Novy-Marx’s screener approach, when analyzed through the lens of multiple regression, reveals how a combined focus on quality and value—augmented by momentum—can help explain how a curated universe of stocks may deliver stronger five-year total returns. The study reinforces that, within a data-driven framework, certain financial metrics provide meaningful signal about future performance, while others contribute less than expected. Taken together, the findings offer a nuanced view of how investors might balance profitability, price discipline, and growth indicators within stock selection processes.
About the Novy-Marx Screener
The Novy-Marx screener represents an integrated screening framework designed to capture two closely related investment philosophies: quality and value. The quality dimension seeks to identify firms with robust profitability characteristics, while the value aspect aims to uncover undervalued assets that, despite their lower current valuation, possess productive capacity and potential for superior returns. This combination—quality plus value—has a long-standing appeal in value investing circles, where the aim is to combine discounted prices with credible, durable earnings potential.
Profitability in the Novy-Marx framework is measured via gross profits relative to assets, a criterion that emphasizes the ability to generate gross profits efficiently from the asset base rather than relying solely on net income or other downstream metrics. This metric indicates how effectively a company can convert its assets into gross profits, providing a lens into the competitive advantages and operating profitability of the business model. The screener then complements this profitability signal with momentum and valuation signals to construct a multi-faceted screening rule set.
Momentum, in this context, is quantified by the one-year price change. This factor adds a temporal element, capturing short- to intermediate-term price dynamics that may reflect market sentiment, growth trajectories, or shifting competitive landscapes. Valuation, as gauged by low price-to-book, serves as the value signal in the trio. By prioritizing stocks with a low price-to-book ratio, the screener aims to flag firms that may be trading at a discount relative to their recorded asset base on the balance sheet, potentially signaling undervaluation in markets where asset valuation has relevance.
Together, these three parameters—quality (gross profits-to-assets), momentum (one-year price change), and value (low price-to-book)—form the core of the Novy-Marx screener. The design also includes filters to avoid stocks that are too small to reliably trade or too illiquid to be meaningfully analyzed. Specifically, the screener excludes firms with very low market capitalizations and those with insufficient trading volume, thereby ensuring the results reflect a reasonably investable universe. The objective of this screening approach is to assemble a cohort that combines durable profitability with price discipline and favorable price mechanics, increasing the likelihood that the five-year total return can be meaningfully linked to the screener’s inputs.
To provide context for interpretation, consider the broader investment maxim attributed to Warren Buffett, which captures the essence of the quality-plus-value philosophy: it is far better to buy a wonderful business at a fair price than to buy a fair business at a wonderful price. The Novy-Marx screener operationalizes this line of thinking by placing emphasis on profitability and value signals while acknowledging the momentum dimension that can reflect how quickly markets are recognizing a firm’s earnings potential. In practical terms, the screener is configured to search for companies that demonstrate robust gross profitability over assets, show positive momentum over the recent year, and present a favorable valuation stance on a price-to-book basis, all while maintaining a commitment to investable liquidity.
Within this framework, the screener uses three parameters to guide the stock search and filtering process. The first parameter, quality, hinges on the ratio of gross profits to total assets, a metric that captures the efficiency and scale of gross profitability relative to asset bases. The second parameter, momentum, is measured by the one-year price change, reflecting the stock’s recent performance trajectory and potential continuation in the near term. The third parameter, value, is captured through the price-to-book ratio, with a lower ratio indicating potential undervaluation of the asset base as registered on the balance sheet. In addition to these primary signals, the screener enforces screening thresholds that remove stocks with insufficient market capitalization or trading activity, thereby preserving the reliability of the results within the analyzed universe.
The universe under consideration for the Novy-Marx screener is constructed to be sizable enough to yield meaningful statistical insights yet constrained to a level where data quality remains robust. In this particular analysis, the screener evaluated a cohort of approximately 250 stocks that spanned various sectors. This sample was filtered to ensure that each stock possessed a market capitalization above a defined threshold and demonstrated adequate daily trading volume, with the goal of maintaining a consistent and investable dataset. As a contextual note, within the top-ranked subset of 250 stocks that passed the screener’s criteria, the ten largest companies, by market capitalization, were used to illustrate the scale and composition of the screening outcomes. The intention was to highlight how the sourcestock mix might influence the observed relationships between screener signals and subsequent five-year performance.
