How to Pick Stocks Using Data, Not Gut Feeling:
A Quantitative Framework.

Why Most Stock Picks Fail

The stock picking industry is built on a paradox: millions of investors spend thousands of hours analyzing companies, yet the overwhelming majority underperform a simple index fund. The S&P Indices Versus Active (SPIVA) scorecard, the definitive study of active management performance, shows that 92.2% of large-cap U.S. funds trailed the S&P 500 over 15 years. This is not a small sample — it covers thousands of professional fund managers with Bloomberg terminals, analyst teams, and decades of experience.

The question is not whether active stock picking fails. The evidence is conclusive. The question is why it fails — and whether a different approach can succeed.

The answer lies in behavioral psychology. Daniel Kahneman and Amos Tversky^[1] identified systematic cognitive biases that distort human judgment under uncertainty. Three biases are particularly destructive for stock pickers:

Barber and Odean^[2] studied 66,465 households at a major discount brokerage between 1991 and 1996. Their finding was devastating: the stocks investors bought underperformed the stocks they sold by 3.3 percentage points per year. Investors were not just failing to add value — they were actively destroying it. The more frequently they traded, the worse they performed. The quintile of most active traders underperformed the least active quintile by 7.1 percentage points annually.

The implication is clear: the human brain is not wired for stock selection. Our pattern-recognition instincts, honed over millennia for survival, produce systematically incorrect judgments in financial markets. The solution is not to become a better stock picker through more research and more conviction. The solution is to remove human judgment from the process entirely and replace it with quantitative factors that have been validated across decades of data.

The 6 Factors That Actually Predict Stock Returns

Academic finance has spent fifty years isolating the characteristics that predict stock returns. From Eugene Fama and Kenneth French's three-factor model in 1993 to the current six-factor models used by quantitative hedge funds, the evidence converges on a set of persistent, economically significant return drivers. These are not statistical artifacts — they have been validated across 90+ years of U.S. data, 40+ international markets, and multiple asset classes.

The BCR framework assigns specific weights to each factor based on the strength of the academic evidence, the decay rate of the signal, and the implementation cost (turnover, market impact, and capacity). Here is the complete breakdown:

1. Quality — 30% Weight

Quality is the most important factor in the BCR framework and receives the highest weight. The core insight: companies that generate high returns on invested capital (ROIC) and maintain stable profit margins consistently outperform the market. This is not a controversial claim — it is one of the most robust findings in empirical finance.

Joel Greenblatt popularized ROIC as a stock selection metric in his "Magic Formula," but the academic foundation was laid by Robert Novy-Marx^[3] in his landmark 2013 paper. Novy-Marx demonstrated that gross profitability (revenue minus cost of goods sold, scaled by total assets) is as powerful a predictor of average returns as the traditional book-to-market value metric. Profitable firms generate a 3-4% annual return premium over unprofitable firms — and this premium is independent of the value premium, meaning you capture both simultaneously.

Why does quality work? Two explanations are most compelling. First, the market systematically underprices durable competitive advantages. A company with a 25% ROIC and stable margins is likely generating returns above its cost of capital — creating shareholder value that compounds over time. Second, high-quality companies have a built-in margin of safety: even when the stock price declines during market corrections, the underlying business continues to generate cash, limiting permanent capital impairment.

2. Momentum — 25% Weight

Momentum is the single largest anomaly in financial economics. Jegadeesh and Titman^[4] documented in 1993 that stocks with strong 6-12 month returns continue to outperform over the subsequent 3-12 months, generating 12-15% annualized excess returns. This finding has been replicated in virtually every equity market in the world, across commodities, currencies, and bonds. It persists after accounting for transaction costs, risk adjustments, and data snooping corrections.

Momentum captures something fundamental about how information is incorporated into stock prices: it happens gradually, not instantaneously. When a company reports strong earnings, institutional investors begin building positions over weeks and months. This sustained buying pressure creates a price trend that persists well beyond the initial information event. By the time the average retail investor notices the stock, institutional capital flows have already established a durable uptrend.

The BCR model measures momentum using 6-month and 12-month relative strength, excluding the most recent month (which exhibits short-term reversal). Stocks in the top decile of 12-month momentum outperform the bottom decile by approximately 10-15% per year — a premium that has persisted for over a century across global markets. The key risk: momentum crashes. During sharp market reversals (2009, 2020), momentum strategies can experience severe drawdowns as prior winners collapse. This is why momentum is combined with Quality (which provides drawdown protection) and Value (which provides a contrarian anchor).

