Rejected thesis
NR7 / inside-day breakout (Toby Crabel 1990)
Thesis (rejected): A day with the narrowest range of the last 7 bars, or an inside day, is a volatility coil — when the next day breaks out of the prior bar's range by ≥ 0.5 %, the move continues for 2-5 trading days.
Code: research/nr7_breakout.py
Report: research/nr7_breakout_report.json (n=655,737 events)
Why it failed: 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 % contrarian edge at 40d — i.e. the textbook-bearish breakout of an NR7 to the downside is on average a weak long opportunity. Same effect as bear-flag and death- cross: bearish technical signals on already-selling names are mean- reverting, not continuation.
| Pattern | n | 5d edge | 20d edge | 40d edge |
|---|---|---|---|---|
| nr7_long | 197,867 | -0.07 % | -0.01 % | -0.19 % |
| nr7_short | 178,777 | +0.17 % | +0.35 % | +0.45 % |
| inside_long | 149,246 | -0.07 % | 0.00 % | -0.17 % |
| inside_short | 129,847 | +0.13 % | +0.31 % | +0.39 % |
Verdict: Skip. Likely needs intraday data to work the way Crabel described it. The short-side mean-reversion is already captured (in much bigger size) by the drawdown-bounce cohort.
| Sample | 655,737 |
|---|---|
| Validator | research/nr7_breakout.py |
| Verdict | Rejected |
Every result here is reproducible from the script named above. Reports are in the repository.
- Volatility squeeze
- Pre-FOMC drift
- Uptrend pullback
- Momentum continuation
- Industry-relative momentum (at 4-week lookback)
- Earnings-reaction reversal (anti-PEAD)
- Beat-persistence pre-earnings drift
- Sector ETF mean-reversion
- 52-week-high breakout
- Volume accumulation (above-200DMA +...
- Cross-sector momentum
- Analyst-upgrade cluster