FoxAlgo → Research → Academic Audit → Reversal During Earnings-Announcements

Academic audit

Failedequity event

Reversal During Earnings-Announcements

The three-gate gauntlet · genuine only if it clears all three and survives adversarial refutation
Gate 1
Survivorship-free
free
clean panel
Eliminated here
Gate 2
Placebo ≥ P95
P26.8
outranked ~54 of 200 baskets
Gate 3
Cost-aware net
RF -0.79
net-negative after costs
Failed
Worst 12-month leg (RF)-1.00
−1.00 floor0
Every strategy here — winners included — loses in its worst 12 months. Depth is honest context, not the verdict.
Rejected at the luck gate — its net ranked no better than random baskets (below the P95 skill line).

The paper studies short-horizon price reversal around scheduled earnings announcements — buying stocks that dropped and selling stocks that jumped into the announcement window, betting the move partly unwinds.

What we found

In our retest the effect did not hold up. The result sits at the 26.8th percentile of random baskets, so it is indistinguishable from noise, and the risk-adjusted RF is negative. The finding is that the earnings-window reversal shows no usable edge once realistic costs and adverse selection are accounted for — the reversal that survives is dominated by trading frictions and by the fact that the names moving most into an announcement tend to be the hardest and most expensive to trade against. It is not a diversifying factor-leg; it failed.

How we tested it
2005–2026 test windowmodelled liquidity-aware costssurvivorship free
  • Tested on a survivorship-free 1077-name US common-stock panel, 2005-2026. Realistic modelled costs.
  • Placebo / robustness test: real result vs random baskets or shuffled signals (real vs the 95th percentile of random)
Source: So & Wang (2014), reversal around earnings announcements
Read the paper ↗
← The Academic Audit — all 54 studies

Research, not investment advice. “Validated” factor-legs are market-neutral diversifying building blocks with a losing worst year — none is a standalone tradeable strategy. Metrics are cost-aware and modelled (not live fills); the 2005–2026 test window is out-of-sample versus the source paper. Dollar figures are not returns and are omitted by design.