short-term mean reversion / overbought

Intraday reversal short

RedundantArchived · short · 2014-2026

Pre-registered 2026-10-23 (commit 6910cb5) · 1 run · data snapshot 2026-09-19T090800Z

Prior: The original 2026 ficha never wrote down a mechanism — it was a pattern kept because it backtested well in May, one of the "tier 1" shorts of a wider search over the same data.

Specification

Signal, universe, exits and sizing are withheld because they are close to a strategy that is still in use.

Gates

Gate 7 — Jensen alphapass

After regressing out market exposure, is there alpha left?

t ≥ 2.13 (Bonferroni, 3 shorts measured together), net of 20%/yr borrow

t (HAC)+4.57alpha / trade %+0.38beta+1.09trades16,489

No ranking factor was ever declared, so on days when the daily cap binds the engine picks trades in alphabetical order — declared as a limitation, not fixed, to keep the definition intact.

Gate 8 — Capture-preserved alphapass

Does the subset of trades a real portfolio can actually take keep the alpha?

t_cap ≥ 0.65 (scaled) and ratio ≥ 0.50

t captured+1.70capture ratio+0.74

Gate F — Frictionpass

Is the alpha per trade large enough to pay for its own execution?

alpha/trade ≥ 0.30% net of borrow

alpha / trade %+0.38

Stabilitypass

Is the alpha present in both halves of history?

t > 0 in both halves

t 2014 2019+1.25t 2020 2026+4.43

Own alphafail

Is the alpha new, or already captured by a strategy that passed?

t ≥ 2.13 on the trades that do not overlap the surviving short, n ≥ 100

t own+0.74n own9,815overlap pct+40.50

Why it died

Real full-history alpha (t +4.57) that passes every other gate, but all of its alpha sits inside the trades of the lab's surviving short strategy, whose details are not published. The 9,815 trades it does not share with that strategy give t +0.74.

What it taught

Three overbought-short definitions written separately in May — different triggers, different horizons — converged on the same core of trades, and outside that core there is nothing. That is evidence for the surviving strategy capturing a mechanism rather than an accident of its definition, and it is why the lab now asks every new short where its alpha lives, not just whether it has any.

Verdict history

2026-05-27Archived at Gate 8: t_captured 1.03 (engine alpha t +12.50), under the capture formula and simulator later found broken Not capturable
2026-09-14Re-examined on unseen data (2026-05-27 to 2026-09-11), with the missing ranking factor added as a bug fix: n 499, t +1.03 vs 2.128 needed, capture ratio 4.59 No alpha
2026-10-23Full-history re-measurement, pre-registered, original definition: passes Gates 7, 8, F and stability; own-alpha t +0.74 Redundant