Week 18 — Trading Plan, Backtesting & Trading as a Business

A complete trading process is a written decision framework that turns market reading into actions you can explain before a trade and review after it. The framework is not complete merely because you can identify a setup. It becomes operational when you can state what market conditions you require, what evidence makes a trade valid, what would invalidate the idea, how much capital is exposed, how the trade will be managed, what makes you wait, and how the decision will later be reviewed.

Week 18 closes The Market Reading Edge by putting the full course into that operating form. The goal is not to create a promise of profit or a perfect set of rules. The goal is to build a process that is defined clearly enough to test, execute, record and improve with evidence.

What Week 18 completes

The earlier weeks built the parts of the decision. You learned to read the market story, identify control and market type, separate primary movement from secondary movement, judge pullbacks, read participation and timing, use lower-timeframe evidence, define invalidation, control capital at risk, adapt to account constraints and review your own execution behaviour. Week 18 joins those parts into one repeatable operating cycle.

That changes the question from “Do I understand this setup?” to “Can I describe the complete decision process, apply it without rewriting it around one outcome, and produce evidence that shows how the process behaves?” A trader who cannot answer that second question does not yet have a testable trading process.

What a complete trading plan should contain

A trading plan should describe the decisions that must be made before, during and after a trade. It should be specific enough that you can later compare what you actually did with what the plan required.

• Market story and context: What is the higher-timeframe story? Who appears to have control? What market type are you dealing with?
• Movement structure: What is the primary move and what is the secondary move? Is the pullback behaving in a way that still supports the original story?
• Participation and timing: Is the market active enough for the idea to make sense now, or is the better decision to wait?
• Setup conditions: What must be present before the idea becomes a candidate trade?
• Evidence and confirmation: What lower-timeframe or other course-defined evidence supports execution?
• Invalidation: What market behaviour would prove that the trade idea is no longer valid?
• Risk decision: How will capital at risk be determined before execution, and are other open exposures or account constraints relevant?
• Execution and management: What conditions allow the order, what conditions keep you out, and how will an open position be managed according to the plan rather than emotion?
• Exit logic: What market or plan condition ends the trade?
• Review process: What evidence will be recorded so that the decision can be evaluated later?

The plan should also define when WAIT or no trade is the correct decision. A process that explains only how to enter but never explains when not to trade is incomplete.

The dedicated trading plan guide owns the deeper mechanics of plan construction. Here, the important point is how the plan connects the entire course into one decision chain.

A trading plan is a decision framework, not a prediction

A written plan does not tell the market what must happen. It tells you what you will do if particular conditions appear.

That distinction matters because market reading is conditional. A higher-timeframe story may be coherent and still fail to produce a valid trade. A setup may begin to form and then lose the evidence that supported it. The account may have enough nominal capital but not enough usable risk capacity for the trade. The correct output can therefore be execution, waiting, or rejection.

A useful plan is written in conditional language: if the story remains valid and the required evidence appears, then the next decision is allowed; if the invalidation condition appears first, the idea is rejected; if important evidence is missing, the trader waits. This keeps the plan connected to observable market behaviour instead of certainty.

Separate strategy rules from business controls

The trading strategy and the trading business are connected, but they are not the same layer.

Strategy rules describe what makes an opportunity valid: market context, setup logic, evidence, invalidation and execution conditions. Business controls describe how that opportunity is allowed to affect capital and how the operation is monitored: acceptable risk, total exposure, account constraints, record keeping, review routines and the process for approving rule changes.

This separation protects the method from pressure. A profit target, a recent loss or an evaluation deadline should not turn a weak setup into a valid setup. Business constraints may reduce size, reduce available exposure or prevent a trade, but they do not improve the market evidence.

Backtesting before risking real capital

Backtesting asks how a defined set of rules would have behaved across relevant historical market conditions. It is a form of evidence gathering, not a certificate that future trades will be profitable.

Before a historical test has meaning, the trader must know what is being tested. If the definition of the setup changes every time an old chart produces an inconvenient result, the test stops measuring one process. It becomes a collection of explanations written after the outcome is already known.

For this course, historical testing should preserve the same decision sequence used in live analysis: market story, control, market type, movement structure, pullback validity, participation, lower-timeframe evidence, invalidation, risk decision and review. The purpose is to see whether the process can be identified consistently, where it becomes ambiguous and under what kinds of conditions its assumptions appear stronger or weaker.

The dedicated backtesting guide owns the deeper historical-testing mechanics. Week 18 does not impose a universal number of trades, a target win rate or an optimization threshold because those numbers are not defined by the course source authority.

What backtesting is meant to answer

A useful historical test should produce questions and evidence such as:

• Can the setup rules be recognized consistently without using the final outcome to redefine them?
• Which market environments repeatedly support the process, and which environments create conflict or ambiguity?
• Are the entry conditions and invalidation conditions clear enough that another review of the same chart would reach a similar classification?
• Does the risk logic remain coherent when losing trades occur in sequence?
• Are there recurring execution assumptions that look easy on a completed chart but would have been unclear in real time?
• Which observations deserve further testing rather than an immediate rule change?

