Skip to main content

NowoIchimoku5mV2 Strategy Analysis

Table of Contents​

  1. Strategy Overview and Technical Framework
  2. Market Theory Foundation
  3. Indicator System
  4. Entry Signal Logic
  5. Exit Mechanism Design
  6. Risk Control System
  7. Parameter Optimization Space
  8. Backtesting Framework Configuration
  9. Live Deployment Advice
  10. Strategy Pros & Cons Analysis
  11. Conclusion and Improvement Directions

1. Strategy Overview and Technical Framework​

1.1 Strategy Positioning​

NowoIchimoku5mV2 is a trend-following strategy designed specifically for the cryptocurrency market, adopting a dual-timeframe analysis method with 5 minutes as the primary trading timeframe and 1 hour as the informative timeframe. The "Nowo" in the name implies innovative improvement, while "Ichimoku" indicates the core logic is built upon the classic Ichimoku Cloud indicator system.

1.2 Technical Architecture​

Built on the Freqtrade quantitative trading framework, an open-source cryptocurrency trading engine, the strategy uses object-oriented design, inheriting from the IStrategy base class and implementing complete lifecycle hook functions.

class NowoIchimoku5mV2(IStrategy):
    timeframe = '5m'
    informative_timeframe = '1h'

1.3 Core Design Philosophy​

The strategy's design philosophy can be summarized as "follow the trend, cloud as the boundary." It identifies market trend direction through the Ichimoku Cloud, dynamically adjusts risk exposure using Bollinger Bands, enters after trend confirmation, and exits when trends reverse or risks accumulate.

⚠️ Important Note: The strategy file header states "This version of the strategy is broken!" — indicating this version may have issues and direct live trading is not recommended. However, the strategy's ideas are valuable for learning.

2. Market Theory Foundation​

2.1 Ichimoku Cloud Theory​

The Ichimoku Cloud (Ichimoku Kinko Hyo) was invented by Japanese journalist Goshu Hosoda in the 1930s as a technical analysis tool whose core value lies in simultaneously displaying support/resistance, trend direction, and momentum strength.

Conversion Line (Tenkan-sen): The fast-reacting line, calculated as the average of the highest and lowest prices over 9 periods. When price is above this line, short-term sentiment is bullish; below indicates bearish sentiment.

Base Line (Kijun-sen): The medium-term reference line, calculated as the average of the highest and lowest prices over 26 periods. Price above this line indicates relatively strong market status.

Leading Span A: Average of Conversion and Base Lines, shifted forward 26 periods. Forms one boundary of the cloud.

Leading Span B: Average of the highest and lowest prices over 52 periods, shifted forward 26 periods. Forms the other cloud boundary.

Chikou Span (Lagging Span): Current closing price shifted backward 26 periods, used to confirm trend direction.

2.2 Bollinger Bands Theory​

Bollinger Bands, developed by John Bollinger in the 1980s, consists of three bands: a middle band (SMA) and upper/lower bands at K× standard deviations. This strategy uses HMA instead of SMA for the middle band, offering lower lag and better smoothness.

2.3 StochRSI Theory​

StochRSI applies stochastic calculation to RSI values, targeting identification of overbought/oversold extremes. Values above 80 indicate extreme overbought conditions; below 20 indicates extreme oversold.

3. Indicator System​

3.1 Timeframe Design​

The strategy uses a dual-timeframe structure: 5-minute primary, 1-hour informative. Benefits:

  • Reduces noise interference
  • Provides forward-looking warning via 1h indicators
  • Low-timeframe high-frequency trading risks reduced by multi-timeframe confirmation

3.2 Hull Moving Average Algorithm​

def hma(series: Series, length: int) -> Series:
    h = 2 * wma(series, math.floor(length / 2)) - wma(series, length)
    hma = wma(h, math.floor(math.sqrt(length)))
    return hma

HMA achieves a balance between low lag and high smoothness through a clever weighted combination.

3.3 Ichimoku Calculation​

ichi_1h = indicators.ichimoku(df_1h)
df_1h['conversion_line'] = ichi_1h['tenkan_sen']
df_1h['base_line'] = ichi_1h['kijun_sen']
df_1h['lead_1'] = ichi_1h['leading_senkou_span_a']
df_1h['lead_2'] = ichi_1h['leading_senkou_span_b']
df_1h['is_cloud_green'] = ichi_1h['cloud_green']

Cloud color: Leading Span A > Leading Span B = green (bullish); reverse = red (bearish).

4. Entry Signal Logic​

4.1 Multi-Timeframe Data Merge​

df = merge_informative_pair(df_5m, df_1h, self.timeframe, self.informative_timeframe, ffill=True)

Uses forward fill (ffill) to ensure each 5-minute candle gets the latest 1-hour indicator values.

4.2 Core Entry Conditions​

Seven conditions must be simultaneously satisfied:

Condition 1: Bullish Candle — close > open Condition 2: Close > shifted_upper_cloud × close_above_shifted_upper_cloud (default 0.603) Condition 3: Close > shifted_lower_cloud Condition 4: is_cloud_green — Leading A > Leading B Condition 5: conversion_line > base_line — short-term momentum upward Condition 6: close > conversion_line.shift(25 × time_factor) — above 25-period-displaced Conversion Line Condition 7: close > upper_cloud.shift(50 × time_factor) — above 50-period-displaced cloud upper

4.3 Cooldown Mechanism​

df['buy_allowed'] = True
for i in range(1, len(df)):
    if df.at[i - 1, 'buy']:
        df.loc[i, 'buy_allowed'] = False
    if not df.at[i, 'is_cloud_green']:
        df.loc[i, 'buy_allowed'] = True
    df.loc[i, 'buy'] = df.at[i, 'buy_allowed'] & df.at[i, 'should_buy']

Logic: Allow buying by default → prohibit after buy → reset when cloud turns red → allow when cloud turns green again. Prevents over-positioning in single trends.

