Mark Douglas: Separating Trade Outcomes From Process

A planned exit at 48.00 executes exactly as intended after an entry at 50.00. The trade loses money, and the trader immediately decides the method is broken. Mark Douglas’s work on risk and probability helps separate that losing result from a different question: did the decision process perform as specified?

Douglas was a trading author and coach. The publisher’s account of Trading in the Zone identifies risk, randomness and probability as central themes and dates The Disciplined Trader to 1990. The publisher lists Trading in the Zone in 2001.

The following examples are hypothetical exercises in evaluating decisions. They aren’t Douglas’s trading record, clinical advice or evidence that mindset alone produces an edge.

Mark Douglas and two columns in the journal

One journal column can record money gained or lost. Another can record whether the action matched the documented plan. The opening example belongs in the losing-result column and, assuming the plan was followed, the compliant-process column.

Now imagine an unplanned position that gains 4.00 per share. Its positive result doesn’t establish that the deviation was sound. The journal can record both the gain and the departure without forcing them into a single judgement of success.

I keep the 48.00 exit separate from the emotional label “bad trade”. The level gives the review a concrete event. The label may conflate a poor outcome, a poor rule, an execution mistake and the normal uncertainty of a process.

The market structure guide helps describe the chart conditions behind a plan. Precise chart language matters because an ambiguous original plan makes later compliance impossible to assess. “Exit when it looks weak” leaves much more room for reinterpretation than an observable condition.

A positive average can include frequent losses

Suppose one hundred hypothetical trades contain forty winners averaging 2R and sixty losers averaging 1R. Gross results total 80R minus 60R, or 20R. Average gross expectancy is 0.20R per trade, despite losses occurring more often than gains.

Here R is a fixed unit of initial planned risk. It standardises the example without making any claim about actual returns. Costs of 0.10R per trade would reduce the average to 0.10R, and larger-than-planned losses could reduce it further.

A trader who expects every qualifying signal to succeed is expecting something inconsistent with this hypothetical distribution. But the distribution itself still needs evidence. Writing down a favourable win rate and payoff ratio doesn’t show that a real method possesses them.

The Ed Seykota profile offers related reading on consistency and risk. A clear process allows the reader to distinguish following a method from establishing that the method deserves to be followed.

The next outcome isn’t owed to the last one

Under a deliberately simplified model with independent trades and a 60% loss probability, five specified consecutive trades all losing has probability 0.60 to the fifth power, or 7.776%. That calculation concerns one specified sequence of five trades.

It isn’t the probability of seeing at least one five-loss run somewhere in a much longer series. There are many possible starting points in the longer series, and those runs can overlap. Confusing those questions can make an ordinary sequence seem more exceptional than it is.

Real trades may also violate the independence assumption. Signals triggered in the same market environment can share risks. Five losses could be partly related to that environment, rather than five unrelated draws.

The useful lesson is limited but practical. An earlier loss doesn’t create an entitlement to a gain, and a winning streak doesn’t remove uncertainty from the next position. A research review still needs to ask whether the method’s assumptions remain credible under the conditions being observed.

Risk acceptance needs a specific amount

Consider a hypothetical position of 100 shares entered at 50.00 with an intended exit at 48.00. The planned price loss is 200 before costs. Increasing the position to 500 shares changes that amount to 1,000 while leaving the chart and exit level unchanged.

I would put 200 and 1,000 on the same review page. A statement such as “I accept the risk” means different things at those two amounts. If the planned loss is intolerable before entry, a verbal commitment to discipline doesn’t resolve the mismatch.

The planned amount also needs an execution caveat. A gap to 45.00 would create a 500 loss on the 100-share position before costs if that were the available fill. The 48.00 instruction cannot guarantee that the market offers that price.

The Ralph Vince profile provides a related sizing perspective. Position size can affect the practical difficulty of following a process, but a smaller position doesn’t transform a method with negative expectancy into a profitable one.

An execution mistake can remain costly after a recovery

Suppose the planned exit is 48.00, but the trader delays until 46.00. If price later recovers to 55.00, the delay may begin to look reasonable in hindsight. Yet the decision at 48.00 was made without knowing that recovery would occur.

A useful review preserves what was known at the time and the reason for the departure. Was there a documented exception? Did new information arrive? Or did the trader simply hope the loss would reverse?

The eventual recovery doesn’t answer those questions. A profitable deviation may suggest a research idea, but testing that idea requires applying the proposed exception to other cases, including those that continued falling.

The same standard applies to early profit-taking. Closing a position before the planned exit may sometimes improve the result. It may also remove the larger winners that sustain the method’s average. A journal can investigate the difference without assuming every deviation is either wisdom or failure.

Consistent execution can faithfully lose money

Imagine a method winning half its trades at an average of 0.8R and losing the other half at an average of 1R. Its gross average is 0.40R minus 0.50R, or minus 0.10R per trade. Perfect adherence reproduces the negative arithmetic.

This is a concrete limit to psychological explanations of trading results. Confidence, calmness and faithful execution cannot establish that the underlying rules have favourable economics. Data, costs, selection effects and changing conditions still deserve examination.

A review can therefore distinguish two tasks. One asks whether the method is sufficiently supported. The other asks whether actual behaviour matches that method. Both may need work, and neither can be diagnosed from the trader’s feelings alone.

When the research record is weak, encouraging greater confidence can make the problem worse by increasing commitment to an untested process. The relevant uncertainty needs to be named before an exercise in consistency has a clear purpose.

A review schedule limits retrospective rule changes

Suppose a hypothetical study records every qualifying signal for a fixed month and reviews deviations at the end of each week. That schedule creates a place for learning while preserving a record of the rules used on each date. It doesn’t require ignoring an immediate operational problem.

If data are wrong or an order was entered incorrectly, the issue can be recorded and addressed as such. If the latest trade merely lost, the review can compare it with the method’s expected range of outcomes before deciding that a rule change is justified.

The Ulcer Index discussion adds a way to examine the account’s path below prior peaks. That is a descriptive measurement, not a diagnosis of the trader’s mindset. A difficult drawdown can arise even when rules are followed.

A record of rule versions, dates and exceptions prevents the latest version from being projected backward onto earlier decisions. It also allows the trader to acknowledge uncertainty without rewriting every loss into an avoidable mistake.

A concrete process score can help expose this distinction. In a hypothetical ten-trade review, eight positions follow their recorded entry and exit conditions, while two involve undocumented changes. The adherence rate is 80%, regardless of whether the account gained or lost money. The two exceptions can then be examined individually. One might involve a practical execution problem; the other might follow an emotional response to a previous loss. Combining them under a single label such as poor discipline would hide the different remedies each requires.

The score also has a limit: a poorly specified plan can make compliance look deceptively high. “Use judgement” isn’t an observable condition against which a later reviewer can reliably assess behaviour.

Books for studying Mark Douglas

Trading in the Zone is linked in hardcover. Its probability and risk themes make it relevant to the distinction between the 48.00 planned exit and the later 46.00 deviation. The examples here are independent. The Disciplined Trader is Douglas’s earlier 1990 work.

As an Amazon Associate, I earn from qualifying purchases.

Trading in the Zone. Hardcover.

(US) Trading in the Zone
(UK) Trading in the Zone
(DE) Trading in the Zone

The 48.00 exit deserves an accurate description

A loss taken according to a tested plan is still a loss. It can also be evidence that execution matched the intended process. Keeping both observations visible makes the review more honest and leaves room to examine the method without treating every adverse outcome as a personal verdict.

Learn the pattern. Ride the trend. Keep the gains.

Educational content only. Not investment advice. Trading involves risk. You are responsible for your decisions.