The objective is not to force historical data to validate the strategy. The objective is to learn how the defined process behaves and where it needs clearer evidence.

Forward testing under current conditions

Forward testing takes the defined process into unfolding market conditions without giving the trader the benefit of seeing the future candles first. It helps answer whether the rules that looked clear in historical review can actually be recognized and followed in real time.

Historical and forward evidence solve different problems. Historical review gives access to more past situations and helps expose patterns. Forward testing exposes recognition, timing, execution and discipline problems that can disappear when a chart is already complete.

The important rule is to carry the same defined process forward. If the trader quietly changes the entry logic after a missed trade, ignores invalidation after a loss or adds a new confirmation only when the old rules fail, the forward test is no longer testing the original process.

The dedicated forward testing guide owns the detailed mechanics of testing in current conditions, while the strategy validation guide explains how different evidence layers contribute to confidence. Neither historical nor forward testing guarantees future performance.

One result does not validate or invalidate a process

A good decision can lose and a poor decision can win. That is why the result of one trade cannot by itself tell you whether the process was sound.

Imagine two trades. In the first, the market story, setup, evidence, invalidation and risk all match the written plan, but the market later invalidates the idea and the trade loses. In the second, the trader ignores the plan, enters late because of fear of missing out, and the trade happens to win. Judging only by money would label the second trade as better. Judging the process shows the opposite.

The first trade may be a valid execution with an unfavourable outcome. The second may be an invalid execution with a favourable outcome. Over time, the journal and testing process must preserve that distinction or the trader will accidentally reward rule-breaking.

Journal process and evidence quality

A trading journal turns memory into evidence. It should preserve enough information to reconstruct the decision without relying on how the trader feels about the result later.

For each reviewed opportunity, useful evidence includes:

• the market story and relevant higher-timeframe context;
• control, market type and primary versus secondary movement;
• the setup and the evidence that made it valid or invalid;
• the planned invalidation and capital-at-risk decision;
• whether the decision was trade, wait or reject;
• what was actually executed and any deviation from the plan;
• how the position was managed and why it was closed;
• the outcome, recorded without allowing the outcome to rewrite the original reasoning;
• post-trade observations that may deserve later review.

Week 17 adds another useful layer: observable behaviour. If the trader felt FOMO, revenge pressure, overconfidence or the urge to overtrade, the journal should record the behaviour and the action it produced rather than merely attaching an emotional label.

The detailed design of journal fields and review routines belongs to the trading journal guide. Week 18 uses the journal as the evidence record for the complete operating process.

Review performance in layers

Review should begin with process quality before isolated profit and loss. A useful order is:

1. Plan adherence: Did the trader follow the written decision rules?
2. Setup and environment: Which valid setups occurred in which market conditions?
3. Risk and exposure: Was capital at risk controlled as planned, including correlated or account-level exposure?
4. Execution behaviour: Were there repeated deviations such as chasing, revenge trading, hesitation or unnecessary activity?
5. Outcome evidence: What do results show when grouped by the conditions and behaviours that produced them?

This order prevents a single winning trade from hiding poor execution and prevents a single losing trade from triggering an unnecessary strategy rewrite. Detailed performance metrics remain the responsibility of the trading performance guide; Week 18 focuses on how metrics fit into the review sequence.

Change rules only with evidence

Rule changes should follow a controlled path rather than a reaction to the latest trade.

1. Observe: Record the issue without immediately changing the plan.
2. Find the pattern: Determine whether the issue appears repeatedly under a definable condition or was simply one outcome.
3. State a hypothesis: Explain what specific rule or assumption may be causing the problem and why.
4. Make one controlled modification: Define the change clearly enough that the old and new versions can be distinguished.
5. Retest: Gather historical and/or forward evidence for the modified version.
6. Record the version: Keep the evidence trail so later results are not mixed across different rule sets.

This creates a simple discipline: observe → identify a repeated pattern → form a hypothesis → change deliberately → retest → document. The course does not prescribe a universal sample size or performance threshold for approving a change. What matters is that the decision is evidence-based and that the test remains tied to a defined version of the process.

Valid case: the process is changed for a documented reason

A trader reviews a body of trades and notices that one clearly defined market condition repeatedly creates the same conflict with a setup rule. The trader does not edit the rule during the review. Instead, the trader documents the pattern, states the proposed change, keeps all other rules stable and retests the revised version before treating it as the new process.

This is a valid improvement cycle because the change can be explained, compared and reviewed.

Invalid case: the latest result keeps rewriting the plan

A trader takes a loss and immediately moves the preferred entry location. After missing the next move, the trader loosens the setup requirement. A winning trade then leads to more risk on the following opportunity. Later, the trader changes timeframe because another setup looked clearer there.

Even if some of those trades win, the trader can no longer tell which strategy produced the results. The problem is not merely emotional. It is a loss of evidence integrity: the operating process keeps changing before it can be tested.

Difficult case: historical and forward evidence disagree

Sometimes a process looks coherent in historical review but becomes much harder to execute in current conditions. The correct response is not to force the forward results to match the backtest, and it is not automatically to declare that the strategy has stopped working.