The conceptual premise behind incorporating these three parameters—quality, momentum, and value—rests on the notion that highly profitable firms with durable asset bases, when combined with favorable momentum and modestly discounted prices, are more likely to deliver sustained outperformance. By blending these dimensions, the screener aims to avoid relying on a single signal that could be cyclical or transient, instead favoring a composite approach that captures a broader set of determinants behind long-run stock returns. In this sense, the Novy-Marx screener is positioned as a practical, data-driven tool that aligns with established investing intuitions about the interplay between profitability, price discipline, and market dynamics.
The development of the screener also accommodates a cautious emphasis on liquidity and tradability. By filtering out stocks with extremely small market capitalizations and low trading volumes, the screener seeks to reduce the risk of sampling bias that could arise from illiquid or sporadically traded securities. This methodological choice helps ensure that any observed relationships between screener signals and five-year total returns are not unduly distorted by participation constraints or thinly traded equities. The result is a curated set of candidates that investors can reasonably consider within a broader strategy, while still enabling empirical exploration of how quality and value factors interact with momentum to forecast mid- to long-term performance.
Ultimately, the Novy-Marx screener’s architecture rests on a balance between theoretical appeal and empirical stability. The inclusion of profitability as a primary signal, complemented by momentum and valuation, reflects an attempt to capture both the enduring competitive advantage of firms and the market’s evolving appraisal of those firms. The screening framework thus offers a structured approach to incorporating the quality-versus-value dichotomy into quantitative stock selection, while also acknowledging real-world considerations such as liquidity constraints and the practicalities of portfolio construction. In the following sections, the analysis details how this screener behaved when subjected to a rigorous regression framework aimed at predicting five-year total returns across a cross-section of stocks.
Data, Variables, and Screening Criteria
The study constructs its analysis around a defined universe of stocks that pass the screener’s initial filters, focusing on a mid-to-large cap segment to ensure adequate liquidity and data reliability. The screening process begins by identifying companies that display robust profitability, as measured by the gross profits-to-assets ratio. This profitability metric is designed to capture the efficiency with which firms convert asset investments into gross profits, offering a perspective on the durable earning power embedded in the business model. By focusing on gross profits relative to total assets, the screener seeks to emphasize the scale and effectiveness of a company’s core operations, rather than relying solely on net income or other downstream profitability measures that can be influenced by tax strategies, extraordinary items, or accounting choices.
In addition to profitability, the screener incorporates momentum as a key signal, operationalized through the one-year price change. The momentum signal is intended to reflect how the stock has performed over the recent year and to capture potential continuation of price strength, market sentiment, and the pace of earnings revelations or strategic developments. Momentum serves as a practical gauge of near-term market dynamics that might interact with longer-term fundamentals to influence five-year returns. The momentum component thus complements the profitability signal by adding a temporal dimension to the screening framework.
Value, another central pillar of the screener, is assessed using the price-to-book ratio. This valuation metric ranks firms by how their share price relates to the accounting value of their equity as reported on the balance sheet. A lower price-to-book ratio suggests a potential discount to asset backing, aligning with classic value investing principles that seek to identify bargains when price reflects insufficient earnings potential, operational efficiency, or growth prospects. The inclusion of a price-to-book filter aims to identify assets that may be undervalued in light of their underlying asset base, while not ignoring the profitability and momentum signals.
Beyond these three signals, the screener enforces practical filters to ensure the resulting stock pool is investable. Stocks with very small market capitalization are excluded, as are those with insufficient trading volumes. These filters help prevent the inclusion of stocks that might be subject to price manipulation, atypical liquidity patterns, or data reliability challenges, thereby increasing the likelihood that the regression results reflect real-world investment dynamics. The concrete outcome is a defined cohort, sized at roughly 250 stocks in this analysis, chosen to represent a cross-section of sectors while maintaining a level of liquidity supportive of practical application.
The analysis highlights a particular subsample feature: for context, among the top-ranked 250 screener candidates, the ten largest companies by market capitalization were identified to illustrate the scale and concentration of the dataset. While the list of specific names is not enumerated in this section, the emphasis remains on understanding how the screener’s signals distribute across a broad, investable universe and how those signals translate into five-year performance expectations after controlling for other factors in a regression framework. The screening methodology thus provides a structured way to examine whether the combination of quality, momentum, and value signals can robustly forecast five-year total return across a representative set of stocks.
