Journaling·9 min read
How to keep a trading journal that actually changes your trading
Most trading journals fail for one reason: they record outcomes and nothing else. A spreadsheet of entries, exits and P&L tells you what happened, but it cannot tell you why, and it cannot show you what to do differently on the next trade. A journal that works records decisions — and then lets the numbers argue with your memory.
Published 9 September 2026 · by the SageTradingJournal team
Why journals fail
A journal is only useful if you can ask it questions later. "Which of my setups makes money?" "Do I trade worse after a loss?" "Is London better than New York for me?" Every one of those questions needs data you must have written down at the time. If you only logged the P&L, the answer to all of them is a shrug.
The second failure is friction. A journal that takes ten minutes per trade gets abandoned in week three. The trick is to separate the two jobs: recording, which should be nearly automatic, and reviewing, which should be deliberate and scheduled. Most of the value is in the review — but the review is impossible without the record.
What to record on every trade
| Field | Why it matters |
|---|---|
| Instrument, side, entry, exit, times | The bare facts. Times matter more than most people think — they give you sessions and time-of-day later. |
| The stop you had at fill | Not the stop you ended with. The initial stop defines your risk, and therefore the R-multiple of everything that follows. |
| Planned target (if any) | Lets you compare what you planned against how you actually managed the trade. |
| Setup / play | The name of the idea you were trading. Without it you cannot group trades into anything meaningful. |
| Session | Asian, London, New York — or whatever divides your market. One of the most common places an edge hides or a leak shows. |
| Checklist confirmed? | Did you check your own rules before entering? Recorded at entry, it becomes evidence; recorded after, it becomes a story. |
| State going in | Calm, rushed, after a loss, tired. Behavioural leaks correlate with state, and state is forgotten by the time you review. |
| A screenshot | The chart as you saw it. A week later you will not remember what the context looked like. |
| Mistakes, in R | "Moved my stop" is a feeling. "Moving my stop cost 0.8R" is a number you can add up over a month. |
Think in R, not in money
R is the distance from your entry to your initial stop — the amount you agreed to lose. Every outcome is then expressed as a multiple of it: a trade that hits its stop is −1R; a trade that makes twice its risk is +2R. This does three useful things. It makes trades of different sizes comparable. It separates the quality of the decision from the size of the bet. And it makes your statistics honest: a strategy that wins 40% of the time at 2.5R is profitable, and you can only see that if the wins are measured against the risk taken.
Trade A: risk $20 (entry 2,315.00, stop 2,313.00) → exit at 2,319.00 = +$40 = +2.0R Trade B: risk $200 (entry 1.0850, stop 1.0830) → exit at 1.0890 = +$400 = +2.0R Same decision quality. In dollars they look nothing alike.
If you want the arithmetic laid out, the risk/reward calculator turns entry, stop and target into a ratio, an R-multiple and the win rate that ratio needs to break even.
Tags: the vocabulary of your own trading
Tags are how a journal answers questions. Keep them few and consistent: your setups ("sweep-into-FVG", "trend-pullback"), your sessions, and the two or three behavioural labels you actually want to track ("took-early", "chased", "no-checklist"). Resist inventing a new tag for every trade — a tag that appears once tells you nothing, and thirty tags that each appear once tell you nothing thirty times.
Better still, write your setups down as a playbook: a thesis, the rules for entry and invalidation, where it applies, and how much it risks. Then tag each trade with the play it belongs to. That turns a pile of tags into a set of hypotheses you can test. Sage's Playbook is built around exactly this, including the discipline of writing down how the idea could be proven wrong before the data answers.
How to review — weekly, by category
Reviewing trade by trade feels productive and mostly is not; you relive each one and learn what you already knew. Review by category instead. Once a week, ask a small fixed set of questions and let the totals answer:
- 1Which setup made and lost the most R this week? Is that in line with the last month?
- 2Which session did I trade best and worst in? Am I trading sessions I have no edge in?
- 3How many trades broke a rule (no checklist, risk over limit, entered within minutes of a loss)? What did they cost in R?
- 4Did my planned exits or my actual management do better? (If you cut early every time, this is where it shows.)
- 5One thing to do differently next week — one, so it actually happens.
Automate the recording, keep the thinking
The fastest way to keep a journal for years is to stop typing what your platform already knows. Import broker history in bulk, or connect the platform so closed trades appear on their own — then your only job is to add the setup, the tags and a sentence of reasoning. Sage does this with a free MetaTrader 5 Expert Advisor that posts closed positions to your journal, including the stop that was active at fill (how it works). If you trade elsewhere, a CSV import gets you most of the way.
What should stay manual is the thinking: why you took the trade, what you saw, how you felt going in. Two sentences at entry beat two paragraphs a week later — memory rewrites the story to fit the result. Sage's journal, The Forge, is built to make that entry fast, and to make the record pay you back immediately with what changed in your stats.
A minimum viable journal, today
- One row per trade with the fields above. Yes, all of them; they take ninety seconds when the trade is fresh.
- One weekly review, twenty minutes, same five questions.
- One change per week, written down, checked the following week.
- After thirty trades, your first honest look at which setups and sessions are carrying you.
Do that for a quarter and the journal stops being a chore and becomes the only opinion about your trading you can fully trust — because it was written down before the outcome was known.