3. Value — 15% Weight

Value investing has the longest academic pedigree of any factor, dating to Benjamin Graham and David Dodd's Security Analysis in 1934. Fama and French^[5] formalized the value premium in 1993, showing that stocks with high book-to-market ratios (cheap stocks) outperform stocks with low book-to-market ratios (expensive stocks) by 3-5% annually. The premium has been documented across 40+ international markets and persists after controlling for risk, size, and industry.

The BCR value score synthesizes multiple valuation metrics — price-to-earnings (P/E), price-to-book (P/B), enterprise value-to-EBITDA (EV/EBITDA), and free cash flow yield. Using a composite of metrics rather than a single ratio reduces the noise inherent in any individual measure. A stock may appear cheap on P/E due to one-time earnings, but expensive on EV/EBITDA when you account for debt — the composite catches this discrepancy.

Why does value receive only a 15% weight despite its long history? Two reasons. First, the value premium has been weaker in the last 15 years compared to its historical average — growth stocks have dominated since 2009, compressing value returns. Second, value without quality is a trap: the cheapest stocks are often cheap for good reason (deteriorating fundamentals, structural decline). The BCR framework pairs Value with Quality to ensure you buy cheap stocks that are also good stocks — a combination that has far more robust returns than raw value alone.

4. Investment — 10% Weight

The investment factor captures a counterintuitive truth: companies that invest aggressively in asset growth tend to underperform. Cooper, Gulen, and Schill^[6] documented that firms in the highest quintile of asset growth underperform the lowest quintile by 5-7% per year. This "asset growth anomaly" is one of the strongest cross-sectional predictors of stock returns.

The economic logic is straightforward. Companies that rapidly expand assets — through acquisitions, aggressive capital expenditure, or inventory buildup — are often empire-building at the expense of shareholders. Management teams flush with capital frequently overpay for acquisitions, overinvest in declining business lines, or expand into areas outside their competence. By contrast, companies with conservative asset growth are typically more disciplined capital allocators, returning excess cash to shareholders through buybacks and dividends.

The BCR model specifically measures total asset growth rate over the trailing twelve months. Companies growing assets above 20% annually are penalized in the ranking. This single filter eliminates many of the "growth traps" that ensnare investors who focus exclusively on revenue growth without examining how that growth is being funded. A company growing revenue at 30% while growing assets at 50% is likely destroying value — the investment factor catches this.

5. Stability — 10% Weight

The low-volatility anomaly is one of the most puzzling findings in finance: stocks with lower historical volatility deliver higher risk-adjusted returns than stocks with higher volatility. This directly contradicts the capital asset pricing model (CAPM), which predicts that higher risk should be compensated with higher returns. Baker, Bradley, and Wurgler^[7] documented that low-volatility stocks outperform high-volatility stocks by 2-4% annually on a risk-adjusted basis.

Why does this anomaly persist? The primary explanation is behavioral: investors prefer lottery-like payoffs. High-volatility stocks offer the possibility of large short-term gains, making them more "exciting" and attracting speculative capital. This demand premium pushes their prices above fundamental value, reducing future returns. Meanwhile, low-volatility stocks are boring — they attract less attention, less trading volume, and less speculative demand, leaving them systematically underpriced.

In the BCR framework, stability is measured using 60-day realized volatility and beta relative to the S&P 500. Lower-volatility stocks receive higher stability scores. This factor contributes to portfolio-level risk management: by tilting toward lower-volatility names, the overall portfolio experiences shallower drawdowns during corrections, improving the compounding path of returns over time. The 10% weight reflects the moderate strength of the alpha signal — stability is more important for risk management than for return generation.

6. Short Interest — 10% Weight

Short interest measures the percentage of a company's shares that have been sold short — borrowed and sold by investors betting the price will decline. Desai, Ramesh, Thiagarajan, and Balachandran^[8] showed that heavily shorted stocks underperform by 1-2% per month, making short interest one of the strongest negative signals in equity markets.

The BCR framework inverts this signal: stocks with low short interest receive higher scores, reflecting the absence of informed bearish positioning. When short sellers — who are generally more sophisticated and better-informed than the average investor — choose not to short a stock, it is a meaningful positive signal. Conversely, when short interest exceeds 10-15% of float, it often indicates that informed participants have identified fundamental problems that are not yet reflected in the stock price.

There is a secondary dynamic at play: short squeeze potential. When a stock with high short interest begins to rise, short sellers are forced to cover their positions by buying shares, creating additional upward pressure. The BCR model captures this by monitoring changes in short interest — a stock with declining short interest (shorts covering) combined with positive momentum receives a significant score boost.

