Ask a trader what they trade and you get a list: the opening drive, the failed breakout, the level fade, the trend pullback. Ask which of those made money last quarter and the answer is usually a feeling. That is a strange gap, because it is the single easiest thing in a journal to measure — provided the trades were named.
The measurement is easy. The tagging is not.
Group by setup, read net, win rate, profit factor and net per trade for each. That is the whole technique.
What defeats it is that a setup assigned afterwards is assigned by somebody who knows how it turned out. Memory is not neutral: a trade that worked becomes a textbook example of the setup you meant to take, and one that failed becomes “not really a proper one of those”. Do that consistently and every setup in your journal has a respectable win rate, because the failures have been quietly reclassified out of it.
The tag has to be attached at the point of the decision, or as near to it as you can manage — at the latest, at the close of the same session. A setup tag written a week later is a record of your memory, not of your trading.
Auto-logged trades arrive unnamed, and that is not a flaw
If your journal fills itself from the broker, the trades come in without a setup, because the fill feed does not know why you took them. It carries the instrument, the size, the price and the time; the idea lives only in your head.
What matters is what the journal does with that gap. Stamping every auto-logged trade with a placeholder produces one enormous bucket labelled with a word nobody chose, sitting at the top of the table looking like your most-used setup. Unnamed should read as unnamed — counted, reported, and left out of the comparison.
What the table looks like when it is honest
| Setup | Trades | Net | Win rate | Profit factor | Net/trade |
|---|---|---|---|---|---|
| Opening drive | 112 | +$7,240 | 61% | 1.62 | +$65 |
| Level fade | 86 | +$1,180 | 52% | 1.11 | +$14 |
| Failed breakout | 74 | −$2,960 | 41% | 0.78 | −$40 |
| Trend pullback | 9 | +$820 | 78% | 3.10 | +$91 |
Three things to read here, in order.
The bottom row is the trap. A 78% win rate and a profit factor of 3.10 look like the best idea on the page, and nine trades cannot support either number. Read the count before the rate, always. That row is an observation, not a finding.
The opening drive is carrying the book. Take it out and this trader is barely above water. That is worth knowing for its own sake — it says where the edge actually lives, and how much of the rest is activity.
The failed breakout is the question. Seventy-four trades and a profit factor under 1 is a large enough sample to take seriously.
Before you retire a setup, check it is the setup
A losing setup is often a sound idea taken in bad circumstances. Cut the same trades again before you decide:
- By time of day. If your failed breakouts are mostly taken in the last half hour, the finding may be about the hour rather than the pattern.
- By position size. If they are consistently taken at twice your median size, you are measuring your sizing, not your setup.
- By sequence. If they are mostly the third or fourth trade of a session, the setup may be what you reach for when the plan has already gone.
That last one is common and worth naming plainly: some setups are not strategies, they are what boredom or frustration looks like when it needs a justification. The numbers cannot tell you which yours is, but they can tell you where to look.
Splitting a journal five ways divides an already modest sample into five small ones. If every bucket is too thin to read, the answer is not more data — it is fewer named setups. Three you can measure beat six you cannot.
What Choptick shows you
Setup is a first-class cut of your journal, alongside twelve others, all computed from the same trades and net of commissions.
- Net, win rate, profit factor and net per trade for every setup — worst first, so the question you should be asking is at the top.
- Unnamed trades reported as unnamed — auto-logged trades carry no setup, and they are counted and left out rather than pooled into a bucket nobody chose.
- The same journal cut twelve other ways — hour, half hour, session, weekday, instrument, direction, hold time, trade of the day, losing runs, re-entry timing, after win/loss and account — so you can check whether a weak setup is the setup or the circumstances.
- The trade count on every row, because a rate over nine trades is not a rate.
- A setup on the manual trade form, and a Setup column read straight from a CSV import, so trades can be named as they arrive.
Figures in the examples are invented and illustrative. Nothing here is financial advice.