How to Review Your Trades: A Weekly Routine That Compounds

You can log every trade for a year and still not get any better. Logging is data collection. Review is where the data turns into a decision. Most traders do the first part religiously and skip the second, then wonder why the journal never seems to help.

This is the routine that closes that gap. It takes about 20 minutes once a week, and it ends with exactly one thing you're going to do differently. Not ten. One.

Logging vs. reviewing

Logging is passive. You dump the trade in — entry, exit, size, a note — and move on. It feels productive because the row count goes up. But a spreadsheet with 300 rows you've never re-read is just a graveyard.

Reviewing is active. You go back into the pile looking for a pattern, with a question in mind: where is my money actually leaking? A good review always produces an answer you can act on. If you finished reviewing and nothing changed about how you'll trade next week, you didn't review — you re-read.

The rest of this is a script you can run in one sitting.

The 20-minute weekly routine

Do this on the same day each week — a slow Sunday works. Same day matters more than which day. You want it to become a reflex, not a decision you have to make.

1. Sort by setup and find what's bleeding (5 min)

Group every trade this week (or the last two weeks if volume is low) by its setup tag — breakout, pullback, reversal, news, whatever taxonomy you use. Look at net P&L per tag, not per trade.

You're not looking at individual winners or losers. You're looking for the tag that's net negative across the group. One losing trade is variance. A tag that's red three weeks running is a leak. If you don't tag setups yet, that's the first fix — everything below depends on it.

2. Sort by time-of-day and weekday to find your graveyard hour (5 min)

Now re-slice the same trades by hour of entry and by day of week. Almost everyone has a graveyard hour — a window where their win rate quietly collapses. Common culprits: the first 15 minutes after the open when you're chasing, the post-lunch chop, the last hour when you're forcing a green day.

You're looking for a time block that's consistently net negative. It's usually not subtle once you sort for it. Same with weekday — plenty of traders are profitable Tuesday through Thursday and give it all back on Monday's noise or Friday's boredom.

3. Read the notes on your losers for the recurring word (5 min)

Open only the losing trades and read the notes you wrote at the time. Don't analyze the charts — read the language. You're hunting for a word that keeps showing up.

It's almost always there. "Felt like it should bounce." "Wanted to get even." "Bored, took it anyway." "Averaged down." When the same word or phrase appears across four losers, that's not four mistakes — it's one habit wearing four different outfits. The recurring word is the real problem; the individual trades are symptoms.

4. Pick ONE thing to change — and write it down (5 min)

You now have three signals: a bleeding setup, a graveyard hour, a recurring word. Pick the single biggest one and write one rule for next week. Literally type it somewhere you'll see it before the session.

Make it a hard constraint, not a vibe:

• "No reversal trades this week" (setup)
• "No new entries between 12:00 and 13:30" (time)
• "If the note would say 'get even,' I close the platform" (habit)

Then stop. Close the journal. You're done.

Why the one-change rule makes it stick

The temptation is to fix everything you found. Resist it. If you change five things and next week is better, you have no idea which change did it — and if it's worse, same problem. You've learned nothing and you can't repeat it.

Changing one thing turns each week into a clean experiment. You isolate the variable, you get a readable result, and you build a genuine, tested list of rules over a season instead of a pile of good intentions you abandoned by Wednesday. One real fix a week is 50 fixes a year, each one proven. That compounds. Ten fixes at once compounds to nothing because none of them survive contact with a bad day.

Daily vs. weekly cadence

Review weekly, not daily. Daily P&L is mostly noise — a single trade or a single volatile session swings it enough that any "pattern" you spot is an illusion. Reacting to it makes you tinker with a working process because of one red day, which is how good systems get abandoned.

Keep a daily habit, but a small one: jot the note at the time of the trade, while the feeling is fresh. That's the raw material. The pattern-finding — the actual review — waits for the week, when you have enough trades for a shape to emerge above the noise.

A worked example

Say you traded 22 times last week and finished down. Painful, but vague.

You sort by setup: breakouts and pullbacks are both slightly green. Reversals are deep red — five trades, four losers, and they account for your entire drawdown. Everything else roughly paid for itself.

You sort by time: those reversal trades cluster between 11:30 and 1:00 — the midday chop, when nothing's really trending.

You read the loser notes: three of the five say something like "looked overextended, faded it." There's your recurring word — faded. You're fighting trends out of boredom in a dead part of the day.

One change, written down: no counter-trend fades between 11:30 and 1:00. That's it. Not "trade better." A specific rule that removes the exact behavior your own data just convicted. Next week you'll know within five trades whether it helped — and either way, you'll have a clean answer.

That's the whole point of a journal. Not the logging — the one honest sentence at the end that tells you what to do Monday.

If you want the mechanics of good notes, start with how to keep a trading journal, and for the numbers to sort by, see the metrics that actually matter. Then run the routine above on top of them.

Sutekka's weekly view is built around exactly this routine — setups, time-of-day, and your own notes, sorted for you so the leak is obvious in a glance. Start free and run your first review this Sunday.