Rejected thesis
Rectangle / horizontal-range breakout
Thesis (rejected): A true rectangle — ≥3 touches at horizontal
support AND ≥3 touches at horizontal resistance over the prior 60
days, range height ≤ 20 % — should add information beyond what
coil_breakout (no touch-count requirement) already finds.
Code: research/rectangle_breakout.py
Report: research/rectangle_breakout_report.json (n=5,132)
| Bucket | n | 5d edge | 10d edge | 20d edge | 40d edge |
|---|---|---|---|---|---|
| all | 5132 | -0.02 % | +0.01 % | -0.47 % | -1.51 % |
| OOS 2020+ | 1661 | +0.11 % | +0.06 % | -0.95 % | -3.34 % |
| dd ≤ -15% | 319 | -0.18 % | -0.76 % | -1.28 % | -2.30 % |
Verdict: Reject. The multi-touch requirement does not improve
on coil_breakout; if anything, the longer 60-day flat structure
selects names that are less primed (older bases that have already
been picked over) than the 10-day coil cohort.
| Sample | 5,132 |
|---|---|
| Validator | research/rectangle_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