A pre-mortem is a planning technique from project management: before starting, you imagine the project has failed and list the reasons why. Applied to trading, it takes about ten minutes on Sunday.
How to run it:
Assume next Friday you are down more than your weekly loss limit.
Write down the three most likely reasons. Be specific: "traded CPI with full size", "kept fading a trend on Tuesday", "too tired after late nights".
For each reason, write one preventive rule for this week only.
Why it works:
It brings known weaknesses into focus before they are triggered.
It ties general rules to this week's actual calendar and circumstances.
On Friday, comparing the real week with the pre-mortem is a quick, honest review.
Over a few months, the same reasons tend to repeat. That list is a better guide to improvement than any general advice.
Would a pre-mortem fit into your weekly routine, and what would be on your first list?
I started doing a Sunday pre-mortem after a week where I blew half the daily budget on a CPI fade I had already banned in my notes. Writing the failure first made the ban stick for that week instead of sitting as a vague intention.
My first list is almost always the same three: full size into US data, revenge re-entry after a scratch, and trading past 16:00 London when I am already tired. For each one I write a single week only rule, like half size until Wednesday if CPI is on the card.
The Friday compare is the part that pays. Last month the pre-mortem named revenge sizing and the real week showed two of those tags, so size dropped the following Monday without a long journal session.
I disagree that the list needs to be creative every week. Repeating the same failure modes is the point; novelty usually means I am inventing fiction.
Do you keep the preventive rules visible on the desk during the week, or only in the Sunday file? I keep a sticky note with the three week only bans next to the platform.
One reason that never made my early lists was something outside my own behaviour. My connection once dropped for 25 minutes with a position open and no stop on the server, because I had been trailing it by hand.
That is exactly the kind of failure a pre-mortem should catch. Now every list has one line for kit, and the rule that goes with it is that every position gets a server side stop before I take my hand off the mouse.
The other change I made is writing each rule as a yes or no check. Half size until Wednesday is easy to review, while be more patient tells me nothing on Friday.
I also run a quick reverse version for green weeks. If the week ends positive but one win came from a trade I had banned, I log it as a rule break anyway, because a lucky violation is the one most likely to repeat.
Does anyone keep score on how often the pre-mortem reasons actually showed up? Mine hit two of three in roughly half the weeks so far.