Your Trading Review Is Probably Only Half Done

Think about the last time you sat down and went through your trades properly. Not a glance at the P&L — an actual look at what you did and why.

It was after a bad day. It almost always is.

That habit has a cost nobody talks about. If the only sessions you examine closely are the ones that hurt, then you are studying a sample made entirely of your worst work. You will get very good at recognising how you fail. You will learn almost nothing about how you succeed — which is the half you were hoping to repeat.

Review is not a post-mortem

A post-mortem happens after something dies. That framing is why the habit only fires after losses, and why it feels like punishment when it does.

The more useful frame is calibration. You are trading two things that never hold still: your own state, and the market’s. Neither sends a notification when it changes. The point of looking back is not to find the guilty party — it is to notice, early and cheaply, that one of them has moved.

That reframing changes the schedule. A post-mortem is triggered by an event. Calibration runs on a cycle, and it runs whether the week was good, bad, or unremarkable. The unremarkable weeks are often the ones with the most to say, because nothing dramatic is drowning out the signal.

Three horizons, three different questions. They are not three reviews — they are one habit at three zoom levels.

Yesterday: where am I standing right now

The first question is the shortest and the one most often skipped: what state am I in this morning?

Not just “did I lose yesterday”. The honest version has two halves. What am I carrying into today — tiredness, a red day I have not processed, an unusual amount of confidence? And what is actually working at the moment — which setup has been paying, which one has quietly stopped?

A trader who ends a losing session and opens the platform the next morning without answering that has not started the day; he has continued yesterday’s. The version of this that costs the most is not tilt in the obvious sense. It is trading the same size, on the same setups, with the same selectivity, on a day when the honest answer to “how am I” was “not great” — and never asking.

The reverse is just as real and gets no attention. Some mornings you are sharp, and the setup you have been trading is in a stretch where it works. Recognising that is also information. It is the difference between taking the good week and letting it pass at half size because you never noticed it was a good week.

The week: is this a tendency or an episode

Zoom out to the last five sessions and the question changes. Not “what happened yesterday” but “what keeps happening”.

This is where patterns surface that no single day reveals. The same mistake in three different names. Entries consistently a few seconds early. Position size creeping up on the days you are already down. Or, on the other side: a setup quietly doing better than it used to, more of your winners coming from one playbook than you would have guessed.

And this is where a weekly review earns its keep, because the output should not be a note. It should be a rule.

• The same mistake three times in a week is not bad luck. It is a rule waiting to be written: a tighter stop, a cap on trades per day, a filter that keeps a whole category of name off the list.
• A setup that has been paying better than the rest deserves the mirror-image decision: more of it, less of everything else, maybe more size.

A note says “I keep chasing”. A rule says “no entry after the first 30 seconds of the move”. Only one of them changes what happens on Monday.

The part that makes this dangerous: signal or noise

Here is where “adjust often” turns bad if you leave it there.

A week is often ten or fifteen trades. That is not enough to conclude much of anything, and a trader who rewrites his stop loss every Friday based on a dozen trades is not calibrating — he is flailing, and adding a new variable each week guarantees he can never tell which change did what.

A few tests that separate the two:

Did it repeat? Once is an episode. Three times in a week, in different names, is a tendency. The repetition is what makes it actionable, not the size of the loss it caused.

Did it show up across setups, or inside one? A problem that appears in every setup you traded is usually you — your state, your execution, your discipline. A problem confined to one setup, while everything else behaves normally, is usually that setup, or the market it depends on.

How many trades is the conclusion standing on? If a playbook is down over eight trades, that is a shrug. Over forty, it is a finding. Any review worth acting on should tell you the sample size next to the number, and if it does not, you should be suspicious of the number.

Change one thing at a time, and give it long enough to tell you something.

The quarter: has the ground moved

The longest horizon asks a question the other two cannot: is the setup still alive?

Markets change character. Volatility regimes shift, the kind of names that run change, liquidity moves. A setup that printed money in the spring can be genuinely dead by autumn — not because you are executing it worse, but because the conditions that made it work are gone. In fast markets that turn can come every few months.

This is the most expensive thing to get wrong, because it looks exactly like a slump. The trader concludes he has lost his edge, tightens up, tries harder at the same thing, and grinds down a setup that stopped working for reasons that have nothing to do with him.

The only way to tell the difference is to look at a long enough window and compare like with like. How did this playbook do over the last three months against the three before it? Is the win rate down, or is the average winner smaller — those mean different things. Has this been true across everything you trade, or only here?

If a setup is genuinely finished, that is not a failure. It is information, and it arrives far cheaper in a review than through six more weeks of losses.

Make it a cycle, not an event

None of this requires a lot of time. It requires that the trigger stop being pain.

Ten minutes on a Friday, whatever the week did. What am I carrying, what is working, what repeated, and does anything need to become a rule. A longer look every month or two at whether the setups you rely on still behave the way they used to.

The traders who stay in this a long time are rarely the ones with the best single week. They are the ones who noticed the change early — in themselves or in the market — and adjusted while it was still cheap. That is not a talent. It is a habit with a calendar attached.

---

This is the review cycle Trandence is built around — the readiness check that starts a session, the weekly view of what repeated, and playbook performance over any window with the sample size next to it, so you can tell a finding from a coincidence. It is free to use.