Valuation methods are the formal techniques for converting facts about a business — its cash flows, earnings, assets, and the prices of comparable firms — into an estimate of what one share, or the whole enterprise, is worth. They exist because price and value are not the same thing: the market quotes a price every second, but value must be reasoned to. This section maps the methods fundamental analysts use to do that reasoning. The unifying tension across all of them is that every method trades one kind of error for another — discounted-cash-flow models demand forecasts no one can know with confidence; relative multiples avoid the forecasting but inherit whatever mispricing is embedded in the comparison group; asset-based methods rest on a balance sheet that records cost, not worth. There is no method that is simply "correct"; mastery is knowing which tool fits which business and which question, and how each one fails.

The two families (and a third approach)

Aswath Damodaran's standard taxonomy splits valuation into two broad families, with a third, narrower approach for option-like assets:

• Intrinsic (absolute) valuation ties value to the asset's own capacity to generate cash and the risk in those cash flows. The flagship is discounted cash flow (DCF): value = present value of expected future cash flows, discounted at a rate reflecting their risk. The dividend discount model, residual income/EVA, and earnings power value are all intrinsic variants that differ in which cash stream they discount and how they handle growth.
• Relative (market) valuation estimates value by pricing an asset against the market prices of comparable assets, standardized by a common variable — earnings (P/E), cash flow (EV/EBITDA), book value (P/B), or sales (EV/Sales). Comparable-company analysis and precedent-transaction analysis live here. Relative valuation dominates real-world practice: Damodaran's survey notes that the large majority of equity research reports and acquisition appraisals are built on multiples, not DCFs.
• A third, asset/cost-based approach values the firm from its balance sheet (assets minus liabilities), and a fourth, contingent-claim (real-options) approach prices option-like assets — distressed equity, undeveloped reserves, patents — with option-pricing models. These matter in specific situations rather than as everyday tools.

A crucial honesty point that runs through the whole section: these approaches routinely give different answers for the same company at the same time, and that is expected, not a bug. As Damodaran puts it, all valuations are biased — the only questions are how much and in which direction — and complex models do not beat simple ones; understanding falls as the number of inputs rises. "Garbage in, garbage out" applies with full force, especially to relative valuation.

When valuation matters — and when it doesn't

Valuation is decisive on a multi-year, business-ownership horizon: deciding whether a long-term holding is cheap or expensive, underwriting an acquisition, setting a margin of safety, or stress-testing a growth narrative against what the price already assumes. It is far less useful on short horizons. None of these methods is a timing tool — a stock can stay mispriced relative to any of them for years, and on a swing-trading horizon (days to weeks) valuation typically functions only as slow-moving context (is the name fundamentally sound, or a value trap?), not as an entry or exit signal. This boundary matters for any downstream consumer that conflates "cheap" with "about to go up."

Map of the sub-topics

This section's children go deep on each method — point to them, do not re-derive them here:

• Discounted Cash Flow (DCF) — the intrinsic flagship, with its own sub-tree: projecting free cash flows, estimating the discount rate (WACC), terminal-value methods, sensitivity & scenario analysis, and common DCF pitfalls. Terminal value commonly dominates total DCF value (often cited at ~60–80%), which is why the sub-nodes on terminal value and sensitivity exist.
• Reverse DCF — inverts the exercise: takes the market price as given and solves for the growth/margins the price implies, turning valuation into a falsifiable question ("what must I believe?").
• Comparable Company Analysis (Multiples) — relative valuation from trading multiples of peers, with sub-nodes on selecting the peer set, choosing the right multiple, and normalizing for comparability — the three judgments that make or break a comp.
• Precedent Transaction Analysis — relative valuation from prices paid in past M&A deals; embeds a control premium and synergies, so it measures acquisition value, usually the highest of the methods.
• Dividend Discount Model (DDM) — the purest intrinsic model: a share is worth the present value of its future dividends. Theoretically clean, but brutally sensitive to growth and required-return inputs and irrelevant to non-payers.
• Residual Income / EVA — charges earnings for the cost of equity (RI) or total capital (EVA); a firm creates value only when it earns above its cost of capital. Theoretically powerful, input-fragile.
• Sum-of-the-Parts (SOTP) — values a diversified firm segment-by-segment with the method best suited to each, then sums; the gap to market cap is the "conglomerate discount."
• Asset-Based Valuation — prices the firm from its balance sheet (going-concern or liquidation premise); the craft is in adjusting recorded book values to economic reality.
• Earnings Power Value (EPV) — Greenwald's no-growth intrinsic method: capitalizes normalized current earnings, deliberately stripping out the speculative growth forecast; the EPV-vs-asset-value spread is a direct test for a moat.

Adoption, debate & evidence

Relative valuation is the most-used family in practice; intrinsic DCF is the academic and textbook standard and the backbone of corporate-finance teaching. The genuine, well-documented debates are not about whether the arithmetic works but about reliability: every intrinsic model is hostage to its discount-rate and terminal assumptions, and small input changes swing the answer materially (a ~1-percentage-point change in WACC commonly shifts DCF value by roughly 10–20%, a widely cited rule of thumb, not a constant). Relative valuation is faster and market-anchored but propagates any sector-wide mispricing and can be quietly gamed through peer selection. The honest stance, shared across Damodaran and the Wall Street training literature, is that valuation is a disciplined framework for forming a differentiated view, not a mechanical edge — there is no peer-reviewed evidence that any single valuation method, run as a stand-alone signal, generates excess returns. (This is distinct from the academic value factor — the cross-sectional tendency of cheap-on-book-or-earnings stocks to outperform, e.g. Fama-French HML — which is a portfolio-sorting phenomenon, not an endorsement of any one DCF or multiples model.)

Sources

• Aswath Damodaran — Valuation Approaches and Metrics: A Survey of the Theory and Evidence (intrinsic vs. relative vs. contingent-claim taxonomy; dominance of relative valuation in practice): https://pages.stern.nyu.edu/~adamodar/pdfiles/papers/valuesurvey.pdf
• Aswath Damodaran — An Introduction to Valuation (lecture notes; three approaches, "time and place for each"): https://pages.stern.nyu.edu/~adamodar/pdfiles/eqnotes/ValIntroSpr24.pdf
• Aswath Damodaran — The Dark Side of Valuation: Bias, Uncertainty and Complexity and Bias in Valuation (all valuations are biased; simpler models beat complex; garbage-in-garbage-out): https://pages.stern.nyu.edu/~adamodar/pdfiles/country/ValuationBermudaTriangleIndia.pdf ; https://aswathdamodaran.blogspot.com/2009/01/bias-in-valuation.html
• Sibling child nodes in this section (DCF, Reverse DCF, Comparables, Precedent Transactions, DDM, Residual Income/EVA, SOTP, Asset-Based, EPV) — for method-level mechanics and per-method evidence.

Flagged dispute: Sensitivity magnitudes (WACC ±1pp ≈ ±10–20%; terminal value ≈ 60–80% of DCF value) are commonly cited heuristics that vary by company duration and growth profile, not universal constants. No academic evidence establishes any single valuation method as a stand-alone return-generating signal; valuation's value is as a framework for a differentiated view.