# Short-Term Reversal Regime-Gated (JP Growth Stocks)

- **Slug**: `short-term-reversal-regime-gated-jp-growth`
- **Verdict**: marginal (no_edge)
- **Asset class**: equity
- **Categories**: mean-reversion, regime, reversal
- **Source**: qbuntu-analysis (computed asof 2026-09-24T02:22:16)
- **Reproducibility**: analysis-engine@1455e77, backtest-lib@17a7674, grid tier=shallow (python 3.12.10, numpy 2.4.2, pandas 3.0.0)
- **HTML**: https://qbuntu.ai/strategies/short-term-reversal-regime-gated-jp-growth
- **JSON**: https://qbuntu.ai/v1/strategies/short-term-reversal-regime-gated-jp-growth
- **License**: CC-BY-4.0

## Hypothesis

Short-term reversal confined to FLAT regime removes UP-regime drag (UP=-0.11) and achieves positive median Sharpe. Phase 1 showed overall median 0.30 (marginal) with FLAT=1.77 but UP=-0.11.

## Why it didn't work

Across 81 grid combinations, median Sharpe=0.46, best=2.65. Primary failure mode: no_edge.

## Performance summary

Headline figures are the **median across the parameter grid**, not the best run.

| Metric | Value |
|---|---|
| Sharpe (median) | 0.4635 |
| Max drawdown | -5% |
| p-value | 0.64 |
| Backtest period | 2023-07-03 to 2024-06-24 |

Sharpe distribution across 81 grid runs: median 0.46, best 2.65 (best-case — not representative), worst -1.04, p25 0.09, p75 1.

## Methodology

- Notes: {"validation": "walk-forward (3 folds, 2022-01-01 onward)", "walk_forward_definition": {"type": "single_pass", "description": "Full period backtest without walk-forward split", "verdict_basis": "full_period"}}
- Slippage model: open-on-open + variable bps fixed
- Position sizing: equal_weight
- Regime definition: N225 SMA20/SMA60 three-class
- Transaction cost bps: 5

