Home/Systematic Trading/Factor Investing & Quant Equity

Factor Investing & Quant Equity

factor-investingLevel 2 — Intermediate

Unlocks

Factor ConstructionFactor CrowdingFactor Risk Model ConstructionMulti-Manager Portfolio ManagementPortfolio Risk AttributionQuantitative CreditSystematic Fixed Income

What It Is

Systematic equity investing based on persistent, empirically documented risk premia (value, momentum, quality, low-volatility, size) that explain cross-sectional variation in stock returns. Includes both pure quant factor strategies and the application of factor analysis to improve fundamental manager portfolios.

Correct Execution

Factors are selected based on economic rationale (not only backtest fit), measured consistently, and combined in a diversified multi-factor portfolio. Practitioner controls for sector/country biases and monitors factor crowding. When applying to discretionary portfolios, factor analysis strips unintended exposures while preserving the idiosyncratic alpha the manager was actually trying to express.

Progression Levels

Diagnostic Tree

Coaching Cues

  • "Factor attribution first — then decide if your alpha is real or accidental." — before claiming any alpha is genuine, run a full factor attribution
  • "The quant's job is to strip what you didn't mean from what you did mean." — when combining fundamental and quant approaches (Omer Cedar)
  • "A new trend becomes a factor when 500+ stocks move together on the same news." — when evaluating whether a thematic trade has become a systematic risk factor (Omer Cedar)
  • "An open mind is a great thing but not so open that your brains fall out." — Cliff Asness; when tempted to abandon a well-grounded factor strategy during a drawdown
  • "Stick with it like grim death — but with as open a mind as you can." — Cliff Asness; the tension between conviction and humility in systematic investing
  • "Objective expected returns and subjective investor sentiment move in opposite directions at extremes. Trust the former." — Antti Ilmanen

Common Errors

  1. Conflating momentum and trend following: Cross-sectional momentum (rank stocks on 12-1m return, long top decile) is different from time-series trend (buy/sell an asset based on its own price path) → different return profiles, different crisis behavior.
  2. Ignoring factor crowding: Many of the documented factor premia have compressed significantly as AUM in factor strategies has grown → expected forward returns are lower than historical → account for this in forecasts.
  3. Using a single factor model for all regimes: Barra-style models are calibrated on long-run data; they do not immediately capture emerging risk factors → supplement with manager knowledge of what fundamental forces are currently operating.
  4. Applying fundamental manager signals without cleaning: Raw analyst signals contain sector bias, recency bias, and risk aversion → cleaning is mandatory before combining with a quant model.
  5. Over-discretizing process: Adding more bells and whistles to factor models because recent backtested tweaks improve fit → each addition risks data mining → the payoff from factor investing is concentrated in the first observation (stocks go up, cheap beats expensive, momentum persists) and marginal complexity yields diminishing returns.
  6. Using analyst EPS growth forecasts for capital market assumptions: Analyst consensus EPS growth is systematically ~10-20% when realistic long-run rate is ~2-3% → their forecasts are inverse market timers → use yield-based objective estimates instead.

Edges

🔑 Hidden Causal Lever

Subjective Investor Sentiment Is An Inverse Return Indicator

factor-investingfactor-investing-quant-equity →

Objective expected returns (CAPE-based, yield-based) and subjective investor sentiment move in opposite directions at extremes. Analyst consensus EPS growth forecasts run at 10-20% during bull markets against a realistic 2-3% long-run rate — they are systematically upward-biased proxies for the recent past, not the future. When subjective sentiment is highest, objective forward expected returns are lowest.

What most people do
Use analyst EPS forecasts and sentiment surveys as inputs to capital market assumptions. Extrapolate recent returns into the future.
What the best do
Use objective, yield-based measures (CAPE, earnings yield) as the primary return estimate. Treat high analyst consensus EPS forecasts as a contrary indicator for forward returns. When subjective and objective measures diverge sharply, trust the objective.
Why it's an edge: Analyst forecasts are widely used as return inputs, but they are systematically wrong at exactly the critical inflection points. Knowing this allows for contrarian tilting at market extremes.
How to exploit: Build a two-signal equity allocation dashboard: (1) current CAPE or earnings yield vs. 10-year average; (2) analyst consensus EPS growth vs. historical 2-3% norm. When both show extremes in the same direction (high sentiment + stretched valuation), reduce equity exposure. When both are at opposite extremes (low sentiment + cheap valuation), add.
Antti Ilmanen, "Understanding Return Expectations," Flirting with Models S7E21, 2025-09-15
⚡ Conventional Wisdom Is Wrong

Value Factor Failure Was A Measurement Problem, Not A Strategy Problem

factor-investingfactor-investing-quant-equity →

The 10+ year underperformance of value (2010-2021) was widely interpreted as the death of value investing. The actual cause was that traditional P/B and P/E metrics systematically mismeasure value for intangible-asset-heavy businesses (tech, platforms) — they look expensive on these metrics while actually being cheaper on economic fundamentals. The strategy is not broken; the measurement tool is.

