Skip to main content

NostalgiaForInfinityV5MultiOffsetAndHO Strategy Analysis

Chapter 1: Strategy Overview

1.1 Strategy Background

NostalgiaForInfinityV5MultiOffsetAndHO is a sophisticated cryptocurrency trading strategy built on the Freqtrade framework. The name encodes: "V5" (5th major version iteration), "MultiOffset" (introducing multi-type moving average offset mechanisms), and "HO" (Hyper-Optimized, with systematically backtested parameters).

The strategy is developed by iterativ, integrating the core logic of NostalgiaForInfinityV5 with the MultiOffsetLamboV0 offset signal system, with hyper-optimization applied to key parameters.

1.2 Core Design Philosophy

Summarized as "Multi-Confirmation, Trend is King, Offset Entry":

  • Multi-Confirmation: Buy signals require simultaneous satisfaction of trend direction, price position, momentum state, and volatility characteristics
  • Trend is King: The strategy always prioritizes trend judgment; most buy conditions require short-term MAs above long-term MAs
  • Offset Entry: The strategy's innovation — applying offset coefficients to moving averages for more precise entry control

1.3 Technical Stack

  • TA-Lib for technical indicators
  • qtpylib for Bollinger Bands and Choppiness
  • NumPy and Pandas for numerical computation

Chapter 2: MultiOffset Parameter System

2.1 Five Moving Average Types

MA TypeBuy OffsetSell OffsetPeriod Default
SMA0.90 (range 0.9-0.99)1.051 (range 0.99-1.1)47 (buy), 34 (sell)
EMA0.931.047
TRIMA0.9731.096
T30.9750.999
KAMA0.9851.07

The strategy buys when price drops below any one of these five MAs at their respective discounted prices, simultaneously satisfying EWO conditions.

2.2 EWO Parameters

EWO = (EMA50 - EMA200) / Close × 100
ewo_high = 3.28 # Positive threshold
ewo_low = -16.004 # Negative threshold

EWO acts as a gatekeeper — price must be at an extreme state (EWO below -16 or above +3.28) for offset MA signals to trigger.


Chapter 3: Technical Indicator System

3.1 Primary Timeframe (5m)

  • BB20/BB40: Bollinger Bands for overbought/oversold and pattern recognition
  • EMA: 12, 20, 26, 50, 100, 200 periods
  • SMA: 5, 30, 200 periods
  • RSI, MFI: 14-period momentum and money flow
  • EWO: 50/200-period Elliott Wave Oscillator
  • Choppiness: 14-period trend strength

3.2 Auxiliary Timeframe (1h)

All indicators with _1h suffix: EMA (15, 50, 100, 200), SMA200, RSI, Bollinger Bands, pump protection flags


Chapter 4: Protection Mechanisms

4.1 Dip Protection (4 levels)

Level1-candle2-candle12-candle144-candle
Strict1.5%6%24%40%
Normal2.0%14%32%50%
Loose2.6%24%42%66%

4.2 Pump Protection (3 strictness levels, 3 time windows)

Each with 9 parameter groups covering Normal, Strict, and Loose variants for 24h/36h/48h windows.


Chapter 5: Buy Signal System (26 triggers)

21 standard conditions + 5 MultiOffset conditions, OR logic (any one triggers a buy).

Key Categories

  1. Conditions 1, 8, 11, 14: Trend-following — requires strong multi-timeframe trend alignment
  2. Conditions 2, 4, 9, 10, 12, 16, 17: BB/EWO oversold — price at/below BB lower band
  3. Conditions 5, 6: EMA divergence — EMA26 above EMA12 (death cross state)
  4. Conditions 3, 13: BB40 compression — BB bandwidth squeeze before breakout
  5. Conditions 19: EMA100_1h re-test — price briefly dips below then reclaims 1h EMA100
  6. Conditions 20, 21: Extreme dual-RSI — both 5m and 1h RSI deeply oversold
  7. MultiOffset: 5 independent MA-type offset buy triggers

MultiOffset Entry Logic

Buy when any one of:

  • Price < SMA(base_period) × low_offset_sma AND EWO condition met
  • Price < EMA(base_period) × low_offset_ema AND EWO condition met
  • Price < TRIMA(base_period) × low_offset_trima AND EWO condition met
  • Price < T3(base_period) × low_offset_t3 AND EWO condition met
  • Price < KAMA(base_period) × low_offset_kama AND EWO condition met

Chapter 6: Sell Signal System

8 Base Sell Conditions

#TriggerKey Thresholds
1BB upper band 6+ consecutive breaksRSI > 79.5
2BB upper band 3+ consecutive breaksRSI > 81
3Pure RSI overboughtRSI > 82
4Dual-timeframe RSI overbought5m > 73.4, 1h > 79.6
6Below EMA200, RSI > 79, between EMAs
71h RSI > 81.7 + EMA death cross
8Price > 1h BB upper × 1.1Extreme overbought

Chapter 7: Custom Sell Mechanism

Tiered Profit-Taking

StageProfit >RSI <Signal
01%33signal_profit_0
13%38signal_profit_1
25%43signal_profit_2
38%48signal_profit_3
425%50signal_profit_4

Trailing Retracement

profit 15%-46% and max_profit - profit > 18% → sell
profit 1%-12% and max_profit - profit > 14% → sell

Trend-Below Exits

When price < EMA200: even tighter profit requirements (profit > 2%, RSI < 56 → sell)


Chapter 8: Risk Control

  • Fixed Stoploss: -10%
  • Trailing Stoploss: Activates at 3% profit, trails 1%
  • ROI: {"0": 0.01} (minimal ROI trigger, actual exits driven by signals)
  • Multi-layer entry protection: Dip + Pump + Trend + Volume + Multi-timeframe + EWO filtering

Chapter 9: Strategy Pros and Limitations

Pros

  • MultiOffset innovation: Five different MA perspectives catch more opportunities
  • 26 buy triggers: Extremely broad coverage of entry scenarios
  • Comprehensive protection: Three-level dip and pump protection
  • Fully optimizable: All parameters support hyperopt

Limitations

  • Extreme complexity: 200+ optimizable parameters, high overfitting risk
  • Trend-dependent: Underperforms in ranging markets
  • Multi-MA calculation overhead: Computationally intensive
  • Requires active management: Needs regular review and adjustment

"timeframe": "5m",
"use_sell_signal": true,
"sell_profit_only": false,
"ignore_roi_if_buy_signal": true
  • Concurrent trades: 4-6
  • Trading pairs: 40-80 stablecoin pairs
  • Exclude: Leveraged tokens (*BULL, *BEAR, *UP, *DOWN)
  • Use Walk-Forward analysis for validation

For the plain-English explanation, please refer to the Strategy Explained version.