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
ReversalHolds up

Short-term reversal (1 month)

Last month's losers beat last month's winners this month.

Jegadeesh; Lehmann · Journal of Finance · published 1990 · original sample 1934–1987 (reported ≈ +20.0%/yr, t ≈ 6)

How our version differs from the paper. Value-weighted here; the effect is far larger equal-weighted and in small caps — and it is a liquidity-provision fee, near zero after costs. 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
+7.4%
1926–2026, per year
Full-sample t
4.2
Sharpe 0.44
Post-publication
+0.0%
since 1991, t = 0.0
Decay
100%
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 (1934–1987)+13.7%16.5%0.875.6-36%
After publication (1991–2026)-2.7%23.6%0.000.0-77%
Full sample (1926–2026)+7.4%21.7%0.444.2-82%

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)+0.3%0.10.09Mkt-RF +0.40
Fama–French 3-factor-0.4%−0.20.10Mkt-RF +0.38, SMB +0.18, HML +0.11
Fama–French 5-factor+0.3%0.10.12CMA -0.49, Mkt-RF +0.34, HML +0.32
FF5 + momentum+3.8%1.70.19Mom -0.39, CMA -0.38, Mkt-RF +0.27

Rolling performance

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

By decade

1920s-16.1%/yrSharpe -0.91
1930s+51.6%/yrSharpe 1.40
1940s+14.3%/yrSharpe 0.78
1950s+11.6%/yrSharpe 1.23
1960s+6.9%/yrSharpe 0.64
1970s+12.1%/yrSharpe 0.74
1980s+8.0%/yrSharpe 0.53
1990s-4.6%/yrSharpe -0.27
2000s+2.2%/yrSharpe 0.08
2010s+4.5%/yrSharpe 0.28
2020s-7.6%/yrSharpe -0.23

The risk explanation

Little risk story — it is usually read as compensation for providing liquidity to forced sellers.

The behavioral explanation

Overreaction to recent news and to price pressure from large trades; the correction comes fast.

Why it might have decayed

Statistically enormous and statistically real, but the turnover is ~1200%/year. After realistic bid/ask and market-impact costs the profit is close to zero for anyone trading size. It is a liquidity-provision fee, not a free lunch.

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.