The ideas we
walked away from.

Tighter compression predicts less forward volatility, not more. The relationship is monotonic and the wrong sign — the super-tight bucket (realised vol <...

The Lucca / Moench finding is intraday — the drift is concentrated in the morning of the announcement day. With only daily close data we have three...

Three filter generations, all weak:

Real but modest signal — IS options edge was +18.6 pp at 90 DTE on the credit put spread, dropping to +3.4 pp at 30 DTE. Most of the absolute return is...

The 4-week lookback is too short. Within-sector ranking at this horizon picks up Jegadeesh 1990 short-term reversal, not momentum. The bottom quintile of...

The bounce doesn't materialize. Edge is essentially zero at every horizon we measure:

Signal exists but the magnitude is too small to support a defined-risk options structure or a meaningful stock-only position.

Hit rate is high (60-66 %) but edge over control is essentially zero - the reversion is generic market drift, not signal-specific alpha.

Edge is consistently negative across every horizon tested in our modern-era (2010+) sample.

Real signal but the magnitude is too small to support an options structure after frictions.

Edge is in the right direction (top beats bottom) but small and only marginally significant.

Edge is negative in the short term and ~zero longer term. The signal is lagging - by the time 3 firms have upgraded, the move is already in.

PEAD edge is invariant to sector context. 20-day forward returns:

The insidertransaction table in db/finance.db has only 44 purchase rows across 19 symbols. The Form-4 ingestion (scripts/buildinsider.py) hasn't been run...

Unconditional edges are tiny and often the wrong sign. All thresholds were ATR-normalised so the test scales across vol regimes, and a trend filter (10-day...

Edges are flat-to-negative across every horizon.

Edges are basically zero in both directions.

On the full US universe (197k NR7 long-breakouts, 178k short-breakouts) the long side is ~0 edge across every horizon. The short side shows a tiny +0.4 %...

Edges are ~ 0 across the full universe on both patterns. Even with a 1 % daily-move pre-filter (selects only meaningful breakout days) and ATR-normalised...

At the unconditional level the gap by itself isn't big enough to time the reversal — only ~1-3 % edge. After options friction these compress to ~0.

Even the strictest combination — all three of RSI / Williams / MFI in extreme oversold simultaneously, on a name already ≥ 30 % off its 252d high, OOS 2020+...

Edges hover around zero across every variant — including the strict "vol ≥ 5x + dd ≤ -30 % + OOS 2020+" subset (n=1,081) which is slightly negative at 5-20 d.

nearly every other "bearish" pattern in the repo is also a long signal (contrarian bounce family). Bearish divergence is the only structural signal that...

Edge is ~0 across every horizon and every variant (fresh, volume-confirmed, 200-DMA-rising).

As a stand-alone breakout-day signal the geometry buys too late — the sequence of contractions ends with a high-volume thrust day, but the day-after...

The same-day reclaim intrabar mechanic is real in microstructure, but our data is daily close — by the time we see the reclaim, the bounce has happened and...

with ntop=10 weekly, the portfolio is highly concentrated. The default turnaround weight fractionally re-ranks names but doesn't break a turnaround...

| horizon | IC | IC t | D10-D1 spread | spread t | |---------|--------|-------|---------------|----------| | 5d | -0.004 | -0.53 | -0.07 % | -0.68 | | 10d |...

it was never tested, in either direction.