In constructing the regression model, the primary dependent variable is the five-year total return. This horizon aligns with the screener’s intended use, offering a long-run performance perspective beyond one-to-two year intervals. The independent variables correspond to the three screener signals: quality, momentum, and value, alongside the practical filters that ensure investability. The analysis recognizes that each variable captures a distinct dimension of stock behavior and fundamental quality, and the model’s goal is to determine the incremental predictive power of each signal when the others are held constant. In this sense, the model attempts to quantify whether profitability relative to assets, one-year price trajectory, and valuation via price-to-book collectively contribute to explaining cross-sectional differences in five-year stock performance.
The data collection process emphasizes consistency and reliability in reporting across the stock universe. The gross profits-to-assets measure requires accurate accounting data for gross profits and total assets, while the one-year price change component relies on precise price data over a recent rolling window. The price-to-book ratio demands careful alignment of market price with book value per share, ensuring that the valuation signal reflects a credible reference point for investors. This rigorous data framework is intended to minimize measurement error and to enhance the stability of the regression estimates, particularly given the cross-sectional nature of the analysis across hundreds of stocks.
In terms of methodological considerations, the choice to use three core predictors—quality, momentum, and value—reflects an intent to balance fundamental efficiency with market-based signals and relative pricing. The model specification aims to guard against overfitting by restricting the number of primary predictors while allowing for interpretation of each signal’s impact on the dependent variable. The evaluation of model fit emphasizes the strength and significance of the multivariate relationship between the screener’s signals and the five-year total return, as well as an assessment of how much of the explained variance is attributable to the 1-year momentum signal, given its close relationship with long-run performance. The following sections present the results of this regression analysis and interpret the statistical and practical significance of the findings.
The data universe comprises roughly 250 stocks drawn from diverse sectors, a range that provides a cross-sectional snapshot of equity markets while avoiding extreme concentration in any one industry. The sector diversity helps ensure that the results are not confined to a single cyclicality or commodity-driven dynamic. The stocks in the sample meet the screener’s thresholds for liquidity and size, making them reasonable candidates for implementation in a research-oriented or practice-oriented investment process. The study notes that such a dataset, while informative, is not without limitations, particularly when considering broader generalizability beyond the defined universe. The intention is to provide a robust evaluation of whether the screener’s signals hold predictive value within a plausible investable population of stocks.
To reiterate the core design, the Novy-Marx screener integrates profitability, momentum, and valuation into a cohesive screening framework aimed at identifying firms with strong gross profitability relative to assets, coupled with favorable one-year price dynamics and a discounted book-value valuation. The regression analysis then scrutinizes how those signals relate to actual five-year outcomes across the selected population. The emphasis is on understanding the extent to which the screener’s inputs are informative about long-run performance, and on interpreting the relative contributions of each signal within the multivariate setting. In this way, the study seeks to provide investors with a structured, data-driven perspective on how quality and value signals interact with momentum to shape predicted five-year total returns, while acknowledging the boundaries of the sample and the conditions under which the results apply.
Regression Results and Key Findings
The regression analysis was applied to approximately 250 stocks spanning various sectors, with the aim of assessing how well the Novy-Marx screener’s three primary signals—quality, momentum, and value—explain the variance in five-year total return. The results indicate that the multivariate relationship among these signals and the five-year total return is statistically significant. Specifically, the regression yields a meaningful positive correlation coefficient (R) of about 0.52 between the screener’s combined signal set and the five-year total return, with high statistical significance denoting a robust relationship. This level of correlation suggests that the screener’s inputs, when combined, have practical explanatory power for forecasting five-year performance across the stock universe studied. The reported p-value of less than 0.001 reinforces the reliability of the observed association, indicating that the likelihood of such a relationship arising by chance is very small given the data.