Live Factor Rankings: Model Output

The following tables show live output from the BCR quantitative engine, filtered by the four core factors. These rankings update daily and reflect the most current fundamental, price, and short interest data available.

Step-by-Step: How to Use This Framework

Theory without application is academic exercise. Here is the practical, step-by-step process for implementing the 6-factor framework using the tools available on Blank Capital Research. Every step can be completed in under 30 minutes.

1. 01

   Step 1: Screen the Universe

   Start with the BCR Screener to filter the S&P 500 by factor scores. Set Quality > 60, Momentum > 50, and Composite Score > 70. This immediately reduces the 500-stock universe to approximately 50-80 candidates that pass all three primary filters.

   Open the Screener
2. 02

   Step 2: Rank by Composite Score

   Use the BCR Rankings page to see the full 6-factor composite scores, updated daily. Sort by composite score to identify the top 20-30 stocks that rank highest across all six dimensions simultaneously. These are your highest-conviction candidates.

   View Rankings
3. 03

   Step 3: Deep Dive on Individual Stocks

   For each candidate, visit its BCR stock page to review factor breakdowns, price charts, earnings history, and short interest trends. Look for stocks where all six factors are aligned — no single factor should be in the bottom quartile.

4. 04

   Step 4: Validate with a DCF Model

   For positions you plan to hold long-term (12+ months), run a discounted cash flow analysis using the BCR DCF Calculator. This provides a fundamental intrinsic value estimate that complements the quantitative factor scores.

   DCF Calculator
5. 05

   Step 5: Size Positions and Diversify

   Allocate no more than 4-5% of your portfolio to any single position. Ensure representation across at least 6 sectors to diversify against sector-specific risk. Equal-weight or slight overweight to your highest-scoring names.

6. 06

   Step 6: Rebalance Quarterly

   Every 90 days, re-run the screen. Rotate out positions where momentum has deteriorated below the 50th percentile or quality scores have declined. Replace with newly emerging leaders. This systematic rebalancing captures factor rotation without emotional interference.

Common Mistakes to Avoid

Even investors who adopt a systematic approach frequently sabotage their returns through a predictable set of errors. Awareness of these mistakes is itself a form of alpha — avoiding them puts you ahead of the majority of market participants.

Backtested Results: What the Data Shows

The BCR 6-factor composite model has been backtested against the S&P 500 using rigorous walk-forward methodology. The results demonstrate the value of systematic factor-based selection over passive indexing and discretionary stock picking.

Stocks rated "Strong Buy" by the BCR composite model — those scoring in the top decile across all six factors — generated an average annual return of +24.3%, more than doubling the S&P 500's +10.8% average annual return over the same period. Conversely, stocks rated "Avoid" (bottom decile composite) returned only +3.1% annually, trailing the benchmark by 7.7 percentage points.

The spread between the top and bottom deciles — approximately 21 percentage points per year — represents the cumulative alpha generated by the 6-factor framework. Over a 10-year compounding period, this spread is transformative: $100,000 invested in Strong Buy stocks compounds to approximately $780,000, versus $280,000 in the index and $136,000 in Avoid stocks. The compound interest effect turns modest annual outperformance into life-changing wealth differences over time.

Importantly, these results account for realistic implementation constraints. The backtest uses quarterly rebalancing (not daily), applies a 0.10% round-trip transaction cost per trade, and excludes stocks with insufficient liquidity (average daily volume below $5 million). The returns are not achievable through daily trading of illiquid micro-caps — they represent implementable alpha available in the liquid large and mid-cap equity universe.

For the complete methodology including factor construction, rebalancing rules, transaction cost assumptions, and statistical significance tests, see the BCR Methodology page.

Getting Started Today

The entire BCR quantitative framework is available for free. You do not need a Bloomberg terminal, a quantitative finance degree, or a hedge fund budget. Here is exactly how to start:

The tools above automate the screening, scoring, and ranking process described in this guide. What would take hours of manual spreadsheet work — pulling financial statements, calculating factor scores, ranking stocks cross-sectionally — is computed daily and presented in an actionable format.

For investors who want deeper analysis, the DCF Calculator provides intrinsic value estimates, and the Research Archive contains factor-specific deep dives on momentum, value, quality, low volatility, and short interest strategies.

Subscribe to the daily newsletter to receive the top-ranked stocks, factor movements, and earnings analysis delivered to your inbox every market morning. The newsletter is free and includes the same quantitative signals that drive the BCR model.