The Trading Journal: What to Log, and What It Tells You
Updated ·8 min read·Reviewed by the StockTools.ai Research Team
- ▸A journal earns its place by answering two questions your broker statement cannot: which setups actually pay, and which repeated behaviors cost you money.
- ▸The minimum useful record is entry, exit, size, stop, and one tag for why you took the trade. Everything else is refinement.
- ▸Tags are what make a journal computable. Untagged trades produce a P&L history; tagged trades produce a per-setup and per-mistake breakdown.
- ▸Review on a fixed schedule, not when you feel like it. A weekly pass over closed trades catches patterns a single-day post-mortem never surfaces.
- ▸Most journals stop at your own numbers. Yours can also show what insiders, Congress, and the earnings calendar were doing around each trade.
A journal is a measurement instrument
Your broker already stores every fill. What it will never tell you is why you took the trade, whether you followed your own plan, or which of your setups is quietly subsidizing the others. A journal exists to capture the parts of a decision that vanish the moment the position closes.
That reframing matters because it changes what you record. A diary captures feelings. An instrument captures variables you can later group by, so that after 60 trades you can say something like "my breakout trades returned $4,200 across 18 trades and my reversal trades lost $900 across 22" and act on it. One of those sentences is worth more than a month of screenshots.
The free StockTools journal runs entirely in your browser and needs no signup, so the cost of starting is a few seconds. Log the next trade you take and the sample builds itself from there.
The fields that earn their keep
Four groups of data cover almost everything a working trader needs. First, the quantitative spine: date, symbol, direction, size, entry price, exit price, planned stop, fees, and realized P&L. The stop is the field traders most often skip and the one that unlocks the most, because entry minus stop times size defines your risk on that trade, which converts every result into an R-multiple.
Second, the reason. One short tag for the setup ("VWAP bounce", "breakout", "gap fill") is enough. The tag is not decoration; it is the grouping key that later splits your results into strategies that work and strategies you like the look of.
Third, trade management. Record any move you made to a stop or a target after entry. Advanced journals also capture Maximum Favorable Excursion and Maximum Adverse Excursion, the best and worst unrealized prices a position reached, which answer whether you exited too early or sat through avoidable heat. The StockTools journal does not compute MFE and MAE today; it logs your planned stop and derives R from it instead.
Fourth, mindset and compliance. A one-word emotional state and an honest answer to "did I follow my written plan" turn psychology into data. Tag a trade "fomo" and it stays a story. Tag 30 of them and the journal can total what fomo cost you.
A worked week: what tagging changes
Take a trader with five closed trades in one week. Two AAPL breakouts made $350 and $300. An NVDA entry chased after the move made minus $200. A TSLA entry, also chased, made minus $250. A GME position averaged down into weakness made minus $200. Net for the week: exactly $0. Win rate: 40%. Read as a P&L line, the week was a waste of five days.
Now read it with tags attached. The "breakout" setup produced $650 across two trades and won both. The "chased" tag appears on two trades and cost $450. "Averaged down" appears once and cost $200. The flat week is actually a profitable strategy carrying two repeatable errors.
The arithmetic is worth doing explicitly. Remove the two chased trades and the same week nets $450 from three trades, an average of $150 per trade rather than $0. Nothing about the winning strategy changed. The improvement came entirely from subtraction, which is the cheapest edge available to any trader, and it was invisible until the trades carried tags.
Two of the five trades had a stop recorded, so stop discipline for the week was 40%. That single percentage is often the most actionable number on the page, because the three trades without a stop are the three that produced the losses.
The three-phase routine
Pre-market, before anything moves: write your watchlist, note the day's scheduled catalysts, and fix a maximum loss for the session. Deciding your stopping point while calm is the entire value of this phase. The earnings calendar tells you which names report before the open and which report after the close, which is the difference between a planned trade and a surprise.
Intra-session, log as you go or immediately after each close. Memory degrades fast and reconstructing a trade an hour later reliably flatters the trader. If typing during the session breaks your focus, log the numbers only and add tags after the bell.
Post-market, do the review that makes the logging worthwhile. Pull up your worst trade and name the cause: was it the market, or was it execution. Check whether your losses cluster in a particular hour. Ask whether each loss came from a broken thesis or a broken rule, because those two failures need opposite fixes.
Weekly, zoom out. A single session is too small a sample to show a pattern, and daily reviews tend to relitigate the last trade rather than the habit. The journal's Weekly review tab assembles net P&L, win rate, average R, stop discipline, your best-performing setup, and your costliest mistake for any week, and it copies to your clipboard so you can paste it into your own notes.