Among the three signals, the one-year total return emerges as the strongest individual predictor for the screener’s five-year performance. The analysis shows a correlation of approximately 0.49 when considering the five-year total return in relation to the one-year return alone. While this is a strong association, the interpretation requires nuance: the one-year return is inherently linked to the five-year horizon since the short-term performance contributes directly to the longer-term outcome. However, the correlation still demonstrates that recent performance encodes valuable information about future trajectories, though the independence between the predictor and the dependent variable is not absolute. The practical takeaway is that historical success, as captured by the one-year momentum, tends to be a good indicator of future performance, but it is not a perfect predictor, and other factors contribute to the longer-run outcome beyond the past year’s gains.
In contrast to expectations, the price-to-book ratio contributed only modestly to the predictive power of the model, showing little to no positive or negative effect on the five-year total return when the other factors were accounted for. This result implies that, within the specified dataset and screening framework, valuation based on price-to-book did not exert a strong independent influence on future five-year returns beyond the signals captured by profitability and momentum. The gross profit-to-total assets ratio, representing profitability, made a small but statistically significant contribution to the model, indicating that profitability has a detectable, though modest, incremental impact on predicting five-year performance when controlling for momentum and value signals. The directional impact and magnitude suggest that profitability plays a supportive role in the long-run forecast, reinforcing the notion that quality signals can complement momentum and valuation in explaining return dynamics.
These findings must be interpreted in the context of the dataset’s characteristics and the screening criteria that shaped the stock universe. The universe comprised roughly 250 stocks with market capitalizations exceeding $100 million and with average daily trading volumes above 20,000 shares. This filtration reduces certain biases associated with micro-cap or illiquid stocks but simultaneously narrows the scope of generalizability to markets and time periods where such liquidity conditions hold. This limitation underscores the importance of understanding that the results reflect relationships within a specific, investable subset rather than a universal law across all equities. The study acknowledges that the sample’s composition—driven by liquidity and size constraints—may influence the observed strength of the relationships and the relative contributions of the screener’s signals. Despite this limitation, the results consistently point to the primacy of the one-year momentum signal as a predictor within the five-year framework, with profitability providing a modest but meaningful additional layer of explanatory power.
A key takeaway concerns the role of the one-year momentum signal as a core driver of predictive strength. The correlation between one-year total return and five-year total return indicates that recent performance is a strong harbinger of longer-term outcomes, although it is not a perfect predictor. The finding aligns with intuitive market dynamics: stocks that have demonstrated strength in the recent period often continue to perform well as positive earnings momentum or strategic advantages unfold, while some reversals can occur as market expectations recalibrate. Nevertheless, the data show that momentum contains the most information about future five-year returns among the three signals, reinforcing the importance of incorporating recent performance trends into screening and investment decisions.
In addition to the primary results, the analysis takes into account the constraints related to the dataset and the selection criteria. The sample’s composition, with a threshold for market capitalization and liquidity, implies that certain segments—such as micro-cap stocks or less liquid firms—are not represented in the analysis. This restriction means that the results should be interpreted with caution when extrapolating to broader markets, particularly in environments characterized by different liquidity regimes or structural shifts in market participation. Nevertheless, the overarching pattern remains informative: the combination of strong short-term momentum and solid profitability, within a screened set that also includes valuation discipline, is associated with higher five-year total returns relative to a broader, less filtered stock universe.
The practical implication of these results for investors and researchers is multifaceted. First, the evidence supports the idea that a quality-and-value framework, augmented by momentum, can help identify stocks with favorable long-run return potential within a defined, investable set. Second, the data underscore the relevance of recent performance signals as a predictor of longer horizons, while also highlighting that valuation signals, at least as captured by price-to-book in this analysis, may have a more muted role when considered alongside profitability and momentum. Third, the results illuminate the importance of sample selection and data quality in interpreting predictive relationships, reminding researchers and practitioners alike to consider the composition of their stock universe and the constraints that accompany any empirical study. Taken together, these findings point toward a nuanced, data-driven approach to stock screening that respects both the enduring appeal of quality and value signals and the practical realities of market dynamics and liquidity.
The regression outcomes also suggest interesting directions for future research and model refinement. One possible avenue involves exploring alternative measures of profitability or different asset bases, such as gross profits relative to invested capital or total assets minus liabilities, to see whether other profitability metrics might yield stronger predictive signals within the same framework. Another potential improvement could be to extend the analysis across multiple time horizons, including longer-term returns beyond five years, to assess whether the observed relationships persist, intensify, or attenuate over extended periods. Researchers might also consider expanding the universe to include micro-cap stocks or higher-turnover portfolios to evaluate whether the observed patterns hold in more diverse contexts, and to determine how liquidity and trading frictions impact the stability of the screener’s predictive power.
