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
Low riskGone

Total volatility

Low-total-volatility stocks beat high-volatility stocks risk-adjusted.

Haugen & Baker; Blitz & van Vliet · Journal of Portfolio Management · published 2007 · original sample 1986–2006 (reported ≈ +6.0%/yr, t ≈ 2.6)

How our version differs from the paper. Value-weighted; the low-volatility literature usually equal-weights or builds minimum-variance portfolios, which look better. 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
-0.1%
1963–2026, per year
Full-sample t
1.1
Sharpe 0.15
Post-publication
-1.1%
since 2008, t = 0.1
Decay
116%
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 (1986–2006)+2.1%31.1%0.231.0-85%
After publication (2008–2026)-6.7%33.1%-0.03−0.1-90%
Full sample (1963–2026)-0.1%28.8%0.151.1-92%

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)+12.0%3.90.34Mkt-RF -1.08
Fama–French 3-factor+10.1%4.00.60SMB -1.32, Mkt-RF -0.74, HML +0.70
Fama–French 5-factor+3.6%1.70.71RMW +1.27, SMB -1.01, CMA +0.85
FF5 + momentum+2.3%1.00.72RMW +1.26, SMB -1.01, CMA +0.81

Rolling performance

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

By decade

1960s-9.8%/yrSharpe -0.45
1970s+7.2%/yrSharpe 0.31
1980s+19.5%/yrSharpe 1.01
1990s-2.7%/yrSharpe -0.10
2000s+10.4%/yrSharpe 0.25
2010s+5.2%/yrSharpe 0.24
2020s-10.0%/yrSharpe -0.24

The risk explanation

Same leverage-constraint story as low beta — total-vol and beta sorts pick up largely the same stocks.

The behavioral explanation

Lottery demand plus benchmark-relative thinking: managers judged against an index won't hold boring low-vol names even when they should.

Why it might have decayed

'Low-vol' became a marketed product category (min-vol ETFs) after ~2011, which arguably crowded the trade and led to a sharp 2016–2020 underperformance.

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.