how to run a personal post-mortem on a losing week

A losing week happens. Sometimes the strategy is broken. Sometimes the market is wrong. Sometimes the trader is wrong. The job is to find out which.

Most retail traders end a losing week by closing the broker screen and not opening it until Monday. The professionals open a different screen — the post-mortem. The post-mortem is one document, one hour, run every Friday after a losing week, with one constraint that makes it actually work.

Here is the template I have used since 2019 and the constraint that makes it useful.

the two failure modes of not doing a post-mortem

Avoidance: you do not look at the week. You convince yourself it was random. You start the next week assuming nothing was learned. This is the slow version of failure — the trader’s process does not improve and the same mistakes repeat.

Revenge: you look at the week, decide it was the market’s fault, and double the position size on Monday to “get it back.” This is the fast version. Revenge trading after a losing week is the single most reliable way I know to turn a 10% loss into a 40% loss.

The post-mortem is the structured process that prevents both.

the template

1. list every trade closed this week
2. for each trade, mark: process error (P) or outcome error (O)
3. compute the process-error rate
4. for each P, identify which rule was violated
5. for each rule violated more than once, decide: enforce harder, or modify
6. write the modification or enforcement in plain language
7. one paragraph: what was happening in your life this week

The first six are mechanical. The seventh matters more than people expect.

process error vs outcome error

A losing trade where you followed your own rules is an outcome error. The strategy expected some losses; this was one of them. Do not change the rule based on a single losing outcome. Do not even feel bad about it.

A losing trade where you violated your own rules is a process error. You entered without a thesis. You held past the stop. You sized larger than your sizing rules permit. The loss is incidental — the violation is the data point.

Almost every retail trader inverts this. They feel terrible about outcome errors (which they should not) and excuse process errors (which is what kills the account). The post-mortem forces the right categorisation.

the modification step

For each rule violated more than once, you have two choices: enforce the rule harder, or admit the rule is wrong.

Enforcing harder means making the rule mechanical. Move the stop to the venue, not your head. Pre-set the size at the order ticket level. Lock the trader out of new positions for a fixed cool-down after a stop.

Admitting the rule is wrong is harder and rarer. Sometimes a rule that worked in a previous regime does not work now. The 50-MA stop that worked in 2021 may not work in 2025. The signal that worked in low-volatility may fail in high-volatility. The post-mortem is the right venue to decide whether a recurring violation is the trader being undisciplined or the rule being obsolete.

The honest answer is usually “trader was undisciplined.” The honest answer is occasionally “rule was obsolete.” Both happen. Distinguishing them requires looking at the data, not just at the emotion.

the life-context paragraph

The seventh step — what was happening in your life this week — is the one I added after years of running these. It catches the meta-patterns the trade-by-trade analysis misses.

Losing weeks correlate with events that are not on the trade journal. A bad week at work. A fight with a partner. A flu. A trip with bad sleep. Multiple times I have looked back at a losing week and realised it was the third sick week of a stretch — and that I had also lost money in the other two without noticing the correlation.

Writing one paragraph about the week’s context surfaces these patterns. Over enough months you learn which life-events predict losing weeks for you. The fix is usually not “fix the life-event.” The fix is “trade smaller during predictable bad-judgment windows.”

the 30-minute rule

The constraint that makes the whole process work: no new positions while writing the post-mortem, and no new positions for 30 minutes after finishing it.

The act of writing the post-mortem stirs up the emotions that caused the bad week. Reading your own process-errors back triggers the revenge instinct that wants to make it back right now. Adding a 30-minute cool-down between the post-mortem and the next trade decision lets the emotion settle.

I lost money for years before I added the cool-down. Within two months of the rule, the post-mortem benefit doubled. The compounding effect of what you do not do is the underrated half of trading.

the compounding benefit

A losing week is a guaranteed cost. The post-mortem is the only mechanism that converts the cost into a lesson. Without the post-mortem, the cost is just a cost.

Over five years of weekly post-mortems, I can show you exactly which rules were modified, which were enforced harder, which were retired. I can also show you the trades I did not take because I was about to write the post-mortem on them and could not articulate the thesis. That subset of “trades not taken” is, by my count, the single largest contributor to net positive PnL since 2020.

The job is not to win every week. The job is to be honest about the losing weeks.

---

→ Earlier: journaling your trades — the meta-layer of feaws → Related: the behaviour gap — measure it in your own trade journal