How to Find Undervalued Stocks — 5 Proven Methods

A stock is undervalued when its market price is significantly below its intrinsic value— the true worth of the underlying business based on its earnings power, cash flows, assets, and growth prospects. The gap between market price and intrinsic value is what Benjamin Graham called the margin of safety.

The margin of safety is not a luxury — it is your protection against estimation errors. No one can calculate intrinsic value precisely. Your estimate of future earnings might be wrong. The economy might enter a recession. A competitor might disrupt the industry. By buying only when the price is well below even a conservative estimate of value, you build a buffer against the inevitable surprises.

Importantly, “undervalued” is not the same as “cheap.” A stock trading at 5x earnings can be overpriced if those earnings are about to collapse. A stock trading at 25x earnings can be undervalued if the business is growing fast enough to grow into that valuation within a few years. Value is always relative to the quality and trajectory of the underlying business.

Benjamin Graham, the father of value investing, developed a formula to estimate intrinsic value based on current earnings and expected growth. The modern adapted version is:

Where EPS is trailing twelve-month earnings per share, g is the expected 5-year earnings growth rate, and Y is the current AAA corporate bond yield.

The formula gives a “no-growth” P/E of 8.5 (what Graham considered fair for a company with zero growth) and adds 2 P/E points for each percentage point of expected growth. The bond yield adjustment accounts for interest rate environments — stocks should be worth more when bond yields are low (money is cheap) and less when yields are high.

Limitations. The formula works best for mature, stable businesses with predictable earnings. It is unreliable for companies with negative earnings, cyclical businesses (where trailing EPS may be at a peak or trough), and high-growth tech companies where the majority of value comes from earnings years into the future.

AlphaStocks automates the Graham Fair Value calculation for all 1,595 stocks, with sector-specific adaptations for banks, REITs, utilities, and other non-standard business models. The Graham model contributes 45% of the composite score's Value axis.

Peter Lynch popularized the PEG ratio as a way to adjust the P/E ratio for growth. The idea is elegant: a stock trading at 20x earnings is cheap if earnings are growing at 25% per year, but expensive if growth is 5%.

Lynch considered a PEG below 1.0 as a signal that a stock might be undervalued relative to its growth. A PEG of 0.5 suggests the stock is trading at half the price that its growth rate would justify. A PEG above 2.0 suggests overvaluation.

When to use it.The PEG ratio is most useful for growth companies where the P/E alone would look expensive. It answers the question: “Am I paying a reasonable price for this company's growth?”

Sector context matters. A PEG of 1.0 means different things in different sectors. Technology companies typically command higher PEGs because their growth is more scalable. Utilities with PEGs above 1.0 are almost always expensive because their growth ceiling is lower. Comparing PEGs within sectors is more meaningful than across sectors.

Learn more about Lynch's approach in our PEG ratio guide. The Lynch PEG model contributes 25% of the AlphaStocks Value axis.

Joel Greenblatt's “magic formula” ranks stocks by two factors: earnings yield and return on capital. Earnings yield is the inverse of the P/E ratio, but Greenblatt uses a more precise version:

Using EBIT (earnings before interest and taxes) instead of net income normalizes for differences in capital structure and tax rates. Using enterprise value (market cap + debt - cash) instead of market cap alone accounts for leverage — a company with heavy debt should look less attractive than one with the same EBIT but no debt.

Compare to bond yields.One of the most powerful uses of earnings yield is comparing it to the risk-free rate. If the 10-year Treasury yields 4.5% and a stock's earnings yield is 9%, the stock offers a 4.5% equity risk premium. If the earnings yield is only 3%, you are being paid less to own a risky stock than a risk-free bond — a clear overvaluation signal.

Greenblatt's magic formula combines earnings yield with return on capital, which measures how efficiently a company converts invested capital into profits. The idea is to buy stocks that are both cheap (high earnings yield) and high quality (high return on capital). This model contributes 30% of the AlphaStocks Value axis.

Earnings can be manipulated through accounting choices — depreciation schedules, revenue recognition timing, one-time charges. Free cash flow (FCF) is harder to fake because it measures actual cash generated by the business after all capital expenditures.

