# Mean Reversion (US S&P 500)

- **Slug**: `mean-reversion-us-sp500`
- **Verdict**: rejected (no_edge)
- **Asset class**: equity
- **Categories**: cross-sectional, factor, mean-reversion
- **Source**: qbuntu-analysis (computed asof 2026-07-27T03:37:13)
- **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/mean-reversion-us-sp500
- **JSON**: https://qbuntu.ai/v1/strategies/mean-reversion-us-sp500
- **License**: CC-BY-4.0

## Hypothesis

Stocks deviating from fair value (measured by z-score, RSI, Bollinger, or MRS) revert to the mean. LONG oversold, SHORT overbought. Applied to S&P 500 constituents.

## Why it didn't work

Across 1024 grid combinations, median Sharpe=-0.38, best=0.97. Primary failure mode: no_edge.

## Performance summary

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

| Metric | Value |
|---|---|
| Sharpe (median) | -0.3843 |
| Max drawdown | -17.9% |
| p-value | 0.35 |
| Backtest period | 2022-02-01 to 2026-06-01 |

Sharpe distribution across 1024 grid runs: median -0.38, best 0.97 (best-case — not representative), worst -1.97, p25 -0.79, p75 0.

## 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: GSPC SMA20/SMA60 three-class: UP (close>SMA20 AND SMA20>SMA60), DOWN (close<SMA20 AND SMA20<SMA60), FLAT (else)
- Transaction cost bps: 5

## Target universe

S&P 500 constituents, ~503 stocks

## Runs (30)

| run_id | params | sharpe | UP | FLAT | DOWN |
|---|---|---:|---:|---:|---:|
| `r-mean-reversion-us-sp500-0000` | cost_bps=5, lookback=10, n_stocks=5, indicator="zscore" | -0.38 | -0.1 | -1.06 | 0.3 |
| `r-mean-reversion-us-sp500-0001` | cost_bps=5, lookback=10, n_stocks=5, indicator="zscore" | -0.46 | -0.22 | -1.06 | 0.3 |
| `r-mean-reversion-us-sp500-0002` | cost_bps=10, lookback=10, n_stocks=5, indicator="zscore" | -0.69 | -0.43 | -1.34 | -0.05 |
| `r-mean-reversion-us-sp500-0003` | cost_bps=10, lookback=10, n_stocks=5, indicator="zscore" | -0.72 | -0.45 | -1.34 | -0.05 |
| `r-mean-reversion-us-sp500-0012` | cost_bps=5, lookback=10, n_stocks=50, indicator="zscore" | -0.88 | -0.81 | -1.31 | -0.44 |
| `r-mean-reversion-us-sp500-0048` | cost_bps=5, lookback=10, n_stocks=5, indicator="zscore" | -0.02 | -0.44 | -0.39 | 1.25 |
| `r-mean-reversion-us-sp500-0053` | cost_bps=5, lookback=10, n_stocks=10, indicator="zscore" | 0.35 | 0.42 | -0.18 | 0.76 |
| `r-mean-reversion-us-sp500-0101` | cost_bps=5, lookback=20, n_stocks=10, indicator="zscore" | -0.06 | -0.09 | -0.47 | 0.26 |
| `r-mean-reversion-us-sp500-0149` | cost_bps=5, lookback=40, n_stocks=10, indicator="zscore" | -0.67 | -0.37 | -1.46 | -0.34 |
| `r-mean-reversion-us-sp500-0192` | cost_bps=5, lookback=60, n_stocks=5, indicator="zscore" | -0.67 | -0.66 | -1.29 | 0.64 |
| `r-mean-reversion-us-sp500-0198` | cost_bps=10, lookback=60, n_stocks=10, indicator="zscore" | -1.16 | -1.44 | -1.24 | -0.31 |
| `r-mean-reversion-us-sp500-0246` | cost_bps=10, lookback=60, n_stocks=10, indicator="zscore" | -0.2 | -0.91 | 0.09 | 1.06 |
| `r-mean-reversion-us-sp500-0294` | cost_bps=10, lookback=10, n_stocks=10, indicator="rsi" | -0.51 | -0.59 | -0.78 | -0.07 |
| `r-mean-reversion-us-sp500-0334` | cost_bps=10, lookback=20, n_stocks=50, indicator="rsi" | -1.97 | -2.57 | -1.47 | -1.53 |
| `r-mean-reversion-us-sp500-0343` | cost_bps=10, lookback=20, n_stocks=10, indicator="rsi" | -0.61 | -0.46 | -1 | -0.05 |
| `r-mean-reversion-us-sp500-0391` | cost_bps=10, lookback=40, n_stocks=10, indicator="rsi" | -1.21 | -1.47 | -1.1 | -0.87 |
| `r-mean-reversion-us-sp500-0439` | cost_bps=10, lookback=40, n_stocks=10, indicator="rsi" | -0.38 | -0.72 | -0.26 | -0.22 |
| `r-mean-reversion-us-sp500-0487` | cost_bps=10, lookback=60, n_stocks=10, indicator="rsi" | 0.22 | -0.29 | 0.15 | 1.32 |
| `r-mean-reversion-us-sp500-0512` | cost_bps=5, lookback=10, n_stocks=5, indicator="bollinger_pct" | -0.38 | -0.1 | -1.06 | 0.3 |
| `r-mean-reversion-us-sp500-0536` | cost_bps=5, lookback=10, n_stocks=20, indicator="bollinger_pct" | -0.57 | -0.38 | -0.89 | -0.67 |
| `r-mean-reversion-us-sp500-0584` | cost_bps=5, lookback=20, n_stocks=20, indicator="bollinger_pct" | -1.12 | -1.11 | -1.47 | -0.44 |
| `r-mean-reversion-us-sp500-0632` | cost_bps=5, lookback=20, n_stocks=20, indicator="bollinger_pct" | -0.44 | -0.69 | -0.69 | 0.33 |
| `r-mean-reversion-us-sp500-0680` | cost_bps=5, lookback=40, n_stocks=20, indicator="bollinger_pct" | -0.39 | -1.42 | 0.11 | 0.64 |
| `r-mean-reversion-us-sp500-0728` | cost_bps=5, lookback=60, n_stocks=20, indicator="bollinger_pct" | -0.59 | -0.5 | -1.12 | -0.16 |
| `r-mean-reversion-us-sp500-0776` | cost_bps=5, lookback=10, n_stocks=20, indicator="mrs" | 0.05 | -0.34 | -0.46 | 1.91 |
| `r-mean-reversion-us-sp500-0817` | cost_bps=5, lookback=10, n_stocks=5, indicator="mrs" | 0.97 | 0.91 | 0.56 | 1.91 |
| `r-mean-reversion-us-sp500-0825` | cost_bps=5, lookback=10, n_stocks=20, indicator="mrs" | 0.43 | 0.85 | -0.1 | 0.21 |
| `r-mean-reversion-us-sp500-0873` | cost_bps=5, lookback=20, n_stocks=20, indicator="mrs" | 0.5 | 0.36 | 0.63 | 0.54 |
| `r-mean-reversion-us-sp500-0921` | cost_bps=5, lookback=40, n_stocks=20, indicator="mrs" | -0.18 | 0.8 | -0.92 | -0.8 |
| `r-mean-reversion-us-sp500-0969` | cost_bps=5, lookback=60, n_stocks=20, indicator="mrs" | -0.39 | -0.17 | -0.95 | -0.08 |

