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
QualityFaded

Accruals

Firms whose earnings are driven by accruals rather than cash underperform.

Sloan · The Accounting Review · published 1996 · original sample 1962–1991 (reported ≈ +10.5%/yr, t ≈ 4.7)

How our version differs from the paper. Sloan's result is an equal-weighted, small-cap phenomenon — his hedge portfolio returned ~10%/yr. Value-weighting (what we have) roughly halves it. The decay figure compares our replication in-sample to our replication after publication, so it isn’t affected by this — but the level gap between “reported” and “our replication” partly is.
Our replication
+3.1%
1963–2026, per year
Full-sample t
2.8
Sharpe 0.35
Post-publication
+1.9%
since 1997, t = 0.9
Decay
57%
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 (1962–1991)+4.1%9.3%0.482.6-26%
After publication (1997–2026)+1.3%11.3%0.170.9-44%
Full sample (1963–2026)+3.1%10.1%0.352.8-44%

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)+4.0%3.20.01Mkt-RF -0.07
Fama–French 3-factor+3.9%3.10.04SMB -0.15, HML +0.06, Mkt-RF -0.04
Fama–French 5-factor+3.9%2.90.04SMB -0.16, CMA +0.12, RMW -0.07
FF5 + momentum+3.2%2.50.06SMB -0.16, CMA +0.10, RMW -0.08

Rolling performance

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

By decade

1960s+8.5%/yrSharpe 0.80
1970s+6.4%/yrSharpe 0.76
1980s+0.9%/yrSharpe 0.10
1990s+5.5%/yrSharpe 0.54
2000s+0.6%/yrSharpe 0.05
2010s+5.5%/yrSharpe 0.66
2020s-3.0%/yrSharpe -0.23

The risk explanation

Very little risk story has ever stuck — accruals do not line up with distress, leverage or covariances.

The behavioral explanation

Investors 'fixate' on the bottom-line earnings number and fail to see that the accrual component is less persistent than the cash component.

Why it might have decayed

A textbook case of post-publication decay: Green, Hand & Soliman (2011) show the accrual strategy's returns fell to roughly zero after ~2003 as hedge funds piled in and as Reg FD and Sarbanes–Oxley changed disclosure.

Long-short = top-decile minus bottom-decile of stocks univariately sorted on the characteristic. Value-weighted, US equities, monthly, gross of transaction and shorting costs. The original-paper figures are approximate and annualized for comparison.