A high FCF yield (above 7–8%) suggests the company is generating substantial cash relative to its share price. This cash can be returned to shareholders through dividends and buybacks, used to pay down debt, or reinvested in growth — all of which create shareholder value.

Why FCF matters more than earnings. A company can report positive earnings while burning cash if it has heavy capital expenditure requirements or is aggressively capitalizing expenses. Conversely, a company with depressed reported earnings might generate robust free cash flow if the earnings depression is driven by non-cash charges like depreciation or amortization.

How to calculate.Free cash flow = Operating Cash Flow − Capital Expenditures. Both figures are found on the cash flow statement. Divide by market cap (or by shares outstanding to get per-share values) to compute the yield. Compare to the stock's historical FCF yield range and to sector peers for context.

FCF yield is particularly powerful for identifying undervalued mature businesses — companies past their high-growth phase that are now generating predictable, repeatable cash flows. The AlphaStocks screener includes FCF-related filters to help isolate these opportunities.

Each of the methods above captures a different dimension of value. Graham focuses on asset-based intrinsic worth. Lynch adjusts for growth. Greenblatt combines cheapness with quality. FCF yield cuts through accounting noise to actual cash generation. But no single method works in all situations.

Why combining methods works. A stock that scores well on one valuation metric could be a false positive. A stock that scores well on three or four independent methods is far more likely to be genuinely undervalued. Each method acts as a cross-check on the others, reducing the false-positive rate dramatically.

Consider a stock trading at 10x earnings (looks cheap on P/E), with a PEG of 0.7 (growth-adjusted cheapness confirmed), an EBIT yield of 12% (well above bond yields), and an FCF yield of 9% (strong cash generation). All four lenses agree: this stock is likely undervalued. The probability of a value trap is much lower when multiple independent methods converge.

AlphaStocks is built on this principle. The composite score integrates all five investment models — Graham, Lynch, Greenblatt, Piotroski, and Buffett — into a single framework. The Value axis alone combines three different valuation approaches (Graham 45%, Greenblatt 30%, Lynch 25%). But value is only part of the story, which is why the composite also weights Quality (40%), Momentum (35%), and Timing (15%).

The biggest risk in value investing is not buying too expensive — it is buying something that looks cheap but is actually deteriorating. Value traps have destroyed more value-investor capital than overvaluation ever has.

Cheap does not equal undervalued.A stock trading at 5x earnings might be cheap because the market correctly anticipates an earnings collapse. Before buying any “cheap” stock, ask: are revenues growing or shrinking? Are margins expanding or contracting? Is free cash flow positive and growing? If the answers are negative, the low multiple is a warning, not an invitation.

The Timing axis. AlphaStocks addresses value traps through its Timing axis: Timing = min(Value, Momentum). A stock must be both undervalued (high Value score) andgaining price momentum (high Momentum score) for the Timing axis to contribute positively. If Value is high but Momentum is low — cheap but still falling — the Timing axis collapses, pulling down the composite score. This is an intentional design choice to prevent the system from recommending stocks the market is actively repricing downward.

Why momentum confirmation matters. Price momentum reflects information that fundamental analysis may miss: insider selling, competitive threats not yet visible in quarterly filings, or industry shifts that the market is pricing in before analysts update their models. Requiring momentum confirmation before acting on a value signal dramatically reduces value-trap exposure.

Applying all five methods manually across hundreds of stocks is impractical for individual investors. Screening tools automate the heavy lifting, letting you focus on the qualitative judgment that machines cannot replicate.

AlphaStocks provides several tools specifically designed for finding undervalued stocks:

Undervalued rankings: The undervalued stocks ranking sorts all 1,595 stocks by the gap between their current price and estimated fair value, highlighting the most deeply discounted names.

Value rankings: The value rankings sort stocks by the Value axis score, which integrates Graham, Lynch, and Greenblatt valuations into a single 0–10 rating.

Stock screener: The screener lets you filter by P/E, PEG, earnings yield, ROE, dividend yield, market cap, sector, and 20+ other criteria to build custom screens matching your specific value criteria.

Model-specific methodology: For deep dives into each valuation approach, see the Graham methodology page, which details how AlphaStocks implements and adapts each model.