Volume Eight · The Fine Print

The Analyst’s Toolkit

Fourteen concepts from a real research screen, each explained in plain words with its own simulation: multiples, PEG, forward earnings, capex J-curves, sum of the parts, concentration, drawdowns, and the opinion machines.

In this series: 1 Foundations · 2 What Moves Prices · 3 Practice Room · 4 Reading a Company · 5 The Dangerous Aisle · 6 Your Portfolio · 7 Case Study · 8 The Toolkit · 9 Crowd Psychology · 10 Crashes · 11 Forecasting · Cheat Sheet

01 · Why These Words Exist

Fourteen tools from one research screen

Volume 7 read a real stock and met fourteen concepts the earlier volumes never needed. This volume gives each one its own full explanation: what it means, a worked example, and where it misleads. Together they turn a broker’s research tab from a wall of numbers into a story you can read.

02 · Multiple Compression and Re-rating

When the crowd changes what it pays per dollar

A stock price is two numbers multiplied: earnings × the multiple (the P/E). Earnings belong to the business. The multiple belongs to the crowd’s mood. Either one can move the price.

Multiple compression (a de-rating) is the crowd deciding to pay less per dollar of earnings, say 34× shrinking to 23×. It can drag a stock down 20% in a year in which earnings grew 23%, exactly the Microsoft case. The reverse, paying more per dollar, is multiple expansion or a re-rating, and it powered much of the great bull markets.

Twelve quarters. Earnings grind up, the multiple drifts. Watch the price.

+23%
23
Practical note. When a stock moves, always ask which engine did it: the earnings or the multiple. Earnings-driven moves rest on facts. Multiple-driven moves rest on mood, and mood mean-reverts. The most dangerous purchases in history were great businesses bought at peak multiples.
03 · The PEG Ratio

The napkin that compares price to growth

PEG = P/E divided by the growth rate. A P/E of 23 with 23% growth is a PEG of 1.0. Peter Lynch popularized the shorthand: around 1, price and growth roughly match; far above, you are paying for growth that has not happened yet.

Slide the two ingredients, read the verdict

23
23%
Where it misleads. PEG assumes the growth continues, treats 10% growth at a bank the same as 10% at a startup, and ignores debt and quality. A PEG of 0.5 on collapsing growth estimates is a trap, not a bargain. It is a first glance, never a verdict.
04 · Trailing vs Forward P/E

Two clocks on the same stock

Trailing P/E divides price by the last twelve months of actual earnings (you will see the tag TTM, trailing twelve months). Forward P/E divides by next year’s estimated earnings. Facts versus forecasts.

Same stock, two P/Es

$100
$5.0
20%
Practical note. Growing companies always look cheaper on forward P/E, because the denominator is a hope. When comparing stocks, use the same clock for both, and remember that estimates get cut exactly when you most need them to be right.
05 · Capex and Free Cash Flow

The construction site and the J-curve

Capex (capital expenditure) is money spent building the future: factories, servers, data centers. It leaves as cash today and returns as revenue over years. Free cash flow is what survives: operating cash minus capex, the money a business truly throws off.

A heavy build creates a J-curve: free cash flow dives during construction, then, if the bet works, recovers above where it started. Markets hate the valley and cannot see the far slope, which is exactly the argument that knocked 20% off Microsoft. Slide the build intensity and watch the valley deepen and the payoff rise.

Ten years. Heavy building for four of them. Payoff assumed to arrive two years behind each dollar spent.

$20B
The honest catch. This simulation assumes the buildings earn their keep. Real capex sometimes never pays: empty factories, obsolete networks. That uncertainty, not the spending itself, is what markets price. Watch the quarter free cash flow stops shrinking; that is the market’s tell that the valley has bottomed.
06 · Price / Book

Price versus the accounting net worth

Book value is what accounting says a company is worth: assets minus liabilities. Price/book compares the market price to that number. At 7.68, Microsoft traded at nearly eight times its accounting net worth.

Where it works: banks, insurers, industrials

Their assets are financial or physical and marked near reality. A solid bank at 0.8× book is trading below its own stated net worth, a genuinely meaningful signal.

Where it breaks: software and brands

Microsoft’s real assets, its code, engineers, ecosystem, and brand, barely appear on the balance sheet. A high P/B here mostly measures how much value accounting cannot see.

Worked example

Two companies both priced at $100/share. A bank with $110 of book value per share: P/B 0.9, priced below its stated worth. A software firm with $13 of book: P/B 7.7, priced almost entirely on invisible assets. Same price tag, completely different meaning.

