How my analyses are made

This page explains how my stock analyses, my Bitcoin analyses and the Alien Analyzer work: where the numbers come from, what is fact and what is my assessment, which assumptions sit behind a fair value, and where I have interests of my own. Every analysis links to this page.

What an analysis is and what it is not

• An analysis is my personal assessment of a company or of Bitcoin, based on publicly available data.
• It is not investment advice. I do not know your financial situation, your goals or your risk tolerance, and I do not take them into account. No analysis says that a stock or Bitcoin suits you.
• Analyses are not updated continuously. The date stated in the analysis applies. Prices, metrics and my assessment may change afterwards.
• Stocks and Bitcoin can fluctuate sharply. Losses up to a total loss are possible.

Facts, estimates, opinion

Every analysis contains three kinds of statements. I keep them apart:

Values I cannot back up with a source are marked as an estimate or as not available, instead of inventing a number.

Sources

• Primary sources: annual reports, annual and quarterly filings and other mandatory disclosures (for US companies via SEC EDGAR), the companies' investor relations pages, official statistics.
• Metrics and price history: my own Alien Analyzer V2, which processes data from Yahoo Finance. This data can contain errors, for example from one-off accounting effects or financial data in a different currency than the share price. I check conspicuous values against the annual reports and say so in the text when doubts remain.
• Research: research tools, including AI-assisted ones, help with collecting and sorting. The primary sources are what counts: numbers and factual statements are checked against them before publication.
• Blog posts and forums are not a source for numbers.

How I value

Fair value

The fair value in my analyses is a model calculation, not a price target. It is based on the fair value indicator of the Alien Analyzer:

The assumptions behind it:

• The market will value the company in the future roughly as it did on average in the past.
• Current earnings are representative of the company's earning power.

Both can be wrong: for cyclicals at an earnings peak or trough, after one-off effects, when growth breaks or when the interest rate environment changes. That is why I check the indicator in every analysis against the P/E history of recent years and put it into context in the text instead of taking it over unchecked. When an analysis names a price scenario, its assumptions are stated in the text.

Metrics and quality check

For context I use historical valuation multiples (P/E, P/CF, P/S, P/B), the dividend history and a quality check covering return on equity, debt, free cash flow, equity ratio, earnings growth, dividend yield and the change in share count. The traffic-light colours in the analyzer follow fixed rules: mostly fixed thresholds, for P/E a comparison with the stock's own average and the sector P/E, and special rules for some sectors. For banks, insurers, REITs, commodity companies and other special cases, some metrics do not fit. The analysis says so.

Bitcoin analyses

Bitcoin is not a stock. There are no earnings and no P/E ratio. My Bitcoin analyses are based on the price history, on the power law model and on publicly available market data.

• Trend: price in USD = 1.0117 × 10^-17 × (days since 3 January 2009)^5.82 (power law model after Giovanni Santostasi, in a widely used calibration; newer estimates give a slightly lower exponent).
• Support: 0.42 × trend (floor from the same calibration). Between support and trend lies the corridor in which the price has often moved so far.
• The model describes the past. No mechanism guarantees that it will keep holding. Price scenarios from the model are estimates.

I hold Bitcoin. Every Bitcoin analysis says so as well.

Conflicts of interest

• Every analysis has a dated disclosure block: whether I or persons closely associated with me hold shares of the company, and whether I have a business relationship with it.
• I receive no money from the companies I analyse. The only exception is Amazon: I receive royalties for my books through Kindle Direct Publishing. The Amazon analysis says so.
• My income: books (Amazon and direct sales), affiliate links (marked with *), donations and services. The YouTube channel is not monetised.
• Requested analyses: A reader pays for an analysis of a stock of his choice. He only chooses the company. He has no influence on content or result, and his personal situation plays no role. A paid analysis is marked as such in the disclosure.

Date and timeliness

The top of every analysis states the data date. Prices and metrics refer to that point in time. The disclosure block states the date and time of publication, after a full revision also the time of that revision, plus the date of the disclosure itself.

Alien Analyzer

The Alien Analyzer V2 calculates automatically for every ticker you enter. It knows neither you nor your goals.

• Data: Yahoo Finance, fetched through my own server. Your browser does not query any third-party data services.
• Fair value indicator: formula as above. If the price is more than 15% above or below fair value, the analyzer shows "above" or "below fair value", otherwise "fairly valued". For comparison, the P/E fan shows earnings per share times 10, 15, 20, 25 and 30. For REITs and mining and metals companies it shows no fair value, because the P/E ratio is of little use there.
• Alien assessment: rule-based text modules depending on sector, special case and metrics. They describe where the metrics stand against their history and what matters in the sector. They are not a recommendation to buy or sell and not a review of the individual case.
• The data can be wrong or incomplete. The analyzer shows a notice for known data problems, but it does not catch every error.