If you ask ChatGPT "how to value a stock", you'll get an answer that mentions 10+ methods and recommends nothing in particular. In practice, professional value investors use four, and only four - because the other six produce results that are either wildly imprecise or wildly precise about the wrong thing.
This post walks through the four methods that actually get used, when each one applies, what their pitfalls are, and the worked example on Apple to show how they triangulate to a defensible price range.
If you want the shortcut: every public stock on invest-like has all four valuations computed and displayed on its /buffett/[ticker]/ page. The numbers below match what you'll see if you open Apple's page.
Method 1: PEG (P/E to Growth)
What it does: Adjusts the P/E ratio by the EPS growth rate. A stock with P/E = 30 and EPS growth = 30% has the same "value" as a stock with P/E = 10 and EPS growth = 10%.
Formula: PEG = P/E ÷ EPS growth rate (expressed as percentage points, not decimal)
Peter Lynch's version (the canonical use): PEG ≤ 1.0 = attractive. PEG between 1.0 and 1.5 = fair. PEG > 1.5 = overpaying relative to growth.
Worked example on Apple:
- Current P/E ratio: ~33
- 5-year EPS growth rate: ~10%
- PEG = 33 / 10 = 3.3
Apple fails Lynch's PEG screen badly because Apple's growth has matured. Lynch would not have bought Apple in 2026 at current price.
When to use PEG: best for growth-quality businesses where EPS growth is genuinely sustainable. Useless for cyclicals (their EPS is volatile and PEG flips wildly). Useless for businesses near zero earnings (PEG becomes meaningless).
Pitfall: PEG uses historical EPS growth. If growth is decelerating, PEG looks attractive on past numbers but bad on forward numbers. Always cross-check with forward EPS estimates.
Method 2: Discounted Cash Flow (DCF)
What it does: Projects the company's free cash flow over the next 10 years, applies a terminal value at year 10, then discounts everything back to present using a discount rate.
Formula (simplified):
Intrinsic Value = Σ (FCF_year_n / (1 + r)^n) + Terminal Value / (1 + r)^10
Where r = discount rate (typically 8-12% for stable businesses)
Terminal Value = FCF_year_10 × (1 + g) / (r − g) (where g = perpetual growth, ~3%)
Worked example on Apple:
Apple's current FCF: $108.8B. Assume:
- FCF grows 7% per year for the next 10 years
- Discount rate: 8% (low because Apple is investment-grade)
- Terminal growth rate: 3%
Year 1-10 FCF: $116B → $214B (growing 7%/yr)
Sum of present-values: $1,166B
Terminal value at year 10: $214B × 1.03 / (0.08 - 0.03) = $4,408B
Present value of terminal: $4,408B / 1.08^10 = $2,042B
Total intrinsic value: ~$3,208B
Apple's current market cap is ~$3,400B. The DCF says Apple is fair-valued to mildly overvalued, depending on which assumptions you choose.
When to use DCF: best for predictable, mature businesses. Toll-bridges, payment networks, mature consumer staples. The math is clean when growth is steady.
Pitfall (the famous one): DCF is hyper-sensitive to the terminal growth rate and discount rate. Change Apple's terminal growth from 3% to 4% and the value jumps 30%+. Change the discount rate from 8% to 9% and the value falls 15%+. DCF is precise but not accurate; use it for sanity-checking, not as the source of truth.
Method 3: Owner-earnings yield (Buffett's preferred)
What it does: Computes the cash the business actually generates per dollar of market cap. Higher yield = cheaper.
Formula: Owner-Earnings Yield = Owner Earnings / Market Cap
Where Owner Earnings = Net Income + D&A − Maintenance Capex (full walkthrough in our owner-earnings explainer)
Worked example on Apple:
- Owner earnings (computed in the earlier post): ~$99.7B
- Market cap: ~$3,400B
- Owner-earnings yield = $99.7B / $3,400B = 2.93%
Buffett's published thresholds:
- ≥ 7-8% absolute = clearly cheap
- 5-7% = fair price for a wonderful business
- 3-5% = stretched, requires extreme business quality to justify
- < 3% = expensive regardless of business quality
Apple at 2.93% falls below the 5% floor. The Buffett scorer marks Apple's valuation pillar as Partial Fit accordingly.
