# Gap Reversal (JP Growth Stocks)

- **Slug**: `gap-reversal-short-jp-growth`
- **Verdict**: rejected (regime_dependent)
- **Asset class**: equity
- **Categories**: gap, intraday, mean-reversion
- **Source**: qbuntu-analysis (computed asof 2026-07-27T03:36:21)
- **Reproducibility**: analysis-engine@bfafe7d, backtest-lib@711b855, grid tier=shallow (python 3.10.11, numpy 2.2.6, pandas 2.3.3)
- **HTML**: https://qbuntu.ai/strategies/gap-reversal-short-jp-growth
- **JSON**: https://qbuntu.ai/v1/strategies/gap-reversal-short-jp-growth
- **License**: CC-BY-4.0

## Hypothesis

Overnight gaps in Japanese Growth stocks tend to reverse within the same or next few trading days. Shorting positive gaps and buying negative gaps captures this mean-reversion effect.

## Why it didn't work

Across 1728 grid combinations, median Sharpe=-0.23, best=2.69. Primary failure mode: regime_dependent.

## Performance summary

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

| Metric | Value |
|---|---|
| Sharpe (median) | -0.2274 |
| Max drawdown | -17% |
| p-value | 0.33 |
| Backtest period | 2022-01-04 to 2026-06-09 |

Sharpe distribution across 1728 grid runs: median -0.23, best 2.69 (best-case — not representative), worst -1.8, p25 -0.58, p75 0.2.

## Methodology

- Notes: {"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: UP (close>SMA20 AND SMA20>SMA60), DOWN (close<SMA20 AND SMA20<SMA60), FLAT (else)
- 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-gap-reversal-short-jp-growth-0000` | side="SHORT", top_n=10, cost_bps=3, hold_days=1 | 2.45 | 1.58 | 1.99 | 0.18 |
| `r-gap-reversal-short-jp-growth-0001` | side="SHORT", top_n=10, cost_bps=5, hold_days=1 | 2.32 | 1.49 | 1.9 | 0.08 |
| `r-gap-reversal-short-jp-growth-0002` | side="SHORT", top_n=10, cost_bps=10, hold_days=1 | 1.99 | 1.26 | 1.67 | -0.17 |
| `r-gap-reversal-short-jp-growth-0006` | side="BOTH", top_n=10, cost_bps=3, hold_days=1 | 1.64 | 0.67 | 0.95 | 1.2 |
| `r-gap-reversal-short-jp-growth-0009` | side="SHORT", top_n=10, cost_bps=3, hold_days=1 | 2.06 | — | 1.99 | 0.18 |
| `r-gap-reversal-short-jp-growth-0018` | side="SHORT", top_n=10, cost_bps=3, hold_days=1 | 2.14 | 3.75 | 2 | 0.18 |
| `r-gap-reversal-short-jp-growth-0081` | side="SHORT", top_n=10, cost_bps=3, hold_days=5 | 0.39 | 0.02 | 0.77 | -2.04 |
| `r-gap-reversal-short-jp-growth-0091` | side="SHORT", top_n=10, cost_bps=5, hold_days=5 | 0.42 | — | 0.73 | -2.07 |
| `r-gap-reversal-short-jp-growth-0108` | side="SHORT", top_n=10, cost_bps=3, hold_days=1 | 2.69 | 2.16 | 2.99 | 1.32 |
| `r-gap-reversal-short-jp-growth-0177` | side="BOTH", top_n=10, cost_bps=3, hold_days=3 | -0.68 | -4.73 | 0.32 | -2.59 |
| `r-gap-reversal-short-jp-growth-0262` | side="SHORT", top_n=10, cost_bps=5, hold_days=2 | 0.2 | 2.85 | 0.32 | -0.79 |
| `r-gap-reversal-short-jp-growth-0324` | side="SHORT", top_n=10, cost_bps=3, hold_days=1 | 2.51 | 3.68 | 4.51 | 3.63 |
| `r-gap-reversal-short-jp-growth-0348` | side="BOTH", top_n=10, cost_bps=3, hold_days=1 | 2.17 | 1.54 | 3.61 | 2.55 |
| `r-gap-reversal-short-jp-growth-0383` | side="LONG", top_n=10, cost_bps=10, hold_days=3 | -1.8 | -8.14 | -2.93 | 1.91 |
| `r-gap-reversal-short-jp-growth-0434` | side="SHORT", top_n=20, cost_bps=10, hold_days=1 | 1.81 | 1.28 | 1.57 | -0.43 |
| `r-gap-reversal-short-jp-growth-0519` | side="BOTH", top_n=20, cost_bps=3, hold_days=5 | -0.34 | -0.6 | -0.09 | 0.21 |
| `r-gap-reversal-short-jp-growth-0604` | side="SHORT", top_n=20, cost_bps=5, hold_days=3 | 0.13 | — | 0.84 | -4.12 |
| `r-gap-reversal-short-jp-growth-0689` | side="LONG", top_n=20, cost_bps=10, hold_days=2 | -1.07 | -9.69 | -1.52 | 4.08 |
| `r-gap-reversal-short-jp-growth-0774` | side="SHORT", top_n=20, cost_bps=3, hold_days=1 | 2.27 | 9.18 | 4.79 | 3.35 |
| `r-gap-reversal-short-jp-growth-0859` | side="LONG", top_n=20, cost_bps=5, hold_days=5 | -0.42 | -6.86 | -1.61 | 6.64 |
| `r-gap-reversal-short-jp-growth-0944` | side="BOTH", top_n=50, cost_bps=10, hold_days=3 | -0.75 | -1.83 | -0.26 | -1.59 |
| `r-gap-reversal-short-jp-growth-1029` | side="LONG", top_n=50, cost_bps=3, hold_days=3 | -0.72 | -4.6 | -0.75 | 0.4 |
| `r-gap-reversal-short-jp-growth-1114` | side="BOTH", top_n=50, cost_bps=5, hold_days=2 | -0.24 | -2.66 | -0.11 | 2.48 |
| `r-gap-reversal-short-jp-growth-1199` | side="SHORT", top_n=50, cost_bps=10, hold_days=1 | 2.01 | — | 4.21 | 2.98 |
| `r-gap-reversal-short-jp-growth-1284` | side="BOTH", top_n=50, cost_bps=3, hold_days=5 | 0.06 | -6.82 | -0.32 | 5.36 |
| `r-gap-reversal-short-jp-growth-1296` | side="SHORT", top_n=100, cost_bps=3, hold_days=1 | 1.48 | 1.6 | 1.78 | -2.49 |
| `r-gap-reversal-short-jp-growth-1370` | side="SHORT", top_n=100, cost_bps=10, hold_days=3 | -0.19 | 2.57 | 0.57 | -5.18 |
| `r-gap-reversal-short-jp-growth-1455` | side="BOTH", top_n=100, cost_bps=3, hold_days=2 | -0.33 | -3.64 | -0.42 | 1.11 |
| `r-gap-reversal-short-jp-growth-1540` | side="SHORT", top_n=100, cost_bps=5, hold_days=2 | 0.12 | 0.07 | 0.72 | -2.56 |
| `r-gap-reversal-short-jp-growth-1625` | side="LONG", top_n=100, cost_bps=10, hold_days=1 | 0.04 | -0.79 | 0.74 | -1.29 |