## Target universe

Tokyo Stock Exchange Growth Market, ~495 stocks, tv >= 30M JPY

## Runs (30)

| run_id | params | sharpe | UP | FLAT | DOWN |
|---|---|---:|---:|---:|---:|
| `r-short-term-reversal-regime-gated-jp-growth-0000` | lookback=3, n_positions=10, regime_gate="ALL" | 0.51 | -0.35 | 1 | 0.31 |
| `r-short-term-reversal-regime-gated-jp-growth-0001` | lookback=3, n_positions=10, regime_gate="ALL" | 1.28 | 0.67 | 2.65 | 0.49 |
| `r-short-term-reversal-regime-gated-jp-growth-0004` | lookback=3, n_positions=15, regime_gate="ALL" | 0.38 | 0.53 | 1.08 | -0.9 |
| `r-short-term-reversal-regime-gated-jp-growth-0006` | lookback=3, n_positions=20, regime_gate="ALL" | 0.17 | -1.34 | 0.92 | 1.23 |
| `r-short-term-reversal-regime-gated-jp-growth-0008` | lookback=3, n_positions=20, regime_gate="ALL" | -0.05 | -0.14 | -0.46 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0011` | lookback=5, n_positions=10, regime_gate="ALL" | -0.04 | -0.67 | 1.53 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0014` | lookback=5, n_positions=15, regime_gate="ALL" | -0.15 | -0.31 | 0.25 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0017` | lookback=5, n_positions=20, regime_gate="ALL" | -0.45 | -0.6 | 0.09 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0018` | lookback=10, n_positions=10, regime_gate="ALL" | -0.49 | -1.12 | -0.23 | 0.05 |
| `r-short-term-reversal-regime-gated-jp-growth-0021` | lookback=10, n_positions=15, regime_gate="ALL" | -0.13 | -0.09 | -0.04 | -0.5 |
| `r-short-term-reversal-regime-gated-jp-growth-0024` | lookback=10, n_positions=20, regime_gate="ALL" | -0.5 | -0.32 | -0.27 | -1.55 |
| `r-short-term-reversal-regime-gated-jp-growth-0027` | lookback=3, n_positions=10, regime_gate="FLAT-only" | 1 | — | 1 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0028` | lookback=3, n_positions=10, regime_gate="FLAT-only" | 2.65 | — | 2.65 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0029` | lookback=3, n_positions=10, regime_gate="FLAT-only" | 1.53 | — | 1.53 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0032` | lookback=3, n_positions=15, regime_gate="FLAT-only" | 0.98 | — | 0.98 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0035` | lookback=3, n_positions=20, regime_gate="FLAT-only" | -0.46 | — | -0.46 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0038` | lookback=5, n_positions=10, regime_gate="FLAT-only" | 1.53 | — | 1.53 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0041` | lookback=5, n_positions=15, regime_gate="FLAT-only" | 0.25 | — | 0.25 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0044` | lookback=5, n_positions=20, regime_gate="FLAT-only" | 0.09 | — | 0.09 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0047` | lookback=10, n_positions=10, regime_gate="FLAT-only" | 2.23 | — | 2.23 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0050` | lookback=10, n_positions=15, regime_gate="FLAT-only" | 2.21 | — | 2.21 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0052` | lookback=10, n_positions=20, regime_gate="FLAT-only" | -1.04 | — | -1.04 | — |
| `r-short-term-reversal-regime-gated-jp-growth-0054` | lookback=3, n_positions=10, regime_gate="FLAT+DOWN" | 0.9 | — | 1 | 0.31 |
| `r-short-term-reversal-regime-gated-jp-growth-0057` | lookback=3, n_positions=15, regime_gate="FLAT+DOWN" | 0.44 | — | 0.74 | -1.23 |
| `r-short-term-reversal-regime-gated-jp-growth-0060` | lookback=3, n_positions=20, regime_gate="FLAT+DOWN" | 0.96 | — | 0.92 | 1.23 |
| `r-short-term-reversal-regime-gated-jp-growth-0063` | lookback=5, n_positions=10, regime_gate="FLAT+DOWN" | -0.59 | — | -0.13 | -2.71 |
| `r-short-term-reversal-regime-gated-jp-growth-0066` | lookback=5, n_positions=15, regime_gate="FLAT+DOWN" | 0.14 | — | 0.37 | -0.95 |
| `r-short-term-reversal-regime-gated-jp-growth-0069` | lookback=5, n_positions=20, regime_gate="FLAT+DOWN" | 0.56 | — | 0.91 | -1.72 |
| `r-short-term-reversal-regime-gated-jp-growth-0072` | lookback=10, n_positions=10, regime_gate="FLAT+DOWN" | -0.17 | — | -0.23 | 0.05 |
| `r-short-term-reversal-regime-gated-jp-growth-0075` | lookback=10, n_positions=15, regime_gate="FLAT+DOWN" | -0.15 | — | -0.04 | -0.5 |

## Facts (5)

Each fact has a permanent citation URL. AI agents quoting qbuntu should use these.

- **Best grid Sharpe ratio is 2.65 (p=0.072).**
  - Value: 2.65 sharpe_ratio
  - Context: period 2023-07-03–2024-06-24; cost=5.0bps
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/short-term-reversal-regime-gated-jp-growth-fact-best-sharpe

- **Median Sharpe across 81 grid runs is 0.46.**
  - Value: 0.46 sharpe_ratio
  - Context: Full grid of 81 parameter combinations.
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/short-term-reversal-regime-gated-jp-growth-fact-median-sharpe

- **Median Sharpe in DOWN regime is -0.34 across 81 runs.**
  - Value: -0.34 sharpe_ratio
  - Context: Regime=DOWN, N225 SMA20/SMA60 classification.
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/short-term-reversal-regime-gated-jp-growth-fact-regime-down

- **Median Sharpe in FLAT regime is 0.83 across 81 runs.**
  - Value: 0.83 sharpe_ratio
  - Context: Regime=FLAT, N225 SMA20/SMA60 classification.
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/short-term-reversal-regime-gated-jp-growth-fact-regime-flat

- **Median Sharpe in UP regime is -0.19 across 81 runs.**
  - Value: -0.19 sharpe_ratio
  - Context: Regime=UP, N225 SMA20/SMA60 classification.
  - Supports verdict: `marginal`
  - Cite: https://qbuntu.ai/cite/short-term-reversal-regime-gated-jp-growth-fact-regime-up

## Related strategies

- same_strategy: [short-term-reversal-jp-growth](https://qbuntu.ai/strategies/short-term-reversal-jp-growth)

---

Cite as: qbuntu (2026). Short-Term Reversal Regime-Gated (JP Growth Stocks). https://qbuntu.ai/strategies/short-term-reversal-regime-gated-jp-growth

Visit qbuntu.ai — the honest scorecard for quant strategies (live, promising, marginal, rejected).