In practical terms, investors considering the Novy-Marx screener should view these results as encouraging but not definitive. The positive association between the screener’s combined signals and five-year total return demonstrates that the quality-plus-value approach, with momentum, can be a meaningful component of a stock-selection process. However, the modest contribution of price-to-book signals and the reliance on one-year momentum as a primary predictor imply that diversification, risk management, and ongoing monitoring remain essential. The findings also underscore the importance of contextualizing results within the specific universe of investable stocks, acknowledging that shifts in market structure, liquidity conditions, and sector dynamics can influence the observed relationships. A prudent application would involve integrating the screener signals into a broader decision-making framework, complemented by regular backtesting and sensitivity analyses that capture changes in market regimes over time.
Context, Validation, and Related Work
Beyond the core analysis, multiple publications and practical adaptations have contributed to the broader understanding of the quality-plus-value approach. The central premise—that profitability (as captured by gross profitability) is a meaningful predictor in its own right—has been explored in publications spanning financial economics and equity research. These works collectively support the notion that profitability dimensions can offer valuable information about relative stock performance, particularly when combined with value signals. In addition to the core empirical findings, a number of researchers and practitioners have highlighted the complementary role of momentum in enhancing screening outcomes, noting that price trends can capture market expectations and dynamic shifts in firm prospects.
Operational variants of the Novy-Marx screener have been developed and deployed within various research libraries and screening platforms. An example is the availability of an operational ranked screener version, designed to support users who wish to apply the same three-signal framework with a structured ranking mechanism. This type of tool allows investors to systematically sort stocks by the composite signal, facilitating more transparent decision-making and easier comparison across time periods and market environments. The existence of such a tool underscores the practical feasibility of translating the screener’s three-signal logic into actionable investment processes, while still honoring the empirical insights derived from the regression analysis.
The broader literature on quality investing has repeatedly emphasized the importance of aligning profitability signals with value considerations. The idea is that firms with strong gross profitability, efficient asset use, or durable competitive advantages can sustain earnings power and shareholder returns, but that stock prices may not fully reflect these attributes immediately. In parallel, the literature on value investing emphasizes patience and the willingness to buy assets at discount to fundamental worth, on the expectation that price discipline and earnings potential will eventually converge toward intrinsic value. When combined with momentum, these strands of research suggest a dynamic approach to stock selection that accounts for the path-dependent nature of markets, where past performance and present fundamentals interact with evolving expectations.
In the context of the Novy-Marx screener’s findings, the emphasis on one-year momentum as a robust predictor aligns with both academic and practitioner observations about short-term price dynamics. While longer horizons often reflect deeper structural improvements in profitability and growth, the one-year signal captures near-term catalysts and market sentiment shifts that can influence price trajectories, thereby shedding light on how momentum can function as a bridge between fundamental quality and valuation signals within a five-year framework. The overall interpretation is that a well-constructed screener blends fundamental profitability with timely price information and prudent valuation, creating a cohesive framework for analyzing potential long-run performance.
Investors who apply the Novy-Marx screener should also consider the practical aspects of implementation. The screener is designed to operate within investable universes—stocks large enough to be reliably traded with sufficient liquidity—so that the results have actionable implications in real portfolios. Even with robust regression evidence, portfolio construction requires attention to risk management, diversification across sectors, and alignment with investment goals and constraints. The evidence base suggests that historical relationships can be informative, but market regimes may evolve, and past patterns do not guarantee future results. Accordingly, practitioners may wish to combine the screener’s signals with additional risk controls, scenario analyses, and continuous performance monitoring to maintain a disciplined investment approach.
The broader validation of the quality-plus-value approach is reinforced by multiple sources that examine profitability, asset efficiency, and valuation as contributors to stock returns. While not every study reaches identical conclusions, the converging thread is that quality-oriented profitability signals can play a meaningful role in predicting cross-sectional return patterns, particularly when paired with valuation discipline and momentum considerations. These insights offer a coherent rationale for incorporating the Novy-Marx screener into systematic screening processes, provided that the limitations of the dataset, the potential for regime shifts, and the need for ongoing validation are acknowledged and addressed in practice.
