Skip to content

RSI Price Divergence Screener

Concept Recap

Divergence highlights momentum disagreement between price action and RSI, hinting at potential reversal or weakening trend. - Bullish: Price lower low, RSI higher low. - Bearish: Price higher high, RSI lower high.

Workflow

  1. Fetch OHLC data.
  2. Compute RSI (default period=14, EWM-based).
  3. Identify swing highs/lows (window default=5).
  4. Compare last two valid price + RSI swing pairs.
  5. Apply tolerance to avoid noise (>=0.5 RSI point difference required).
  6. Output structured results + optional CSV.

Command-Line Usage

stockcharts-rsi-divergence                  # all signals
stockcharts-rsi-divergence --type bullish   # only bullish
stockcharts-rsi-divergence --type bearish   # only bearish
stockcharts-rsi-divergence --min-price 10 --max-price 100
stockcharts-rsi-divergence --rsi-period 21 --lookback 90 --period 6mo
stockcharts-rsi-divergence --min-volume 2_000_000

Key options:

--type {bullish,bearish,all}
--min-price N --max-price N
--min-volume N
--rsi-period N
--swing-window N
--lookback N  (bars for divergence search)
--period {3mo,6mo,1y}  (data fetch span)
--output FILE
--input-filter FILE (restrict tickers to prior screener results)

Result Object (CSV Columns)

  • Ticker, Company, Price, RSI, Divergence Type
  • Bullish, Bearish (booleans)
  • Details (string describing swing comparison)
  • Precomputed indices for chart overlay

Tolerance Logic

BEARISH_RSI_TOLERANCE = 0.5
BULLISH_RSI_TOLERANCE = 0.5
# Bearish: RSI2 + tol < RSI1
# Bullish: RSI2 - tol > RSI1

Prevents false positives (e.g., PLTR slight higher RSI not flagged bearish).

Programmatic Use

from stockcharts.screener.rsi_divergence import screen_rsi_divergence
results = screen_rsi_divergence(divergence_type='bullish', min_price=10.0, period='6mo')
for r in results:
    print(r.ticker, r.divergence_type, r.rsi, r.details)

Plotting

stockcharts-plot-divergence --ticker NVDA --period 6mo --period 1d

Overlays price + RSI + detected divergence markers (using precomputed indices if provided).

Note: stockcharts-plot-divergence now uses --period for aggregation (1d, 1wk, 1mo) for consistency with other commands. The legacy --interval flag remains as a deprecated alias.

Enhancements Implemented

  • Input filtering to intersect signals with HA screener output.
  • Volume filter to restrict to liquid setups.
  • Precomputed divergence indices to guarantee chart marker alignment.
  • RSI tolerance to remove marginal noise.

Future Improvements

See docs/roadmap.md.

Extended Guide (Merged)

The legacy comprehensive guide has been merged here for single-source documentation. It includes detailed trading workflows, expanded CLI examples, Python API usage, interpretation guidelines, strategy integrations (Heiken Ashi + RSI), and troubleshooting. For full historical content refer to repository history; redundant sections have been consolidated to avoid duplication.

Detailed Concepts

Bullish divergence: Price forms lower low while RSI forms higher low (momentum improvement). Bearish divergence: Price forms higher high while RSI forms lower high (momentum deterioration).

Expanded CLI Examples

stockcharts-rsi-divergence --type bullish --min-price 10 --period 6mo --output results/swing_trades.csv
stockcharts-rsi-divergence --period 1mo --rsi-period 9
stockcharts-rsi-divergence --type bearish --min-price 100 --period 1y
stockcharts-rsi-divergence --rsi-period 21 --lookback 90 --period 1y

Python Usage Pattern

from stockcharts.screener.rsi_divergence import screen_rsi_divergence, save_results_to_csv
results = screen_rsi_divergence(divergence_type='bullish', min_price=10.0, period='6mo')
for r in results:
    print(r.ticker, r.close_price, r.rsi, r.divergence_type, r.details)
save_results_to_csv(results, 'bullish_divergences.csv')

CSV Output Columns

Ticker, Company, Price, RSI, Divergence Type, Bullish, Bearish, Details.

Strategy Intersection (RSI + Heiken Ashi)

Use divergence screener first, then feed tickers to HA color change screener via --input-filter for higher confidence setups.

Troubleshooting Tips

  • No results: increase --lookback, decrease --swing-window.
  • Too many results: raise --min-volume, tighten price filters.
  • Slow performance: restrict universe with price/volume filters or custom ticker list.

Best Practices

Confirm divergence with price action (reversal candle patterns), volume expansion, and multi-timeframe alignment. Apply disciplined risk management (stops below swing lows for bullish, above swing highs for bearish).