How Stoic Edge Researches a Market Question

Market research starts with a question narrow enough to answer. “Is this stock good?” mixes the business, valuation, timing, and the reader's circumstances. “What changed in the company's reported margins, and what explains it?” gives the work a useful boundary.

Stoic Edge's research process separates reported facts, interpretation, and a possible trading decision. Each needs different evidence.

State the question and the time horizon

Write what you are investigating, the period it covers, and the decision it might inform. The quarterly business change and a five-minute entry are different subjects. Evidence for one does not automatically qualify the other.

Record the date of the work. The conclusion based on an earlier filing should not read like an update after new results have been released.

Begin with primary material

For a company question, use the filing, earnings release, or original presentation behind the claim. For a product feature, use the current documentation. For a price study, record the data source, symbol, session, adjustment policy, and coverage.

Secondary coverage can reveal useful questions and disagreements. Keep the original source attached where a precise number or claim matters. If access is limited to an abstract, preview, or summary, state that limit.

Separate the evidence from the explanation

Keep a short evidence ledger: the observation, its source and date, the interpretation, and what would weaken that interpretation. Watch for denominators, accounting changes, small samples, and time periods selected because they make a story look convincing.

For example, higher revenue alone does not establish stronger unit economics. The rising price after an announcement does not isolate the announcement as the cause. Describe the uncertainty instead of assigning a motive you cannot observe.

Look for a competing explanation

Ask which available evidence conflicts with the thesis. Consider changes in the broader market, sector conditions, one-time events, and measurement choices. You do not need a contrarian position on everything; you need to know what the preferred explanation leaves out.

Before testing a chart idea, write the classification and execution assumptions. Then retain rejected cases as well as the attractive examples. The later revision should preserve what changed and why.

Bring the research back to the decision

The output should state the current view, supporting evidence, important uncertainty, and conditions that would change it. The trading action still needs its own setup, risk, and room.

The conditional planning article shows how to move from a view to observable scenarios. The System guide describes the separate chart framework.

Keep the record open to correction

Date substantial updates and correct factual errors in the article itself. Do not refresh a date just to make an unchanged page appear new. If a result is hypothetical, identify it near the result. If there is no verified performance record, do not imply one through selected charts.

Our editorial standards explain attribution, commercial interests, and how to request a correction. This article describes the research method; it is not a claim that every question produces an actionable trade.