the KOSPI is up 1.2% with low volatility is a different bet from an A-grade signal on a day when the KOSPI is down 0.8% with elevated volatility. The journal records make it possible to analyze signal outcomes conditioned on market context — which is exactly how I identified the regime-specific performance patterns described in the post on market regimes.
Category Six: Emotional State Note
This is the shortest category and the one I was most resistant to adding. Once a day — not per trade, but once per session — I write one sentence describing my emotional state during the session. This isn't therapy journaling; it's pattern recognition.
After about three months of adding this category, the pattern became clear: sessions where I noted feeling frustrated, anxious, or unusually eager to trade correlated with worse rule compliance and worse average outcomes. Sessions where I noted feeling calm or neutral correlated with better compliance and better outcomes. The specific emotional states that preceded deviations from my rules were identifiable in advance — which means the daily note is providing a real-time calibration input that I can use to adjust my behavior before problems occur.
Category Seven: One-Line Post-Session Assessment
At the end of every session, I write one sentence summarizing what I would do differently. Not a lengthy review — a single sentence. This forces a specific observation rather than a vague reflection.
Over twelve months, these one-line assessments have accumulated into one of the most useful diagnostic tools I have. Reviewing the last thirty of them before any significant system change reveals whether the same issues keep appearing — which indicates a structural problem worth addressing — or whether the issues are varied and non-recurring — which indicates normal session-to-session variation that doesn't require intervention.
What the Journal Has Revealed Over Time
Three findings from the journal data stand out as having produced concrete changes to how I trade.
First, my win rate on discretionary exits is meaningfully lower than my win rate on mechanical exits. Trades that hit defined targets outperform trades I exit manually, on average, by a margin large enough that reducing discretionary exits is clearly worth the psychological discomfort of holding longer.
Second, my performance deteriorates measurably in the second half of a session where I've already taken two or more losses. The journal makes this visible by allowing me to stratify outcomes by "position in session" and "prior loss count." The deterioration isn't catastrophic, but it's consistent enough that I've adopted a rule of reducing position sizing after two losses in a single session, regardless of subsequent signal quality.
Third, the stocks I trade from tier-one of the watchlist (institutional participation present) outperform tier-two stocks (retail-dominated) by a margin that justifies the additional morning preparation time to build the watchlist properly. I had suspected this was true. The journal confirmed it quantitatively.
Today's Investing Insight — The Role of Deliberate Practice in Trading Skill Development
Research on expertise development, particularly the work of psychologist Anders Ericsson on deliberate practice, suggests that performance improvement comes not from experience alone but from experience combined with specific, structured feedback on what went wrong and what could be improved. Simply trading for years doesn't automatically produce skill — traders who plateau often do so because they're accumulating experience without the structured review that converts experience into learning. A trading journal is, in this framework, the mechanism that converts trading experience into deliberate practice. The daily record, the post-session review, and the periodic analysis of patterns across many sessions create the feedback loop that translates raw experience into measurable skill improvement. This is why experienced traders who keep detailed journals tend to improve faster than equally experienced traders who don't, even when their underlying strategies are similar.
---
This post documents a personal journey of building and running an algorithmic trading system and reflects personal experience and perspective. Trading journal practices described here are based on personal observation and are not a recommendation for any specific approach. All investment decisions and their outcomes are the sole responsibility of the investor.
Comments
Post a Comment