Confusing profit with edge is the first problem for a reason.

Two different objects

One is a property of the method. The other is a property of a sample.

Edge is a characteristic of a trading method: the expected value it produces per unit of risk, net of costs, over the population of trades it will generate. Profit is a characteristic of one finite, path-dependent sample drawn from that population. They are different kinds of object, and the relationship between them is statistical rather than definitional — which is precisely why a method with edge can produce a losing quarter and a method with none can produce a winning one without either outcome being surprising.

Why the confusion survives

Profit is measured continuously and edge is not measured at all.

Every brokerage platform displays profit permanently, in currency, updating in real time. Almost none displays expectancy. A trader is therefore surrounded by one number and would have to construct the other deliberately, from records they may not keep, using a formula nobody handed them. Under those conditions treating profit as the scoreboard is not sloppy thinking — it is the only reading available. The confusion persists because the infrastructure of retail trading makes it the default, not because traders have failed to notice a distinction. It also survives because profit is socially legible in a way expectancy is not — an account balance is a number other traders understand immediately, and R-denominated expectancy requires an explanation before it can be discussed.

What it makes impossible

Every downstream decision inherits the error.

Because so much depends on the answer, an unstable reading here propagates into every other decision the operator makes, and each inherited error looks like an independent mistake:

• Sizing — a profitable sequence reads as a stronger edge, so risk rises.
• Retirement — a losing stretch reads as a broken method, so it is abandoned.
• Attribution — the profitable branch reads as the good one regardless of R.
• Validation — promotion decisions get made on the least stable evidence.

The separation

MARS reports both and never lets one stand in for the other.

The system's answer is not to demote profit — profit is the point — but to refuse to let it answer questions it cannot answer. Expectancy is computed by branch, in R, net of friction, with the sample size displayed beside it, and it is the figure that governs sizing and retirement decisions. Profit is reported as the outcome it is. When the two disagree, the disagreement is treated as information about sample position rather than as a contradiction requiring one of them to be wrong. Displaying both side by side has a secondary effect worth noting: over a few months the operator develops an intuition for how far apart the two can legitimately drift, which is itself a form of statistical education no explanation delivers as well.

The tell

Ask a trader what their edge is and listen for the units.

There is a quick diagnostic. Asked to describe their edge, a trader who answers in currency or percentage return is describing a sample; one who answers in R per trade with a trade count attached is describing a method. The distinction sounds pedantic and is not — the second answer can be compared across time, across instruments, and against a simulation, while the first cannot be compared to anything except itself. The units a trader instinctively reaches for reveal which object they have actually been tracking.

Why it heads the list

It is the only problem that corrupts the diagnosis of the others.

The remaining five problems can be diagnosed and repaired somewhat independently. This one cannot be left in place while the others are worked on, because it corrupts the evidence used to assess the repairs. A trader who fixes their exposure accounting and then evaluates the fix by whether the next month was profitable has learned nothing, and may well reverse a correct change on the strength of a losing sample. Establishing an honest expectancy reading is not the first item because it is the most damaging. It is first because everything after it is unmeasurable until it exists.