Owner track 🧭 · six sessions, no code
For anyone who wants to think about a business the way an owner does. What a share is, why value is a range, the margin of safety, the edge of what you know, Mr Market, and the mistakes made in your own head. The running case is Halden Tiles, a constructed family business; every Halden figure is invented and labelled so.
You are buying a business
A share as a slice of the owners' future cash. Profit versus what is left after the spend to stand still. Value as cash that has not arrived yet, and the four questions Buffett asked in 1977 before any number.
📐 Owner session 2Value is a range, not a number
Three honest views of one business, the input that widens the answer most, and why nearly half the value sits beyond year ten. Halden's range: 8.2 to 15.8 million.
🌉 Owner session 3The margin of safety
The gap between value and price, measured against value. Buffett's bridge for heavier trucks, a margin that is positive against one view and negative against another, and how big a gap to ask for.
⭕ Owner session 4The edge of what you know
The circle of competence, where its boundary is, how to test it before it tests you, and the too-hard tray that costs nothing to fill.
🎭 Owner session 5Mr Market quotes, you decide
Graham's moody partner as Buffett told it in 1987: a price that swings while the business barely moves, and why a quote is there to serve you, not to guide you.
🧠 Owner session 6The mistakes made in your own head
Anchoring, forecasts that are too narrow, loss aversion, and Buffett's own list of twenty-five years of mistakes, from cigar butts to the institutional imperative.
Analyst track 🔬 · ten sessions, browser and Python
For the people who build the number and have to defend it. Owner earnings from tagged 10-K facts, normalising through a cycle, return on capital, a discounted cash flow by hand and in reverse, sensitivity, base rates of growth, a moat as a valuation input, and the memo that says what would prove it wrong. Four filers throughout: Caterpillar, Exxon Mobil, Johnson & Johnson and Procter & Gamble.
Owner earnings from the filings
Tagged facts, the sieve, net income plus D&A minus capex per share, and the three places the approximation is wrong. Johnson & Johnson's owner earnings beat its profit; Exxon's sat far below for five years.
🔄 Analyst session 2Normalising through a cycle
Five- and ten-year averages, telling a cycle from a one-off, and Exxon's 2020 loss next to its 2022 record. Where you start decides what you see.
🏭 Analyst session 3Return on invested capital
Returns against the cost of capital, why growth only adds value above it, and growth as reinvestment times return.
🧮 Analyst session 4A discounted cash flow, by hand
Two stages, a discount rate you choose, a terminal value, and why 57 percent of the value sits beyond year ten at the default inputs.
🔁 Analyst session 5What the price already assumes
The reverse discounted cash flow: a price you type turned into the growth it needs. Double the value needs 14.17 percent a year.
🎚️ Analyst session 6Which input moves the answer
Sensitivity grids, the range from one point of discount rate and two of growth, and the input you must argue hardest.
📉 Analyst session 7How often fast growth lasts
Base rates from the Base Rate Book: forecasts too narrow by half, earnings growth that does not persist, and the growth fade.
🏰 Analyst session 8How long excess returns last
A moat as a valuation input: how many years returns stay above the cost of capital, and returns that halve their lead in ten years.
📝 Analyst session 9A memo that says what would prove it wrong
The investment memo as a test you set yourself: the thesis, the range, the kill criteria, and the disputed evidence on value as a strategy.
🎛️ Analyst session 10 · bench nightThe intrinsic value bench
Four filers, five rules checked against the filings, the anti-lever that misses by 19,614.8 percent, the break button, and a margin of safety against a price you type.
Your passport 🎫
Clear all three quiz questions in a session and it gets stamped. Sixteen stamps means you can describe a business as its owner, write its value as a range, say what a price assumes, and name the input that would prove you wrong. Stamps live in this browser only.
What this is not, so you land in the right place 🧭
| If your question is | Go here instead |
|---|---|
| How do I read an income statement, a cash flow statement and a balance sheet, or why cash and profit differ? | Finance: Reading the Numbers ↗ |
| Why do some businesses keep their customers and their returns: network effects, switching costs, moats as economics? | Economics, Games and moats ↗ |
| Why is money later worth less than money now? | Economics, Money, inflation and interest ↗ |
| What should I buy, when should I sell, how should I build a portfolio? | Nowhere on these pages. Educational, not investment advice: no tips, no prices, no tickers. |
| I want to reason about what a business is worth, say it as a range, and know what would prove me wrong. | you are in the right place |
Educational, not investment advice. A reasoning discipline taught on one constructed business and four real companies' filings. Whether cheap-looking stocks beat dear ones as a strategy is genuinely disputed in the research, and the pages teach that disagreement rather than taking a side.
What is honest here 🧾
The Analyst spine is real public filings, recomputed for these pages and recomputed again on the bench. Where a figure comes from a letter, a book or a paper, the page names it and says whether it was read at source; where nothing citable exists, the page says that instead.
| Claim | Status |
|---|---|
| Every filer figure and every bench number: owner earnings, the ladder, values, ranges, implied growth | Really computed from SEC EDGAR XBRL company facts for Caterpillar, Exxon Mobil, Johnson & Johnson and Procter & Gamble; the SEC refused the build machine, so the records were read from a public mirror and spot-checked against the SEC's own data; the bench recomputes every number in your browser and a separate Python reference agrees with it |
| Owner earnings | An approximation, labelled everywhere: net income plus depreciation and amortisation minus all capital spending, because filings do not separate the spend to stand still from the spend to grow |
| Values, ranges, margins of safety, implied growth | Modelled under four inputs you choose; no value is a recommendation |
| Buffett's letters, Damodaran's terminal-value rule, the Base Rate Book | Read at source for these pages, named on each page that uses them |
| Graham's books, Expectations Investing, the behavioural papers | Reported, paraphrased, never quoted |
| Whether value investing beats the market as a strategy | Not claimed. The research genuinely disagrees, and the pages teach the disagreement |
| Halden Tiles, its cousin, its prices | Constructed, and the page says so each time |
The knowledge map 🧠
Both tracks at a glance - hover a session to spotlight its concepts, click any node to jump in.