In sum, the body of work around the Novy-Marx screener and its related quality-plus-value framework supports a nuanced perspective on stock selection. The regression findings demonstrate that a three-signal approach can explain a notable portion of five-year total return variation within a carefully defined universe, while also revealing the dominant role of short-term momentum as a predictor within that framework. The collaboration between profitability signals, momentum, and valuation reflects a balanced strategy that leverages different dimensions of stock behavior, and the interpretation of these results should remain anchored in a careful understanding of data limitations, sample composition, and the practicalities of portfolio management.
Practical Implications for Investors and Portfolio Strategy
From an investor’s standpoint, the Novy-Marx screener provides a structured methodology for integrating quality and value signals with momentum to forecast five-year total returns within a focused, investable stock universe. The practical takeaway is that incorporating a profitability measure (gross profits-to-assets) alongside momentum and valuation signals can yield a meaningful predictive signal when evaluating stocks over a multi-year horizon. The observed strength of the one-year momentum signal also suggests that recent performance remains an informative input for long-run projections, though it is essential to interpret this signal within the context of the other signals and the broader market environment.
A prudent approach for practitioners would be to implement the screener as part of a broader investment framework. This could entail creating a ranking system that weights the three signals in a way that reflects historical performance and risk tolerance, followed by rigorous backtesting across multiple market regimes to examine stability and robustness. Diversification across sectors and careful risk controls are important, given that the results derive from a defined dataset with liquidity constraints. Investors may also consider periodic rebalancing and stress testing to ensure the screener’s signals continue to align with evolving fundamentals and price dynamics.
The limitations highlighted by the study warrant thoughtful consideration by practitioners. The findings are based on a dataset of approximately 250 stocks with market capitalization above $100 million and adequate trading volume, which means that the results may not generalize to smaller, less liquid firms or to markets that exhibit distinct liquidity patterns. Consequently, while the screener contributes valuable signals within its defined universe, extrapolating beyond those bounds should be done with caution. The modest contribution of price-to-book in the multivariate setting indicates that value signals, when combined with profitability and momentum, may have less incremental explanatory power than anticipated in some contexts. This nuance emphasizes the importance of focusing on the overall signal mix and the quality of the dataset rather than relying on a single metric as a universal predictor.
For investors who prioritize long-run performance and disciplined screening, the Novy-Marx screener offers a practical framework to explore the intersection of profitability, valuation, and momentum. Its empirical results illustrate how a multi-factor approach can capture meaningful cross-sectional variation in five-year returns within a chosen universe. The approach underscores the value of combining robust operating profitability with favorable price discipline and momentum signals, while also illustrating the limits of any single factor in isolation. In practice, this means that a well-designed screener should be part of an integrated investment process that includes risk assessment, portfolio construction considerations, and ongoing performance evaluation to adapt to changing market conditions and to maintain alignment with investment objectives.
The broader validation of quality investing and its integration with value and momentum signals further supports an ongoing exploration of how these factors can complement each other in portfolio construction. Investors who adopt the screener should remain mindful of regime shifts, potential biases arising from the chosen stock universe, and the need for continuous refinement as new data and market conditions emerge. The empirical evidence from the Novy-Marx screener thus contributes to a larger conversation about how best to combine fundamental profitability signals with market-driven and valuation-based indicators to inform long-run investment decisions.
Limitations, Assumptions, and Areas for Enhancement
While the results are informative, they come with important caveats about generalizability and robustness. The analysis acknowledges that the dataset—approximately 250 stocks with specific liquidity and size characteristics—limits the extent to which the findings can be extended to broader markets or other time periods. The stock universe’s composition, by virtue of the liquidity threshold and capitalization floor, inherently shapes the observed relationships. If the universe were expanded to include smaller or less liquid stocks, the dynamics of the screener’s signals and their predictive power might differ, potentially altering both the strength of the correlations and the relative contributions of each signal.
