Trade Journaling: How to Record, Review, and Learn from Investment Decisions

What to record in a trade journal

An effective trade journal captures two categories of information: the decision inputs (what was known and believed at the time of entry) and the decision rationale (how those inputs led to the specific decision made). The distinction is important: the journal should capture what the investor believed at the time, not a post-hoc reconstruction influenced by knowing what happened afterward. This is why journal entries should be written before or immediately at trade entry, not after the position has already developed.

At entry, the journal should record: the investment thesis in 2-3 sentences (what edge does this position have, why is it mispriced, what must be true for this to work); the key assumptions that the thesis depends on (specific, falsifiable claims: revenue growth will exceed current consensus, margins will recover as one-time costs roll off, the multiple will expand as the market recognizes the quality of the business); the entry price and the reasoning for the specific entry timing; the position size and the reasoning for that specific size relative to normal sizing (why this conviction level justifies this allocation); the predefined stop-loss price and the reasoning for that level (what would it mean if price reached that level, and why would that disprove the thesis); and a preliminary exit target or conditions (what price or development would represent thesis confirmation or completion).

During the holding period, the journal should be updated when material events occur: earnings releases, management guidance changes, competitive developments, macro events that affect the thesis. Each update should assess: does this event support, challenge, or leave unchanged the key assumptions identified at entry? If assumptions are being challenged, is the position still justified at its current size? If assumptions are being confirmed, does the thesis now support a larger position or an accelerated exit timeline?

At exit, the journal records: the exit price and the reason for exiting at that specific time (stop-loss triggered, thesis completed, better opportunity required the capital, time-based exit, change in thesis); the profit or loss in dollar and percentage terms; and the retrospective assessment of the thesis. Was the thesis correct? Were the key assumptions correct? Did the position perform as expected and for the reasons expected? What would the investor do differently if given a second chance at the same entry point with the same information?

Decision quality versus outcome quality in journal reviews

The most valuable use of a trade journal is the periodic retrospective review, in which the investor reads back through past journal entries and assesses decision quality independently of outcome quality. A good decision made with strong process can produce a bad outcome due to bad luck or unforeseeable events. A bad decision made with poor process can produce a good outcome due to good luck. Reinforcing good-process decisions and improving poor-process decisions is the path to durable performance; reinforcing whatever produced the last good outcome is the path to overconfidence and inconsistency.

Decision quality is assessed by asking: was the thesis coherent and specific? Were the key assumptions falsifiable (could they have been wrong, and was there a clear signal that would indicate they were wrong)? Was the position size consistent with the conviction and the process, or was it sized emotionally (too large in excitement, too small in fear)? Was the stop-loss placed at a price that would actually disprove the thesis, or was it placed where the investor would feel uncomfortable? Was the entry timing based on a systematic criterion or on impulse?

Outcome quality is assessed separately: did the position make or lose money, and why? This assessment should then be compared to the decision quality assessment. The four combinations reveal different things about the investor's process. Good process and good outcome: the process is working, reinforce it. Good process and bad outcome: bad luck or unforeseeable event, the process should not change based on this single data point. Bad process and good outcome: lucky this time, identify and correct the process flaw before it causes a large loss. Bad process and bad outcome: expected result of poor process, identify and correct the process flaw.

Systematic error patterns emerge over many trades in the journal. Common patterns include: consistently oversizing positions in sectors the investor finds exciting (overconfidence in a domain of interest); consistently undersizing positions in sectors the investor finds unfamiliar (underconfidence in areas of comparative advantage); exiting winning positions too early (loss aversion applied to gains); holding losing positions too long (loss aversion applied to losses, the disposition effect); entering too early on contrarian ideas (underestimating how long the market can sustain a narrative before reversing). Each pattern is correctable once identified; the journal provides the data needed to see it.

Journal formats and review cadence

Trade journal formats range from simple spreadsheets to dedicated software. The minimum viable format is a spreadsheet with one row per position and columns for each entry and exit field. This format is easy to search, filter, and analyze statistically after accumulating enough trades. Narrative-focused investors prefer a structured document with more free-text space for thesis articulation and the retrospective narrative; this format captures nuance better but is harder to aggregate for statistical analysis.

The review cadence for journal entries typically follows a three-tier structure. Weekly: review any exits from the prior week and complete the exit entry while the reasoning is fresh. Monthly: read back through the last month's entries and assess for decision quality versus outcome quality; look for any early pattern signals across 4-8 trades. Quarterly: a deeper statistical review of entries from the last quarter; calculate average decision quality scores, look for sector-level or conviction-level patterns, and compare the retrospective thesis accuracy across different thesis types.

The journal review should be conducted in a calm, non-trading period with adequate time for reflection. Reading past journal entries during active market hours or when the investor is under emotional stress from current position movements produces biased assessments. The goal of the review is honest self-evaluation, which requires a degree of psychological distance from the current portfolio and the emotional state attached to it. Some investors conduct journal reviews on weekends or after market close on a non-news day specifically to create this separation.

Frequently asked questions