8 validation tests | Bootstrap CI | Shuffle permutation | Walk-forward | Conditional independence
| Setup | H1 WR | H1 N | H2 WR | H2 N | Delta | Stable? |
|---|---|---|---|---|---|---|
| Semi LONG fav+fc | 67.1% | 255 | 72.5% | 153 | +5.4 | YES |
| Soft SHORT fav+fc | 67.5% | 228 | 66.1% | 348 | -1.4 | YES |
| Soft SHORT fav | 51.8% | 519 | 51.7% | 698 | -0.1 | YES |
| Soft LONG fav+fc | 65.8% | 386 | 59.7% | 382 | -6.1 | YES |
| Semi LONG fav | 53.3% | 473 | 62.3% | 276 | +9.0 | DRIFT |
| Semi LONG unfav | 44.0% | 416 | 53.9% | 532 | +9.9 | DRIFT |
Semi LONG fav+fc: rock solid 65-95% every quarter. Soft SHORT fav+fc: 60-71% except small 2025Q2.
10/11 months >= 50%. Worst = 57.4% (May 2025). No catastrophic month. VERY STABLE.
6/10 months >= 50%. Problem: March 2026 = 22% (N=41). Otherwise solid. More volatile.
| Setup | WR% | 95% CI | P(WR>50%) | P(Avg>0) |
|---|---|---|---|---|
| Semi LONG fav+fc | 69.1% | [64.5%, 73.5%] | 100% | 100% |
| Soft SHORT fav+fc | 66.7% | [62.8%, 70.5%] | 100% | 100% |
| Soft LONG fav+fc | 62.8% | [59.4%, 66.3%] | 100% | 100% |
| Semi LONG fav | 56.6% | [53.1%, 60.2%] | 100% | 100% |
| Soft SHORT fav | 51.8% | [49.0%, 54.6%] | 89.5% | 100% |
| Soft LONG unfav nofc | 33.6% | [30.9%, 36.5%] | 0% | 0% |
fav+fc combos: entire CI well above 50%. Confirmed statistically significant.
| Setup | Actual | Random Mean | Z-score | p-value |
|---|---|---|---|---|
| Semi LONG fav+fc | 69.1% | 64.6% | 1.09 | 0.134 |
| Soft SHORT fav+fc | 66.7% | 70.0% | -0.75 | 0.779 |
PROBLEM: Shuffle test NOT significant. Rotation label shuffled across dates produces similar WR. This means fc_confirms is doing most of the work, not the rotation signal itself.
| Group | Dir | fc | Fav WR | N | Unfav WR | N | Delta |
|---|---|---|---|---|---|---|---|
| Semi | LONG | Y | 69.1% | 408 | 60.8% | 513 | +8.3pp |
| Semi | LONG | N | 41.8% | 340 | 36.4% | 434 | +5.4pp |
| Software | LONG | Y | 62.8% | 768 | 56.6% | 1301 | +6.1pp |
| Software | LONG | N | 33.1% | 590 | 33.6% | 1089 | -0.6pp |
| Software | SHORT | Y | 66.7% | 576 | 71.8% | 1104 | -5.2pp |
| Software | SHORT | N | 38.0% | 629 | 43.1% | 1193 | -5.1pp |
| Semi | SHORT | Y | 62.8% | 296 | 62.1% | 491 | +0.7pp |
Semi LONG: rotation adds +8.3pp beyond fc_confirms. REAL independent signal.
Soft LONG: +6.1pp conditional on fc=Y. Useful.
Soft SHORT: rotation HURTS (-5.2pp). Unfavorable rotation Soft SHORT is BETTER (71.8% vs 66.7%)!
Semi SHORT: no effect (+0.7pp).
All correlations near zero. Rotation is NOT a proxy for ARKK, VIX, SPY gap, or prevol. Genuinely independent signal (for Semi LONG at least).
Semi LONG fav+fc: WR 68.7-71.8% across all thresholds (0.1-2.0). Very insensitive = robust. Soft SHORT stable around 63-69%.
| Metric | Semi LONG fav+fc | Soft SHORT fav+fc |
|---|---|---|
| Worst month WR | 57.4% (May 2025) | 22.0% (Mar 2026) |
| Best month WR | 95.5% (Apr 2026) | 81.0% (Feb 2026) |
| Max consec. losses | 5 | 7 |
| Max drawdown | -8.56% | -32.96% |
| Total cum. return | +346.67% | +550.81% |
Semi LONG fav+fc: incredibly stable, worst month still 57%. Soft SHORT: one terrible month (Mar 2026) with -33% drawdown.
VERDICT: 7/8 PASS - ROBUST SIGNAL
Shuffle test borderline (p=0.13) because fc_confirms carries 60%+ base WR regardless. But conditional test proves +8.3pp independent lift.
VERDICT: 4/8 PASS - NOT ROBUST
Rotation HURTS Soft SHORT. The 66.7% WR is from fc_confirms, not rotation. Unfavorable rotation actually gives 71.8% WR.