Skip to main content

NormalizerStrategyHO2 Strategy Analysis

Strategy ID: #284 (284th of 465 strategies)
Strategy Type: Price Normalization Mean Reversion Strategy + Hyperopt Optimized Version
Timeframe: 1 Hour (1h)


I. Strategy Overview

NormalizerStrategyHO2 is the Hyperopt-optimized version of NormalizerStrategy (HO2). Building on the original normalization mean reversion strategy, it systematically tunes key parameters through hyperparameter optimization techniques. The strategy retains the core concept of normalization technology — identifying extreme deviations by mapping prices to a standardized range — while pursuing an improved risk-reward ratio on historical data through optimized multi-tier ROI configuration and parameter adjustments.

Core Features

FeatureDescription
Entry ConditionsExtreme value detection based on normalized price
Exit ConditionsMulti-tier ROI exit + mean reversion exit
Protection MechanismsExtremely relaxed stop-loss + tiered take-profit
Timeframe1 Hour
Optimization MethodHyperopt Hyperparameter Optimization
DependenciesTA-Lib, technical, qtpylib

II. Strategy Configuration Analysis

2.1 Basic Risk Parameters

# ROI Exit Table - Multi-tier time ladder
minimal_roi = {
"0": 0.35, # Immediate exit: 35% profit
"405": 0.248, # After 6.75 hours: 24.8% profit
"875": 0.091, # After 14.6 hours: 9.1% profit
"1585": 0 # After 26.4 hours: break-even exit
}

# Stop-Loss Settings
stoploss = -0.99 # -99% hard stop-loss

Design Philosophy:

  • Multi-Tier ROI Design: Unlike the original single-tier 18% ROI, the HO2 version uses a four-tier declining ROI, embodying a "make big money quickly, make small money slowly" dynamic take-profit philosophy
  • Extremely High Initial ROI (35%): Pursues capturing large mean reversion opportunities, indicating high confidence in signal quality
  • Break-Even Exit Timepoint: Maximum holding period of ~26 hours allows break-even exit, preventing long-term capital occupation
  • Extremely Relaxed Stop-Loss (-99%): Inherited from the original design, grants price ample room to fluctuate

2.2 Order Type Configuration

order_types = {
'entry': 'limit',
'exit': 'limit',
'stoploss': 'market',
'stoploss_on_exchange': False
}

Configuration Notes:

  • Entry and exit both use limit orders for better fill prices
  • Stop-loss uses market orders to ensure timely exit during extreme market conditions

2.3 Hyperopt Optimization Identifier

The "HO2" suffix indicates this is a second version经过 Hyperopt 优化. Hyperopt is a Bayesian optimization technique that searches the parameter space to find the best-performing parameter combination on historical data.


III. Entry Conditions Details

3.1 Core Principle: Price Normalization

The strategy inherits NormalizerStrategy's core methodology, using price normalization technology to identify trading opportunities:

Normalization Formula:

Normalized_Price = (Current_Price - Lowest_Price) / (Highest_Price - Lowest_Price)

Normalized Value Interpretation:

Normalized ValuePrice PositionTrading Signal
Close to 0Price near lower range boundaryPotential buy opportunity
Close to 1Price near upper range boundaryPotential sell opportunity
Close to 0.5Price in middle of rangeNo clear signal

3.2 Entry Condition Logic

# Typical entry condition (Hyperopt-optimized)
def populate_entry_trend(self, dataframe, metadata):
# Buy when normalized price is below threshold
conditions.append(
(dataframe['normalized'] < self.buy_threshold.value)
)

Trigger Conditions:

  • Normalized price drops below the set buy threshold
  • Indicates current price is at a low point relative to historical range
  • Expects price to revert upward to the mean

3.3 Entry Threshold Parameters

ParameterOriginal Typical ValueHO2 Optimized ValueDescription
buy_threshold0.15-0.25OptimizedNormalized price buy threshold
lookback_period20-50OptimizedHistorical period for normalization

IV. Exit Logic Details

4.1 Multi-Tier ROI Take-Profit System

Unlike the original single 18% ROI, the HO2 version uses a four-tier declining ROI:

Timepoint           Minimum Profit Rate    Description
───────────────────────────────────────────────────────
0 minutes 35% Immediate take-profit (highest target)
405 minutes (6.75h) 24.8% Medium-term target
875 minutes (14.6h) 9.1% Later-stage target
1585 minutes (26.4h) 0% Break-even exit