Another crucial consideration is the interdependence among the predictor variables, especially between momentum and the five-year horizon outcome. Since the one-year return directly contributes to the five-year return, the independence assumption is not strictly satisfied. This interdependence does not invalidate the findings, but it does temper the interpretation of the results. In practice, it means that while momentum is a strong predictor, it is not entirely independent of the target measure, and thus the incremental information provided by momentum should be viewed as part of a broader predictive framework rather than as an isolated cause of long-run performance.
The modest impact of price-to-book on five-year returns in the multivariate context suggests that value signals, while meaningful in some settings, may be less potent in this particular model when paired with profitability and momentum signals. This outcome invites further investigation into alternative valuation metrics or the weighting of valuation signals within a multi-factor framework. For example, other price valuation indicators, such as price-to-earnings, price-to-cash-flow, or price-to-sales, could be explored to determine whether different valuation constructs offer incremental predictive value when combined with quality and momentum.
Additionally, the study’s reliance on a cross-sectional regression across a fixed stock set raises questions about time-series dynamics and potential non-stationarity. Future work could examine the stability of the screener’s predictive signals over different market regimes, including periods of elevated volatility, shifting interest rates, or structural changes in corporate profitability and capital allocation. A time-series extension or a rolling-window analysis could shed light on how the screener’s signals perform under varying macroeconomic conditions and how their predictive power evolves through different cycles.
From a practical standpoint, even with a positive and statistically significant relationship between the screener’s signals and five-year returns, investors must recognize that past performance is not a guarantee of future results. The long-run predictive value of the screener depends on the persistence of the underlying factors that drive profitability and market expectations, which can shift as industries mature, competitive dynamics change, and macroeconomic forces influence corporate earnings and valuations. As such, a disciplined approach to backtesting, risk management, and ongoing validation remains essential for practitioners who apply the Novy-Marx screener in real-world portfolios.
In sum, while the analysis provides robust evidence of a meaningful relationship between the Novy-Marx screener’s three signals and five-year total returns within a defined universe, it also highlights important boundaries and potential avenues for enhancement. The stated limitations underscore the need for continued research, diversification of signal sets, and careful consideration of data quality and universe selection when drawing broader conclusions about stock-picking effectiveness. The ongoing exploration of quality, value, and momentum remains a fertile area for quantitative investing, inviting refinements that can further illuminate how best to combine these signals in practice.
The March Midness Parallel and Related Analyses
In an earlier exploratory effort, a separate blog post applied a similar multiple regression approach to Alex Reisman’s six “March Midness” screeners, with the aim of identifying optimal weightings for predicting stock prices across multiple time horizons. That prior work served as a comparative backdrop for evaluating the Novy-Marx screener by using an analogous methodological framework. The central idea was to assess how different screener inputs, when weighted together, can explain variations in stock prices over short- to medium-term spans, thereby enriching the understanding of how screeners perform under varying predictive objectives.
The March Midness analysis highlighted the potential for growth-oriented signals to contribute meaningfully to price forecasts, particularly when combined with momentum and valuation considerations. In this sense, the two lines of inquiry—Novy-Marx’s quality-plus-value framework and Reisman’s March Midness approach—converge on a shared theme: multi-factor screening can potentially capture a broader spectrum of drivers behind stock price movements and returns than any single signal alone. By examining both approaches side by side, researchers and practitioners gain insight into how different emphasis—whether on profitability, growth dynamics, or momentum—affects predictive performance across time horizons.
The parallel analyses illuminate a broader methodological takeaway: that multi-factor screening, when properly calibrated, provides a more nuanced view of expected performance than univariate strategies. The comparative approach highlights the value of cross-validation across screeners and time frames, enabling a more robust understanding of which signals tend to dominate under certain market conditions and how combinations of signals interact to forecast price trajectories and total returns. The insights drawn from these parallel analyses reinforce the broader conclusion that incorporating multiple dimensions of stock quality, momentum, and value can yield a richer picture of potential future performance.
From a practical perspective, the Reisman and Novy-Marx comparisons suggest that investors and researchers should consider employing diversified screening frameworks that blend growth-oriented signals with profitability and valuation metrics, rather than relying exclusively on one dimension. The idea is to exploit complementary strengths across different signals and to recognize that the most informative combination may shift with market regimes. By adopting a versatile screening toolkit that accommodates multiple perspectives on what constitutes a sound investment, practitioners can improve their ability to identify stocks that exhibit favorable long-run prospects while maintaining appropriate risk controls and liquidity considerations.
