Method
How I research.
I research by starting with the question, then the data, then the framework. Each note opens with what the market price already assumes, using a reverse DCF, before I state what I think.
Question and framework
The reverse DCF sets the starting point for the analysis: what expectations are embedded in the current price. I then assess the evidence against those expectations and build a framework for the call, keeping sourced facts, assumptions and opinions separate.
Sources and estimates
Quarterly financials come from the company’s filings on SEC EDGAR: 8-K, 10-Q and 10-K. I do not use summaries in place of those filings. I use a sell-side target only when I can cross-check it in at least two independent places, and I attribute industry data, such as memory pricing, to the firm that published it. Macro data comes from primary releases by the BEA, BLS and Federal Reserve. If I cannot trace a number to a primary source, I label it as an estimate in the text rather than state it as fact. The sources behind the site are listed on the data sources page.
Ratings and valuation
Every rating has a target and a kill condition: the specific number that would prove the call wrong, written down before the result is known. I value cyclical businesses on mid-cycle earnings, not the current quarter, and build a scenario tree with probabilities. The target is therefore a probability-weighted value, not a single guess.
Position discipline
I publish position limits, pre-committed trim levels and exits with the call. Setting these rules in advance makes the decision before the price moves.
Public record
When the deciding number prints, the following note scores the call against the line set earlier. I name misses in the note itself, and they stay on the record; superseded ratings remain visible in the rating history. I date corrections and leave them in place. The coverage table is generated from the notes, so it cannot show a rating that the research does not.
Models
The valuation calculators and the DCF, comps and LBO templates behind the notes are free in the toolkit. The models are open for readers to review and use.
The track record page shows how calls have been graded, and the coverage table lists ratings generated from the research notes.
I am a publisher of research, not a registered investment adviser, and nothing on this site is investment advice. See the disclaimer.