The 20 Minute Weekly Trading Review

A journal you never read is a diary. Here is a six step weekly review that fits in twenty minutes and actually changes how you trade next week.

Daniël Vermes

Tradeflow Editorial Team

Aug 4, 2026

10 min read

A weekly trading review is a structured examination of your last week of trades to identify which patterns made money, which cost money, and what single change to make next week. It is the step that converts a journal from a record into an edge, and it is the step almost everyone skips.

Most traders who keep a journal have never properly reviewed one. They log diligently for months, accumulate several hundred trades, and open the data perhaps twice. The logging feels productive, which is precisely the problem, because logging on its own changes nothing about how you trade.

Here is a review that fits in twenty minutes and produces one decision.

Why reviews get skipped

Not laziness. Cost.

A proper review by hand means filtering a few hundred trades by setup, grouping the results by session and by state, cross referencing what you did in the days following your worst losses, then labelling weeks of historical trades against a pattern you noticed five minutes ago. In a spreadsheet that is an afternoon.

So it gets postponed to next Sunday, then the Sunday after, and by the time it happens the data is three months stale and the patterns have already been paid for.

The second reason is that reviews are unpleasant. You are voluntarily returning to the trades that did not work, on a weekend, to be told something unflattering about yourself. Any friction in that process is enough to stop it happening, which is why the review has to be short. A thorough review you do four times a year is worth less than a shallow one you do every week.

The rule that makes it work

One change per week.

The temptation after a review is to fix everything: adjust your stop distance, tighten your entry criteria, cut a setup, change your session hours. Do four things at once and you learn nothing, because you cannot attribute the result to any of them.

One change, written down, tracked the following week. It feels slow. Fifty two changes a year compound into a different trader.

Step 1: scan the week

Do not analyse yet. Just look.

Total P&L, number of trades, and the shape of the week on a calendar view. You are looking for the obvious: green days followed by red days suggests giving back profit, clusters of red suggests something went wrong and kept going, an unusually high trade count suggests overtrading regardless of the result.

Note anything that looks odd. You will come back to it.

Step 2: rank your setups

The most valuable four minutes in the review.

Filter by setup and rank by profit factor. Then compare that ranking against the frequency of each setup. The question you are answering: are the setups I take most often the ones that pay?

They usually are not. Most traders have one or two patterns generating almost all their profit, hidden inside five or six they trade regularly out of habit. The habit ones are rarely catastrophic, which is why they survive. They just quietly break even while consuming your attention and your risk budget.

If a setup has been below break even across thirty or more trades, it is not unlucky. It is negative expectancy and it needs a rule attached or removed.

Step 3: check rule adherence separately from p&l

This is the step that separates a useful review from a scoreboard.

What percentage of the week's trades followed your plan? Log this independently of whether they made money, because the four combinations mean completely different things.

Right for the right reasons is a system working. Wrong for the right reasons is a system working and variance doing its job. Wrong for the wrong reasons is expensive but at least legible. And right for the wrong reasons is the dangerous one, because a profitable rule break teaches you to break the rule again, and the lesson arrives with a reward attached.

Traders who track adherence separately often discover their strategy was never the problem.

Step 4: look at what happened after losses

The single most predictive data in any trading history, and close to nobody checks it.

Find your three worst losses of the week. Look at the next two or three trades after each one. You are checking for three specific things: did position size increase, did the time between trades shorten, and did the setup quality drop.

That sequence, loss to frustration to re-entry to oversizing, is the cascade that ends accounts. It is invisible in any individual trade and obvious across a week once you look for it directly.

If you find it, that is your one change.

Step 5: group by state

If you tagged how you felt, group performance by it. If you did not, this is the argument for starting.

Nearly every trader has a state that reliably costs them money and cannot name it without data. Tired, rushed, bored, and the day after a big win are the usual suspects, and the answer is rarely the one you would have guessed. The value here is not psychological insight, it is a rule: if a state costs you money consistently, you now have a reason not to trade in it.

Step 6: write one rule

Specific, testable, about process rather than outcome.

