I kept telling myself I was a disciplined trader, right up until I tried to prove it. The problem with discipline is that it lives in your head as a feeling, and feelings are terrible at accounting. You remember the three times you followed your plan perfectly and forget the eleven times you nudged a stop, added to a loser, or took a setup that was not really your setup. So the honest thing to do is stop treating discipline as a character trait and start treating it as a measurement, which means you have to define what a rule is, count when you break it, and put a P&L number next to the breaks.
None of this is hard. It is just tedious in the way most useful things are tedious. And the payoff is worth it, because when you finally have the data, almost everyone finds the same thing. The violations are not spread evenly across your trading. They pile up in a handful of situations, and those situations tend to form your single worst-performing bucket. Which is oddly good news, because it means the fix is narrow.
Define your rules before you can break them
You cannot log a violation against a rule you never wrote down. This sounds obvious and it is the step people skip. Before anything else, list the specific, checkable rules you actually trade by. Not aspirations, not "be patient," but things a stranger could verify from your trade log. Something like: I only enter on my two named setups, I risk a fixed fraction per trade, I set the stop before I enter and I do not move it wider, I do not add to a losing position, and I stop trading for the day after a set number of losses.
The test for a good rule is whether you could argue about whether you broke it. If you moved a stop from where you set it to somewhere worse, that is not a judgment call, it is a yes. If a rule requires interpretation every time, it is too vague to measure and you should either sharpen it or drop it. Five to eight hard rules is plenty. More than that and you will stop logging within a week.
Log the deviation and tag its type
For every trade, you record two things you probably already track, entry and exit and P&L, plus one thing you probably do not, which is whether you followed the rules and if not, which one you broke. That last field is the whole exercise. I keep it as a simple tag on each trade, and a trade can carry more than one.
A tag vocabulary that has held up for me looks roughly like this:
- Unplanned entry. The setup was not one of your named setups. You took it anyway.
- Oversize. You risked more than your fixed fraction, usually because you felt sure.
- Stop moved. You widened the stop after entry, or deleted it entirely.
- Averaging down. You added to a position that was already underwater.
- Revenge. The trade existed only because the previous one lost.
- Overtrading. You kept going past your daily loss limit or your trade count cap.
- Early exit. You closed a winner well before your plan said to, out of fear.
You will notice most of these are about the exit and the sizing, not the entry. That matches what actually costs people money. Getting into a trade is the easy part. The damage tends to happen in how you manage risk once you are in it, and the tags are where that shows up.
Compute an adherence rate, then attribute P&L to the breaks
Once you have a few weeks logged, two numbers fall out of it. The first is your adherence rate, which is just the share of trades that carry no violation tag. If you took forty trades and eleven had at least one tag, you followed your rules on roughly seventy percent of them. That single number is your discipline KPI. Track it week over week and it moves, and you can see it move, which is more than you could ever do with a feeling.
The second number is the one that changes behavior. Split your P&L into two buckets, clean trades and rule-breaking trades, and total each. What most people find is uncomfortable in a useful way. The clean bucket is quietly positive, sometimes very positive, and the rule-breaking bucket is where most of the losses live. Your edge was fine. It was the deviations dragging the whole account sideways.
You can go one level deeper and attribute P&L by tag. Sum the result of every trade carrying "averaging down," then every trade carrying "revenge," and so on. This is where it gets specific, because usually one or two tags own the damage. It is rarely all seven. You might find that oversizing barely hurts you but that moving stops is a slow bleed you never noticed, or that revenge trades are individually small but you take so many that they add up to your worst category by far.
Fix the cluster, not the character
The reason to do all this is that the fix becomes obvious and narrow. If your P&L attribution says averaging down is eighty percent of your rule-breaking losses, you do not need to become a more disciplined person in general. You need one mechanical guardrail against one behavior. Turn off the ability to add to a losing position, or make the rule a hard block rather than a soft intention, and you have recovered most of the leak.
A few things worth knowing before you start. Log the tag at the moment of the trade or immediately after, never at the end of the week, because by then you have quietly rewritten the story in your favor and your adherence rate will look better than it was. Do not punish yourself for a low rate early on. The first honest number is almost always worse than you expected, and that is the point. And do not let the tags multiply into a taxonomy so fine that logging becomes its own chore. A rough, consistent tag beats a precise one you abandon.
If you already keep your trades somewhere structured, a spreadsheet works fine for this and so does the journal inside most serious tools. On Blockcircle we surface P&L attribution against your execution history, which makes the clean-versus-broken split easy to run, but the method matters more than the tool. You could do the whole thing on paper. What you cannot do is skip the counting and still claim to know how disciplined you are, because until you have the adherence rate and the two buckets, discipline is just a story you are telling yourself, and the story is usually kinder than the ledger.