5. Exit Mechanism Design​

5.1 Fixed Stop-Loss​

stoploss = -0.293

A generous -29.3% hard stop — the designer wanted sufficient price fluctuation room.

5.2 Custom Stop-Loss Function​

Four dynamic stop-loss/exit conditions:

Trigger 1: StochRSI Overbought

if (last_candle['srsi_k'] > 80) & (current_profit > srsi_k_min_profit.value):
    return -0.0001

When K > 80 AND profit > 3.6% (default), lock in profits.

Trigger 2: Bollinger Upper Break

if (previous_candle['close'] < previous_candle['upper']) & \
   (current_rate > last_candle['upper']) & \
   (current_profit > above_upper_min_profit.value):
    return -0.0001

When price breaks above upper band AND profit > 1.1% (default), exit.

Trigger 3: Dynamic Limit

limit = trade.open_rate + ((trade.open_rate - trade_candle['shifted_lower_cloud']) * limit_factor.value)
if current_rate > limit:
    return -0.0001

Target based on entry price minus displaced cloud lower × limit_factor (default 1.918).

Trigger 4: Cloud Lower Support Failure

if current_rate < (trade_candle['shifted_lower_cloud'] * lower_cloud_factor.value):
    return -0.0001

When price drops below 97.1% (default) of displaced cloud lower, exit.

6. Risk Control System​

6.1 Multi-Layer Filter​

Seven-entry-condition combination builds a multi-layer risk filter. Each condition verifies trend validity from different angles.

6.2 Trailing Stop​

trailing_stop = True

Trailing stop moves upward with price, locking in profits.

6.3 Time-Decaying ROI​

minimal_roi = {
    "0": 0.10,
    "30": 0.05,
    "60": 0.02
}

Immediate target 10% → 5% after 30 min → 2% after 60 min.

6.4 Five Optimizable Parameters​

srsi_k_min_profit = DecimalParameter(0.01, 0.99, default=0.036, space="sell")
above_upper_min_profit = DecimalParameter(0.001, 0.5, default=0.011, space="sell")
limit_factor = DecimalParameter(0.5, 5, default=1.918, space="sell")
lower_cloud_factor = DecimalParameter(0.5, 1.5, default=0.971, space="sell")
close_above_shifted_upper_cloud = DecimalParameter(0.5, 2, default=0.603, space="buy")

7. Parameter Optimization Space​

close_above_shifted_upper_cloud (0.5-2.0, default 0.603): Lower = looser entry; higher = stricter, fewer but higher-quality signals.

srsi_k_min_profit (0.01-0.99, default 0.036): Higher = more holding, bigger potential gains but more drawdown risk.

limit_factor (0.5-5, default 1.918): Higher = higher targets, longer holding; lower = shorter trades.

lower_cloud_factor (0.5-1.5, default 0.971): Lower = tighter stop; higher = more room.

8. Backtesting Framework Configuration​

8.1 Startup Candles​

startup_candle_count = int(100 * time_factor)  # 500 candles for 5m

Requires 500 5-minute candles minimum.

8.2 Time Factor​

time_factor = int(60 / 5)  # = 12

Each 1-hour candle = 12 5-minute candles.

8.3 Backtesting Notes​

  • Ensure complete historical data with no gaps
  • Set reasonable slippage parameters (0.5%-1%)
  • Set appropriate fee rates (0.1%+)
  • Ensure capital management can handle consecutive losses

9. Live Deployment Advice​

9.1 Exchange Selection​

High-liquidity pairs only: BTC/USDT, ETH/USDT, etc.

9.2 Position Management​

With -29.3% stop, risk per trade should be 1-3% of total capital.

9.3 Key Metrics to Monitor​

  • Signal frequency (too frequent = adjust parameters)
  • Average holding time
  • Maximum drawdown
  • Win rate and profit/loss ratio

10. Strategy Pros & Cons Analysis​

Pros​

  • Strong trend-following capability (Ichimoku is time-tested)
  • Dynamic risk management (custom stop-loss)
  • Multi-timeframe confirmation (filters noise)
  • Flexible parameters (tunable)
  • Clean code structure

Cons​

  • Stop-loss too wide (-29.3%) — aggressive for most traders
  • Entry conditions too strict (7 conditions) — may miss opportunities
  • No explicit profit-taking mechanism
  • No market state adaptation
  • File header explicitly warns: "strategy is broken"

11. Conclusion and Improvement Directions​

NowoIchimoku5mV2 has a clear Ichimoku Cloud trend-following design, combining Bollinger Bands volatility analysis with StochRSI overbought/oversold judgment. The dual-timeframe and multi-layer filter mechanisms reflect emphasis on signal quality.

The wide stop-loss and strict entry conditions make it more suitable for traders with higher risk tolerance. The file header warning indicates it may still be in development — caution advised before live use.

Improvement suggestions:

  1. Tighten stop to -10% ~ -15%
  2. Add market state filter (ADX)
  3. Optimize entry conditions (5-out-of-7 rather than all-7)
  4. Add partial profit-taking mechanism
  5. Fix known bugs before use

This document is for Freqtrade strategy learning and research. Please fully understand strategy logic and conduct sufficient backtesting validation before use.