Skip to content

CDL_CDL_RISEFALL3METHODS (Rising/Falling Three Methods)

Overview

The Rising/Falling Three Methods is a continuation candlestick pattern used in technical analysis to identify potential trading opportunities. The pattern consists of Long candle, three small counter-trend candles, long continuation candle. This pattern provides traders with insights into market psychology and potential price movements when properly confirmed.

Parameters

Parameter Type Default Description
open Array Required Array of opening prices
high Array Required Array of high prices
low Array Required Array of low prices
close Array Required Array of closing prices

Parameter Details

Note: Array elements should be ordered from oldest to newest (chronological order)

OHLC Arrays - All four price arrays must have the same length - Requires sufficient historical data for pattern recognition - More data provides better trend context for signal validation

Usage

Basic Usage

require 'sqa/tai'

open =  [50.0, 49.5, 48.0, 47.5, 48.5]
high =  [50.5, 50.0, 48.5, 48.0, 49.5]
low =   [49.5, 48.5, 47.5, 47.0, 48.0]
close = [49.5, 48.5, 48.0, 47.5, 49.0]

pattern = SQA::TAI.cdl_risefall3methods(open, high, low, close)

puts "Rising/Falling Three Methods signal: #{pattern.last}"
# Output: 100 or -100 (pattern detected) or 0 (no pattern)

With Confirmation

require 'sqa/tai'

# Get historical data
open, high, low, close = get_ohlc_data('AAPL')

# Calculate pattern
pattern = SQA::TAI.cdl_risefall3methods(open, high, low, close)

# Add confirmation with volume and RSI
volume = get_volume_data('AAPL')
rsi = SQA::TAI.rsi(close, period: 14)

if pattern.last != 0
  puts "Rising/Falling Three Methods detected!"
  puts "Signal strength: #{pattern.last}"
  puts "RSI: #{rsi.last.round(2)}"
  puts "Consider entry with proper risk management"
end

Understanding the Pattern

What It Measures

The Rising/Falling Three Methods pattern measures: - Market Psychology: Brief pause in strong trend before continuation - Trend Dynamics: Potential shift or continuation in price direction - Momentum: Balance of power between buyers and sellers

Pattern Characteristics

  • Type: Continuation
  • Context: Best used in strong trend
  • Signal: 100 or -100
  • Reliability: High - classic continuation pattern
  • Timeframes: Most reliable on daily and 4-hour charts

Interpretation

Signal Values

  • +100: Bullish pattern detected - potential upward reversal or continuation
  • -100: Bearish pattern detected - potential downward reversal or continuation
  • 0: No pattern detected - continue monitoring

Pattern Recognition

The Rising/Falling Three Methods is identified by: - Specific candle body and shadow relationships - Position within the prevailing trend - Appears primarily in strong trend

Trading Signals

Entry Strategy

Pattern Confirmation: 1. Wait for pattern completion 2. Confirm with next candle moving in signal direction 3. Verify with volume increase 4. Check RSI for overbought/oversold conditions

Entry Points: - Aggressive: Enter on pattern completion - Conservative: Wait for next candle confirmation - Safe: Combine with support/resistance levels

Risk Management

Stop Loss Placement: - Place beyond the pattern's extreme point - Use ATR to set appropriate distance - Never risk more than 1-2% of capital per trade

Take Profit Targets: - Target 1: Recent swing high/low - Target 2: Key support/resistance levels - Target 3: Fibonacci extensions

Best Practices

Optimal Use Cases

Rising/Falling Three Methods works best when: - Pattern appears in appropriate trend context (strong trend) - Volume confirms the price action - Other technical indicators align (RSI, MACD, Moving Averages) - Pattern forms at key support or resistance levels

Combining with Other Indicators

Recommended Combinations: - RSI: Confirm momentum and overbought/oversold conditions - MACD: Verify trend direction and strength - Volume: Validate buying/selling pressure - Moving Averages: Identify overall trend direction - Support/Resistance: Enhance signal reliability at key levels

Common Pitfalls

Avoid These Mistakes: 1. Trading pattern without confirmation 2. Ignoring overall market trend 3. Not using proper stop losses 4. Over-leveraging based on single pattern 5. Forgetting to check volume confirmation

Parameter Optimization

  • Use daily timeframe for highest reliability
  • Combine with 4-hour for entry timing
  • Avoid very short timeframes (increases false signals)
  • Test on historical data before live trading

Practical Example

Complete trading scenario:

require 'sqa/tai'

# Historical price data
historical_data = get_market_data('AAPL', days: 50)

# Calculate pattern and indicators
pattern = SQA::TAI.cdl_risefall3methods(
  historical_data[:open],
  historical_data[:high],
  historical_data[:low],
  historical_data[:close]
)

rsi = SQA::TAI.rsi(historical_data[:close], period: 14)
macd, signal, _ = SQA::TAI.macd(historical_data[:close])

# Trading logic
if pattern.last == 100  # Bullish signal
  current_price = historical_data[:close].last
  pattern_low = historical_data[:low].last(3).min

  if rsi.last < 70 && macd.last > signal.last
    puts "STRONG BUY SIGNAL"
    puts "Entry: $#{current_price}"
    puts "Stop: $#{pattern_low * 0.98}"
    puts "Target: $#{current_price * 1.05}"
    puts "Risk/Reward: 1:2.5"
  end
elsif pattern.last == -100  # Bearish signal
  current_price = historical_data[:close].last
  pattern_high = historical_data[:high].last(3).max

  if rsi.last > 30 && macd.last < signal.last
    puts "STRONG SELL SIGNAL"
    puts "Entry: $#{current_price}"
    puts "Stop: $#{pattern_high * 1.02}"
    puts "Target: $#{current_price * 0.95}"
    puts "Risk/Reward: 1:2.5"
  end
end

Similar Patterns

  • Doji - Indecision candle patterns
  • Hammer - Bullish reversal pattern
  • Engulfing - Strong reversal pattern

Complementary Indicators

Pattern Family

The Rising/Falling Three Methods belongs to the candlestick pattern family used for: - Identifying potential reversals - Confirming trend continuations - Analyzing market psychology - Timing entry and exit points

Advanced Topics

Multi-Timeframe Analysis

Use Rising/Falling Three Methods across multiple timeframes: - Weekly: Identify major trend direction - Daily: Spot pattern formation - 4-Hour: Fine-tune entry timing - 1-Hour: Manage position and stops

Market Regime Adaptation

Pattern reliability varies by market: - Trending Markets: Higher reliability - Ranging Markets: Moderate reliability - High Volatility: Use wider stops and smaller position sizes - Low Volatility: Tighten stops but expect smaller moves

Statistical Considerations

  • Success rate varies by market conditions (typically 50-65%)
  • Confirmation significantly improves reliability
  • Volume validation adds 10-15% to success rate
  • Pattern at support/resistance adds 15-20% reliability

References

  • Nison, Steve. "Japanese Candlestick Charting Techniques" (1991)
  • Bulkowski, Thomas. "Encyclopedia of Candlestick Charts" (2008)
  • Morris, Gregory L. "Candlestick Charting Explained" (2006)
  • StockCharts Candlestick Patterns

See Also