Logging without the friction
The most common reason a journal dies is data entry. Typing 40 trades by hand after a busy week is a chore nobody sustains, and a half-filled journal produces conclusions worse than no journal, because the missing trades are rarely a random sample. Traders skip the embarrassing ones.
Your broker already exports the raw record. Import a broker CSV and the journal rebuilds your closed trades from the individual fills, pairing buys with sells in order, handling scale-ins, scale-outs, and short positions. Positions still open when the file ends are reported and skipped, because a journal logs what you closed.
Step-by-step export instructions exist for Charles Schwab, Fidelity, E*TRADE, Webull, Robinhood, Interactive Brokers, and tastytrade. Importing the same file twice adds nothing, so you can safely re-import after each week.
What no import can supply is the reason and the emotion. Backfill the numbers automatically, then spend your review time adding the tags, which is where the value actually lives.
Reading the numbers once they exist
Win rate on its own is close to meaningless. A trader winning 70% of the time with 1-point winners and 4-point losers goes broke, and a trader winning 40% with 3R winners and 1R losers compounds steadily. Judge the system by expectancy, the average result per trade, which combines how often you win with how much you win when you do.
Profit factor, gross wins divided by gross losses, is the fastest sanity check. Below 1.0 the system loses money by definition. Between 1.0 and 1.3 it is fragile and any increase in fees or slippage can flip it. Maximum drawdown, the largest peak-to-trough fall in your cumulative P&L, tells you what the strategy demands of you emotionally, which is usually the constraint that actually breaks people.
Beyond the headline stats, the journal runs deterministic checks for four expensive patterns: sizing up straight after a loss, stacking trades on a day that is already red, trading while tagged anxious or tilted, and entering without a stop. These are stated as arithmetic over your own log rather than opinion, and each one names what the pattern cost.
What most journals cannot tell you
Every journal on the market can compute your win rate, because it only needs your own trades. None of them can tell you what the rest of the market was doing on the day you traded, because they never collected it.
The StockTools journal can, because the filings are already there. Reconcile your log against the tape and each trade is annotated with the objective activity around it: insider Form 4 buying and selling in the days either side, congressional trades in the same name, and how close your trade sat to an earnings date. A trade you took long in a week when an officer sold and three members of Congress sold is not automatically a bad trade, and the journal does not say it was. It shows you the record so you can judge your own thesis against what was disclosed.
Only the ticker and the date leave your browser when you run that check. Your position sizes, prices, P&L, and notes stay local, which means the tool could not offer you a recommendation even if the house rules allowed it. It reports what happened. What to do about it stays your call.
New to the routine? The day trading playbook lays out the daily sequence and the risk rules that make the journal worth keeping.
FAQ
How many trades before a journal tells me anything real?
Expect noise below roughly 30 closed trades and treat anything under 20 as a story rather than a statistic. The same strategy can show a 65% win rate over its first 20 trades and 45% over the next 20 purely by chance. Per-setup breakdowns need the same sample within each tag, so a setup you have traded four times is not yet evidence.
Should I journal every trade or only the interesting ones?
Every closed trade. Selective logging is the fastest way to a flattering and useless dataset, because the trades people skip are disproportionately the impulsive ones, which is exactly the population you need to measure. Importing a broker CSV sidesteps the temptation entirely, since the file contains everything.
What is the difference between MFE/MAE and a stop?
A stop is your planned exit, decided before entry. Maximum Adverse Excursion is how far the trade actually went against you before it resolved, and Maximum Favorable Excursion is the best unrealized gain it reached. MAE tells you whether your stops sit too tight; MFE tells you whether you leave money on the table. The StockTools journal records planned stops and R-multiples rather than MFE and MAE.
Is my journal data private?
The StockTools journal stores everything in your own browser using local storage, with no account required and nothing uploaded. The one feature that touches the network is the optional reconciliation against filings, which sends only each trade's ticker and date. Clearing your browser data deletes the journal, so export the CSV or JSON periodically.
How often should I review, and what am I looking for?
Daily for execution errors while the session is fresh, weekly for patterns. In the daily pass, name the cause of the worst trade. In the weekly pass, look at three numbers: net by setup, net by mistake tag, and the share of trades that had a stop. Those three answer what to do more of, what to cut, and whether your risk process is holding.
Does a journal help swing traders, or only day traders?
The fields change slightly, the value does not. Swing traders log fewer trades, so the sample takes longer to build and per-setup conclusions arrive more slowly, but the same grouping by setup and mistake applies. Holding periods spanning earnings make the earnings proximity in the reconciliation more useful, not less.
Put it to work
Related guides
Sources & further reading
- ▸ SEC Form 4 insider filings (EDGAR)
- ▸ STOCK Act congressional trade disclosures
More to learn
Educational only — not financial advice. Concepts simplified for clarity; markets are messier than definitions.