← Research Library
New to these reports? Start here
  • Dotted-underlined words have a plain-English definition — hover or tap them. Every term is also on the glossary page.
  • "Null" means we found nothing, not that something broke. Most reports here are negative results, on purpose — knowing an idea doesn't work is the point.
  • Two questions get asked separately. First, is the effect real? Second, is it already priced into the betting odds? An effect can be completely real and still useless to bet on.
  • A "calibration" row is a self-check. It runs the same method on something already known to be true. If that fails, the whole report is unreliable — so it's reported alongside the findings.
  • If a confidence interval includes zero, the real effect might be nothing at all, so no claim gets made.

Strategy 007: Short-Term Reversal (1-week loser-winner)

Source

Jegadeesh (1990), Lehmann (1990) — the original short-horizon reversal papers. The mirror image of 12-1 momentum: short-term returns reverse at 1-week/1-month horizons, while medium-term (12-1) returns continue.

Thesis

Stocks that fall hard in a single week tend to bounce the following week. Mechanism: short-term selling pressure (forced liquidations, panic, tax-loss harvesting) creates temporary mispricing that gets corrected as buyers step in.

This is the most-tested microstructure-level anomaly in equities.

Formula

For each week: * Compute past 5-day return for every stock * Sort universe descending * Hold bottom decile (~10% — last week's losers), equal-weight * Hold for 1 week * Rebalance weekly

Why it might still work

  • Behavioral mechanism (overreaction to short-term news) is durable
  • Forced selling continues to exist (margin calls, ETF rebalances, redemptions)
  • Volatility is information; the rebound captures information decay

Why it might NOT work

  • Slippage is brutal for weekly turnover. Most of the academic alpha comes from <1bps profit per trade × thousands of trades.
  • HFT firms have arbitraged the obvious form away since 2010
  • Mid-week intraday reversals are even faster than weekly; this weekly version may catch the wrong side of intraweek reversal
  • In our long-only implementation, we miss the more powerful long- short version (long losers, short winners)

Implementation

  • Universe: sp500
  • Filter: bottom 10% by past-5-day return (around 50 names)
  • Hold: equal-weight, top 30 (deepest losers)
  • Rebalance: weekly

Data requirements

  • Daily prices only

Expected outcome

  • Academic literature: ~0.3-0.5%/week excess return long-short before costs; long-only is ~half that = ~+10%/yr gross
  • After realistic slippage on weekly turnover (~5% annual cost in our model): probably 0 to slightly negative alpha
  • In recent regimes (2020+), high vol has made the reversal signal bigger in absolute terms but more correlated with crash dynamics

Pass/fail criteria

  • Pass: CAGR within 2pp of SPY with lower volatility (mean-revert signal often dampens drawdowns)
  • Fail: CAGR < SPY by 3+pp, OR worse SharpeSharpe ratioReturn relative to how much it bounced around. Higher means smoother returns for the same profit. than SPY
  • Note: this is mainly a test of "does the harness catch high-turnover strategies correctly" — if it kills this one due to slippage, that's the harness working as intended.

On this page

Terms in this report

Sharpe ratio

Related

Strategy 002: 12-1 Cross-Sectional Momentum Strategy 009: Insider Cluster Buying Strategy 008: Sell-in-May Calendar Effect Strategy 006: Net-Net (Graham Deep Value) Strategy 005: VIX Term Structure Strategy 004: Piotroski F-Score Strategy 003: Post-Earnings Announcement Drift (PEAD) Strategy 001: Magic Formula (Greenblatt)

Source

backtests/docs/007_short_term_reversal.md
updated 2026-05-24 23:23