Design Philosophy:

  • Pursuing High Returns: The 35% initial ROI indicates the strategy targets capturing large mean reversion opportunities
  • Time Decay: As holding period increases, lower profit expectations, accelerate capital turnover
  • Avoid Long-Term Occupation: Break-even exit after 26 hours prevents capital being trapped long-term

4.2 Mean Reversion Exit

Exit triggered when normalized price reverts near the mean:

# Sell when normalized price reverts to mean
conditions.append(
(dataframe['normalized'] > self.sell_threshold.value)
)

4.3 Sell Signal Summary

Signal TypeTrigger ConditionPriorityDescription
ROI Take-Profit 1Profit >= 35% (immediate)HighestTarget achieved, exit immediately
ROI Take-Profit 2Profit >= 24.8% (after 6.75h)HighTime decay target
ROI Take-Profit 3Profit >= 9.1% (after 14.6h)MediumLater-stage target
ROI Take-Profit 4Profit >= 0% (after 26.4h)LowBreak-even exit
Mean ReversionNormalized value > sell threshold-Price reverted to high
Stop-Loss TriggerLoss >= 99%LastExtreme protection

V. Technical Indicator System

5.1 Core Indicators

Indicator CategorySpecific IndicatorsPurpose
Normalized IndicatorPrice normalizationIdentify extreme price positions
Trend IndicatorsEMA/SMAAssist trend direction judgment
Volatility IndicatorsATRCalculate price range

5.2 Hyperopt-Optimized Parameters

As a Hyperopt-optimized version, the following parameters may have been tuned:

Parameter CategoryPossible Optimization ItemsDescription
Normalization Parameterslookback_periodNormalization calculation period
Entry Thresholdbuy_thresholdEntry trigger condition
Exit Thresholdsell_thresholdExit trigger condition
ROI Parametersminimal_roiTake-profit time ladder
Stop-Loss ParametersstoplossAlthough currently -99%, can be optimized

VI. Risk Management Features

6.1 Extremely Tolerant Stop-Loss Strategy

The strategy adopts a -99% hard stop-loss:

Design Considerations:

AdvantageDisadvantage
Grants price ample room to fluctuateLacks downside protection
Avoids being stopped out by normal volatilityExtreme market conditions cause huge losses
Fits mean reversion logicRequires strong risk tolerance

6.2 Multi-Tier ROI Risk Control

The four-tier ROI design provides more flexible risk management:

# Time-return tradeoff table
Holding Time Minimum Return Rate Capital Efficiency
─────────────────────────────────────────────────────────────
0 minutes 35% Highest
6.75 hours 24.8% High
14.6 hours 9.1% Medium
26.4 hours 0% Break-even

VII. Strategy Pros & Cons

Advantages

  1. Parameter Optimization: Through Hyperopt technology, parameters systematically optimized, theoretically better performance on historical data
  2. Multi-Tier Take-Profit: Four-tier ROI design more flexible than single target, balancing return and time efficiency
  3. High Return Target: 35% initial ROI indicates pursuit of high-quality signals, reducing low-quality trades
  4. Time Decay Design: Lower profit expectations as holding period increases, accelerates capital turnover
  5. Inherits Normalization Advantages: Retains original's simple logic and universality

Limitations

  1. Overfitting Risk: Hyperopt-optimized parameters may overfit historical data, live performance may not match backtest
  2. Extremely Wide Stop-Loss Risk: -99% stop-loss may cause enormous losses in extreme market conditions
  3. Low Signal Frequency: High 35% target means even scarcer signals, possible long periods without trades
  4. Depends on Historical Similarity: Optimized parameters assume future resembles past, market structural changes may invalidate
  5. Lacks Trailing Stop: No trailing stop mechanism, may miss larger profits

VIII. Applicable Scenarios

Market EnvironmentRecommended ConfigurationDescription
Wide RangingStandard configurationOptimal environment, price fluctuates repeatedly within range
Weak TrendUse with cautionMay produce false signals, recommended to pair with trend filter
Strong TrendNot recommendedPrice continuously deviates from mean, strategy logic fails
High VolatilityAdjust parametersExpand normalization period, lower buy threshold

IX. Applicable Market Environment Details

NormalizerStrategyHO2 is a typical mean reversion strategy with Hyperopt parameter optimization. Based on its strategy architecture and ROI design, it is best suited for ranging markets and performs poorly in strong trending markets.