What most people do
Conclude that value investing no longer works and reduce or eliminate value factor exposure based on decade-long underperformance.
What the best do
Update the measurement: capitalize R&D, include brand value, adjust P/B for intangibles. Combine value with quality to avoid value traps. The economic rationale for value (cheap assets eventually mean-revert) is intact; the lens needs updating.
Why it's an edge: The practitioners who correctly diagnosed the measurement problem vs. the strategy problem maintained or increased value exposure at the trough and captured the subsequent recovery.
How to exploit: Use intangibles-adjusted valuation metrics (e.g., enterprise value to adjusted EBITDA including capitalized R&D) alongside traditional P/B. Compare the factor portfolio composition — if "cheap" names are structurally weaker businesses, it's a measurement problem. Fix the measurement before abandoning the strategy.
Omer Cedar, Flirting with Models S3E11, 2021-04-10; Clayton Gillespie, Flirting with Models S7E5, 2024-02-05
🔑 Hidden Causal Lever

Most Fundamental Manager Alpha Is Unintended Factor Exposure In Disguise

factor-investingfactor-investing-quant-equity →

When fundamental managers outperform, they typically attribute it to stock selection skill. Factor attribution frequently reveals that a large portion of the outperformance came from inadvertent factor tilts (high-volatility names, sector concentrations, momentum tilts) — not from idiosyncratic insight. Stripping these unintended exposures reveals whether genuine stock selection alpha exists.

What most people do
Evaluate fundamental manager performance on total return vs. benchmark. Accept manager attribution narratives about which stock picks drove performance.
What the best do
Run factor attribution (Barra/Axioma) on the fundamental portfolio before assessing manager skill. Identify the top 3-5 unintended factor bets. Hedge or reweight to remove them. What remains is the idiosyncratic alpha the manager was actually trying to express.
Why it's an edge: Separates genuine stock selection alpha (scalable, worth paying for) from inadvertent factor exposure (can be replicated cheaply through factor ETFs). Enables accurate manager evaluation and proper compensation design.
How to exploit: For any fundamental portfolio, run monthly factor attribution. Calculate what % of total return variance is explained by unintended factor loadings. If >50%, the manager is primarily a factor exposure vehicle. Target reducing unintended factor exposure to <30% of return variance to isolate genuine idiosyncratic alpha.
Omer Cedar, Flirting with Models S3E11, 2021-04-10
💎 Elite-Only Behavior

Stick With It Like Grim Death — 99/100 Times The World Hasn't Changed

factor-investingfactor-investing-quant-equity →

When a theoretically grounded factor strategy is in extended drawdown (2-3 years of underperformance), maintaining conviction 99% of the time is the correct response. The 2019-2021 value collapse and 1998-2000 AQR underperformance both rewarded those who held. The diagnostic: check economic rationale, check crowding, check if drawdown is within historical range — if all pass, hold.

What most people do
Reduce allocation after 18-24 months of underperformance. Rotate to what's recently worked. "This time it's different."
What the best do
Run the three-question diagnostic: (1) Has the economic rationale changed? (2) Is the factor dangerously crowded? (3) Is the drawdown within the 95th percentile of historical drawdowns? If all three pass, maintain or increase allocation.
Why it's an edge: Factor strategies produce their best returns immediately after their worst drawdowns. The investor who exits at month 24 misses the recovery that rewards those who stayed. The asymmetry is enormous — but only accessible to those with structural conviction.
How to exploit: Before deploying any factor strategy, document the three-question diagnostic in writing. During drawdowns, run the diagnostic quarterly. Only reduce allocation if at least one of the three questions fails. Otherwise, hold — or increase if the drawdown has created a valuation opportunity.
From Diagnostic Tree symptom 5, Cliff Asness quote: "Stick with it like grim death — 99 out of 100 times the world hasn't changed."

Sources

  • Omer Cedar, Flirting with Models S3E11 (2021-04-10) — unintended bets, idiosyncratic alpha extraction, quant/fundamental intersection
  • Clayton Gillespie, Flirting with Models S7E5 (2024-02-05) — fundamental insights in quant equity, regime-change factor identification
  • Jay Rajamony, Flirting with Models S7E23 (2025-10-13) — modern era quant equity, beyond traditional factors
  • Cliff Asness, "But Not So Open Your Mind Falls Out," Flirting with Models S3E13 (2021-04-10) — sticking with factors through drawdowns, value spread as monitoring tool, within-industry vs. cross-industry factor construction
  • Antti Ilmanen, "Understanding Return Expectations," Flirting with Models S7E21 (2025-09-15) — objective vs. subjective return expectations, analyst over-optimism, CAPE as valuation anchor, rear-view mirror bias