## Facts (7)

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

- **Average monthly return correlation with other strategies in DOWN regime: 0.14**
  - Value: 0.1368 correlation
  - Context: Pairwise Pearson correlation of monthly returns across 42 strategies, filtered to DOWN regime months (N=3, N225 < SMA20 < SMA60)
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/mean-reversion-us-sp500-fact-avg-corr-down

- **Average monthly return correlation with other strategies (full period): 0.02**
  - Value: 0.019 correlation
  - Context: Pairwise Pearson correlation of monthly returns across 42 strategies, full backtest period
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/mean-reversion-us-sp500-fact-avg-corr-full

- **Best grid Sharpe ratio is 0.97 (p=0.045).**
  - Value: 0.97 sharpe_ratio
  - Context: period 2022-02-01–2026-06-01; cost=5bps
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/mean-reversion-us-sp500-fact-best-sharpe

- **Median Sharpe across 1024 grid runs is -0.38.**
  - Value: -0.38 sharpe_ratio
  - Context: Full grid of 1024 parameter combinations.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/mean-reversion-us-sp500-fact-median-sharpe

- **Median Sharpe in DOWN regime is -0.14 across 1024 runs.**
  - Value: -0.14 sharpe_ratio
  - Context: Regime=DOWN, N225 SMA20/SMA60 classification.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/mean-reversion-us-sp500-fact-regime-down

- **Median Sharpe in FLAT regime is -0.64 across 1024 runs.**
  - Value: -0.64 sharpe_ratio
  - Context: Regime=FLAT, N225 SMA20/SMA60 classification.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/mean-reversion-us-sp500-fact-regime-flat

- **Median Sharpe in UP regime is -0.49 across 1024 runs.**
  - Value: -0.49 sharpe_ratio
  - Context: Regime=UP, N225 SMA20/SMA60 classification.
  - Supports verdict: `rejected`
  - Cite: https://qbuntu.ai/cite/mean-reversion-us-sp500-fact-regime-up

## Related strategies

- same_strategy: [mean-reversion-jp-all](https://qbuntu.ai/strategies/mean-reversion-jp-all)
- same_strategy: [mean-reversion-jp-growth](https://qbuntu.ai/strategies/mean-reversion-jp-growth)

---

Cite as: qbuntu (2026). Mean Reversion (US S&P 500). https://qbuntu.ai/strategies/mean-reversion-us-sp500

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