Anomaly Replication Lab
  • The zoo
  • Screener
  • Multiple testing
  • Correlations
  • Composite
  • Methodology
  • Guides
  • Resources
Sign in
Anomaly Replication LabSign in
  • The zoo
  • Screener
  • Multiple testing
  • Correlations
  • Composite
  • Methodology
  • Guides
  • Resources

Educational replication, not advice. Long-short returns are decile spreads or published factor series from the Kenneth R. French Data Library and AQR, value-weighted, US equities, and gross of transaction and shorting costs — which for high-turnover signals like short-term reversal is the whole story. Reproducing a premium in-sample is not evidence it will persist. Past performance does not predict future results.

© 2026 Austin Levenberg. All rights reserved. This site’s code, design, written analysis, and computed results are not licensed for reuse or redistribution — terms of use · privacy.

Data: Kenneth R. French Data Library, AQR Capital Management public data sets.

Part of austinlev.dev — four research tools, and a note on where this is headed.

← the zoo
QualityHolds up

Quality minus junk

Profitable, growing, safe, well-managed firms beat 'junk' — and the market underpays for quality.

Asness, Frazzini & Pedersen · Review of Accounting Studies · published 2013 · original sample 1957–2016 (reported ≈ +5.0%/yr, t ≈ 5.9)

Our replication
+3.7%
1957–2026, per year
Full-sample t
3.6
Sharpe 0.49
Post-publication
+3.3%
since 2014, t = 1.1
Decay
23%
of the premium is gone

The whole history

Growth of $1 in the long-short strategy, gross of trading and shorting costs. The dashed line marks the year the paper was published — everything to its right is out-of-sample.

$1 in the long-short strategy, gross of costs. Grey = the sample the paper had; blue = everything since.

Before and after publication

The comparison the site is built around.

WindowReturn/yrVolSharpet-statMax DD
Paper's sample (1957–2016)+4.1%7.4%0.583.8-28%
After publication (2014–2026)+2.8%10.7%0.311.1-36%
Full sample (1957–2026)+3.7%8.1%0.493.6-36%

Against the factor models

Is there a risk-adjusted alpha left once you control for the standard factors? Alpha is annualized; t-stats use Newey–West standard errors.

ModelAlpha/yrAlpha tR²Biggest loadings
CAPM (market only)+6.1%6.10.26Mkt-RF -0.28
Fama–French 3-factor+6.7%7.00.37SMB -0.26, Mkt-RF -0.24, HML -0.08
Fama–French 5-factor+4.0%5.60.68RMW +0.64, Mkt-RF -0.20, HML -0.17
FF5 + momentum+3.5%5.30.69RMW +0.63, Mkt-RF -0.19, HML -0.14

Rolling performance

Trailing 3-year Sharpe ratio. Long stretches below zero are common.

By decade

1950s+3.2%/yrSharpe 0.76
1960s+2.1%/yrSharpe 0.42
1970s+1.4%/yrSharpe 0.19
1980s+6.7%/yrSharpe 1.31
1990s+5.2%/yrSharpe 0.82
2000s+6.5%/yrSharpe 0.56
2010s+4.3%/yrSharpe 0.58
2020s+0.4%/yrSharpe 0.03

The risk explanation

Hard to argue quality firms are riskier — that is the point. AFP frame the small premium as evidence the market only partially prices quality.

The behavioral explanation

Investors chase glamour and story stocks; safe compounders are 'boring' and left slightly cheap.

Why it might have decayed

Broad, low-turnover, and a clean hedge to value — it tends to do well exactly when value does badly. Out-of-sample it has held, though the premium is modest.

Long-short = AQR's published factor series. Value-weighted, US equities, monthly, gross of transaction and shorting costs. The original-paper figures are approximate and annualized for comparison.