Rule of thumb

Reach for P/B on asset-heavy businesses, and reach for earnings and cash flow measures on idea-heavy ones. Using P/B on a software company is measuring a poet by weight.

07 · Who Owns It, Who Opines On It

Institutional ownership and the three opinion machines

Around every stock sits an audience. Knowing who is in the seats, and how each critic scores, keeps their opinions in proportion.

Institutional ownership

The share held by funds, pensions, and institutions: 71% in the Microsoft case, typical for a giant. High means professionals have done homework, prices are efficient, and when they head for the exit together, the fall is fast. Their quarterly holdings are public in 13F filings on SEC EDGAR.

Analyst price targets and consensus

Human 12-month forecasts, averaged. In the case study the mean target sat about 45% above the price with dozens of buys and zero sells. Useful as a mood gauge, but targets notoriously chase prices: they get raised after rallies and cut after crashes.

Quant ratings

Machine scores like StarMine, built mostly from price momentum and whether analysts are raising or cutting estimates. After a brutal month they will read Bearish almost by construction. They describe the recent tape, not the decade ahead.

Social sentiment scores

A meter of chatter mood on social platforms. Microsoft’s read Neutral even mid-crash, meaning retail noise carried no signal either way. Extreme readings are more interesting as contrarian flags than as instructions.

The reconciliation, again. These sources disagreeing is normal, because they answer different questions on different clocks: value eventually (models), direction lately (quant), hope officially (targets), noise currently (sentiment). Your job is to know which question each one answered before letting it vote.
08 · Sum of the Parts

Valuing the divisions, then adding

Sum of the parts (SOTP) values each division as if it were its own company, adds the pieces plus net cash, and compares the total to the market price. It is exactly how that $558 Microsoft estimate was built: productivity software worth so much, cloud so much, consumer so much, plus cash.

Build your own estimate, per share, then compare to the market

$258
$250
$44
$390
Where it misleads. Every slider above is an assumption wearing a suit. Nudge each division 15% and “undervalued by 43%” becomes “roughly fair”. SOTP is honest about structure and silent about certainty; treat its output as a range, never a target.
09 · Concentration Risk

When one customer is the whole story

Concentration risk means too much revenue, backlog, or exposure hanging on one customer, product, supplier, or partner. In the case study, nearly half the company’s contracted future revenue traced to a single AI partner.

How much of the business rides on the biggest customer?

Practical note. Companies must disclose customers above 10% of revenue in their filings; search the 10-K for “concentration”. The risk is not that the big customer is bad, it is that their renegotiation leverage, their stumbles, and their strategy changes all become yours.
10 · Drawdown From Peak

A measure of pain, not of value

Drawdown is the distance below the highest point so far: “27% off its peak”. It describes how much it hurt to hold. It says nothing about whether the stock is now cheap, because the peak itself may have been madness.

A price path, its running peak, and the drawdown between them

The anchoring trap, one last time. “Down 50% from the high” described both a bargain-priced bank in 2009 and a still-absurd dot-com in 2001. Same drawdown, opposite meanings. Value lives in earnings and cash, never in the distance from a memory.
11 · Delayed vs Real-Time Quotes

You may be aiming where the price was

Free app quotes are often delayed, typically by about 15 minutes, unless you enable real-time data. Most brokers offer the upgrade as a free toggle buried in settings, often requiring a re-login, and many even display a warning banner about it. Checking that your quotes are real-time is a one-time task worth doing before your first order. For a long-term investor the lag is trivia. For anyone placing a trade today, it means the number on screen is history.

The dark line is the live price. The rust dot is what a delayed app shows.

Practical note. Enable real-time quotes before your first real order, and remember Volume 1’s rule: a limit order protects you from stale prices in a way a market order never can.
12 · Practical Notes and Further Reading

Where to go deeper

These fourteen tools are the standard vocabulary of valuation and market plumbing. The originals below cover them properly.

The Little Book of Valuation · Aswath Damodaran

Multiples, cash flows, and sum-of-the-parts by the professor who literally teaches Wall Street. His full NYU courses and spreadsheets are free online.

One Up on Wall Street · Peter Lynch

The home of the PEG shorthand and the six categories of stocks, told in plain stories.

Berkshire Hathaway letters · Warren Buffett

The best writing in existence on why cash flow, not reported profit, is what an owner actually receives.

Company filings · SEC EDGAR

The 10-K and 10-Q behind every number on a research tab, plus 13F filings showing exactly what institutions hold. All free at sec.gov.

The routine. Next time a research screen confronts you, run this page’s tools in order: which engine moved the price, what the PEG whispers, which clock the P/E uses, where the cash is going, whose opinions you are reading, and how concentrated the story is. Fourteen words, one habit.