9.1 Strategy Core Logic

  • Mean Reversion Assumption: Price will revert to the mean after extreme deviations
  • Range Trading Mindset: Buy low and sell high at price range boundaries
  • Extreme Value Identification: Quantifies how extreme a price is through normalization
  • Optimized Parameter Application: Uses Hyperopt-found historical optimal parameters

9.2 Performance in Different Market Environments

Market TypePerformance RatingAnalysis
Slow BullishStarsStarsStars (Moderate)Continuous upward price may miss buy opportunities but catches pullbacks
Slow BearishStarsStars (Poor)Continuous downward price may cause premature buying
Wide RangingStarsStarsStarsStarsStars (Best)Optimal environment, normalization strategy performs best
Narrow RangingStarsStarsStars (Moderate)Small swings may struggle to reach 35% target
Extreme VolatilityStarsStars (Poor)Price breaks original range, normalization parameters may fail
Sideways ConsolidationStarsStarsStarsStars (Good)Secondary optimal environment

9.3 Key Considerations for Hyperopt Parameters

Hyperopt optimization has the following potential issues:

IssueDescription
OverfittingParameters optimized for specific historical data, may not apply to future
Data Peeking BiasOptimization process may use future information that shouldn't be known
Market Structural ChangesHistorical optimal parameters may fail in new market environments
Parameter SensitivitySmall parameter changes may cause large performance fluctuations

X. Important Reminder: The Cost of Optimization

10.1 Hyperopt Optimization: A Double-Edged Sword

Hyperopt optimization is a double-edged sword:

Advantages:

  • Finds optimal parameter combinations on historical data
  • Systematic optimization, avoids subjective guesswork
  • Can test large numbers of parameter combinations

Risks:

  • Overfitting to historical data, live performance decreases
  • "Perfect" historical performance may be an illusion
  • Parameters may be optimized for specific market phases, weak generalization

10.2 Learning Curve

Understanding Hyperopt-optimized strategies requires:

  • Understanding normalization concept and its dynamic characteristics
  • Knowing Hyperopt optimization principles and limitations
  • Ability to judge whether strategy is overfitting
  • Mastering out-of-sample testing methods

10.3 Hardware Requirements

Number of PairsMinimum RAMRecommended RAM
1-5 pairs1GB2GB
5-20 pairs2GB4GB
20+ pairs4GB8GB

Note: Strategy computational demand is low, hardware requirements are minimal.

10.4 Backtesting vs. Live Trading Differences

Hyperopt-optimized strategies may exhibit significant differences between backtesting and live trading:

DifferenceBacktest PerformanceLive Performance
Parameter OptimizationUses historical dataUnknown environment
Overfitting RiskExcellentMay drop significantly
Extreme Value IdentificationHindsight biasReal-time judgment more difficult
LiquidityIgnores liquidity issuesMay be unable to fill in extreme markets

XI. Summary

NormalizerStrategyHO2 is a parameter-optimized mean reversion strategy. On the basis of the original NormalizerStrategy, it introduces multi-tier ROI design and Hyperopt-optimized parameters. Its core value lies in:

  1. Systematic Optimization: Through Hyperopt technology, key parameters optimized for improved historical performance
  2. Flexible Take-Profit: Four-tier declining ROI design more flexible than single target
  3. Methodological Innovation: Normalization technology provides a unique method for quantifying price extreme degrees

For quantitative traders, NormalizerStrategyHO2 provides a valuable case study: how to optimize parameters on a simple strategy. However, one must recognize:

  • Hyperopt-optimized parameters may overfit historical data
  • Extremely relaxed stop-loss requires strong risk tolerance
  • High return target (35%) means fewer trading opportunities

Final Recommendation: Before using this strategy, conduct sufficient out-of-sample testing to verify optimized parameter performance on new data. Moderately lowering the initial ROI target (e.g., 15%-25%) may improve trading frequency and practical usability. Remember, historical optimum does not equal future optimum, and risk management is always the top priority.

Risk Warning: This strategy uses an extremely relaxed stop-loss setting (-99%) and Hyperopt-optimized parameters, carrying overfitting risk and potential for enormous losses in extreme market conditions. Please adjust parameters according to your own risk tolerance and conduct sufficient out-of-sample testing before use.