What Is a Composite Stock Score?

No single metric captures the full picture of a stock. A company with excellent profitability can still be overpriced. A deeply discounted stock might be cheap for good reason. And a stock with strong momentum could be riding a bubble with no fundamental support.

The composite score solves this by requiring a stock to perform well across multipledimensions simultaneously. It is the intersection of quality, value, and momentum that produces the strongest long-term returns — not any single factor in isolation.

Question: Is this a fundamentally strong, durable business?

The Quality axis combines two complementary models. The Piotroski F-Score (35% of Quality) compares nine binary signals — profitability, leverage and liquidity, operating efficiency — in the latest fiscal year with the year before. The Buffett criteria (65%of Quality) test ten years of returns on equity and tangible capital, owner earnings, debt and capital allocation against thresholds from Buffett's shareholder letters.

Quality receives the largest weight: a great business bought at a fair price is the core idea behind both models. A model that does not apply to a company (Piotroski for banks, for example) is left out of the axis, never counted as zero.

Question: Is the stock priced below what the business is worth?

Three models contribute to the Value axis: Graham's defensive-investor criteria (45% of Value), Lynch's categories and PEG (25%), and Greenblatt's Magic Formula (30%). Each approaches cheapness from a different angle — Graham's price tests, price against growth, and earnings yield ranked with return on capital. The site's fair value estimate is shown separately and is not an input to the composite.

Value receives a deliberately modest weight. A stock can appear cheap for years while the underlying business deteriorates. The Value axis matters, but it is constrained to prevent the formula from chasing value traps.

Question: Is the stock price moving in the right direction?

The Momentum axis is the 12-1 month price return — the close about 30 days ago over the close about 365 days ago — ranked across the universe: 10 × the share of stocks with a lower return. The definition follows Jegadeesh & Titman (1993), Journal of Finance 48(1), pp. 65-91; the most recent month is skipped because one-month returns reverse (Jegadeesh 1990; Carhart 1997). Stocks that have risen over the past year tend to keep rising over the next few months — one of the most replicated findings in empirical finance.

Momentum serves two purposes. First, it captures market intelligence: rising prices often reflect information not yet visible in quarterly filings. Second, it acts as a timing filter — a fundamentally excellent stock that the market is actively selling may have problems the models cannot see yet.

Question: Is the stock simultaneously undervalued and gaining momentum?

This formula is the composite score's value-trap killer. If a stock has a high Value score but low Momentum, the Timing axis collapses to the lower of the two. The stock cannot score well overall until the market confirms the value thesis by actually bidding the price higher.

The sweet spot is where both Value and Momentum are elevated: cheap AND rising. This is where the Timing axis adds the most points to the composite.

A bank and a software company cannot be evaluated with the same yardstick. AlphaStocks uses seven sector profiles — General, Bank, REIT, Insurer, Utility, Holding, and Asset Manager. A profile decides which models apply (a model that does not fit is left out, never scored as zero), which fair value method is used, and the axis weights. The models themselves keep their sources' thresholds.

A company whose Quality or Value axis has no model with a score gets no composite at all, and the page says why instead of inventing one. Insurers, for example, have only Buffett's insurer adaptation on their Quality axis; one without enough data for it has no composite.

Each model contributes to one or more axes of the composite score:

Not as a return predictor. AlphaStocks backtests the composite point-in-time on the S&P 500, using only data available on each date. A portfolio of the top 10 highest-scoring stocks, rebalanced monthly from 2021 to early 2026, ended slightly ahead of the index, but the excess is not statistically significant and is mostly explained by market exposure. That backtest was run on the previous (v22) formula; the current engine was not re-backtested. The composite is useful for building a consistent research shortlist; it does not reliably pick market-beating stocks.

See the full backtest results for methodology and disclaimers.