AlphaStocks

Scores are algorithm-generated research tools, not investment recommendations.

Graham-Style Fair Value

A systematic intrinsic value calculation inspired by the father of value investing, estimating what a stock is actually worth based on its earnings and book value.

Axis:Value(15%)
Weight in axis:45%
Effective composite weight:6.75%

Score output: Strong Buy Zone / Attractive / Fair Value / Overvalued / Reject

What is the Graham Score?

The Graham Score on AlphaStocks is a data-driven fair value estimate inspired by Benjamin Graham's intrinsic value framework from The Intelligent Investor. It calculates what a stock is worth using the Graham Number formula -- sqrt(22.5 x EPS x BVPS) -- and compares the result to the current market price. When a stock trades significantly below its calculated fair value, it has a "margin of safety," meaning there is a buffer protecting the investor even if the estimate is somewhat wrong. AlphaStocks extends Graham's original formula by computing three scenarios (conservative, base, and optimistic), applying sector-specific adaptations for banks, REITs, utilities, and insurers, and attaching a confidence indicator so you know how reliable the estimate is. The Graham Score contributes 45% of the Value axis, which carries 15% of the composite score, giving it an effective weight of 6.75% in the final 0-10 rating.

How does the Graham Score work?

The core of the Graham model is the Graham Number, a simplified intrinsic value formula: Intrinsic Value = sqrt(22.5 x EPS x BVPS). The constant 22.5 is the product of Graham's maximum acceptable P/E ratio of 15 and maximum acceptable P/B ratio of 1.5. The geometric mean of these two boundaries produces a single fair value estimate that deliberately avoids projecting future growth and instead values the company based on what it has already earned and accumulated.

AlphaStocks calculates three scenarios rather than a single number. The conservative estimate uses lower-bound earnings and tighter multipliers, assuming recent earnings may be near a peak. The base estimate closely follows Graham's original formula using trailing twelve-month earnings and book value. The optimistic estimate uses normalized earnings and allows for moderate growth with higher but still disciplined multipliers. The range between conservative and optimistic is itself informative -- a narrow range suggests high confidence, while a wide range means the valuation is sensitive to assumptions.

Sector-specific adaptations ensure fair treatment across industries. Banks use book value-based valuation with tangible book value adjusted for loan-loss reserves. REITs use Funds From Operations (FFO) per share instead of traditional EPS. Utilities receive adjusted multipliers reflecting their regulated return environment. Insurers use adjusted book value accounting for investment portfolio quality and underwriting reserves.

A confidence indicator is computed based on data completeness, earnings stability, and sector fit. A high-confidence estimate with a narrow range is much more actionable than a low-confidence estimate with a wide range. The model then calculates the discount or premium versus the current market price, scores the stock based on the magnitude of the discount, and applies the confidence weighting.

What does Graham look for in a stock?

The Graham model evaluates stocks based on the relationship between market price and estimated intrinsic value. Here are the key criteria and what each status means:

Strong Buy Zone

Pass

Stock trades more than 30% below base fair value with reasonable confidence. Deep discount represents a significant margin of safety.

Fail

Stock trades above the base fair value estimate, offering no margin of safety by Graham's standards.

Attractive

Pass

Stock trades 10-30% below base fair value. Moderate discount offers some cushion against estimation error.

Fail

Stock trades within 10% of fair value or above it, not meeting the discount threshold.

Fair Value

Pass

Stock trades within 10% of the base fair value estimate. Approximately fairly priced by Graham metrics.

Fail

Stock trades at a significant premium or discount, falling into a different category.

Positive Earnings

Pass

The company has positive trailing twelve-month earnings, allowing the Graham Number to be computed.

Fail

Negative earnings make the formula impossible to compute. The stock receives an "Insufficient Data" status.

Reasonable Book Value

Pass

Book value per share is meaningful relative to the company's business model and sector.

Fail

Asset-light businesses (software, consulting, brands) have book values that significantly underestimate true economic value, producing artificially low fair value estimates.

How AlphaStocks uses Graham

The Graham-Style Fair Value model contributes 45% of the Value axis, which carries 15% of the composite score, giving it an effective weight of 6.75%. It is the highest-weighted component within the Value axis, reflecting the fundamental importance of intrinsic value assessment in any value-investing framework.

It pairs with Lynch-Style PEG Analysis (25% of Value), which specifically accounts for growth that Graham's formula ignores, and Greenblatt-Style Magic Formula (30% of Value), which evaluates earnings yield relative to return on capital. Together, these three models answer the Value question from different angles: Is the stock cheap relative to its earnings and assets? Is it cheap relative to its growth rate? Is it cheap relative to its return on capital?

A high-growth company that scores poorly on Graham-Style but well on Lynch-Style will still receive a balanced Value axis score. The Quality axis (Buffett and Piotroski) can also recognize that a growth company has strong fundamentals even if Graham considers it expensive. The composite score integrates all perspectives rather than letting any single model dominate.

Graham effective weight = 45% × 15% = 6.75% of the composite score

Current top stocks by Graham

The Graham-Style Fair Value model is evaluated for all 1,595 stocks in our coverage. You can filter the full rankings table to see which stocks currently score highest on this model and compare them across all five investment frameworks.

Frequently Asked Questions

What is the Graham Number and how is it calculated?

The Graham Number is a formula for estimating intrinsic value: sqrt(22.5 x EPS x BVPS). The constant 22.5 comes from multiplying Graham's maximum acceptable P/E ratio (15) by his maximum acceptable P/B ratio (1.5). It produces a single fair value estimate based on current earnings and book value, without projecting future growth.

Why does the Graham model show three scenarios instead of one number?

A single fair value estimate creates false precision. AlphaStocks calculates conservative (lower-bound earnings, tighter multipliers), base (standard Graham Number), and optimistic (normalized earnings with moderate growth) scenarios. The range between them indicates confidence -- a narrow range means high confidence, while a wide range signals sensitivity to assumptions.

Does the Graham Score work for technology and growth stocks?

The Graham model has a known blind spot for high-growth companies. It systematically undervalues them because it does not project future growth. A fast-growing software company might appear overvalued by 200% when its growth rate justifies the price. AlphaStocks addresses this by pairing Graham with Lynch-Style PEG analysis and Greenblatt-Style Magic Formula, ensuring growth stocks receive balanced composite scores.

How does AlphaStocks adapt the Graham formula for banks and REITs?

Banks use a book value-based valuation with tangible book value per share adjusted for loan-loss reserves, since bank assets are primarily financial instruments carried near market value. REITs use Funds From Operations (FFO) per share instead of traditional EPS, because REITs distribute most earnings as dividends, making standard EPS misleading. Utilities and insurers also have sector-specific adjustments.

What does "margin of safety" mean in the Graham Score?

Margin of safety is Graham's most important principle: never pay full price. If the estimated fair value is $100, Graham would wait until the price drops to $70-75 -- a 25-30% discount. This buffer protects you even if your estimate is wrong. AlphaStocks calculates the percentage difference between market price and base fair value, with larger discounts earning higher scores.

See all 1,595 stocks scored

Every stock evaluated by Graham-Style Fair Value and four other proven investment models. Updated with every SEC filing.

Explore the rankings

Explore Other Models

Model names reference the published investment methodologies of their respective authors. AlphaStocks is not affiliated with, endorsed by, or sponsored by Benjamin Grahamor any related entities. Our implementation is an interpretation of publicly described principles and may differ from the original author's approach.

AlphaStocks provides investment research and analysis, not investment advice. Past performance does not guarantee future results. Always do your own due diligence before making investment decisions. Data sourced from SEC EDGAR and Alpaca Markets.