Validation, Reproducibility, and Operational Considerations
The body of work surrounding the Novy-Marx screener emphasizes reproducibility and operational practicality. Researchers who replicate the analysis can expect to encounter similar patterns in the relationships among quality, momentum, and value signals and five-year returns, provided that consistent data definitions and universe constraints are maintained. Reproducibility rests on clearly defined metrics for gross profits, total assets, price-to-book, and one-year price change, as well as on the careful construction of the stock universe, including filters for market capitalization and trading liquidity. The emphasis on consistent observational rules ensures that results are comparable across time and across different research settings, enabling practitioners to validate findings within their own portfolios or simulation environments.
An important operational consideration relates to the availability of screener configurations and ranking tools. In practical use, investors may employ a ranked or scoring approach that aggregates the three signals into a composite score, which can then be used to construct a portfolio or to guide stock selection. The ranking framework should incorporate appropriate weighting, which could be derived from historical performance, risk-adjusted considerations, or bespoke strategic objectives. The repeatability of the screener depends on transparent, well-documented parameters and on the ability to maintain data quality across time. When integrating such a screener into live investment processes, it is crucial to implement governance around data updates, signal recalibration, and performance reporting to ensure ongoing reliability.
The empirical findings also highlight the potential benefits of periodic revalidation. Markets evolve, and the interplay between profitability, valuation, and momentum can shift as capital allocation patterns change, firms’ competitive dynamics adapt, and macroeconomic conditions transform the drivers of stock prices. Revalidating the screener’s performance at regular intervals—such as quarterly or semi-annually—helps ensure that the approach remains aligned with current market realities and continues to provide meaningful information for decision-making. This ongoing validation fosters a disciplined investment approach that can adapt to new data while preserving the core logic of combining quality, value, and momentum signals.
From a broader perspective, the Novy-Marx screener fits into a larger tradition of evidence-based investing that seeks to quantify relationships between financial fundamentals and market prices. The combination of profitability signals, valuation discipline, and price momentum reflects a pragmatic stance toward stock selection, balancing durable earning power with market-driven price movements. The practical implications of this approach are clear: investors who adopt the screener should be prepared to integrate it into a comprehensive investment process, maintain vigilance regarding data quality and universe composition, and remain attentive to changes in market conditions that could alter the predictive relationships observed in historical analysis.
Conclusion
The analysis of the Novy-Marx screener through a rigorous regression framework demonstrates that a combined quality-plus-value approach, augmented by momentum, can explain a meaningful portion of five-year total return within a carefully defined, investable stock universe. The results underscore that the one-year momentum signal stands out as a particularly strong predictor for longer-horizon performance, while profitability (gross profits-to-assets) offers a smaller yet significant incremental contribution. Price-to-book, as a valuation metric within this multivariate setup, contributes little to the predictive power when considered alongside the other signals, highlighting the nuanced interplay among quality, value, and momentum. The dataset’s characteristics—approximately 250 stocks with market caps above $100 million and adequate trading volume—shape the observed relationships and emphasize the importance of context when generalizing beyond the sample.
These findings carry practical implications for investors seeking to enhance long-run return potential through disciplined screening. A three-signal framework that integrates profitability, momentum, and valuation can be a valuable component of stock-selection processes, provided that risk management, diversification, and data integrity are maintained. The results also point to the enduring relevance of momentum, consistent with market dynamics in which recent price trends reflect evolving expectations about earnings potential and strategic developments. However, investors should treat the results as part of a broader toolkit rather than as a standalone predictor, recognizing that past performance does not guarantee future outcomes and that market conditions can change in ways that affect the strength and stability of the observed relationships.
The broader body of research and practical implementations related to the quality-plus-value approach—including parallel analyses like the March Midness framework—supports a nuanced, multi-factor perspective on stock selection. This perspective emphasizes the value of combining fundamental profitability signals with market-based momentum and valuation considerations, while acknowledging the limitations imposed by sample selection, data quality, and regime shifts. As investors continue to refine screening methodologies, ongoing validation and careful calibration will be essential to maintaining robust, actionable insights that can contribute to informed, disciplined investment decisions.