Valuation basics

Valuation starts from an uncomfortable truth: a share of stock is worth the cash it will deliver to its owner over its lifetime, discounted for time and risk — and nobody knows what that is. Every valuation tool is a way of approximating that unknowable number or of sidestepping the need to know it. Multiples, yields, anchors, and reverse-engineering are the four approaches this site uses, and each is best understood as a different compression of the same underlying question: what are you paying, and what are you getting?

A multiple divides price by a measure of what the business produces. P/E is market cap over net income: pay 20 times earnings and, if earnings never changed and were fully paid out, you would recoup the price in 20 years. P/FCF replaces earnings with free cash flow. P/S uses revenue — meaningful only in light of what margins that revenue can eventually carry. P/B compares price to accounting net worth, mostly useful for banks and asset-heavy companies whose book values approximate reality. Enterprise-value multiples like OCF yield on EV go one step further and price the whole firm, debt included: EV is market cap plus debt minus cash — what it would truly cost to own the business outright — and it makes companies with different balance sheets comparable, which market-cap multiples quietly fail to do.

Every multiple has an inverse, and the inverse is often more intuitive: a P/E of 20 is an earnings yield of 5%; a P/FCF of 25 is an FCF yield of 4%. Yields put stocks in the same units as every other asset. A 10-year Treasury paying 4% is the risk-free alternative; a stock offering a 5% FCF yield pays you one point more, plus whatever growth materializes, minus whatever goes wrong. That spread is the whole equity bargain in one line. When earnings yields on stocks fall below bond yields, you are betting entirely on growth; when they sit well above, you are being paid to wait. This comparison — equity yield versus bond yield — is the oldest sanity check in investing and still among the best.

Why do we lean on cash flow rather than earnings? Net income is an opinion — a careful, audited opinion, but an opinion: it depends on depreciation schedules, revenue-recognition timing, reserve estimates, and one-time items. Cash is a fact. Free cash flow — operating cash flow minus capital expenditure — is the money genuinely available to owners after the business has paid its bills and maintained itself. Companies can report growing profits for years while cash stagnates (the gap usually hiding in receivables, inventory, or aggressive accruals), and that divergence is one of the most reliable early warnings in accounting. When earnings-based and cash-based multiples disagree, believe the cash. One honest caveat: because stock-based compensation is added back in operating cash flow, FCF flatters heavy issuers of stock — which is why we track SBC as a share of FCF alongside it.

Anchors are deliberately crude rules that resist the sophistication trap. Ours is 10 times operating cash flow: a company priced at or below 10x OCF is generating cash at roughly a 10% rate against the whole purchase price — pre-capex — a level at which durable businesses have historically been scarce and worth examining one by one. We display this as a ratio: market cap divided by 10x OCF, where 1.0 or below means the anchor is met. The anchor's blind spots are the price of its simplicity — it ignores capex, so capital-hungry businesses look cheaper than they are, and it ignores debt, which the EV-based OCF yield corrects. Its value is that it cannot be argued into saying whatever you want, which is more than can be said for a spreadsheet of forecasts.

Multiples also mean nothing without a reference point, and there are two honest ones. The first is the company's own history: a business that traded between 15 and 25 times FCF for a decade and now sits at 12 is being repriced — the market believes something changed, and your job is to decide whether it is right. This is reversion logic, and it works exactly when the business itself has not deteriorated, which is the judgment the multiple cannot make for you. The second is peers: comparing a railroad's multiple to a software company's is meaningless, but comparing it to other railroads isolates what you are actually paying for. Cross-sector multiple comparisons are the most common valuation error beginners make.

High multiples are not automatically wrong — they are claims about the future. A stock at 35 times FCF growing that FCF at 20% a year is cheaper, correctly measured, than a stock at 12 times with shrinking cash flow: within a few years the fast grower's cash flow overtakes the price difference. The discipline is to make the claim explicit rather than vibes-based — which growth rate, for how long, justifies this price? That is precisely the question the reverse DCF answers, and it is covered in the expectations-investing guide.

A workable routine, using this site's numbers: start with FCF yield and the 10x-OCF ratio to place the company on the cheap-to-expensive spectrum. Check OCF yield on EV so leverage cannot hide. Compare the current multiples to the company's own range and its sector's medians on the screener. Then ask what the price assumes — the implied-growth figure — and whether the company's record makes that assumption modest or heroic. No single number decides anything; the pattern across all of them usually speaks clearly.