Everything so far — plan, capture, grade, compare, measure, slice — produces data. And data, by itself, changes nothing. A trader who captures meticulously but never reviews has built an instrument they never read; the compounding that a journal promises happens entirely in the review loop, and this chapter is about running it with the discipline of science rather than the good intentions that die by February.
The science underneath the loop: skill improvement in any complex domain requires deliberate practice — a tight cycle of attempt → feedback → adjustment → repeat, where the feedback is specific and the adjustment is concrete. Trading's natural feedback is delayed, noisy, and emotional (Chapter 1) — the worst possible learning conditions. The review loop's entire purpose is to manufacture good feedback out of bad: to take the noisy stream of outcomes and, through structured review, extract the specific, process-focused signal that deliberate practice requires. The journal is the data; the review is the practice.
The cadence — four loops at four timescales, each answering a different question:
Daily (5–10 min) — the discipline check. Right after the session, while memory is warm: did I follow my plans today? Any unplanned trades, moved stops, or emotional entries? This loop isn't about P&L — it's about catching discipline breaks the same day, before they harden into habits. Its output is awareness, not analysis.
Weekly (20–30 min) — the pattern scan. Once a week: rule-adherence rate, the week's emotional tags, any recurring behaviour, the stand-down/plan-break ledger. This is where week-scale patterns surface — the FOMO cluster, the Friday leak — while they're still small. Its output is one or two specific things to watch next week.
Monthly (1 hour) — the metrics review. Once a month, the numbers of Chapters 13–14: expectancy, profit factor, R-multiple, sliced by setup and regime. This loop asks is my edge real and where does it live? and produces decisions — which setup to cut, which to concentrate on, whether a strategy's live results still match its promise. Its output is changes to what you trade.
Quarterly (half day) — the strategy audit. Every quarter, the highest loop: are my strategies still working (your Legendary Traders school's Asness question — has the evidence changed, or just my comfort)? Should my rules, my risk limits, my playbook itself evolve? This is where the system amends — deliberately, in writing, never mid-drawdown (your Behavioural Finance school's constitution clause). Its output is changes to the rules themselves.
The reframe that makes the loop stick: capture is the cost; review is the return. Traders lavish effort on capturing perfect data and then never read it — spending the cost and skipping the return. The disciplines that make the loop survive are the same ones that saved the journal itself: the reviews must be calendared (an unscheduled review is a skipped review), time-boxed (so they're sustainable), and their outputs must be written back into the system as new rules or new fields (a lesson noted only in your head is scheduled for re-purchase, at market rates). Run this loop for a year and the journal stops being a record of your trading and becomes the engine of its improvement — which was always the point.
Key Takeaway
Captured data changes nothing; the compounding is entirely in the review. Trading's feedback is delayed, noisy, and emotional — the worst conditions for learning — so the review loop exists to manufacture the specific, process-focused feedback that deliberate practice requires. Run four calendared loops: daily (discipline check), weekly (pattern scan), monthly (metrics and edge-location), and quarterly (strategy audit) — each producing a concrete written output. Capture is the cost; review is the return, and most traders pay the cost and skip the return.
Think About It
Run four review loops at four timescales: daily (5–10 min, checking discipline and catching unplanned trades or moved stops same-day), weekly (20–30 min, scanning for behavioural patterns and rule-adherence), monthly (1 hour, reviewing expectancy and metrics sliced by setup to decide what to trade), and quarterly (half a day, auditing whether your strategies and rules still work). Each loop answers a different question and produces a concrete written action. Calendar them like appointments — an unscheduled review is a skipped review, and the review is where all the improvement actually happens.
Journal Lab — Install the Loop
Put four recurring appointments in your calendar right now: a 10-minute daily review (end of session), a 30-minute weekly (weekend), a 1-hour monthly (month-end), and a half-day quarterly. For each, write the one question it answers (discipline / patterns / edge-location / strategy-validity) and commit to writing one concrete action per review back into your journal or rules. Run it for one month and check: are the actions from week 1 actually changing week 4's behaviour? That closing of the loop — action written, behaviour changed — is the entire return on journaling.