The 10-year Treasury anchor: Buffett's 2025 letter explicitly invoked the 10-year Treasury yield as the discount-rate floor. Apple's owner-earnings yield of 2.93% is below the 10-year yield of ~4.5%, which is a flashing red flag for absolute valuation.
When to use owner-earnings yield: the cleanest single number for "am I overpaying". Especially useful for high-quality businesses where the DCF assumptions are too sensitive.
Pitfall: estimating maintenance capex is the judgment call. We use a 5-year average to smooth, and we publish the methodology so users can audit.
Method 4: Reverse-DCF (the most useful method most retail investors don't know)
What it does: Inverts DCF. Instead of computing intrinsic value from assumptions, you compute what assumptions are required to justify the current price. The result tells you whether the market's implied assumptions are realistic.
Worked example on Apple:
Apple's current market cap is $3,400B. Working backward:
If discount rate = 8% and terminal growth = 3%, then current price requires 10-year FCF CAGR of ~6.5% (math: solve for FCF growth that makes DCF equal current price).
So at the current price, the market is implicitly betting Apple's FCF grows 6.5% per year for the next decade. Question: is that realistic?
- Apple's last 5-year FCF CAGR: ~10%
- Apple's last 3-year FCF CAGR: ~4% (slower lately)
- Forward analyst consensus for FCF growth: ~5-7%
The market's 6.5% implied requirement is plausible but tight. If FCF growth disappoints by even 1-2 percentage points over the decade, Apple becomes meaningfully overvalued.
Reverse-DCF is the most useful method because it eliminates the "garbage in, garbage out" problem of forward DCF. Instead of guessing at growth assumptions and producing a precise-but-wrong value, you work backward from the actual price and ask "is this implied future believable?"
How to use all four together - the triangulation method
The honest answer to "what is this stock worth": never one number, always a range.
For Apple, the four methods produce:
- PEG: severely overvalued (PEG 3.3 vs Lynch's 1.0 threshold)
- DCF (8%, 7% growth, 3% terminal): fair value
- Owner-earnings yield 2.93%: stretched
- Reverse-DCF: requires 6.5% FCF growth - plausible but tight
The triangulation: Apple is fair-to-stretched at current price. Not a screaming bargain. Not catastrophically overvalued either. The business is wonderful; the price is not cheap.
This is why our 5-pillar Buffett-Brain verdict on Apple scores Strong Fit on Moat/Durability/Management/Health but Partial Fit on Valuation. Three of four valuation methods agree the price is rich.
How invest-like uses these methods
Every stock page on invest-like surfaces all four computed valuations:
- The PEG appears in the Lynch framework section
- The DCF appears with the discount rate and growth assumptions clearly stated
- The Owner-earnings yield drives the Buffett scorer
- The Reverse-DCF appears as an "Implied growth required" line, contextualised against historical and forward consensus
The point of computing four: any single method can mislead. Four together triangulate. The synthesis is what makes the verdict robust.
Pitfalls that kill new investors
-
Anchoring on one method: DCF alone produces precise wrong answers. Owner-earnings yield alone misses growth runways. PEG alone breaks on cyclicals. Use multiple.
-
Ignoring quality entirely: a 20% PEG-passing stock with a deteriorating moat is a value trap. Quality (pillars 1-3 in the Buffett 5-pillar framework) is co-equal with valuation.
-
Ignoring price entirely: a wonderful business at any price is also wrong. The whole point of the value-investing framework is that price is half the math.
-
Confusing DCF precision with accuracy: the result has 4 decimal places, but the inputs are guesses. Always sanity-check with reverse-DCF.
-
Using stale numbers: P/E from January is not P/E in May after a 30% rally. The numbers must be current.
Disclosure
Educational tool. The four valuation methods described are standard practice in value investing. Applying them to Apple in this post is for illustration; the numbers are current as of fiscal-year 2024 fundamentals and may move quarter-to-quarter. Past valuation does not predict future returns.
Author: Zaid Ghazal, founder of invest-like, Kiel, Germany. Not a registered investment adviser. The DCF and reverse-DCF methodologies are documented in detail at /methodology/.