Setup Miss and Execution Error: Why the Split Matters

Two trades go wrong on the same day for opposite reasons. In the first, a qualified setup printed and the trader sat on their hands, then watched the move run without them. In the second, a qualified setup printed, the trader took it, and entered late at twice the planned size. Both cost money. Both feel like the same vague failure at the end of the session. But one is an error of leaving a valid trade on the table, and the other is an error of handling a valid trade badly, and they need different fixes. This post defines a setup miss and an execution error, shows why they have to be measured in different places, and lays out how to log each so a losing stretch points at the right cause.

By Imperial Analytics

What a setup miss is

A setup miss starts with a valid signal. The conditions the trader defined in advance lined up, the entry rule was satisfied, and the trade qualified to be taken. The trader then did not take it, whether from hesitation, distraction, a flinch after a prior loss, or a rule they overrode in the moment. The setup was real; the action was absent.

What makes this category slippery is that it produces no record. A trade that is never placed has no fill, no P&L, and no row in the journal. The trader's log shows the trades they took, which means it shows a tidy and incomplete version of their decisions. The missed setups, which can carry more cost than any single bad fill, sit entirely outside the data unless the trader deliberately writes them down. The hardest cost to manage is the one that never appears on the page.

What an execution error is

An execution error also starts with a valid setup, but here the trade gets placed. The fault is in how it was carried out. The trader entered three ticks late and shrank the trade's reward before it began, or doubled the planned size and turned a normal loss into an outsized one, or widened the stop mid-trade and let a small loss grow, or closed early and cut the move short. The signal was sound. The hands were not.

This category is dangerous in the opposite way to a setup miss. The execution error is fully present in the log, dressed as an ordinary trade with a real entry, a real exit, and a real result. Its P&L gets pooled in with cleanly executed trades, and the average that comes out blends two different things: how good the strategy is and how well the trader ran it. A losing trade that was executed perfectly and a losing trade that was botched look identical in a column of P&L. Only a tag that marks the execution separates them.

Why the two are measured in different places

The two mistakes occupy different parts of the record, and that is the whole reason they need separate handling. A setup miss is defined by absence. To measure it, the trader has to build a small register of setups that qualified and were skipped, because nothing in the normal trade log will ever surface them. An execution error is defined by presence. To measure it, the trader tags a trade that already exists in the log with a marker that says the setup was valid but the handling was not.

This is why a single P&L column cannot carry both. The column has rows only for trades that were placed, so it can never contain a miss. And it shows the result of each placed trade without saying whether the result came from the strategy or from the execution, so it cannot, on its own, isolate an error of commission. Measuring the two means working in two places at once: a register for what was skipped, and a tag for what was mishandled. This extends the same logic as separating plan trades from improvised trades, which adds a second axis to the trades a journal already holds.

How conflating them corrupts an edge estimate

A strategy's edge is a statement about its qualified setups, taken and executed as designed. The moment the measurement mixes in setups that were skipped or trades that were mishandled, it stops describing the strategy at all. Consider a qualified setup taken across a sample. Suppose the trader skipped eight of those setups and botched the handling of five more. The eight misses never enter the count, so any winners among them are credited to nobody, and the strategy looks weaker than it is. The five botched trades enter the count as losses or shrunken wins, so the strategy is charged for mistakes it did not make.

Cleaned up, the picture inverts. Measured on qualified setups that were actually taken and executed to plan, the strategy's hit rate and average result describe the edge itself. The skipped setups move to a separate count that measures the trader's willingness to pull the trigger. The botched trades move to a third count that measures execution quality. The same raw session now answers three questions instead of muddying one, which is the difference between a behavioral claim a trader can trust and a number that quietly mixes three causes together.

How to log each one

The missed-setup register is the harder of the two because it asks the trader to record a non-event. The discipline is to note, at the moment a qualified setup is recognized and passed up, a single row: the date, the setup name, and a brief note on what happened next. It does not need a P&L, because the trade never existed. It needs only enough to count how often valid setups are being left on the table and, over a sample, whether those skipped setups would have worked. Without this register the misses stay invisible, the same way an untracked habit of taking unqualified trades stays invisible until overtrading is defined precisely enough to measure.

The execution-quality tag attaches to trades that are already in the log. Each placed trade carries one extra field: executed to plan, or not, plus a short reason code when the answer is no. The reason codes stay small and fixed, so they can be counted: late entry, oversized, stop moved, exited early. A trade tagged not-to-plan is still a real trade with a real result, but the tag lets the trader pull those trades out of the edge calculation and into a separate execution-quality count. The point is not to relabel every loss as an error, but to mark only the trades where the handling, not the strategy, broke the rule.

What the split tells you to fix

When the missed-setup register grows, the strategy is not the problem; the trader is not acting on it. The repair lives in the pre-trade process: a clearer entry rule, a smaller size that lowers the fear of pulling the trigger, or a circuit breaker that keeps a prior loss from freezing the next valid setup. Reworking the strategy itself would be a misdiagnosis, because the strategy was never given the chance to fail or succeed on those setups.

When the execution-error tag grows, the opposite is true. The setups are being taken, so the trigger is not the issue; the handling is. The repair lives at the point of execution: entering at the planned price instead of chasing, sizing to the plan instead of the mood, leaving the stop where it was placed, and exiting on the rule rather than the feeling. A trader who reads a flat week as a strategy failure, when the real cause is a cluster of botched executions, will tear up an edge that was working. Separating the two counts is what keeps the diagnosis honest, and it sits one level above the trade-by-trade tagging that feeds it by telling the trader which repair the numbers are actually asking for.

Frequently asked questions

• q: What is the difference between a setup miss and an execution error? a: A setup miss is a qualified setup that was never taken, an error of omission that leaves no trade in the log. An execution error is a qualified setup that was taken but handled incorrectly, an error of commission that appears in the log as a real trade with a misleading result.
• q: Why does a setup miss not show up in my trade journal? a: A trade that is never placed has no fill, no P&L, and no row to record. Unless the trader keeps a separate register of qualified setups they skipped, missed setups stay entirely outside the data, which is what makes them easy to ignore and expensive to leave unmeasured.
• q: How do I tag an execution error without relabeling every loss? a: Mark a trade as not-to-plan only when the handling broke a rule, such as a late entry, oversized position, a stop moved off its plan, or an early exit. A loss on a setup that was taken and executed correctly is not an execution error; it is the strategy doing what strategies sometimes do.
• q: Why does mixing the two corrupt my strategy's numbers? a: Skipped setups drag the apparent hit rate down with trades the strategy never got credit for, and botched trades drag the average result down with losses the strategy did not cause. Pooled into one P&L column, they describe neither the strategy nor the execution cleanly.
• q: What do I actually fix when each count rises? a: A rising missed-setup count is a trigger-pulling problem fixed in the pre-trade process and in managing hesitation. A rising execution-error count is a handling problem fixed at order entry and discipline. The split tells you which repair the data is asking for.
• q: How many trades do I need before these counts mean anything? a: Per the Imperial Analytics sample-size discipline, a per-setup hit rate, expectancy, or error rate is held back as a claim until that setup has at least twenty cleanly executed matching trades, and it is shown with its sample size attached once it clears that floor.