Return to the evidence. Were historical decisions influenced by hindsight? Are the same market conditions actually being compared? Are execution assumptions different? Is the setup definition too vague to recognize consistently? Did the market environment change in a way that affects the original assumptions?

Document the disagreement, tighten the question and gather more relevant evidence. A difficult result is information about the process. It is not permission to invent a conclusion.

Treating trading as a business process

Treating trading as a business does not mean demanding a profit from the market every day. It means operating with defined decisions, known risk, records, review routines and controlled changes.

A business-process mindset asks:

• What process are we operating?
• What conditions allow us to take risk?
• What evidence shows whether we followed the process?
• Where are errors or repeated weaknesses occurring?
• What should remain unchanged until more evidence exists?
• What specific change should be tested next?

This is different from trading to satisfy a daily target or trying to recover a loss immediately. The market does not owe the business a trade. A professional operating decision can be to do nothing when the evidence is incomplete or the risk constraints are not satisfied.

The complete operating loop

The full course can now be expressed as one operating loop:

1. Read: Build the higher-timeframe story, identify control, classify the market and understand primary versus secondary movement.
2. Qualify: Judge pullback validity, participation, timing and whether the environment is ON, WAIT or OFF for the idea.
3. Confirm: Use the required lower-timeframe or course-defined evidence without letting one signal replace the wider story.
4. Invalidate: State what market behaviour would prove the idea wrong before execution.
5. Risk: Define capital at risk and account-level exposure before placing the trade, including any relevant prop-firm constraints.
6. Execute: Trade only when both the market process and the operating constraints allow it. Otherwise wait or reject the opportunity.
7. Manage: Follow the pre-decided management and exit logic rather than changing it because of fear, hope or a target.
8. Record: Preserve the reasoning, execution, behaviour and outcome in the journal.
9. Review: Separate process quality from individual outcomes and look for repeated evidence.
10. Test and refine: Change the documented process only when a specific evidence-based hypothesis is ready to be retested.

The loop ends where the next decision begins. Review improves the quality of the next reading; it does not create certainty.

Week 18 capstone exercise

Your final course exercise is to produce one operating package that another version of you could follow without relying on memory.

Part 1: Write the one-page decision framework

Summarize the sequence from market story to review. State what must be true before an idea is valid, what evidence permits execution, what invalidates it, how risk is decided, and what creates a WAIT or no-trade decision.

Part 2: Define the test protocol

Explain how you will review the same rules historically and then under current conditions. State what evidence you will preserve and how you will prevent hindsight or outcome-fitting from changing the setup definition during the test.

Part 3: Build the journal evidence record

Create fields that capture market context, setup logic, invalidation, risk, the actual decision, execution deviations, outcome and post-trade observations. Include an observable-behaviour field so Week 17 discipline can be reviewed as part of process adherence.

Part 4: Define the review questions

• Did I follow the plan?
• Was the market environment correctly classified from the evidence available at the time?
• Was the setup genuinely valid or did I lower the standard because of pressure?
• Was invalidation known before execution?
• Was capital at risk consistent with the plan and current account constraints?
• Did management follow the planned logic?
• Which observation is repeated enough to deserve deeper testing?

Part 5: Create a rule-change log

For every proposed change, record the old rule, the observed problem, the evidence supporting the hypothesis, the new rule being tested, and the testing status. Do not mix results from different rule versions without identifying the version.

Graduation checklist

By the end of Week 18, you should be able to answer yes to the following:

• I can explain the higher-timeframe market story before talking about an entry.
• I can identify control, market type, primary movement and secondary movement in the context of that story.
• I can explain why a pullback is valid, invalid or still uncertain.
• I can distinguish useful participation and timing evidence from activity that does not support the wider story.
• I can state the lower-timeframe evidence I require without allowing it to replace higher-timeframe context.
• I can define invalidation before execution.
• I know the capital at risk before the trade and can account for overlapping exposure or external account constraints.
• I accept WAIT and no trade as valid outputs when the evidence is incomplete.
• I can identify behaviour that causes me to violate the process and review the action it produced.
• I have a written trading plan that can be tested without changing its meaning after each result.
• I can separate a good decision from a favourable outcome.
• I can record enough evidence to review what happened without rewriting the original reasoning.
• I can propose a rule change as a testable hypothesis rather than an emotional reaction.
• I understand that historical and forward evidence can inform a process but cannot guarantee future performance.

From course knowledge to operating discipline

The final skill is not finding one more entry technique. It is being able to read the market, explain the evidence, define what would make you wrong, control the risk, execute only when the conditions are present and review the decision honestly afterward.

When that sequence is written, tested and recorded, trading becomes a process that can be examined rather than a collection of isolated memories. Backtesting gives historical evidence. Forward testing exposes the process to current conditions. The journal preserves what actually happened. Review separates decisions from outcomes. Controlled revisions allow the process to evolve without losing its identity.

That is the operating standard at the end of The Market Reading Edge: read → qualify → confirm → invalidate → risk → execute or wait → manage → record → review → test → refine.

Educational note: This course teaches a structured decision and review process. Trading involves risk, and historical or simulated results do not guarantee future performance.