Relative vs Absolute Valuation: When Each One Lies to You

---

title: "Relative vs Absolute Valuation: When Each One Lies to You" description: "Absolute valuation says what a stock is worth. Relative valuation says what it's worth compared to peers. Neither is always right — here's when to trust which." publishedAt: "2026-03-30" updatedAt: "2026-03-30" keywords: ["relative valuation", "absolute valuation", "stock valuation methods", "dcf vs multiples"]

TL;DR. Absolute valuation (DCF, intrinsic value) and relative valuation (multiples versus peers) answer different questions. Absolute asks: what is this business worth on its own terms? Relative asks: is this business cheaper than other businesses that look like it? They are complements, not substitutes. Tessera Alpha screens on relative — ranking stocks by P/E discount versus sector median across thousands of names — and then uses absolute valuation, via an integrated DCF calculator, to validate the top candidates before any entry. Skipping either step has a failure mode. Here's the map of where each one breaks.

Absolute valuation in one paragraph

A discounted cash flow (DCF) asks what price a rational owner would pay today in exchange for the future free cash flows the business will produce. You project cash flows over some explicit window (usually 5–10 years), discount them back at a cost of capital, then append a terminal value that captures everything after the explicit window. Its strength is that it is grounded in the business itself, not in whatever the rest of the peer group happens to be trading at this quarter. Its weakness is that it is a forecast, and forecasts are wrong. The structural problem: in a typical 10-year DCF, roughly 70–80% of the computed value sits in the terminal value — the bucket representing year 11 onward, which nobody can observe, validate, or audit. You are stacking the majority of your answer on top of two inputs (terminal growth rate, cost of capital) that are effectively assumptions with error bars wider than the spread you are trying to measure.

Relative valuation in one paragraph

Relative valuation compares a stock's multiple — P/E, EV/EBITDA, P/B, P/S — against the multiples of its peers. "Peers" usually means the same sector, sometimes the same sub-industry. The strengths are that multiples are observable today (no forecast required), comparable across companies, and fast to compute at scale. The weakness is the embedded assumption: the peer group is correctly priced. If the whole sector is mispriced, a stock that looks cheap within that sector is still expensive in absolute terms. A 15x P/E in a sector trading at 22x looks like a discount; if the sector should trade at 12x, you've bought into a cheap slice of an overvalued pie.

When absolute wins

There are a few regimes where relative valuation is structurally unreliable and absolute is the better tool.

Entire sector mispriced. This is the textbook case. Tech in 1999 traded at multiples that in hindsight were unmoored from any plausible cash-flow trajectory. Financials in 2007 traded at single-digit P/Es that looked cheap until the invisible balance-sheet leverage became visible. In both cases, a relative screen would have said "this stock is cheap versus its peers" — because it was — while every peer was collectively overvalued (tech) or collectively understating its risk (financials). A disciplined DCF, with conservative assumptions, would have flagged the whole group.

Structural regime change. When the forward picture genuinely differs from the trailing picture — a pandemic, a rate regime shift, an energy shock, an AI-driven margin restructuring — trailing multiples aggregate noise. Peer comparisons are a way of saying "assume next year looks like last year, relatively." In regime changes that assumption breaks everywhere at once. A DCF, even a rough one, forces you to write down what you actually think the next five years will look like — which is more honest than pretending the peer average knows.

Idiosyncratic catalysts. Companies with known, unique cash-flow events — a large litigation settlement, a pending spin-off, a one-time tax benefit, a known contract ramp — don't have peers in any meaningful sense. The "peer" multiple prices a different asset. DCF is the only reasonable tool.

When relative wins

Within-sector mispricing. One SaaS company trades at 12x revenue while the peer median is 18x. A DCF on a growth-stage SaaS company is almost useless: the terminal value dominates, and the terminal value depends on margin assumptions a decade out that are speculative fiction. Relative is cleaner here — you are explicitly saying "given that the market is pricing this category at 18x, this name at 12x is an anomaly worth investigating," without pretending to know the 2035 free cash flow margin.

Forecast uncertainty too high for DCF. When five-year cash flows are genuinely a guess — early-stage companies, commodity-sensitive businesses, cyclicals deep in a cycle — a DCF is pseudo-precision. It looks rigorous because there are numbers, but the numbers are wet clay. A multiple at least exposes its assumption honestly: the market thinks this category is worth X today. You can argue with that; you cannot argue with your own made-up 2031 number in a spreadsheet.

Scale. You cannot run a disciplined DCF on 5,000 stocks every week. Forecast labor alone makes it uneconomic, and automated DCFs drift into garbage quickly because the assumptions aren't being thought about. Relative valuation is the only approach that scales to a full investable universe.

Why Tessera defaults to relative for screening

At-scale screening requires metrics that are (1) robust, (2) fast, and (3) observable. Relative is all three. Tessera's pipeline is deliberately two-stage:

1. Relative screen — rank the universe by P/E discount versus sector median, generating roughly 50 candidates per week. The detailed mechanics are covered in Relative P/E Sector Analysis.
2. Quality grade filter — the 24 quality factors (profitability, balance sheet health, growth consistency, earnings quality) typically cut those 50 candidates to ~25 that pass the quality floor.
3. Absolute validation — for the top 3–5 candidates ranked by combined score, Tessera's integrated DCF calculator produces an intrinsic-value range. If relative says "cheap" but DCF says "the terminal-growth assumption you need to justify this price is 6%," that's a flag, not a buy.

The DCF calculator is wired into the same quality-factor inputs that drive the screen, so the forecast starts from observed business fundamentals rather than from a blank cell.

The two-step discipline

Screen with relative — broad, fast, peer-aware. Validate with absolute — slow, deep, peer-independent.

Most retail investors skip step two and end up in value traps: stocks that were genuinely cheap versus peers because the peer group was rerating downward and this one was just ahead of the line.

Most fundamental-only investors skip step one and miss setups entirely. Running a DCF on a short list of "names I follow" is not the same as noticing that a name you don't yet follow has quietly become the cheapest thing in its sector. Without a systematic cross-sectional screen, the DCF-only investor's universe is whatever happens to be in their head — which is a subset biased by narrative, not by mispricing.

Common mistakes

Treating multiples as fundamentals. A P/E ratio is shorthand for "price relative to current earnings power." It is not a measure of anything intrinsic to the business. A company can have a 15 P/E because it's cheap, or because the market correctly expects earnings to fall 40%. The multiple does not tell you which.

Cross-sector multiple comparisons. A 20x P/E in consumer staples and a 20x P/E in semiconductors are not the same valuation. Different capital intensity, different growth profiles, different cyclicality, different cost of capital. This is why Tessera compares within sector, never across. More on this in Relative P/E Sector Analysis.

DCF terminal-value gymnastics. If moving your terminal growth rate from 2% to 3% changes your DCF result by 30%, you have not valued a business. You have written down an opinion about terminal growth and decorated it with five years of explicit cash-flow projections. Stress-test every DCF by asking what happens when the terminal assumptions shift by a plausible amount; if the conclusion flips, you don't have a conclusion.

Anchoring on historical multiples. "This company traded at 20x P/E for a decade; now it trades at 15x, so it's cheap." This argument implicitly assumes that whatever made 20x appropriate still holds. If growth has decelerated, if competitive position has weakened, if the industry structure has changed, the historical multiple is a distraction. The right question is: what multiple is appropriate for the business as it exists today? — which is back to either a forward DCF or a cross-sectional peer comparison.

Related reading

• Relative P/E Sector Analysis
• Quality Screening Factors
• Backtesting Pitfalls