mean-reversion-screener

name	mean-reversion-screener
description	Screen for oversold stocks with high mean reversion probability using RSI, Bollinger Band position, and distance from moving averages. Identifies quality stocks in Stage 2 uptrends that have pulled back to support and show statistical reversion setups. Use when user asks for oversold stocks, mean reversion setups, pullback entries, RSI dip buys, or stocks bouncing off support levels.

Mean Reversion Screener

Find oversold quality stocks in uptrends showing statistically significant pullbacks with high reversion probability.

When to Use

• User asks for oversold stocks or pullback buy setups
• User wants RSI dip entries in uptrending stocks
• User asks for Bollinger Band lower-band touches with reversal signals
• User wants to buy dips in quality names that have pulled back to support

Prerequisites

• FMP API key (FMP_API_KEY environment variable or --api-key)
• Free tier (250 calls/day) sufficient for watchlist screening

Workflow

Step 1: Run the Screener

# Screen watchlist (default 30 curated stocks)
python3 skills/mean-reversion-screener/scripts/screen_mean_reversion.py \
  --output-dir reports/

# Custom symbols with tighter RSI threshold
python3 skills/mean-reversion-screener/scripts/screen_mean_reversion.py \
  --symbols AAPL MSFT NVDA AMD META GOOGL AMZN TSLA \
  --rsi-max 35 \
  --min-pullback-pct 8.0 \
  --output-dir reports/

# Aggressive: RSI < 40 within 5% of 50-day MA
python3 skills/mean-reversion-screener/scripts/screen_mean_reversion.py \
  --rsi-max 40 \
  --max-distance-from-50d 5.0 \
  --output-dir reports/

Step 2: Evaluate Candidates

Quality filters (applied automatically):

• Price > 200-day SMA (Stage 2 uptrend)
• RSI(14) < 40 (oversold relative to own history)
• Price within 2σ of lower Bollinger Band
• Volume declining on down days (healthy pullback, not distribution)

Reversion score (0–100):

• RSI component (35%): lower RSI = higher score
• BB position (30%): closer to lower band = higher score
• Distance from 50-day MA (20%): further below = higher score
• Volume profile (15%): low-volume pullback = higher score

Step 3: Entry and Risk

• Entry: At current price or limit at Bollinger Band lower
• Stop: Below the recent swing low (1–3% below pullback low)
• Target: 50-day SMA (typical mean reversion target)
• Risk/Reward: Should be minimum 2:1 before entry

Step 4: Avoid Traps

• Reject if: news-driven selloff (fundamental change)
• Reject if: broader market in downtrend (S&P 500 below 200-day)
• Reject if: earnings within 5 trading days (binary event risk)

Output

• mean_reversion_YYYY-MM-DD.json — Ranked candidates with scores
• mean_reversion_YYYY-MM-DD.md — Trade setup table

Resources

• references/mean_reversion_theory.md — Statistical basis for reversion signals