## Facts (5)

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

- **Best grid Sharpe ratio is 2.69 (p=1.6e-08).**
  - Value: 2.69 sharpe_ratio
  - Context: period 2022-01-04–2026-06-09; cost=3bps; side=SHORT
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/gap-reversal-short-jp-growth-fact-best-sharpe

- **Median Sharpe across 1728 grid runs is -0.23.**
  - Value: -0.23 sharpe_ratio
  - Context: Full grid of 1728 parameter combinations.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/gap-reversal-short-jp-growth-fact-median-sharpe

- **Median Sharpe in DOWN regime is 0.33 across 1728 runs.**
  - Value: 0.33 sharpe_ratio
  - Context: Regime=DOWN, N225 SMA20/SMA60 classification.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/gap-reversal-short-jp-growth-fact-regime-down

- **Median Sharpe in FLAT regime is 0.10 across 1728 runs.**
  - Value: 0.1 sharpe_ratio
  - Context: Regime=FLAT, N225 SMA20/SMA60 classification.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/gap-reversal-short-jp-growth-fact-regime-flat

- **Median Sharpe in UP regime is -1.93 across 1728 runs.**
  - Value: -1.93 sharpe_ratio
  - Context: Regime=UP, N225 SMA20/SMA60 classification.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/gap-reversal-short-jp-growth-fact-regime-up

## Related strategies

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

---

Cite as: qbuntu (2026). Gap Reversal (JP Growth Stocks). https://qbuntu.ai/strategies/gap-reversal-short-jp-growth

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