Weak: trade better in the afternoons. Strong: no new positions after 2pm, because afternoon trades cost me money in six of the last eight weeks.

Weak: stop revenge trading. Strong: after any loss over 1R, no new position for thirty minutes.

Write it where you will see it before the open. Track compliance next week. That number, not your P&L, is what you are measuring.

Doing this without spending an afternoon

Every step above is filtering, grouping and cross referencing. It is exactly the work that makes manual review too slow to sustain, and it is the reason most journals go unread.

This is where Tradeflow is built differently from a dashboard. Rather than filtering through views, you ask. Show me profit factor by setup for the last thirty days. Which trades came within an hour of a loss. Tag those as tilt. Group last quarter by state and show me where I lose money.

It reads your synced trade history, works out what it is looking at, applies the tags, and gives you the answer. The four minute step stays four minutes instead of becoming an afternoon, which is the entire difference between a review you plan to do and one you actually do.

That capability is on every plan from $19 a month, which is worth saying plainly because the norm in this category is to put AI behind a top tier at two to four times that price, keeping the most useful part of the product away from the traders most likely to give up.

Monthly and quarterly

The weekly review catches behaviour. Longer reviews catch drift.

For the monthly review, compare this month's setup rankings to last month's. Check whether your average winner is shrinking while your average loser holds steady, which is expectancy eroding quietly before it appears as a drawdown. Review your rule compliance trend across four weeks rather than one.

Quarterly, cut something. A setup that has not made money in three months goes, or gets a rule attached. A journal that never removes anything is not being used.

Quarterly is also the point to look at your record as an outsider would. Sort by loss size, check sizing consistency, look at the shape of the drawdowns. This is what a prop firm risk manager will do with your track record if you ever apply, and it is considerably better to find the problems yourself first.

What a good review feels like

Slightly uncomfortable and quite short.

You are not looking for reassurance, and a review that consistently makes you feel good about your trading is one that is not looking hard enough. You are looking for one thing you did not know, and most weeks you will find one.

The trades you want to skip past are the ones with the information in them. That is the entire skill.

Key takeaways

A journal you never review is a record, not an edge. Logging alone changes nothing. Reviews get skipped because manual filtering is slow, not because traders lack discipline. One change per week means multiple simultaneous changes make attribution impossible. Track rule adherence separately from P&L, because a profitable rule break is the most expensive trade you can take. The trades following your worst losses are the most predictive data in your history and the least examined. Weekly catches behaviour, monthly catches drift, quarterly is when you cut something. Tradeflow lets you run the whole review by asking rather than filtering, on every plan from $19.

Frequently asked questions

How often should I review my trading journal? Weekly for behaviour, monthly for drift, quarterly for structural changes. Weekly is where nearly all actionable insight appears, because a month is long enough that a costly pattern runs for four weeks before you catch it.

How long should a weekly review take? Twenty minutes if the filtering is handled for you, considerably longer by hand. Keep it short deliberately. A brief review you complete every week beats a thorough one you abandon after three.

How many trades do I need before a review is meaningful? Around thirty in a single category for a basic conclusion, fifty or more when looking at combinations of setup, session and state. Below that you are reading noise, though the habit is worth building before the data supports conclusions.

What if I did not tag my trades? Start with what the data gives you: setup, time, size and result are all in your synced history. State and reasoning require you to have logged them, which is the argument for the twenty seconds per trade. With Tradeflow you can also tag historically, so a pattern you notice this week can be applied to trades from three months ago.

Should I review winning weeks too? Especially winning weeks. Rule breaks that happened to work are invisible when you only review losses, and those are the ones that teach you the worst habits. Check adherence on green weeks specifically.

What is the single most useful thing to look at? What you did immediately after your worst loss. If size went up or the gap between trades shortened, you have found the pattern that costs most traders the most money.

Make the review something you actually do

Tradeflow syncs every trade automatically from 600+ brokers and platforms, and lets you run your review by asking questions instead of building filters. Twenty minutes, once a week, one change.

Start free for 7 days. From $19 a month afterwards, AI included on every plan.