Guide
How to track a trading plan
A trading plan you do not track is a wish. Tracking means comparing what you actually did against what the plan said — trade by trade — and catching the drift while it is still cheap.
By Daniel Kapadia, founder of Mettle · Published June 12, 2026
The problem
Plans do not fail loudly — they drift
Almost nobody abandons their trading plan on purpose. It erodes: a stop moved "just this once", an entry taken a tick early, a setup traded in conditions the plan excluded. Each deviation is small enough to excuse, and three months later the trader is running a strategy nobody ever wrote down or tested.
The drift is invisible precisely because nothing measures it. The P&L cannot tell you — it mixes plan trades and improvisations into one number. Memory cannot tell you — it remembers the plan, not the deviations.
Tracking a plan means making the comparison explicit: every trade tagged to the play it was supposed to follow, every deviation recorded at the moment it happens, and a weekly look at the gap between the written plan and the actual record.
Step by step
- 1
Write the plan as discrete, gradeable plays
A trackable plan is a set of named plays, each with an explicit trigger, stop placement, sizing rule, and exit logic. "Trade breakouts with good risk-reward" cannot be tracked; "Breakout-retest: enter on the retest hold, stop under the level, fixed risk per trade" can.
- 2
Define what "followed the plan" means per play
For each play, write the three or four binary checks that make a trade compliant: entered at trigger, stop at the defined level and never widened, size per the rule, exit per the logic. Binary checks make grading honest — "mostly followed it" is where drift hides.
- 3
Tag every trade to a play at journal time
When you journal a trade, the first question is: which play was this? Every trade gets an answer, including "no play". Do not backfill this at week's end — the whole value is recording the intent while you still remember it.
- 4
Record deviations the moment you notice them
Moved the stop, added size, took an early exit: note it on the trade with a consistent tag. A deviation recorded honestly is data; a deviation rationalized is the first step of drift.
- 5
Review plan-vs-actual weekly, per play
Once a week, look at each play's record: how many trades, what compliance rate, what the compliant trades did versus the deviations. This is where the plan stops being theory — you find out which plays you actually trade, which you trade badly, and which you quietly avoid.
- 6
Change the plan deliberately, never silently
Plans should evolve — but as written, dated edits, not as accumulated exceptions. If you keep deviating the same way and it keeps working, promote the deviation into the play. If it keeps costing you, the tracking just paid for itself.
Worked examples
Finding stop-drift in the numbers
Twelve trades on one play this month; the journal shows the stop was widened on four of them. Those four account for most of the play's losses. Without per-trade compliance tags, this reads as "the play is getting worse". With them, it reads as what it is: the trader is, on that one rule.
The no-play count
A week's journal shows nine trades: six tagged to plays, three tagged "no play". That ratio is one of the most honest discipline metrics that exists, and it is invisible unless every trade gets the which-play question at journal time.
The play you keep skipping
Plan-vs-actual review shows a play with a strong record — that you have traded twice in six weeks while its trigger fired far more often. The plan is fine; something about pulling the trigger is not. That finding routes to your discipline work, not to plan edits.
Common mistakes
Writing a plan too vague to grade
If a reasonable person could not look at a trade and the plan and say "compliant or not", the plan is a mood board. Specificity is what makes tracking possible at all.
Editing the plan mid-week to match what you did
That is not updating the plan, it is laundering deviations. Collect the evidence all week; edit the plan once, deliberately, at review time.
Tracking only P&L per setup
Per-play P&L without compliance data tells you which plays made money, not whether the plays were actually traded as written. The drift hides inside the aggregate.
Treating every deviation as a moral failure
Deviations are data. Some reveal a leak; some reveal the plan is wrong. Punishing yourself for each one just teaches you to stop recording them.
Keeping the plan in your head
An unwritten plan cannot be compared against anything, conveniently. Writing it down is the act that makes drift visible — which is exactly why it feels uncomfortable.
Tools for the job
The workflow needs three things: a written, versioned plan; a journal that tags trades to plays; and a weekly view of plan-vs-actual. A document and a spreadsheet can do it. A purpose-built playbook removes the bookkeeping that makes most traders quit by week three.
A plan document plus a spreadsheet
Write the plays in a doc, add a "play" and a "deviation" column to the trade log, and build the weekly per-play summary by hand. Entirely workable — the failure mode is the manual upkeep, not the method.
Mettle — a playbook your journal is wired to
Every setup you trade gets a written play page, and every journaled trade gets tagged back to the play it was supposed to follow — so plan-vs-actual is a built-in view, not a weekend spreadsheet project. Deviations are captured as behavioral tags at journal time.
See how it works in MettleExecution analytics on top of the tags
Once trades carry play and compliance tags, Mettle's analytics can show each play's record with the deviations separated out — the view where drift becomes visible.
See how it works in MettleFAQ
What if I do not have a trading plan yet?
Start with one play — your most familiar setup, written with an explicit trigger, stop, size, and exit. Track just that one for two weeks. A one-play plan you actually track beats a ten-play plan that lives in a drawer.
How often should I update the plan?
Review weekly, edit deliberately — typically no more than once a week, with the change written down and dated. If you find yourself wanting to edit it mid-session, that urge is itself worth journaling.
Is a trade outside the plan always a mistake?
No — it is an untracked experiment. The problem is not the occasional exploratory trade; it is not labeling it. Tag it "no play", and if the same experiment keeps appearing, that is a candidate for becoming a written play.
Is Mettle free to start?
Yes. You get full access free for 14 days with no card. We only ask for a card once you have reviewed three sessions — after the product has proven it earns a place in your routine.
Put your plan where your trades are
Write your plays in Mettle's playbook, tag every journaled trade back to its play, and see plan-vs-actual every week without building a spreadsheet.
Start free — no cardMore guides
Picking a journal? Read our honest roundup of trading journal software.