Factor Scores

Factor scores describe a stock's exposure to the systematic drivers of return that academic finance research has documented over decades of data. A factor, in the financial economics sense, is a characteristic that explains differences in returns across stocks and is robust enough to have persisted through multiple market regimes. The five factors most widely used in modern practice are value (cheap stocks on fundamental multiples tend to outperform expensive ones), quality (highly profitable stocks with strong balance sheets tend to outperform), momentum (stocks with strong recent price performance tend to continue), size (smaller stocks have historically earned a premium over larger ones), and low volatility (stocks with lower price volatility have earned higher risk adjusted returns than the market as a whole). Each factor represents a compensated risk or behavioral bias that has been documented across international markets, asset classes, and time periods.

The modern factor framework traces back to the 1992 Eugene Fama and Kenneth French paper "The Cross Section of Expected Stock Returns," which demonstrated that a three factor model (market, size, value) explained the cross sectional variation in US equity returns far better than the single factor capital asset pricing model. Mark Carhart extended the model in 1997 to include momentum. Quality was added formally by Asness, Frazzini, and Pedersen in 2013. Low volatility has been documented since the 1970s but became a recognized factor in the 2000s after research showed that the low risk anomaly was not explained by traditional factors. Today, factor investing is the single largest quantitative category in global asset management, with trillions of dollars allocated to factor based products across passive index funds, systematic active managers, and multi strategy hedge funds.

TickerXray's factor scores are computed individually for each factor using sector neutral rankings. Each stock is scored from 0 (lowest) to 100 (highest) within its sector on each factor, and a composite score blends the five into a single number. The reason for sector neutral scoring is that factor exposures have very different meanings across sectors. A consumer staples company with a value score of 90 is genuinely cheap relative to its peers. A technology company with a value score of 90 is only cheap relative to other technology companies, which may or may not translate to cheap on an absolute basis. TickerXray reports both the sector neutral and absolute versions so you can see both readings.

Where factor scores add the most value is in portfolio construction and in diagnostic analysis. A portfolio that is unknowingly concentrated in a single factor is taking risk the investor may not be aware of. Looking at the factor decomposition of existing holdings can reveal hidden bets. Similarly, when evaluating an individual name, knowing its factor profile tells you what kinds of market regimes it is likely to do well in and what kinds of regimes it is likely to struggle. A high quality, low momentum stock will outperform in a quality led market and underperform in a momentum led one; knowing that changes how you interpret its recent price action.

Each factor score is reported on a 0 to 100 scale, where 0 is the lowest exposure to the factor within the stock's sector and 100 is the highest. For example, a value score of 80 means the stock is in the top 20 percent of its sector on the value metrics. A quality score of 30 means it is in the bottom 30 percent of its sector on the quality metrics.

There is no universal "best" factor score. Different factors outperform in different market regimes, and no single factor has produced monotone outperformance across every period. The practical reading is that a high composite score indicates a stock that aligns with the style factors historically associated with outperformance, subject to the caveat that factor based outperformance comes with significant dispersion year to year and often requires long holding periods to show up.

As a portfolio diagnostic: decompose the factor exposure of the portfolio overall, not just of individual holdings. A long only portfolio with a composite factor score of 55 is mildly tilted toward factor exposure; one with a composite of 75 is strongly tilted. The decomposition reveals whether the tilt is deliberate (you set out to be a quality investor) or incidental (you chose names individually and ended up with a concentrated factor bet).

As a screening criterion: filter the investable universe to names with a composite factor score above a minimum threshold, or to names with a specific factor profile (high value and high quality together, for instance). The screen narrows the universe to names that align with the factors historically associated with outperformance.

As a risk management tool: understanding the factor composition of positions helps explain drawdowns. When a stock sells off and you can attribute the move to momentum factor underperformance, that is a different kind of drawdown from one caused by a stock specific earnings miss. The former is likely to reverse with the factor cycle; the latter requires fundamental analysis.

As a regime awareness input: the historical record shows that different factors lead in different regimes. Value has led strongly in rising rate environments and lagged in growth heavy ones. Quality has been more stable across regimes but slower to compound than the others. Tracking factor performance at the market level gives context for individual stock moves.

Factors exhibit long periods of underperformance. The value factor underperformed dramatically from 2007 through 2020 before staging a comeback. Investors who tilted heavily to value during that period suffered for more than a decade. Factor based investing requires either a systematic approach that can hold through cycles, or a valuation overlay that times factor exposure, neither of which is easy.

Factor definitions vary across practitioners. TickerXray's value factor uses four metrics; Morningstar uses three; AQR uses a different four. The choice of metrics affects the resulting score. The TickerXray choice is standard academic and consistent with most practitioner frameworks, but the reader should understand that other factor scores in the industry may disagree modestly.

Factors can be redundant. A stock with high quality and low debt often also scores well on low volatility, because safer balance sheets produce less volatile stock prices. Double counting of related factors is a known issue in multi factor products, and one reason the composite score uses equal weights rather than attempting to orthogonalize the five.

Finally, factor investing works on average but not on every stock. A high composite score does not guarantee outperformance of any individual name; the factor premium is a statistical feature of diversified portfolios, not a deterministic feature of single positions.

The academic foundation began with the capital asset pricing model (CAPM) in the 1960s, developed by William Sharpe, John Lintner, and Jack Treynor. The CAPM said that a stock's expected return depended only on its beta to the market. The single factor model was challenged by empirical work in the 1970s and 1980s, culminating in Eugene Fama and Kenneth French's 1992 paper, which introduced size and value factors and showed that the three factor model was a much better description of the data. Mark Carhart's 1997 paper added momentum. Later research formalized quality and low volatility as distinct, robust factors. Dimensional Fund Advisors, founded in 1981, became the first major practitioner application of the Fama French framework. In the 2000s, BlackRock, Vanguard, and State Street launched factor based ETFs that made the factor approach available to retail investors. Today, AQR Capital Management, Research Affiliates, and many hedge funds run systematic factor strategies at scale. The academic research continues to evolve, with recent work on the interaction between factors, the role of machine learning in factor selection, and the question of whether the classic factors have been arbitraged away.